| quarta-feira, 11 de novembro de 2009
The Generation of Macromolecular
Complexity in the Brain
Neurobiologists, like other intellectuals, often divide themselves on ideological
grounds into two groups: reductionists and holists. Reductionists (called
cellular connectionists in neurobiology) tend to think that the brain is best
studied on the cellular level because 1) the cell is the fundamental signaling
unit of the nervous system, and 2) the cells of the brain are not identical (as
are the parenchymal cells of the liver): one nerve cell often differs quite remarkably
from the next.

The Generation of Macromolecular
Complexity in the Brain
Neurobiologists, like other intellectuals, often divide themselves on ideological
grounds into two groups: reductionists and holists. Reductionists (called
cellular connectionists in neurobiology) tend to think that the brain is best
studied on the cellular level because 1) the cell is the fundamental signaling
unit of the nervous system, and 2) the cells of the brain are not identical (as
are the parenchymal cells of the liver): one nerve cell often differs quite remarkably
from the next. Holists think the brain should be studied only as a
whole because the whole is much more than the sum of its parts. New principles
emerge by looking at the brain as a whole that cannot be inferred by
looking at its constituent cells.
Over the last two decades, much of what is interesting in neurobiology,
and all that I have so far reviewed in the symposium, has been done within
a conceptual framework of cellular connectionism. But, surprisingly, biochemists
first coming into the field often resort to a more global approach,
not philosophically—because biochemists are reductionists and clearly
want to understand how molecules and cells work—but for expedience, because
to work effectively biochemists need lots of starting material, which
can most readily be obtained by grinding up whole brain or regions of the
brain. However, in most cases, as they begin to appreciate that the brain is
not the liver but a remarkably heterogeneous organ made up of many cell
types and subtypes, most biochemists typically move either toward improving
the purity of their system or toward finding a better one.
This symposium has shown that under some (I would think rare) cirNeurobiology
and Molecular Biology 187
cumstances, valuable information can come from studies of the brain as a
whole tissue. First, these studies have shown that the brain expresses more
genes than any other tissue of the body (Hahn et al.; Sutcliffe et al.). The kidney
or liver expresses between 10,000 and 20,000 distinct mRNA sequences;
the brain is thought to express at least four times as many.
In addition to expressing a greater number of genes, there is some evidence
that the cells of the brain make extensive use of an unusual class of
mRNA that is encountered only infrequently in other cells. Whereas in other
tissues the mRNA that is translated into protein almost invariably is polyadenylated
(poly[A]+), Hahn and his colleagues and Chikaraishi et al. have
found that the brain contains a large number of mRNAs lacking this poly(A)
tail (poly[A]– mRNA). Of further interest is the finding that poly(A)– messages
are not abundant at birth but become prominent only during postnatal
development, suggesting that they have a special role in late stages of development.
Thus, despite its preliminary nature, the finding of abundant
poly(A)– mRNA in the brain and the possibility that it codes for proteins
other than those coded by poly(A)+ messages are intriguing and potentially
important. But to demonstrate in a compelling manner the functional significance
of this unusual form of mRNA, it will be essential to show that these
poly(A)– mRNAs encode different proteins than do poly(A)+ mRNAs.
I hasten to add an obvious cautionary comment: Not everything that is
brain-specific need be important. The mammalian genome is thought to
contain about 100,000 genes. Perhaps 30% of these genes may be specific to
brain. One would like to think that all of them will prove to be equally interesting,
but I doubt it. Some brain-specific genes will certainly prove much
more important than others. Moreover, we may never be able to understand
the brain completely, in all of its detail. We will want therefore to focus on
some key problems and to explore them thoroughly. A corollary to this argument
is that although the brain is the organ of mentation, not every brainspecific
mRNA need code (or, indeed, is likely to code) for a protein involved
in a higher mental function. On the other hand, many mental functions
that we consider fascinating probably utilize the same proteins that
other cells of the body use for different purposes.
The possibility that an unusual type of mRNA processing may be common
in the brain raises the question of whether other specialized processes
are exploited by the brain more extensively than elsewhere. Is there, for example,
genomic rearrangement in the brain? In immune systems, several
functional gene domains are recombined to generate a diverse set of antibody
genes. Rearrangement of DNA in neurons—were it found to be nonrandom
and combinatorial—could also be important for expanding the
informational potential of the brain’s genome.
The brain expresses more genes than other organs, but, like other organs,
188 Psychiatry, Psychoanalysis, and the New Biology of Mind
it can also achieve macromolecular complexity in posttranscriptional ways.
Good evidence now exists for at least three classes of additional mechanisms:
1) alternative processing of mRNA, 2) alternative processing of protein
precursors, and 3) covalent modifications of mature proteins (phosphorylation,
methylation, glycosylation, etc.).
Evans and his colleagues have found that a single calcitonin gene is capable
of generating two mRNAs and two gene products by differential processing
of the mRNA. In the cells of the thyroid gland, the calcitonin gene
transcribes an mRNA in which a poly(A) site is selected that is part of the
calcitonin exon. In brain cells, another polyadenylation site is recognized,
and this site is adjacent to the exon from a peptide (the CGR peptide) so that
in the brain only this peptide gene and not the calcitonin gene is transcribed.
We have already considered alternative precursor processing as a way to
produce different families of opioid peptides in different cells (Herbert et
al.). The same peptide can also exist in native and covalently modified form.
A common modification is phosphorylation. At least 10% of all brainspecific
proteins can be phosphorylated on serine or threonine residues.
Some of these (such as protein 1 or synapsin I) are common to all vertebrate
nerve cells (DeGennaro et al.). This protein is composed of two peptides of
86 kD and 80 kD and is associated with synaptic vesicles (DeGennaro et al.).
It may be a protein involved in vesicle mobilization or retrieval and is especially
interesting because its state of phosphorylation is modulated by several
chemical transmitters and by impulse activity. Other phosphoproteins
are common to broad classes of cells (for example, cells receiving dopaminergic
input), and some are specific to a single cell type, such as the Purkinje
cell of the cerebellum (Lewis et al.). In contrast to the ubiquity of phosphorylation,
methylation and adenylation are used much less frequently.
Neuronal diversity and neuronal recognition
The complexity of the nervous system emerges from many developmental
steps that encompass determination, differentiation, formation of connections,
cell death, synapse elimination, and fine-tuning of the remaining
synapses. These steps reflect lineage, early developmental history, later competitive
interactions, and other epigenetic modifications. The resulting heterogeneity
of cellular properties in the brain is often referred to as neuronal
diversity. Although descriptively correct, the phrase is in some ways unfortunate
because it cannot help but bring to mind that much better understood
instance of biological diversity, antibody diversity. The term antibody diversity
refers to the fact the B lymphocytes make millions of different antibodies
that bind to different antigenic determinants. Antibody diversity involves a
change in a single product, immunoglobulin. The genetic mechanism that
Neurobiology and Molecular Biology 189
generates antibody diversity operates on a common family of proteins with
a common function. In contrast, neurons are not diverse in only one way but
have a family of diversities. Let me illustrate this by indicating the different
classes of apparently independent functions that contribute to differences
between neurons.
First, neurons differ in their signaling capabilities. Some neurons generate
action potentials; others do not. Neurons that generate action potentials
may be silent, or they may be spontaneously active even in the absence of
input (like the pacemaker cells of the heart). Cells that are spontaneously active
can fire either regularly or in bursts. All these (and other, more subtle)
differences can be traced to the family of specific ion channels present in the
membrane of a given neuron and, as we have seen, neurons differ in their
ion channels.
Second, neurons differ in the chemical transmitters that they synthesize.
Some neurons synthesize one of a large family of small molecule transmitters;
other neurons synthesize one and often several peptide transmitters;
still others synthesize combinations of these transmitters. Expression of any
particular transmitter, whether it is a small molecule or a peptide, commits
the cell to a whole pattern of differentiation. This pattern includes not only
the biosynthetic enzymes for the transmitter but also characteristic membrane
systems consisting of vesicles and uptake mechanisms, both within
the vesicles and in the external membrane.
Third, neurons differ from one another in the connections they make
with their target cells.
Fourth, neurons differ in the connections they receive and therefore in
the receptors they have to small molecule transmitters, peptides, and hormones.
Not only do cells respond differently to different transmitters, but
they can respond differently to the same transmitter. For example, some receptors
to a given transmitter molecule are linked to adenylate cyclase; others
are not (Schramm et al.).
Fifth, neurons differ in structure, in the size of the cell body (which can
vary from 5 μm to 100 μm in vertebrates and from 5 μm to 1,000 μm in invertebrates),
in the presence or absence of axons, and in the number and
shape of dendrites (Lasek et al.; Matus et al.). The shape of the neuron is an
external manifestation of the molecular structure of its cytoskeleton (Baitinger
et al.; Ginzburg et al.; Lasek et al.; Matus et al.; Weber et al.), and this
too differs between cells.
Finally, neurons are thought to have distinctive recognition molecules that
distinguish neurons in the various regions of the brain from each other during
development and that also distinguish the position of each cell or cell grouping
within each region. Surface recognition molecules are thought to be important
for allowing cells to interconnect with their appropriate targets.
190 Psychiatry, Psychoanalysis, and the New Biology of Mind
It is in the context of the last distinction that the concept of neuronal diversity
is currently used most often. This usage derives from the fact that
some developmental neurobiologists were struck by the similarity between
the recognition problem faced by antibodies and that faced by nerve cells
during development. Studies ranging from those of Cajal at the beginning of
the century to the recent work of Sperry have indicated that neurons are connected
to each other in a precise way. This precision suggested to Sperry and
to other (but by no means all) developmental biologists that each neuron, or
at least small groups of postsynaptic cells, may have a molecular individuality
that allows recognition by appropriate outgrowing presynaptic neurons.
Even if we assume that this type of diversity in recognition molecules actually
exists (and there is as yet no hard evidence to support that assumption),
most probes that explore differences between neurons cannot distinguish
between diversity based on cell recognition and diversity due to any one of
the other factors, genetic and epigenetic, that affect the generation and modification
of specific proteins that cause neurons to differ from one another.
Thus, insofar as differences between neurons are explored, it will be important
for the time being to specify the dimension of diversity to which those
differences belong. In particular, recognitional diversity should, for the
present, be distinguished from other forms of diversity, and it should be
sought primarily in the developing brain. Now, it is conceivable that the
other forms of diversity are secondary to recognition. Perhaps nerve cells develop
specific surface molecules after birth for the purpose of recognition,
and perhaps other differences derive from this early marking. But these possibilities
must all be demonstrated.
How do we define a cell type in the brain?
The extent of neuronal diversity raises one of the deepest questions in neurobiology:
What constitutes a cell type in the nervous system? Unlike muscle
or liver, not only does the vertebrate brain consist of a very large number
of nerve cells but even the most cursory examination of histological sections
with a Golgi stain suggests that the brain contains many different types of
cells. The current estimate is that there are about 1,000–10,000 different cell
types, but there may be many more. For example, nerve cells of the same differentiated
set do not connect with the same targets. The millions of
Purkinje cells of the cerebellum connect not to one but to several quite different
deep cerebellar nuclei. Should these subpopulations be considered
different types of cells? Is there any cellular feature that correlates with output
connections? Are the cells also distinguished by their inputs? Their location
in the cerebellum? Is any of these features correlated with other
cellular properties?
Neurobiology and Molecular Biology 191
Here a family of molecular questions needs to be probed: How does one
distinguish a particular cell type from another on the molecular level? How
many different proteins are necessary to generate a cell type? At what level
of regulation (transcriptional or posttranscriptional) is this difference established?
How directly is cell type related to cell lineage in vertebrates? How
directly is cell type related to particular recognition molecules? How many
cell types are there actually in the vertebrate nervous system?
This last question is nicely illustrated with an example from the vertebrate
retina. Anatomical studies using Golgi stains (until recently the major
method for distinguishing cell type) have long encouraged the belief that
there is only one type of amacrine cell in the retina (although Cajal early on
pointed out that amacrine cells could be distinguished from one another by
subtle differences in the pattern of their dendritic branching). By adding immunocytochemical
and other criteria to those of Cajal, we are now able to
recognize about 25 distinct classes of amacrine cells (Brecha et al. 1983).
Similarly, there are several types of receptor, bipolar, and horizontal cells,
and some 20 classes of ganglion cells. All told, the retina, once thought to be
a simple structure consisting of only five cell types, is now known to contain
about 75 cell types. The differences are based in each case on at least two independent
By analyzing subtle but fundamental differences between cells, we might
be able to arrive at a fingerprint of a cell—a multifactorial definition of what
constitutes a cell type. A particularly good strategy would be to seek genes
and gene products that regulate the differentiation of neuronal cell types.
This could perhaps best be done by examining otherwise similarly appearing
cells such as Purkinje cells, or the granule cells of the cerebellum with two
questions in mind: 1) Can reliable differences be recognized on the molecular
level? 2) What functional consequences do these differences have? Monoclonal
antibodies, which have primarily been used to explore distinctions
between very different cells, could now be used to detect differences between
otherwise similarly appearing cells. Moreover, plus/minus screening
of a cDNA library made from particular regions in the brain (such as the cerebellum)
against labeled cDNA made from mRNA derived from subgroups
of Purkinje cells could lead to the discovery of differences in gene expression
among Purkinje cells. With this approach, one might also uncover regulatory
proteins that determine differences between cell types.
By analogy to other cells of the body, it would appear very likely that different
cell types express different sets of genes and that a wide variety of cell
types might be specified by a few genes coding for regulatory proteins,
which in turn are expressed in various combinations. To give but one dramatic
example, alterations in regulatory genes are thought to result in homeotic
mutants of Drosophila, where wholesale transformations in body
192 Psychiatry, Psychoanalysis, and the New Biology of Mind
parts—the formation of a complete and complex alternative structure—occur
following a single gene mutation. Thus, a single gene mutation converts
a part of the body that normally makes an antenna into one that makes a leg
by switching precursor antenna cells into making leg proteins. The proteins
encoded by these mutant genes are each thought to control a large number
of other genes. A number of simple schemes have been outlined whereby
many different cell types can be specified by combinations of a few regulatory
proteins (see Gierer 1974). The notched locus has homeotic properties
and is known to grossly alter neuronal development. One possible outcome
of cloning the notch locus (Kidd et al. 1983) might be an insight into a class
of genes with a particularly important role in controlling neuronal differentiation.
As is evident from this discussion of cell types, the questions posed about
neurons are not unique to molecular studies of brain development but are
being asked in all areas of developmental biology. Molecular techniques are
extremely powerful for analyzing the structure of individual molecules, but
we are only beginning to develop techniques that examine how these molecules
interact during development to produce a functioning cell. For a
deeper understanding of the mechanisms that give rise to the different cell
types of the vertebrate brain, we will need to be able to define and subtly manipulate
these molecular systems.
Molecular heterogeneity can be detected
with monoclonal antibodies
Since neurons differ from one another in a variety of ways, molecular heterogeneity
should be readily demonstrable with hybridoma technology, which
can generate specific antibodies using a complex antigen. This has indeed
proved to be the case. Screens of the nervous system with libraries of monoclonal
antibodies raised to the nervous system show a remarkable antigenic
heterogeneity (Barnstable et al.; Hockfield et al.; McKay et al.; Zipser et al.;
S. Benzer, personal communication). Some antigens are specific for neurons,
some specific for glial cells, some for axons, some for cell bodies; some are
unique to a specific cell type, others are shared by several cell types; the cell
types may in turn prove to be related. These antibodies have therefore provided
superb markers for certain cell types (see, for example, Raff et al.), and
they should make possible the isolation and ultimate purification of many
neuron-specific proteins. This step is badly needed. Our understanding of
neurons on the molecular level is seriously limited by the fact that we know
very few of their important proteins. The discovery of many new antigenic
determinants, and the handle they provide to uncovering new proteins, is
only the first and probably the easiest step. It will be more difficult to relate
Neurobiology and Molecular Biology 193
these proteins to function. One strategy for accomplishing this goal is to cast
a wide net. We need to know the biochemical properties of these antigens
and their distribution in the cell. The new molecular techniques make it easier
to gather this information, but we also must have assay systems with
which we can explore directly whether a particular gene or gene product alters
a neuronal property of interest.
Interest in the biochemical functions of antigens related to neuronal differentiation
should not blind us to other potentially interesting outcomes of
the growing availability of monoclonal antibodies. For example, monoclonal
antibodies raised against the cat spinal cord identify subsets of neurons organized
in columns in the primate visual cortex (Hockfield et al.). Markers
of this kind might allow us to analyze the early development of cortical columns,
which current techniques have been unable to address because they
require that connections between the eye and the cortex already exist. Similarly,
the availability of several antibodies that distinguish different neuronal
types in the Drosophila eye (S. Benzer, personal communication) may
allow us to probe retinal development.
Perception, Behavior, and Learning
An ultimate aim of neuroscience is to provide an intellectually satisfying set
of explanations, in molecular terms, of normal mentation, perception, motor
coordination, feeling, thought, and memory. In addition, neuroscience
would ultimately like to account for the disorders of function produced by
neurological and psychiatric diseases.
I have so far outlined how the papers of this symposium and the recent
progress in molecular neurobiology have contributed to our understanding
of the signaling of nerve cells and their patterns of development and interconnection.
The patterns of interconnection established by genetic and
developmental processes, in turn, determine the capability for motor coordination
and perception. Thus, the molecular information that we are beginning
to obtain on how neurons interconnect will provide details essential for
explaining behavior.
Take the simplest case of an elementary reflex—a sensory stimulus producing
a motor response. We need to understand how information flows
from the transduction of the sensory stimulus to the initiation of the movement.
As I have indicated, we are gaining a fairly good appreciation of the
molecular mechanisms of central synaptic transmission and neuromuscular
transmission. By contrast, we have not, until recently, had comparable insights
into sensory transduction. But substantial progress has now been
made in understanding transduction by visual and chemical stimuli.
Advances in elucidating the molecular basis of visual transduction
194 Psychiatry, Psychoanalysis, and the New Biology of Mind
(Dunn et al.; Stryer) have come from two sources: from studies of bacteriorhodopsin
and of rhodopsin in the vertebrate retina. Bacteriorhodopsin is
contained in the purple membrane, a specialized patch in the membrane of
Halobacteria. A protein of 248 amino acids, it contains a light-absorbing
prosthetic group (or chromophore) called retinal, which is identical to that
found in the rod photoreceptor cells of the vertebrate eye. The complete
amino acid sequence of bacteriorhodopsin is now known from both protein
and gene sequencing (Dunn et al.). Three-dimensional reconstruction of the
protein by electron microscopy and low-angle electron diffraction analysis
suggests that the protein transverses the membrane seven times, forming an
α helix each time. The exact length of each helix is not known, but it is
thought to be about 30 amino acids (Dunn et al.). A single photon of light
excites the chromophore, causing it to change in conformation; in so doing,
the chromophore transfers two H+ ions out of the cell. The key question now
being addressed is: What path does the ion take through the membrane in
moving from the inside to the outside of the cell? To approach this question,
Khorana and his colleagues are using site-directed mutagenesis (replacing
the natural nucleotide sequence with specifically synthesized oligonucleotides)
to produce specific amino acid replacements.
In rod photoreceptor cells of vertebrates, the rhodopsin is not located in
the external membrane but in the membrane of the disk, an intracellular organelle.
Nonetheless, photoisomerization of the 11-cis retinal chromophore
of rhodopsin to the all-trans form gives rise to a potential change (a hyperpolarization)
in the external (plasma) membrane of the photoreceptor neuron
that is essential for signaling. How is this accomplished? Presumably, a
chemical messenger—a transmitter or an ion—must carry information from
the disk membrane to the external membrane. Evidence points strongly to
the participation of two messengers: Ca++ and cGMP. In the dark, cGMP depolarizes
the membrane by keeping Na+ channels open. Light activates a
phosphodiesterase, which hydrolyzes cGMP. This is thought to have a role
in closing the Na+ channels and hyperpolarizing the cell (although here
more quantitative data are still needed). A single photo-excited rhodopsin
molecule activates several hundred molecules of phosphodiesterase. The
photoactivation of the phosphodiesterase is mediated by transducin, a peripheral
membrane protein whose activation requires the exchange of GDP
for GTP (Stryer). In its dependence on GTP and its sensitivity to inhibition
by cholera toxin, transducin resembles the G protein of the adenylate cyclase
system. This similarity suggests that the transducin and the adenylate cyclase
systems of other tissues belong to the same family of signal-coupling
proteins and that the activation of cGMP phosphodiesterase by light resembles
the activation of adenylate cyclase by hormones or transmitters.
A remarkably good understanding is now also being achieved into
Neurobiology and Molecular Biology 195
chemotransduction from studies of bacteria (Adler; Koshland et al.). Bacteria
(E. coli, Salmonella) have different chemoreceptors for different attractant
and repellent sugars. A few of these receptors are methyl-accepting (chemotaxis)
proteins whose degree of covalent modification is proportional to
stimulus intensity. They generate an excitatory signal—the nature of which
is still not known—which determines frequency of tumbling: the changes in
the direction of rotation of the flagella that move the bacterium. In response
to a positive gradient of attractant, the tumbling is suppressed; the flagella
rotate counterclockwise for long periods, moving the bacterium in a straight
path. For an escape response to a repellent, the flagella rotate clockwise,
causing the bacterium to tumble. The response of the bacterium can adapt
over time, even though the attractant or repellent is still present. This adaptation
results from a change in the methylation of the methyl-accepting
chemotaxis proteins.
Thus, as in the adenylate cyclase and transducin systems, chemoreception
in bacteria involves more than sensing and recognition of the ligand by
the receptor. In each case, the receptors are part of a complex of molecules
that initiate a cascade of events both in series and in parallel. In the case of
the aspartate receptor (Koshland et al.), the three key functions—recognition,
signal transduction, and adaptation—can be separated from each other
by the techniques of in situ mutagenesis.
Recent studies have indicated that in the multicellular nervous systems
of invertebrates and vertebrates there is, imposed upon the network of nerve
cells and interconnections that control a behavior, a set of regulatory processes
that can alter the excitable properties of nerve cells and modify the
strength of their connections. These regulatory processes are activated by
experience, such as learning, and result in the modification of behavior.
Learning refers to the modification of behavior by the acquisition of new
information about the world; memory refers to the retention of the information.
A given learning process can produce both long- and short-term memory.
We are beginning to see in invertebrates how simple neural circuits give
rise to elementary forms of behavior and how these behaviors can be modified
(Aceves-Piña et al.; Kandel et al.; Schwartz et al.). Insights have come
from genetic studies in Drosophila and from cell-biological studies in Aplysia
and other opisthobranch mollusks into simple forms of learning and the
short-term memory for each. In the three forms that have been studied, habituation,
sensitization, and classical conditioning, the learning has been
pinpointed to specific neurons and has been shown to involve changes in
both cellular properties and synaptic strength. In the instances of short-term
memory so far analyzed, the changes in synaptic strength lead to a change
in the amount of transmitter released. Altered transmitter release in turn is
caused by a modulation of ion channels in the presynaptic terminal. In both
196 Psychiatry, Psychoanalysis, and the New Biology of Mind
Drosophila and Aplysia, sensitization and classical conditioning seem to involve
aspects of the same molecular machinery. Short-term memory has
been shown to be independent of new protein synthesis and to involve covalent
modification of preexisting protein by means of cAMP-dependent
protein phosphorylation (Aceves-Piña et al.; Camardo et al.; Kandel et al.;
Schwartz et al.). In classical conditioning, this cascade is amplified, whereas
in sensitization it is not. It is noteworthy that covalent modification of preexisting
proteins also produces behavioral adaptation (this time by methylation)
in bacteria (Adler; Koshland et al.).
Although we are beginning to understand aspects of the molecular
changes underlying short-term memory, we know little about long-term
memory. An important clue has been provided by Craig Bailey and Mary
Chen (1983), who have found that long-term memory in Aplysia is associated
with structural changes in the synapses. It is therefore possible that new
protein synthesis is required to produce these changes (Schwartz et al.).
With recombinant DNA techniques, one should be able to explore the question,
Does learning produce long-term alterations in behavior by regulating
gene expression?
As this last question and the many earlier questions that I have posed illustrate,
we will be confronting in the nervous system some of the most difficult
and profound problems in biology. The early émigrés from molecular biology
were overly optimistic in 1965 in thinking that all but the biology of the
brain could be inferred from the principles at hand. But they were correct in
thinking that the nervous system is one of the last frontiers of biology and
that insights into its cellular and molecular mechanisms are likely to be particularly
penetrating and unifying. For in studying the molecular biology of
the brain, we are taking another important step in a philosophical progression
to which experimental biology has become almost inexorably committed
since Darwin. In Darwin’s time, it was difficult to accept that the human
form was not uniquely created but evolved from lower animals. More recently,
there has been difficulty with the narcissistically even more disturbing
notion that the mental processes of humans have also evolved from those
of animal ancestors and that mentation is not ethereal but can be explained
in terms of nerve cells and their interconnections. The next challenge, which
this symposium and modern neurobiology have opened up for us, is the possibility—
indeed, the likelihood—that many molecules important for the
higher nervous functions of humans may be conserved in evolution and
found in the brains of much simpler animals, and, moreover, that some of
these molecules may not even be unique to the cells of the brain but may be
Neurobiology and Molecular Biology 197
used generally by cells throughout the body. The merger of molecular biology
and neurobiology that the two encounters have accomplished is therefore
more than a merger of methods and concepts. Ultimately, molecular
neurobiology, the joining of the disciplines, represents the emerging conviction
that a coherent and biologically unified description of mentation is possible.
I have benefited from the comments on earlier drafts of this summary by
James H. Schwartz, Sally Muir, Arthur Karlin, and Richard Axel.
Anderson CR, Stevens CF: Voltage clamp analysis of acetylcholine produced endplate
current fluctuations at frog neuromuscular junction. J Physiol 235:655–
691, 1973
Bailey CH, Chen M: Morphological basis of long-term habituation and sensitization
in Aplysia. Science 220:91–93, 1983
Brecha N, Eldred W, Kuljis RO, et al: Identification and localization of biologically
active peptides in the vertebrate retina, in Progress in Retinal Research. Edited
by Osborne NN, Chader GJ. New York, Plenum, 1983
Fatt P, Katz B: An analysis of the end-plate potential recorded with an intra-cellular
electrode. J Physiol 115:320–370, 1951
Gierer A: Molecular models and combinatorial principles in cell differentiation and
morphogenesis. Cold Spring Harb Symp Quant Biol 38:951–961, 1974
Hille B: Ionic basis of resting and action potentials, in Handbook of Physiology; the
Nervous System, Part 1, Vol 1. Edited by Kandel ER. Bethesda, MD, The American
Physiological Society, 1977, p 261
Hodgkin AL, Huxley AF: A quantitative description of membrane current and its application
to conduction and excitation in nerve. J Physiol 117:500–544, 1952
Hodgkin AL, Huxley AF, Katz B: Measurement of current-voltage relations in the
membrane of the giant axon of Loligo. J Physiol 116:424–448, 1952
Hökfelt T, Johansson O, Ljungdahl Å, et al: Peptidergic neurones. Nature 284:515–
521, 1980
Hughes J, Smith TW, Kosterlitz HW, et al: Identification of two related pentapeptides
from the brain with potent opiate agonist activity. Nature 258:577–580, 1975
Katz B, Miledi R: Membrane noise produced by acetylcholine. Nature 226:962–963,
Kidd S, Lockett TJ, Young MW: The notch locus of Drosophila melanogaster. Cell
34:421–433, 1983
Letourneau PC: Cell-to-substratum adhesion and guidance of axonal elongation. Dev
Biol 44:92–101, 1975
198 Psychiatry, Psychoanalysis, and the New Biology of Mind
Mayeri E, Brownell P, Branton WD: Multiple, prolonged actions of neuroendocrine
bag cells on neurons in Aplysia, I. effects on bursting pacemaker neurons. J Neurophysiol
42:1165–1184, 1979
Nachmansohn D: Chemical and Molecular Basis of Nerve Activity. New York, Academic
Press, 1959
Neher E, Sakmann B: Single-channel currents recorded from membrane of denervated
frog muscle fibres. Nature 260:799–802, 1976
Rothman BS, Mayeri E, Brown RO, et al: Primary structure and neuronal effects of α-
bag cell peptide, a second candidate neurotransmitter encoded by a single gene
in bag cell neurons of Aplysia. Proc Natl Acad Sci USA 80:5753–5757, 1983
Rubin LL, Schuetze SM, Weill CL, et al: Regulation of acetylcholinesterase appearance
at neuromuscular junctions in vitro. Nature 283:264–267, 1980
Takeuchi A, Takeuchi N: On the permeability of the presynaptic terminal of the crayfish
neuromuscular junction during synaptic inhibition and the action of γ-
amino-butyric acid. J Physiol 183:433, 1966
Steven E. Hyman, M.D.
The work of Eric Kandel stands as an inspiration to psychiatry because it
connects the experiential and biological levels of analysis with each other
(Kandel 1998). In so doing, this work suggests a serious forward path for
an eventual understanding of the mechanisms by which psychiatric treatments—
especially psychotherapies—might act. That there might be such a
connection seems uncontroversial today, but at the time when Kandel began
his psychiatric training, links between psyche and brain could only be imagined,
and were occasionally denied. Indeed, throughout the mid-twentieth
century, many important figures in psychiatry treated neuroscience as almost
irrelevant to understanding either illness or treatment. Partly as a result,
the typical career path for a person interested both in serious academic
psychiatry and in fundamental neuroscience was to give up one or the other.
As evidenced by the papers collected here, Kandel never abandoned psychiatry.
Although he devoted his career to the bench, not the ward or consulting
room, he reached out to psychiatry at regular intervals to remind its practitioners
of the important connections that could be established (Kandel 1998).
While openly confessing Cartesians (who would declare mind and brain
to be completely different substances requiring special mechanisms to interact)
were rare in late-twentieth century psychiatry, all too many psychiatrists
behaved day to day as if Descartes had been right in his dualism. While by
200 Psychiatry, Psychoanalysis, and the New Biology of Mind
no means a universal view, many psychiatrists in the middle and even the
end of the twentieth century divided disorders into those that were “biological”
and others that resulted from experiences during development. For “biological”
disorders, medication would be the treatment, whereas for those
based on life experience, the answer would lie in psychotherapy. To some degree,
this distinction remains enshrined in the Diagnostic and Statistical
Manual of Mental Disorders, Text Revision (American Psychiatric Association
2000), in its categorical separation of personality disorders (thought to be
experiential in origin) from other psychiatric disorders on its own diagnostic
axis. While such a diagnostic structure would not be agreed to today de
novo, it exists as a fossil record of the thinking of the 1970s. The group of
colleagues who we might describe as “crypto-Cartesians” might have agreed
that a brain is required either to administer psychotherapy or to benefit from
it, but viewed the brain as a rather general substrate about which detailed
understandings might at best serve as a distraction from clinical matters at
hand (very much as Kandel describes the training environment at the Massachusetts
Mental Health Center in the introduction to this volume).
The implication for psychiatry in Kandel’s work and that of others who
have worked on brain plasticity is that life experience and indeed all types
of learning, including psychotherapy, influence thinking, emotion, and behavior
by modifying synaptic connections in particular brain circuits. Moreover,
as many scientists have shown, these circuits are shaped over a lifetime
by multiple complexly interacting factors including genes, illness, injury, experience,
context, and chance.
Clearly, we have a long way to go before we can claim understanding of
the precise cellular mechanisms and neural circuits involved in psychopathology
and its treatment, but substantial progress has been made in understanding
the fundamental mechanisms by which memories are inscribed in
neural circuits, as the following essay shows. This type of progress in basic
neuroscience combined with the rise of cognitive neuroscience, brain imaging,
progress in genetics (albeit slow), and, above all, open-minded pragmatism
about treatment modalities in a younger generation of psychiatrists, has
led to the steady, if not yet complete, emergence of a post-Cartesian psychiatry.
In some sense, psychiatry as a field is now ready to grapple with the
work of Kandel and other scientists who have elucidated the mechanisms by
which the brain is altered by experience in health and in disease.
Besides the undercutting of dualist approaches to mind and brain that is
at the core of Kandel’s experimental work, there is an additional take-home
message for psychiatry in the following essay, “Neural Science: A Century of
Progress and the Mysteries That Remain,” in which the authors take on no
less ambitious a task than summarizing the highlights of neuroscience from
its very beginnings to the present with some predictions as to its most fruitNeural
Science 201
ful future directions. Beginning with the first page of the essay, the authors
distinguish two approaches to neuroscience: a top-down, or holistic, approach
to problems versus a bottom-up, or reductionist, approach to problems.
The essay makes it compellingly clear not only that both approaches
are needed but that they must interact if progress is to be made in understanding
cognition, emotion, the control of behavior, and the underpinnings
of psychiatric illness. That should not be a very controversial point. It must
be added, however, that progress comes only when the right approach is
taken to the problem at hand. The kind of reductionism to which the essay
refers is a scientific approach that is appropriate at a certain stage of problem
solving; it is not a philosophical goal or a worldview. In other words, the experimental
reductionism of Kandel does not represent the goal of explaining
all of human behavior in terms of more and more fundamental components,
such as individual cells, genes, molecules, atoms, or quarks. Rather, the
point is to break down problems into tractable components, with the ultimate
goal of understanding how all of the components come together—in
full recognition of the fact that identifying and characterizing the individual
parts does not explain higher-level phenomena. (Here we have to credit Descartes,
who recommended this approach to science.) As the following essay
illustrates, perhaps most clearly in its extensive discussion of the visual system,
it is not possible to make progress without effective reductionist approaches,
but ultimately, purely reductionist explanations will not answer
our most fundamental questions.
Psychiatry has too often treated holism and reductionism as if they must
be opposed to each other instead of being necessarily complementary approaches
to be wielded wisely as a particular problem dictates. Taking a reductionist
approach to understanding a psychiatric illness through genetics
or neuropathology is not a denial of the importance of the whole person or
the psychosocial context in which he or she functions but an effective route
toward understanding. Kandel’s career illustrates the success that comes
from a disciplined approach to science. Had he taken a prematurely holistic
approach to learning and memory, the results would likely have been superficial
and ultimately unsatisfying. Knowing Eric as I do, I am quite certain
that what he was and is most interested in are the highest integrated aspects
of thought and emotion and how memory contributes to them. However, he
disciplined himself to ask the most penetrating questions that were still tractable.
Kandel was courageous enough to select as a model organism for the
initial stage of his career Aplysia californica, a creature neither well known
nor attractive—and presumably not even tasty (others interested in the neurobiology
of behavior chose to work on the lobster). He chose Aplysia for the
best of reductionist reasons: the organism was complex enough to exhibit
simple forms of learning, but its nervous system was simple enough to be
202 Psychiatry, Psychoanalysis, and the New Biology of Mind
thoroughly analyzed. This organism provided a platform from which to gain
a mechanistic understanding of memory, especially simple forms such as
sensitization. Through years of painstaking investigation, Kandel and his
colleagues were able to provide information that proved relevant to higher
organisms, and indeed, through their more recent efforts on a mammalian
model, the mouse, they have been able to apply what was initially learned
from Aplysia.
It should be noted that even in disciplines that from the point of view of
a psychiatrist might seem inherently fully reductionist, such as cell biology,
the dialectic between reductionism and holism is playing itself out today. It
turns out that the important protein building blocks of cells do not work in
isolation nor can their function within even an individual cell be understood
one molecule at a time. What has become clear is that the molecular components
of cells function within complexly interacting networks that exhibit
compensation, redundancy, and adaptation. We cannot understand the
brain—or individual cells—without knowing the building blocks and their
properties, but we cannot understand cells, organs, the brain, or behavior by
just knowing their component parts.
American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders,
Fourth Edition. Washington, DC, American Psychiatric Association,
Kandel ER: A new intellectual framework for psychiatry. Am J Psychiatry 155:457–
469, 1998
C H A P T E R 6
A Century of Progress and the
Mysteries That Remain
Thomas D. Albright, Ph.D.
Thomas M. Jessell, Ph.D.
Eric R. Kandel, M.D.
Michael I. Posner, Ph.D.
The goal of neural science is to understand the biological mechanisms that
account for mental activity. Neural science seeks to understand how the neural
circuits that are assembled during development permit individuals to
perceive the world around them, how they recall that perception from memory,
and, once recalled, how they can act on the memory of that perception.
Neural science also seeks to understand the biological underpinnings of our
emotional life, how emotions color our thinking, and how the regulation of
emotion, thought, and action goes awry in diseases such as depression, mania,
schizophrenia, and Alzheimer’s disease. These are enormously complex
This article was originally published in Cell, Volume 100, and Neuron, Volume 25,
2000, pp. S1–S55.
204 Psychiatry, Psychoanalysis, and the New Biology of Mind
problems, more complex than any we have confronted previously in other
areas of biology.
Historically, neural scientists have taken one of two approaches to these
complex problems: reductionist or holistic. Reductionist, or bottom-up, approaches
attempt to analyze the nervous system in terms of its elementary
components, by examining one molecule, one cell, or one circuit at a time.
These approaches have converged on the signaling properties of nerve cells
and used the nerve cell as a vantage point for examining how neurons communicate
with one another, and for determining how their patterns of interconnections
are assembled during development and how they are modified
by experience. Holistic, or top-down, approaches focus on mental functions
in alert, behaving human beings and in intact experimentally accessible animals
and attempt to relate these behaviors to the higher-order features of
large systems of neurons. Both approaches have limitations, but both have
had important successes.
The holistic approach had its first success in the middle of the nineteenth
century with the analysis of the behavioral consequences following selective
lesions of the brain. Using this approach, clinical neurologists, led by the pioneering
efforts of Paul Pierre Broca, discovered that different regions of the
cerebral cortex of the human brain are not functionally equivalent (Ryalls
and Lecours 1996; Schiller 1992). Lesions to different brain regions produce
defects in distinctively different aspects of cognitive function. Some lesions
interfere with comprehension of language, others with the expression of language;
still other lesions interfere with the perception of visual motion or of
shape, with the storage of long-term memories, or with voluntary action. In
the largest sense, these studies revealed that all mental processes, no matter
how complex, derive from the brain and that the key to understanding any
given mental process resides in understanding how coordinated signaling in
interconnected brain regions gives rise to behavior. Thus, one consequence
of this top-down analysis has been initial demystification of aspects of mental
function: of language perception, action, learning, and memory (Kandel
et al. 2000).
A second consequence of the top-down approach came at the beginning
of the twentieth century with the work of the Gestalt psychologists, the forerunners
of cognitive psychologists. They made us realize that percepts, such
as those that arise from viewing a visual scene, cannot simply be dissected
into a set of independent sensory elements such as size, color, brightness,
movement, and shape. Rather, the Gestaltists found that the whole of perception
is more than the sum of its parts examined in isolation. How one perceives
an aspect of an image, its shape or color, for example, is in part
determined by the context in which that image is perceived. Thus, the Gestaltists
made us appreciate that to understand perception we needed not
Neural Science 205
only to understand the physical properties of the elements that are perceived,
but more importantly, to understand how the brain reconstructs the
external world in order to create a coherent and consistent internal representation
of that world.
With the advent of brain imaging, the holistic methods available to the
nineteenth-century clinical neurologist, based mostly on the detailed study
of neurological patients with defined brain lesions, were enhanced dramatically
by the ability to examine cognitive functions in intact, behaving normal
human subjects (Posner and Raichle 1994). By combining modern cognitive
psychology with high-resolution brain imaging, we are now entering an era
when it may be possible to address directly the higher-order functions of the
brain in normal subjects and to study in detail the nature of internal representations.
The success of the reductionist approach became fully evident only in
the twentieth century with the analysis of the signaling systems of the brain.
Through this approach, we have learned the molecular mechanisms through
which individual nerve cells generate their characteristic long-range signals
as all-or-none action potentials and how nerve cells communicate through
specific connections by means of synaptic transmission. From these cellular
studies, we have learned of the remarkable conservation of both the longrange
and the synaptic signaling properties of neurons in various parts of the
vertebrate brain—indeed, in the nervous systems of all animals. What distinguishes
one brain region from another and the brain of one species from
the next is not so much the signaling molecules of their constituent nerve
cells but the number of nerve cells and the way they are interconnected. We
have also learned from studies of single cells how sensory stimuli are sorted
out and transformed at various relays and how these relays contribute to perception.
Much as predicted by the Gestalt psychologists, these cellular studies
have shown us that the brain does not simply replicate the reality of the
outside world but begins at the very first stages of sensory transduction to
abstract and restructure external reality.
In this review, we outline the accomplishments and limitations of these
two approaches in attempts to delineate the problems that still confront neural
science. We first consider the major scientific insights that have helped
delineate signaling in nerve cells and that have placed that signaling in the
broader context of modern cell and molecular biology. We then go on to consider
how nerve cells acquire their identity, how they send axons to specific
targets, and how they form precise patterns of connectivity. We also examine
the extension of reductionist approaches to the visual system in an attempt
to understand how the neural circuitry of visual processing can account for
elementary aspects of visual perception. Finally, we turn from reductionist
to holistic approaches to mental function. In the process, we confront some
206 Psychiatry, Psychoanalysis, and the New Biology of Mind
of the enormous problems in the biology of mental functioning that remain
elusive, problems in the biology of mental functioning that have remained
completely mysterious. How does signaling activity in different regions of
the visual system permit us to perceive discrete objects in the visual world?
How do we recognize a face? How do we become aware of that perception?
How do we reconstruct that face at will, in our imagination, at a later time
and in the absence of ongoing visual input? What are the biological underpinnings
of our acts of will?
As the discussions below attempt to make clear, the issue is no longer
whether further progress can be made in understanding cognition in the
twenty-first century. We clearly will be able to do so. Rather, the issue is
whether we can succeed in developing new strategies for combining reductionist
and holistic approaches in order to provide a meaningful bridge
between molecular mechanisms and mental processes: a true molecular biology
of cognition. If this approach is successful in the twenty-first century,
we may have a new, unified, and intellectually satisfying view of mental processes.
The Signaling Capabilities of Neurons
The Neuron Doctrine
Modern neural science, as we now know it, began at the turn of the century
when Santiago Ramón y Cajal provided the critical evidence for the neuron
doctrine, the idea that neurons serve as the functional signaling units of the
nervous system and that neurons connect to one another in precise ways
(Ramón y Cajal 1894, 1906/1967, 1911/1955). Ramón y Cajal’s neuron doctrine
represented a major shift in emphasis to a cellular view of the brain.
Most nineteenth-century anatomists—Joseph von Gerlach, Otto Deiters,
and Camillo Golgi, among them—were perplexed by the complex shape of
neurons and by the seemingly endless extensions and interdigitations of
their axons and dendrites (Shepherd 1991). As a result, these anatomists believed
that the elements of the nervous system did not conform to the cell theory
of Schleiden and Schwann, the theory that the cell was the functional
unit of all eukaryotic tissues.
The confusion that prevailed among nineteenth-century anatomists took
two forms. First, most were unclear as to whether the axon and the many
dendrites of a neuron were in fact extensions that originated from a single
cell. For a long time they failed to appreciate that the cell body of the neuron,
which housed the nucleus, almost invariably gave rise to two types of extensions:
to dendrites that serve as input elements for neurons and that receive
information from other cells, and to an axon serves as the output element of
Neural Science 207
the neuron and conveys information to other cells, often over long distances.
Appreciation of the full extent of the neuron and its processes came ultimately
with the histological studies of Ramón y Cajal and from the studies
of Ross Harrison, who observed directly the outgrowth of axons and dendrites
from neurons grown in isolation in tissue culture.
A second confusion arose because anatomists could not visualize and resolve
the cell membrane and therefore they were uncertain whether neurons
were delimited by membranes throughout their extent. As a result, many believed
that the cytoplasm of two apposite cells was continuous at their points
of contact and formed a syncytium or reticular net. Indeed, the neurofibrils
of one cell were thought to extend into the cytoplasm of the neighboring
cell, serving as a path for current flow from one cell to another. This confusion
was solved intuitively and indirectly by Ramón y Cajal in the 1890s and
definitively in the 1950s with the application of electron microscopy to the
brain by Sanford Palay and George Palade.
Ramón y Cajal was able to address these two questions using two methodological
strategies. First, he turned to studying the brain in newborn animals,
where the density of neurons is low and the expansion of the dendritic
tree is still modest. In addition, he used a specialized silver staining method
developed by Camillo Golgi that labels only an occasional neuron, but labels
these neurons in their entirety, thus permitting the visualization of their cell
body, their entire dendritic tree, and their axon. With these methodological
improvements, Ramón y Cajal observed that neurons, in fact, are discrete
cells, bounded by membranes, and inferred that nerve cells communicate
with one another only at specialized points of appositions, contacts that
Charles Sherrington (1897) was later to call synapses.
As Ramón y Cajal continued to examine neurons in different parts of the
brain, he showed an uncanny ability to infer from static images remarkable
functional insights into the dynamic properties of neurons. One of his most
profound insights, gained in this way, was the principle of dynamic polarization.
According to this principle, electrical signaling within neurons is unidirectional:
the signals propagate from the receiving pole of the neuron—the
dendrites and the cell body—to the axon, and then along the axon to the
output pole of the neuron—the presynaptic axon terminal.
The principle of dynamic polarization proved enormously influential
because it provided the first functionally coherent view of the various compartments
of neurons. In addition, by identifying the directionality of information
flow in the nervous system, dynamic polarization provided a logic
and set of rules for mapping the individual components of pathways in the
brain that constitute a coherent neural circuit (Figure 6–1). Thus, in contrast
to the chaotic view of the brain that emerged from the work of Golgi,
Gerlach, and Deiters, who conceived of the brain as a diffuse nerve net in
208 Psychiatry, Psychoanalysis, and the New Biology of Mind
FIGURE 6–1. Ramón y Cajal’s illustration of neural circuitry of the
A drawing by Ramón y Cajal based on sections of the rodent hippocampus, processed
with a Golgi and Weigert stain. The drawing depicts the flow of information from the
entorhinal cortex to the dentate granule cells (by means of the perforant pathway)
and from the granule cells to the CA3 region (by means of the mossy fiber pathway)
and from there to the CA1 region of the hippocampus (by means of the Schaffer collateral
Source. Based on Ramón y Cajal 1911/1955.
Neural Science 209
which every imaginable type of interaction appeared possible, Ramón y Cajal
focused his experimental analysis on the brain’s most important function:
the processing of information.
Sherrington (1906) incorporated Ramón y Cajal’s notions of the neuron
doctrine, of dynamic polarization, and of the synapse into his book The Integrative
Action of the Nervous System. This monograph extended thinking
about the function of nerve cells to the level of behavior. Sherrington pointed
out that the key function of the nervous system was integration; the nervous
system was uniquely capable of weighing the consequences of different
types of information and then deciding on an appropriate course of action
based upon that evaluation. Sherrington illustrated the integrative capability
of the nervous system in three ways. First, he pointed out that reflex actions
serve as prototypic examples of behavioral integration; they represent coordinated,
purposeful behavior in response to a specific input. For example in
the flexion withdrawal and cross-extension reflex, a stimulated limb will flex
and withdraw rapidly in response to a painful stimulus while, as part of a
postural adjustment, the opposite limb will extend (Sherrington 1910). Second,
since each spinal reflex—no matter how complex—used the motor
neuron in the spinal cord for its output, Sherrington developed (1906) the
idea that the motor neuron was the final common pathway for the integrative
actions of the nervous system. Finally, Sherrington discovered (1932)—
what Ramón y Cajal could not infer—that not all synaptic actions were excitatory;
some could be inhibitory. Since motor neurons receive a convergence
of both excitatory and inhibitory synaptic input, Sherrington argued
that motor neurons represent an example—the prototypical example—of a
cellular substrate for the integrative action of the brain. Each motor neuron
must weigh the relative influence of two types of inputs, inhibitory and excitatory,
before deciding whether or not to activate a final common pathway
leading to behavior. Each neuron therefore recapitulates, in elementary
form, the integrative action of the brain.
In the 1950s and 1960s, Sherrington’s last and most influential student,
John C. Eccles (1953), used intracellular recordings from neurons to reveal
the ionic mechanisms through which motor neurons generate the inhibitory
and excitatory actions that permit them to serve as the final common pathway
for neural integration. In addition, Eccles, Karl Frank, and Michael
Fuortes found that motor neurons had a specialized region, the initial segment
of the axon, which served as a crucial integrative or decision-making
component of the neuron (Eccles 1964; Fuortes et al. 1957). This component
summed the total excitatory and inhibitory input and discharged an action
potential if, and only if, excitation of the motor neuron exceeded
inhibition by a certain critical minimum.
The findings of Sherrington and Eccles implied that each neuron solves
210 Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 211
the competition between excitation and inhibition by using, at its initial segment,
a winner takes all strategy. As a result, an elementary aspect of the integrative
action of the brain could now be studied at the level of individual
cells by determining how the summation of excitation and inhibition leads
to an integrated, all-or-none output at the initial segment. Indeed, it soon became
evident that studies of the motor neuron had predictive value for all
neurons in the brain. Thus, the initial task in understanding the integrative
action of the brain could be reduced to understanding signal integration at
the level of individual nerve cells.
The ability to extend the analysis of neuronal signaling to other regions
of the brain was, in fact, already being advanced by two of Sherrington’s contemporaries,
Edgar Adrian and John Langley. Adrian (1957) developed
methods of single unit analysis within the central nervous system, making it
possible to study signaling in any part of the nervous system at the level of
single cells. In the course of this work, Adrian found that virtually all neurons
use a conserved mechanism for signaling within the cell: the action potential.
In all cases, the action potential proved to be a large, all-or-none,
regenerative electrical event that propagated without fail from the initial segment
of the axon to the presynaptic terminal. Thus, Adrian showed that
FIGURE 6–2. The action potential (opposite page).
(A) This historic recording of a membrane resting potential and an action potential
was obtained by Alan Hodgkin and Andrew Huxley with a capillary pipette placed
across the membrane of the squid giant axon in a bathing solution of seawater. Time
markers (500 Hz) on the horizontal axis are separated by 2 msec. The vertical scale
indicates the potential of the internal electrode in millivolts; the seawater outside is
taken as zero potential.
(B) A net increase in ionic conductance in the membrane of the axon accompanies
the action potential. This historic recording from an experiment conducted in 1938
by Kenneth Cole and Howard Curtis shows the oscilloscope record of an action potential
superimposed on a simultaneous record of the ionic conductance.
(C) The sequential opening of voltage-gated Na+ and K+ channels generates the action
potential. One of Hodgkin and Huxley’s great achievements was to separate the
total conductance change during an action potential, first detected by Cole and Curtis
(Figure 6–2B), into separate components that could be attributed to the opening
of Na+ and K+ channels. The shape of the action potential and the underlying conductance
changes can be calculated from the properties of the voltage-gated Na+ and
K+ channels.
Source. (A) From Hodgkin AL, Huxley AF: “Action Potentials Recorded From Inside
a Nerve Fiber.” Nature 144:710–711, 1939. (B) Modified from Kandel et al. 2000.
(C) From Kandel ER, Schwartz JH, Jessell T: Principles of Neural Science, 4th Edition.
New York, McGraw-Hill, 2000.
212 Psychiatry, Psychoanalysis, and the New Biology of Mind
what made one cell a sensory cell carrying information of vision and another
cell a motor cell carrying information about movement was not the nature
of the action potential that each cell generated. What determined function
was the neural circuit to which that cell belonged.
Sherrington’s other contemporary, John Langley (1906), provided some
of the initial evidence (later extended by Otto Loewi, Henry Dale, and Wilhelm
Feldberg) that, at most synapses, signaling between neurons—synaptic
transmission—was chemical in nature. Thus, the work of Ramón y Cajal,
Sherrington, Adrian, and Langley set the stage for the delineation, in the second
half of the twentieth century, of the mechanisms of neuronal signaling—
first in biophysical (ionic), and then in molecular terms.
Long-range signaling within neurons: the action potential
In 1937, Alan Hodgkin found that an action potential generates a local flow
of current that is sufficient to depolarize the adjacent region of the axonal
membrane, in turn triggering an action potential. Through this spatially interactive
process along the surface of the membrane, the action potential is
propagated without failure along the axon to the nerve terminal (Figure 6–
2A). In 1939, Kenneth Cole and Howard Curtis further found that when an
all-or-none action potential is generated, the membrane of the axon undergoes
a change in ionic conductance, suggesting that the action potential reflects
the flow of ionic current (Figure 6–2B).
Hodgkin, Andrew Huxley, and Bernhard Katz extended these observations
by examining which specific currents flow during the action potential.
In a landmark series of papers in the early 1950s, they provided a quantitative
account of the ionic currents in the squid giant axon (Hodgkin et al.
1952). This view, later called the ionic hypothesis, explained the resting membrane
potential in terms of voltage-insensitive (nongated or leakage) channels
permeable primarily to K+ and the generation and propagation of the
action potential in terms of two discrete, voltage-gated conductance pathways,
one selective for Na+ and the other selective for K+ (Figure 6–2C).
The ionic hypothesis of Hodgkin, Huxley, and Katz remains one of the
deepest insights in neural science. It accomplished for the cell biology of
neurons what the structure of DNA did for the rest of biology. It unified the
cellular study of the nervous system in general, and in fact, the study of ion
channels in general. One of the strengths of the ionic hypothesis was its generality
and predictive power. It provided a common framework for all electrically
excitable membranes and thereby provided the first link between
neurobiology and other fields of cell biology. Whereas action potential signaling
is a relatively specific mechanism distinctive to nerve and muscle
cells, the permeability of the cell membrane to small ions is a general feature
Neural Science 213
shared by all cells. Moreover, the ionic hypothesis of the 1950s was so precise
in its predictions that it paved the way for the molecular biological explosion
that was to come in the 1980s.
Despite its profound importance, however, the analysis of Hodgkin,
Huxley, and Katz left something unspecified. In particular, it left unspecified
the molecular nature of the pore through the lipid membrane bilayer and the
mechanisms of ionic selectivity and gating. These aspects were first addressed
by Bertil Hille and Clay Armstrong. In the late 1960s, Hille devised
procedures for measuring Na+ and K+ currents in isolation (for a review, see
Hille et al. 1999). Using pharmacological agents that selectively block one
but not the other ionic conductance pathway, Hille was able to infer that the
Na+ and K+ conductance pathways of Hodgkin and Huxley corresponded to
independent ion channel proteins. In the 1970s, Hille used different organic
and inorganic ions of specified size to provide the first estimates of the size
and shape of the pore of the Na+ and the K+ channels. These experiments led
to the defining structural characteristic of each channel—the selectivity filter—
the narrowest region of the pore, and outlined a set of physical-chemical
mechanisms that could explain how Na+ channels are able to exclude K+ and
conversely, how K+ channels exclude Na+.
In parallel, Armstrong addressed the issue of gating in response to a
change in membrane voltage. How does an Na+ channel open rapidly in response
to voltage change? How, once opened, is it closed? Following initial
experiments of Knox Chandler on excitation contraction coupling in muscle,
Armstrong measured minute “gating” currents that accompanied the
movement, within the transmembrane field, of the voltage sensor postulated
to exist by Hodgkin and Huxley. This achievement led to structural predictions
about the number of elementary charges associated with the voltage
sensor. In addition, Armstrong discovered that mild intracellular proteolysis
selectively suppresses Na+ channel inactivation without affecting voltage-dependent
activation, thereby establishing that activation and inactivation involve
separate (albeit, as later shown, kinetically linked) molecular
processes. Inactivation reflects the blocking action of a globular protein domain,
a “ball,” tethered by a flexible peptide chain to the intracellular side of
the channel. Its entry into the mouth of the channel depends on the prior
activation (opening) of the channel. This disarmingly simple “mechanical”
model was dramatically confirmed by Richard Aldrich in the early 1990s. Aldrich
showed that a cytoplasmic aminoterminal peptide “ball” tethered by a
flexible chain does indeed form part of the K+ channel and underlies its inactivation,
much as Armstrong predicted.
Until the 1970s, measurement of current flow was carried out with the
voltage-clamp technique developed by Cole, Hodgkin, and Huxley, a technique
that detected the flow of current that followed the opening of thou214
Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 215
sands of channels. The development of patch-clamp methods by Erwin
Neher and Bert Sakmann revolutionized neurobiology by permitting the
characterization of the elemental currents that flow when a single ion channel—
a single membrane protein—undergoes a transition from a closed to an
open conformation (Neher and Sakmann 1976) (Figure 6–4A). This technical
advance had two additional major consequences. First, patch clamping
could be applied to cells as small as 2–5 μm in diameter, whereas voltage
clamping could only be carried out routinely on cells 50 μm or larger. Now,
it became possible to study biophysical properties of the neurons of the
mammalian brain and to study as well a large variety of nonneuronal cells.
With these advances came the realization that virtually all cells harbor in
their surface membrane (and even in their internal membranes) Ca2+ and K+
channels similar to those found in nerve cells. Second, the introduction of
patch clamping also set the stage for the analysis of channels at the molecular
level, and not only voltage-gated channels of the sort we have so far considered
but also of ligand-gated channels, to which we now turn.
Short-range signaling between neurons: synaptic transmission
The first interesting evidence for the generality of the ionic hypothesis of
Hodgkin, Huxley, and Katz was the realization in 1951 by Katz and Paul Fatt
that, in its simplest form, chemical synaptic transmission represents an extension
of the ionic hypothesis (Fatt and Katz 1951, 1952). Fatt and Katz
found that the synaptic receptor for chemical transmitters was an ion channel.
But rather than being gated by voltage as were the Na+ and K+ channels,
the synaptic receptor was gated chemically, by a ligand, as Langley, Dale,
FIGURE 6–3. The conductance of single ion channels and a preliminary
view of channel structure (opposite page).
(A) Recording of current flow in single ion channels. Patch-clamp record of the current
flowing through a single ion channel as the channel switches between its closed
and open states.
(B) Reconstructed electron microscope view of the ACh receptor-channel complex in
the fish Torpedo californica. The image was obtained by computer processing of negatively
stained images of ACh receptors. The resolution is 1.7 nm, fine enough to visualize
overall structure but too coarse to resolve individual atoms. The overall
diameter of the receptor and its channel is about 8.5 nm. The pore is wide at the external
and internal surfaces of the membrane but narrows considerably within the
lipid bilayer. The channel extends some distance into the extracellular space.
Source. (A) Courtesy of B. Sakmann. (B) Adapted from studies by Toyoshima and
Unwin; from Kandel ER, Schwartz JH, Jessell T: Principles of Neural Science, 4th Edition.
New York, McGraw-Hill, 2000.
216 Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 217
Feldberg, and Loewi had earlier argued. Fatt and Katz and Takeuchi and
Takeuchi showed that the binding of acetylcholine (ACh), the transmitter
released by the motor nerve terminal, to its receptors leads to the opening of
a new type of ion channel, one that is permeable to both Na+ and K+ (Takeuchi
and Takeuchi 1960) (Figure 6–3). At inhibitory synapses, transmitters,
typically γ-aminobutyric acid (GABA) or glycine, open channels permeable
to Cl– or K+ (Boistel and Fatt 1958; Eccles 1964).
In the period 1930 to 1950, there was intense controversy within the neural
science community about whether transmission between neurons in the
central nervous system occurred by electrical or chemical means. In the early
1950s Eccles, one of the key proponents of electrical transmission, used intracellular
recordings from motor neurons and discovered that synaptic excitation
and inhibition in the spinal cord was mediated by chemical synaptic
transmission. He further found that the principles of chemical transmission
derived by Fatt and Katz from studies of peripheral synapses could be readily
extended to synapses in the nervous system (Brock et al. 1952; Eccles 1953,
1964). Thus, during the 1960s and 1970s the nature of the postsynaptic response
at a number of readily accessible chemical synapses was analyzed, including
those mediated by ACh, glutamate, GABA, and glycine (see, for
example, Watkins and Evans 1981). In each case, the transmitter was found
to bind to a receptor protein that directly regulated the opening of an ion channel.
Even prior to the advent in the 1980s of molecular cloning, which we shall
consider below, it had become clear, from the biochemical studies of Jean-
Pierre Changeux and of Arthur Karlin, that in ligand-gated channels the trans-
FIGURE 6–4. The membrane topology of voltage- and ligand-gated
ion channels (opposite page).
(A) The basic topology of the α subunit of the voltage-gated Na+ channel, and the
corresponding segments of the voltage-gated Ca2+ and K+ channels. The α subunit of
the Na+ and Ca2+ channels consists of a single polypeptide chain with four repetitions
of six membrane-spanning α helical regions. The S4 region, the fourth membrane-
spanning α helical region, is thought to be the voltage sensor. A stretch of
amino acids, the P region between the fifth and sixth α helices, dips into the membrane
in the form of two strands. A fourfold repetition of the P region is believed to
line the pore. The shaker type K+ channel, by contrast, has only a single copy of the
six α helices and the P region. Four such subunits are assembled to form a complete
(B) The membrane topology of channels gated by the neurotransmitters ACh, GABA
glycine, and kainate (a class of glutamate receptor ligand).
Source. (A) Adapted from Catterall 1988 and Stevens 1991. (B) From Kandel ER,
Schwartz JH, Jessell T: Principles of Neural Science, 4th Edition. New York, McGraw-
Hill, 2000.
218 Psychiatry, Psychoanalysis, and the New Biology of Mind
mitter binding site and the ionic channel constitute different domains within
a single multimeric protein (for reviews see Changeux et al. 1992; Cowan and
Kandel 2000; Karlin and Akabas 1995).
As with voltage-gated channels, the single channel measurements of Neher
and Sakmann (1976) brought new insights into ligand-gated channels.
For example, in the presence of ligand, the ACh channel at the vertebrate
neuromuscular junction opens briefly (on average for 1–10 msec) and gives
rise to a square pulse of inward current, roughly equivalent to 20,000 Na+
ions per channel per msec. The extraordinary rate of ion translocation revealed
by these single channel measurements confirmed directly the idea of
the ionic hypothesis—that ions involved in signaling cross the membrane by
passive electrochemical movement through aqueous transmembrane channels
rather than through transport by membrane carriers (Figure 6–3A).
Following the demonstration of the chemical nature of transmission at
central as well as peripheral synapses, neurobiologists began to suspect that
communication at all synapses was mediated by chemical signals. In 1957,
however, Edwin Furshpan and David Potter (1957) made the discovery that
transmission at the giant fiber synapse in crayfish was electrical. Subsequently,
Michael Bennett (1972) and others showed that electrical transmission
was widespread and operated at a variety of vertebrate and invertebrate
synapses. Thus, neurobiologists now accept the existence of two major
modes of synaptic transmission: electrical, which depends on current
through gap junctions that bridge the cytoplasm of pre- and postsynaptic
cells; and chemical, in which pre- and postsynaptic cells have no direct continuity
and are separated by a discrete extracellular space, the synaptic cleft
(Bennett 2000).
The Proteins Involved in Generating Action Potentials and
Synaptic Potentials Share Features in Common
In the 1980s, Shosaku Numa, Lily Yeh Jan, Yuh Nung Jan, William Catterall,
Steven Heineman, Peter Seeburg, Heinrich Betz, and others cloned and expressed
functional voltage-gated Na+, Ca2+, and K+ channels, as well as the
ligand-gated receptor channels for ACh, GABA, glycine, and glutamate
(Armstrong and Hille 1998; Green et al. 1998; Numa 1989). Prior biophysical
studies already had taught us much about channels, and as a consequence
molecular cloning was in a position rapidly to provide powerful new
insights into the membrane topology and subunit composition of both voltage-
gated and ligand-gated signaling channel proteins (Armstrong and Hille
1998; Colquhoun and Sakmann 1998). Molecular cloning revealed that all
ligand-gated channels have a common overall design and that this design
shares features with voltage-gated channels.
Neural Science 219
Based on sequence identity, ligand-gated channels can be divided into
two superfamilies: 1) receptors for glutamate (of the NMDA [N-methyl-D-aspartic
acid] and non-NMDA classes) and 2) receptors for other small molecule
transmitters: nicotinic ACh, 5-hydroxytryptamine, GABA, glycine, and
ATP (Green et al. 1998) (Figure 6–6). Of these, the most detailed information
is again available on the nicotinic ACh receptors of skeletal muscle (Figure
6–3B). This receptor is made up of four distinct subunits, α, β, γ, and δ,
with the α subunit represented twice in a five-subunit channel (α2βγδ).
Three-dimensional images reveal a channel made up of the five subunits surrounding
the water-filled channel pore (Figures 6–3B and 6–4). Much as
predicted by Hille, the channel appears to be divided into three regions: a
relatively large entrance region on the external surface; a narrow transmembrane
pore, only a few atomic diameters wide, which selects for ions on the
basis of their size and charge; and a large exit region on the internal plasma
membrane surface.
The first of the voltage-sensitive channels to be cloned, the brain Na+
channel, was found to consist of one large (α) and two smaller (β) subunits.
The α subunit is widely distributed and is the major pore-forming subunit
essential for transmembrane Na+ flux, whereas the smaller subunits are regulatory
and are expressed only by subsets of cells (where they participate in
channel assembly and inactivation). The α subunit consists of a single peptide
of about 2,000 amino acids with four internally repeated domains of
similar structures. Each domain contains six putative membrane-spanning
segments, S1 to S6, which are thought to be α helical, and a reentrant P loop.
The P loop connects the S5 and S6 segments and forms the outer mouth and
selectivity filter of the channel.
The voltage-gated Ca2+ channels are similar to the Na+ channel in their
overall design. However, each of the cloned K+ channels encodes only a single
domain, of about 600 amino acids, containing the six putative transmembrane
regions and the P loop. As might be predicted from this structure,
four of these subunits are required to form a functional channel (either as
homo- or as heterotetramers).
The wealth of sequence information that emerged from molecular cloning
illustrated the remarkable conservation of channel molecules, and in
turn demanded information on the structure of these channels. One of the
recent successes of ion channel biology has been the first steps in the elucidation
of ion channel structure. The first ion channel structure to be
revealed was that of a K+ channel (called KcsA) from the bacterium, Streptomyces
lividans. The amino acid sequence of KcsA shows it to be most similar
to the inward rectifier type of K+ channel that contributes to the regulation
of the resting membrane potential. The amino acid sequence of these channels
predicts only two transmembrane domains connected by a P loop, in
220 Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 221
contrast to the more familiar voltage-gated K+ channels, which have six
transmembrane domains. When reconstituted in lipid bilayers, KcsA forms
a tetramer. The 3.2 Å resolution crystal structure reported by Roderick
MacKinnon and his colleagues revealed that the tetramer has two transmembrane-
spanning α helices connected by the P region (Doyle et al. 1998) (Figures
6–5A and 6–5B).
In retrospect it was remarkable how accurately this structure had been
anticipated by the earlier biophysical studies of Hille and Armstrong. Hille
and Armstrong had, for example, correctly predicted the selectivity filter to
be a narrow region near the outer face of the membrane lined by polar residues.
One surprise, however, is that the channel pore is not lined by hydrophilic
amino acid side chains but by the carbonyl backbone of conserved
amino acids, containing glycine-tyrosine-glycine residues that are characteristic
of nearly all K+-selective channels. The narrow channel in the selectivity
filter rapidly broadens in hourglass fashion to form a “lake” roughly halfway
through the membrane, in which 60–100 water molecules diffuse the
charges of K+ ions residing in this cavity. Four short α helices in the P loops
have their helix dipole negative electrostatic fields focused on the cavity, further
stabilizing the K+ ion poised at the selectivity filter. Finally, a long waterfilled
hydrophobic channel tunnels to the cytoplasm.
MacKinnon’s compelling images even visualized two K+ ions within the
selectivity filter. Thus, a total of three K+ ions are positioned at distinct sites
within the pore, each separated from the other by about 8 Å. This view of a
single pore capable of accommodating three K+ ions was precisely as predicted
by Hodgkin some 50 years earlier. MacKinnon’s structure thus pro-
FIGURE 6–5. The crystal structure of a bacterial inward-rectifying
K+ channel and a glutamate receptor (opposite page).
(A1) A view of the bacterial K+ channel in cross section in the plane of the membrane.
The four subunits are shown, with each subunit depicted in a different color. The
membrane-spanning helices are arranged as an inverted teepee.
(A2) A side view of the channel illustrating three K+ ions within the channel. The pore
helices contribute a negative dipole that helps stabilize the K+ ion in the water-filled
inner chamber. The two outer K+ ions are loosely bound to the selectivity filter
formed by the P region.
(B) Schematic depiction of a bacterial ligand-gated glutamate receptor channel with
a K+ channel pore. The extracellular regions of the channel show sequence similarity
to the ligand-binding domains of glutamate receptors (red in the figure here). The
pore region resembles an inverted potassium channel pore (blue).
Source. (A2
) From Doyle DA, Morais Cabral J, Pfuetzner RA, et al: “The Structure
of the Potassium Channel: Molecular Basis of K+ Conduction and Selectivity.” Science
280:69–77, 1998. (B) Image courtesy of E. Gouaux; see Chen et al. 1999.
222 Psychiatry, Psychoanalysis, and the New Biology of Mind
vided explanations for K+ channel selectivity and conduction. What we lack,
however, is an insight into the mechanisms of voltage-dependent gating.
The membrane subunits of many voltage-dependent potassium channels
associate with additional proteins known as the β subunits (Isom et al.
1994). One function of β subunits is to modify the gating of K+ channels.
MacKinnon and his colleagues have now gone on to provide the structure of
the β subunit of a voltage-dependent K+ channel from eukaryotic cells (Gulbis
et al. 1999). Like the integral membrane components of the potassium
channel, the β subunits have a fourfold symmetrical structure. Surprisingly,
each subunit appears similar to an oxidoreductase enzyme, complete with a
nicotinamide cofactor active site. Several structural features of the enzyme
active site, including its location with respect to the fourfold axis, imply that
it may interact directly or indirectly with the K+ channel’s voltage sensor.
Thus, the oxidative chemistry of the cell may be intrinsically linked to
changes in membrane potential by the interaction of the α and β subunits of
the voltage-dependent K+ channels.
The expression of ligand-gated receptors also is not limited to multicellular
organisms. For example, it has become evident recently that even
prokaryotes have functional ligand-gated glutamate receptors. Eric Gouaux
and his colleagues (Chen et al. 1999) have cloned and expressed a
glutamate-gated channel from the cyanobacterium Synechocystis PCC 6803,
and in so doing have provided a further surprise: the receptor has a transmembrane
structure similar to that of KcsA and forms a K+ -selective pore.
Thus, this receptor is related both to the inward rectifier K+ channels and to
eukaryotic glutamate receptors (Figure 6–5B). The extracellular region
bears sequence homology to the ligand-binding domains of glutamate receptors,
whereas the pore region bears resemblance to an inverted K+ channel.
This finding has led Gouaux and his colleagues to propose a prokaryotic
glutamate receptor as the precursor of eukaryotic receptors. In addition, this
receptor provides a missing link between K+ channels and glutamate receptors,
and indicates that both ligand- and voltage-gated ion channels have a
similar architecture, suggesting that they both derive from a common bacterial
Synaptic Receptors Coupled to G Proteins
Produce Slow Synaptic Signals
In the 1970s, evidence began to emerge from Paul Greengard and others that
the neurotransmitters that activate ligand-gated (ionotropic) channels to produce
rapid synaptic potentials lasting only milliseconds—glutamate, ACh,
GABA, serotonin—also interact with a second, even larger class of receptors
(termed metabotropic receptors) that produce slow synaptic responses that
Neural Science 223
persist for seconds or minutes (for a review, see Nestler and Greengard
1984). Thus, a single presynaptic neuron releasing a single transmitter can
produce a variety of actions on different target cells by activating distinct
ionotropic or metabotropic receptors.
Molecular cloning revealed that these slow synaptic responses are transduced
by members of a superfamily of receptors with seven transmembranespanning
domains, which do not couple to ion channels directly but do so
indirectly by means of their coupling to G proteins. G proteins couple this
class of receptors to effector enzymes that give rise to second messengers
such as cAMP, cGMP, diacylglycerol, and metabolites of arachidonic acid.
G proteins and second messengers can activate some channels directly. More
commonly, these messengers activate further downstream signaling molecules,
often a protein kinase that regulates channel function by phosphorylating
the channel protein or an associated regulatory protein (for review see
Nestler and Greengard 1984). The family of G protein–coupled seven transmembrane-
spanning receptors is remarkably large, and its members serve
not only as receptors for small molecule and peptide transmitters, but also
as the sensory receptors for vision and olfaction.
The study of slow synaptic potentials mediated by second messengers
has added several new features to our understanding of chemical transmission.
Four of these are particularly important. First, second messenger systems
regulate channel function by acting on cytoplasmic domains of
channels. This type of channel regulation can be achieved in three different
ways: 1) through the phosphorylation of the channel protein by a second
messenger-activated protein kinase, 2) through the interaction between the
channel protein and a G protein activated by the ligated receptor, or 3) by
the direct binding to the channel protein of a cyclic nucleotide, as is the case
with the ion channels of photoreceptor and olfactory receptor cells gated by
cAMP or cGMP. Second, by acting through second messengers, transmitters
can modify proteins other than the channels, thereby activating a coordinated
molecular response within the postsynaptic cell. Third, second messengers
can translocate to the nucleus and modify transcriptional regulatory
protein, in this way controlling gene expression. Thus, second messengers
can covalently modify preexisting proteins as well as regulate the synthesis
of new proteins. This latter class of synaptic action can lead to long-lasting
structural changes at synapses. Finally, we are beginning to appreciate functional
differences in slow synaptic actions. Whereas fast synaptic actions are
critical for routine behavior, slow synaptic actions are often modulatory and
act upon neural circuits to regulate the intensity, form, and duration of a
given behavior (Kandel et al. 2000).
224 Psychiatry, Psychoanalysis, and the New Biology of Mind
Chemical Transmitter Is Released From the
Presynaptic Terminal in Multimolecular Packets
In addition to providing initial insights into the structure and function of the
ligand-gated postsynaptic receptors responsible for postsynaptic transmission,
Katz and Fatt also provided the groundwork for a molecular analysis
of transmitter release from the presynaptic terminals with the discovery of
its quantal nature (reviewed in B. Katz 1969). Katz, with Fatt and Jose del
Castillo, discovered that chemical transmitters, such as ACh, are released
not as single molecules but as multimolecular packets called quanta. At the
neuromuscular junction each quantum comprises about 5,000 molecules of
transmitter (del Castillo and Katz 1954; Fatt and Katz 1952). Each quantum
of ACh (and of other small molecule transmitters such as glutamate or
GABA) is packaged in a single small organelle, the synaptic vesicle, and is released
by exocytosis at specialized release sites within the presynaptic terminal
called the active zones. In response to a presynaptic action potential, each
active zone generally releases 0 or 1 quantum, in a probabilistic manner
(Figure 6–6). Synapses that release large quantities of transmitter to evoke a
large postsynaptic response, such as the synapse between nerve and muscle,
contain several hundred active zones (Heuser 1977) (Figures 6–8A and 6–
8B). In the central nervous system, however, many presynaptic terminals
contain only a single active zone.
Fatt and Katz (1952) discovered that synapses release quanta spontaneously,
even in the absence of activity, giving rise to spontaneous miniature synaptic
potentials. For a single active zone, the rate of spontaneous release is
quite low, around 10–2 per second. In response to a presynaptic action potential,
the rate of release is dramatically, but transiently, elevated to around
1,000 per second. Within a few milliseconds, the quantal release rate then
decays back to its low resting level. We know from the work of Katz and
Ricardo Miledi as well as from the studies of Rodolfo Llinas that intracellular
Ca2+ is the key signal that triggers the increase in release. When the action
potential invades the terminal, it opens voltage-gated Ca2+ channels that are
enriched near the active zone. The resultant influx of Ca2+ produces localized
accumulations of Ca2+ (to >100 μM) in microdomains of the presynaptic
terminal near the active zone release site. The local increase in Ca2+
concentration greatly enhances the probability of vesicle fusion and transmitter
release. Many presynaptic terminals also have ionotropic and metabotropic
receptors for transmitters, and these, in turn, modulate Ca2+ influx
during an action potential and thus modify transmitter release.
Kinetic analyses suggest that the exocytotic release of neurotransmitter
from synaptic vesicles involves a cycle composed of at least four distinct
steps: 1) the transport (or mobilization) of synaptic vesicles from a reserve
Neural Science 225
FIGURE 6–6. The quantal nature of neurotransmitter release.
Neurotransmitters are released in fixed unitary increments, or quanta. Each quantum
of transmitter produces a postsynaptic potential of fixed amplitude. The amplitude
of the postsynaptic potential depends on the quantal unit amplitude multiplied by
the number of quanta of transmitter.
(A) Intracellular recordings show the change in potential when eight consecutive
stimuli of the same size are applied to a motor nerve. To reduce transmitter output
and to keep the end-plate potentials small, the tissue was bathed in a Ca2+-deficient
(and Mg2+-rich) solution. The responses to the stimulus vary. Two impulses produce
complete failures, two produce unit potentials, and the others produce responses that
are approximately two to four times the amplitude of the unit potential. The spontaneous
miniature end-plate potentials (S) are similar in size to the quantal unit potential.
(B) The quantal nature of neurotransmitter release. After recording many end-plate
potentials, the number of responses at each amplitude was counted and plotted. The
distribution of responses falls into a number of peaks. The first peak, at 0 mV, represents
release failures. The first peak at 0.4 mV represents the unit potential, the smallest
elicited response. This unit response is the same amplitude as the spontaneous
miniature potentials (inset). The other peaks in the histogram occur at amplitudes
that are integral multiples of amplitude of the unit potential. The solid line shows a
theoretical Gaussian distribution fitted to the data of the histogram. Each peak is
slightly spread out, reflecting the fact that the amount of transmitter in each quantum
varies randomly about the peak. The distribution of amplitudes of the spontaneous
miniature potentials, shown in the inset, also fits a Gaussian curve (solid line).
Source. (A) Adapted from Liley 1956. (B) Adapted from Boyd and Martin 1956.
226 Psychiatry, Psychoanalysis, and the New Biology of Mind
pool (tethered to the cytoskeleton) to a releasable pool at the active zone; 2)
the docking of vesicles to their release sites at the active zone; 3) the fusion
of the synaptic vesicle membrane with the plasma membrane during exocytosis,
in response to a local increase in intracellular Ca2+; and 4) the retrieval
and recycling of vesicle membrane following exocytosis.
A major advance in the analysis of transmitter release was provided by
the biochemical purification and molecular cloning of the proteins that participate
in different aspects of the vesicle release cycle (Figure 6–7). Paul
Greengard’s work on the synapsins and their role in short-term synaptic
plasticity, the work of Thomas Südhof and Richard Scheller on vesicleassociated
proteins, and the work of Pietro De Camilli on membrane retrieval
have each contributed seminally to our current view of the dynamics
of synaptic vesicle mobilization, docking, and release (for reviews, see Bock
and Scheller 1999; Fernandez-Chacon and Südhof 1999). Although we now
know most of the molecular participants, at present we still do not have a
precise understanding of the molecular events that control any of the four
kinetic stages of release. In some instances, however, we have a beginning.
By reconstituting the vesicle cycling system in a test tube, James Rothman
and his colleagues have succeeded in identifying proteins that are essential
for vesicle budding, targeting, recognition, and fusion (Nickel et al.
1999; Parlati et al. 1999; Söllner et al. 1993). Based on these studies, Rothman
and colleagues have proposed an influential model, according to which
vesicle fusion requires specialized donor proteins (vesicle snares or v-snares)
intrinsic to the vesicle membrane that are recognized by and bind to specific
receptor proteins in the target membrane (target snares or t-snares).
Rothman, Scheller, and their colleagues have found that two proteins located
in the nerve terminal plasma membrane—syntaxin and SNAP-25—
appear to have the properties of plasma membrane t-snares, whereas synaptobrevins/
VAMP (vesicle-associated membrane protein), located on the
membrane of the synaptic vesicles, have the properties of the donor proteins,
or v-snares. The importance of the three snare proteins—VAMP, syntaxin,
and SNAP-25—in synaptic transmission was immediately underscored by
the findings that these three proteins are the targets of various clostridial
neurotoxins, metalloproteases that irreversibly inhibit synaptic transmission.
Subsequent reconstitution studies by Rothman and his colleagues
showed that fusion could occur with liposomes containing v- and t-snares
(Weber et al. 1998). Finally, structural studies by Reinhard Jahn and his colleagues
based on quick-freeze/deep-etch electron microscopy and X-ray
crystallography demonstrated that VAMP forms a helical coiled-coil structure
with syntaxin and SNAP-25 that is thought to promote vesicle fusion by
bringing the vesicle and plasma membrane into close apposition (Hanson et
al. 1997; Sutton et al. 1998). From these studies it would appear that vesicle
Neural Science 227
FIGURE 6–7. Some vesicle terminal membrane–associated proteins.
This diagram depicts characterized synaptic vesicle proteins and some of their postulated
receptors and functions. Separate compartments are assumed for 1) storage
(where vesicles are tethered to the cytoskeleton), 2) the trafficking and targeting of
vesicles to active zones, 3) the docking of vesicles at active zones and their priming
for release, and 4) release. Some of these proteins represent the targets for neurotoxins
that act by modifying transmitter release. VAMP (synaptobrevin), SNAP-25, and
syntaxin are the targets for tetanus and botulinum toxins, two zinc-dependent metalloproteases,
and are cleaved by these enzymes. α-Latrotoxin, a spider toxin that
generates massive vesicle depletion and transmitter release, binds to the neurexins.
1) Synapsins are vesicle-associated proteins that are thought to mediate interactions
between the synaptic vesicle and the cytoskeletal elements of the nerve terminal. 2)
Rab GTPases appear to be involved in vesicle trafficking within the cell and also in
the targeting of vesicles within the nerve terminal. 3) Vesicle docking, fusion, and release
appear to involve distinct interactions between vesicle proteins and proteins of
the nerve terminal plasma membrane: VAMP (synaptobrevin) and synaptotagmin
(p65) are located on the vesicle membrane, and syntaxins and neurexins on the nerve
terminal membrane. Arrows indicate potential interactions suggested on the basis of
in vitro studies. 4) The identity of the vesicle and plasma membrane proteins that
comprise the fusion pore remains unclear. Synaptophysin, an integral membrane
protein in synaptic vesicles, is phosphorylated by tyrosine kinases and may regulate
release. Vesicle transporters are involved in the concentration of neurotransmitter
within the synaptic vesicle.
Source. From Kandel ER, Schwartz JH, Jessell T: Principles of Neural Science, 4th
Edition. New York, McGraw-Hill, 2000.
228 Psychiatry, Psychoanalysis, and the New Biology of Mind
fusion uses a helical coiled-coil mechanism analogous to that used for viral
fusion proteins (Bock and Scheller 1999; Nickel et al. 1999; Parlati et al.
1999; Söllner et al. 1993). Indeed, VAMP resembles a viral fusion peptide.
One of the most important insights to emerge from research on synaptic
vesicle–associated proteins is that sets of molecules similar to those involved
in mediating evoked transmitter release are also important for constitutive
release. Indeed, homologs of the v- and t-snares participate in many aspects
of membrane trafficking and constitutive vesicle fusion, including the trafficking
of vesicles from the endoplasmic reticulum to the Golgi. Thus, the
properties of v- and t-snares do not by themselves explain the specific tight
Ca2+-dependent regulation of vesicle fusion characteristic of evoked transmitter
release from nerve terminals. Südhof has presented evidence that this
calcium-dependent step in synaptic vesicle fusion is mediated by the synaptic
vesicle proteins, the synaptotagmins (or p65). The synaptotagmins contain
two domains (C2 domains) homologous to the Ca2+ and phospholipidbinding
regulatory region of protein kinase C. This property suggested to
Südhof that the synaptotagmins might insert into the presynaptic phospholipid
bilayer in response to Ca2+ influx, thus serving as the Ca2+ sensor for
exocytosis. Indeed, as shown by Charles Stevens, mice lacking the synaptotagmin-
1 gene lack the fast synchronized Ca2+-dependent phase of synaptic
transmitter, although spontaneous release (which does not depend on
Ca2+ influx) occurs normally (Fernandez-Chacon and Südhof 1999).
Neurotransmitter is taken up by membrane transporters
Acetylcholine was the first transmitter substance to be identified. In the
course of studying its function, it soon became apparent that the action of
ACh was terminated by the enzyme acetylcholinesterase. This enzyme is located
in the basal membrane in close apposition to the ACh receptor and
regulates the amount of ACh available for interaction with the receptor and
the duration of its action. Thus, drugs that inhibit the acetylcholinesterase
potentiate and prolong the synaptic effects of ACh.
Based upon this set of findings in the cholinergic system, most neurobiologists
in the 1950s assumed that all neurotransmitter systems would similarly
be inactivated by enzymatic degradation. Thus, when norepinephrine
was discovered to be an autonomic transmitter, it was expected that there
would be enzymes with a dedicated degradative function. But in 1959, Julius
Axelrod and his colleagues found that actions of norepinephrine were terminated
not by enzymatic degradation but by a pump-like mechanism that
transports norepinephrine back into the presynaptic nerve terminal (Hertting
and Axelrod 1961; Iversen 1967). Similar uptake mechanisms were
soon found for serotonin and for other amine and amino acid neurotransNeural
Science 229
mitters. The mechanism of enzymatic degradation that inactivates ACh, in
fact, turned out to be an exception rather than a rule. Reuptake pumps now
have been shown to represent the standard way in which the nervous system
inactivates the common amino acid and amine neurotransmitters after they
have been released from the synapse. Many therapeutically important drugs,
among them antidepressants, are powerful inhibitors of the uptake of norepinephrine
and serotonin. Indeed, effective antidepressants such as Prozac
are selective inhibitors of the uptake of serotonin.
Peptide transmitters
In addition to small molecules, it is now clear from the work of Thomas
Hokfelt and his colleagues that neurons also release small peptides as transmitters.
The number of peptides that act in this way exceeds several dozen
and raised the question, How do their actions relate to classical neurotransmitters?
Originally it was thought that the peptide-containing neurons represented
a separate class of cells: neuroendocrine cells. However, Hokfelt
and his colleagues showed that peptides and classical small molecule transmitters
such as ACh, norepinephrine, and serotonin coexist in individual
neurons. Insight into the functional significance of co-transmission has
emerged over the last two decades. In the salivary gland, for example, parasympathetic
cholinergic neurons contain VIP-like peptides. In contrast,
sympathetic norepinephrine neurons contain neuropeptide Y (NPY). In
both cases these peptides act to augment the action of the classical transmitter.
Thus, VIP induces a phase of vasodilatation and enhances the secretory
effects of ACh, while NPY causes phasic vasoconstriction, like norepinephrine
(Hokfelt 1991). Gene targeting studies in mice are now beginning to reveal
many additional functions for neuropeptide transmitters within the
central nervous system.
The Plastic Properties of Synapses
Ramón y Cajal first introduced the principle of connection specificity: the idea
that a given neuron will not connect randomly to another but that during
development a given neuron will form specific connections only with some
neurons and not with others. The precision of connections that characterizes
the nervous system posed several deep questions: How are the intricate
neural circuits that are embedded within the mature nervous system assembled
during development? How does one reconcile the properties of a specifically
and precisely wired brain with the known capability of animals and
humans to acquire new knowledge in the form of learning? And how is
knowledge, once learned, retained in the form of memory?
230 Psychiatry, Psychoanalysis, and the New Biology of Mind
One solution to this problem was proposed by Ramón y Cajal in his 1894
Croonian Lecture, in which he suggested that “mental exercise facilitates a
greater development of the protoplasmic apparatus and of the nervous collaterals
in the part of the brain in use. In this way, preexisting connections
between groups of cells could be reinforced by multiplication of the terminal
branches of protoplasmic appendix and nervous collaterals. But the preexisting
connections could also be reinforced by the formation of new collaterals
and protoplasmic expansions.”
An alternative solution for memory storage was formulated in 1922 by
the physiologist Alexander Forbes. Forbes suggested that memory was sustained
not by plastic changes in synaptic strength of the sort suggested by
Ramón y Cajal but by dynamic reverberating activity within a closed, interconnected
loop of self-reexciting neurons. This idea was elaborated by
Ramón y Cajal’s student Rafael Lorente de Nó (1938), who found examples
in his own analyses of neural circuitry and in those of Ramón y Cajal that
neurons were often interconnected in the form of closed chains, circular
pathways that could sustain reverberatory information.
This view of synaptic plasticity also was seriously challenged by B.
Deslisle Burns in his influential book of 1958, The Mammalian Cerebral Cortex.
Adopting a dynamic view, Burns wrote critically of plasticity mechanisms:
The mechanisms of synaptic facilitation which have been offered as candidates
for an explanation of memory. ..have proven disappointing. Before any
of them can be accepted as the cellular changes accompanying conditioned
reflex formation, one would have to extend considerably the scale of time on
which they have been observed to operate. The persistent failure of synaptic
facilitation to explain memory makes one wonder whether neurophysiologists
have not been looking for the wrong kind of mechanisms. (Burns 1958,
pp. 96–97)
The distinction between these two ideas—of dynamic as opposed to plastic
changes for memory storage—was first tested experimentally in invertebrates,
where studies of nondeclarative memory storage in the marine snail
Aplysia showed that memory is stored as a plastic change in synaptic
strength, not as self-reexciting loops of neurons. These studies found that
simple forms of learning—habituation, sensitization, and classical conditioning—
lead to functional and structural changes in synaptic strength of
specific sensory pathways that can persist for days and that these synaptic
changes parallel the time course of the memory process (Castellucci et al.
1970; Kandel and Spencer 1968). These findings reinforced the early ideas
of Ramón y Cajal, which have now become one of the major themes of the
molecular study of memory storage: Even though the anatomical connecNeural
Science 231
tions between neurons develop according to a definite plan, the strength and
effectiveness are not entirely predetermined and can be altered by experience
(Squire and Kandel 1999).
Modern cognitive psychological studies of memory have revealed that
memory storage is not unitary but involves at least two major forms: declarative
(or explicit) memory and nondeclarative (or implicit) memory. Declarative
memory is what is commonly thought of as memory. It is the conscious
recall of knowledge about facts and events: about people, places, and objects.
This memory requires the medial temporal lobe and a structure that
lies deep to it: the hippocampus. Nondeclarative memory such as habituation,
sensitization, classical and operant conditioning, and various habits reflect
the nonconscious recall of motor and perceptual skills and strategies
(Squire and Zola-Morgan 1991). In invertebrates these memories are
often stored in specific sensory and motor pathways. In vertebrates these
memories are stored, in addition, in three major subcortical structures: the
amygdala, the cerebellum, and the basal ganglia (B. Milner et al. 1998).
Behavioral studies of both simple nondeclarative and more complex declarative
memories had earlier shown that for each of these forms of memory
there are at least two temporally distinct phases: a short-term memory lasting
minutes and a long-term memory lasting days or longer (B. Milner 1965;
B. Milner et al. 1998). These two phases differ not only in their time course,
but also in their molecular mechanism: long-term but not short-term memory
requires the synthesis of new proteins. Molecular studies in Aplysia and
in mice have revealed that these distinct stages in behavioral memory are reflected
in distinct molecular phases of synaptic plasticity (Abel et al. 1997;
Bourtchouladze et al. 1994; Montarolo et al. 1986). In Aplysia, these stages
have been particularly well studied in the context of sensitization, a form of
learning in which an animal strengthens its reflex responses to previously
neutral stimuli, following the presentation of an aversive stimulus (Byrne
and Kandel 1996; Carew et al. 1983; Squire and Kandel 1999). The shortand
long-term behavioral memory for sensitization is mirrored by the shortand
long-term strengthening of the synaptic connections between the
sensory neuron and the motor neuron that mediate this reflex. In this set of
connections, serotonin, a neurotransmitter released in vivo by interneurons
activated by sensitizing stimuli leads to a short-term synaptic enhancement,
lasting minutes, which results from a covalent modification of preexisting
proteins mediated by the cAMP-dependent protein kinase A (PKA) and by
protein kinase C (PKC). By contrast, facilitation lasting several days results
from the translocation of PKA and mitogen-activated protein kinase
(MAPK) to the nucleus of the sensory neurons, where these kinases activate
CREB-1 and derepress CREB-2, leading to the induction of a set of immediate
response genes and ultimately resulting in the growth of new synaptic
232 Psychiatry, Psychoanalysis, and the New Biology of Mind
connections (Bartsch et al. 1995, 1998).
A similar cascade of gene induction is recruited for nondeclarative memory
storage in Drosophila (Dubnau and Tully 1998; Yin and Tully 1996; Yin
et al. 1995) and for spatial and object recognition memory, forms of declarative
(explicit) memory storage that can be studied in mice (Abel et al. 1997;
Bourtchouladze et al. 1994; Impey et al. 1996, 1998, 1999; Silva et al. 1998),
indicating that this set of mechanisms may prove to be quite general. In both
Aplysia and mice, experimental manipulations that reduce the level of the repressor
CREB-2 or enhance the level of the activator CREB-1 act to enhance
synaptic facilitation and amplify memory storage (Bartsch et al. 1995; Yin et
al. 1995). Thus, this set of mechanisms may prove to be quite general and to
apply to instances of both declarative and nondeclarative memory in both
vertebrates and invertebrates.
The requirement for transcription provided a provisional molecular explanation
for the behavioral observation that long-term memory requires the
synthesis of new proteins. This requirement, however, posed a cell-biological
problem: how can the activation of genes in the nucleus lead to long-lasting
changes in the connectivity of those synapses that are active and not in inactive
synapses? Recent studies have shown that this synapse-specific, spatially
restricted plasticity requires both the activity of the activator CREB-1
in the nucleus as well as local protein synthesis in those processes of the sensory
cell exposed to serotonin (Casadio et al. 1999; Martin et al. 1998).
This synapse-specific facilitation can be captured by another synapse of
the neuron. Once synapse-specific long-term facilitation has been initiated,
stimuli which per se induce only transient facilitation are able to recruit
long-term facilitation and the growth of new connections when applied to a
second branch (Casadio et al. 1999; Martin et al. 1998). A similar capture of
long-term synaptic plasticity has been found in the hippocampus by Frey
and Morris (1997). As we have seen, the hippocampus, a region essential for
declarative memory, is involved in the storage of memory for objects and
space (B. Milner et al. 1998). In 1973, Tim Bliss and Terje Lømo made the
remarkable discovery that major synaptic pathways in the hippocampus, including
the Schaffer collateral pathway, undergo a long-term form of synaptic
plasticity (long-term potentiation, or LTP) in response to a burst of highfrequency
stimulation (Figure 6–8). Subsequent studies by Graham Collingridge,
Roger Nicoll, and others found that LTP in the Schaffer collateral
pathway depends on activation of an NMDA receptor to glutamate in the
postsynaptic cell (the pyramidal cell of the CA1 region), resulting in an influx
of Ca2+ and an activation of the Ca2+ calmodulin-dependent protein kinase
IIα (CaMKIIα) (for a review, see Collingridge and Bliss 1995).
The correlation between LTP in the Schaffer collateral pathway and spatial
memory is not perfect (see, for example, Zamanillo et al. 1999 for an imNeural
Science 233
FIGURE 6–8. The phenomenon of long-term potentiation.
Long-lasting posttetanic potentiation of the hippocampus.
(A) 1) A diagrammatic view of a parasagittal section of the hippocampus showing a
stimulating electrode placed beneath the angular bundle (ab) to activate perforant
pathway fibers (pp) and a recording microelectrode in the molecular layer of the dentate
area (AD). Hipp fiss=hippocampal fissure; Stim=stimulatory electrode; Rec=recording
electrode; Fim=fimbria. 2) Arrangement of electrodes for stimulation of the experimental
pathway and the control pathway (in the contralateral hippocampus).
(B) Amplitude of the population of excitatory postsynaptic potential (EPSP) for the
experimental pathway (filled dots) and ipsilateral control pathway (open dots) as a
function of time and of conditioning impulse trains (15/s for 10 s) indicated by arrows.
Each value is a computed average of 30 responses. Values are plotted as a percentage
of the mean preconditioning value of the population (pop) EPSP.
Source. From Bliss TVP, Lømo T: “Long-Lasting Potentiation of Synaptic Transmission
in the Dentate Area of the Anaesthetized Rabbit Following Stimulation of the
Perforant Path.” The Journal of Physiology 232:331–356, 1973.
234 Psychiatry, Psychoanalysis, and the New Biology of Mind
portant dissociation). Nevertheless, a variety of experiments have found that
interfering with LTP in this pathway (by means of gene knockouts of the
NMDA receptor or by the expression of dominant-negative transgenes)
commonly interferes both with the representation of space by the neurons
of the hippocampus (place cells) and with memory for space in the intact animal
(Mayford and Kandel 1999; Tsien et al. 1996) (Figure 6–9). Moreover,
enhancing LTP in the Schaffer collateral pathway enhances memory storage
for a variety of declarative tasks (Han and Stevens 1999; Tang et al. 1999).
Despite these initial attempts to link LTP to behavioral memory storage,
we still lack a satisfactory knowledge about most key facets of hippocampal
synaptic plasticity in relationship to memory storage. For example, the facilitation
used experimentally to induce LTP involves frequencies of firing that
are unlikely to be used normally. The form of LTP used in most experiments
therefore is best viewed as a marker for a general capability for synaptic plasticity.
How the animal actually uses this capability is not yet known. In addition,
although there is agreement that LTP is induced postsynaptically (by
the activation of the NMDA receptor and consequent Ca2+ influx), there is
no consensus on whether the mechanisms of expression are postsynaptic or
presynaptic. The persistence of this lack of consensus suggests, as one possibility,
that the mechanism for expression of LTP is complex and involves a
coordinated pre- and postsynaptic mechanism. Finally, the hippocampus is
only one component of a larger medial temporal cortical system. How the
components of this system interact and how they relate to neocortical sites
of storage is entirely unknown.
A Future for the Study of Neuronal Signaling
Molecular structure, molecular machines, and
the integration of signaling pathways
During the last four decades, we have gained great insight from the reductionist
approach to neuronal signaling and synaptic plasticity. The molecular
characterization of voltage- and ligand-channels and of the many G protein–
coupled receptors that we have gained has dramatically advanced the initial
insights of Hodgkin, Huxley, and Katz and has revealed a structural unity
among the various molecules involved in neural signaling. Elucidation of
the primary sequence of these proteins also immediately revealed a commonality
in the signaling functions of proteins in neurons and those of other
cells. For example, many of the proteins involved in synaptic vesicle exocytosis
are used for vesicle transport and for secretion in other cells including
yeast. Conversely, bacteriorhodopsin, a bacterial membrane protein, has
proven to be the structural prototype for understanding G protein–coupled
seven transmembrane–spanning receptors such as those that are activated
Neural Science 235
FIGURE 6–9. The detection of place field cells in the mammalian
(A) A recording chamber used to record the firing patterns of place cells. The head
of a mouse inside the chamber is attached to a recording cable that is attached to a
device able to resolve the timing of action potentials (“spikes”) from one or more
CA1 pyramidal (place) cells. As the mouse explores the chamber, the location of a
light attached to its head is recorded by an overhead TV camera. Its output goes to a
tracking device that detects the position of the mouse. The occurrence of spikes as a
function of position is extracted and used to form two-dimensional firing-rate patterns
that can be analyzed quantitatively or visualized as color-coded firing-rate
(B) The firing patterns from a recording session of a single CA1 hippocampal pyramidal
place cell. Darker colors (violet or red) indicate high rates of firing and lighter
color (yellow) indicates a low firing rate. Before the recording session the animal was
moved and then reintroduced into the circular enclosure. During the recording session,
the mouse explores all areas of the enclosure equally. However, each place cell
fires only when the mouse is in a specific location. Each time the mouse is returned
to the chamber, place cells fire when the animal occupies the same locations that
fired those cells previously. The firing pattern for a given cell from a wild-type mouse
is stable.
Source. Courtesy of R. Muller.
236 Psychiatry, Psychoanalysis, and the New Biology of Mind
by light, odorants, and chemical transmitters. Receptors of this class come
into play during certain forms of learning and memory, and may even be important
in primates for aspects of arousal and attention.
Although we are now only beginning to enter the era of the structural biology
of voltage and of ligand-gated channels, we already appreciate that the
existing molecular understanding of receptors and of ion channels is remarkably
good. In retrospect, however, the obstacles confronted in the study of
channels and receptors were comparatively straightforward. The essential
properties of receptors and channels are contained within a single molecular
entity, and these functions had been well characterized by earlier biophysical
and protein chemical studies. Thus, the initial information about primary protein
sequence was immediately informative in generating models of transmembrane
protein topography and in defining domains that represent the
voltage sensor, the ligand-binding domain, the pore, and the inactivation gate.
Subsequent site-directed mutagenesis permitted rapid tests of these early predictions,
tests that proved surprisingly informative because the structure of
channels and receptors predicted the existence of distinct modular domains.
But we now know that many of these receptors, such as the NMDA and
AMPA receptors for glutamate, do not function alone but possess specialized
cytoplasmic protein domains that serve as platforms for assembling protein
machines important for signaling. Thus, in shifting the focus of analysis
from the ion channel to cytoplasmic signaling, we are entering a more complex
arena of protein-protein interaction and in the interaction between different
intracellular signaling pathways where function depends less on the
properties of single molecules and intramolecular rearrangement and more
on the coordination of a series of molecular events.
Fortunately, in the search for some of the components of these multimolecular
machines, such as the presynaptic proteins important for the targeting
and docking of vesicles at release sites and the assembly of the molecular
machinery for fusion and exocytosis, the study of synaptic transmission will
be aided by parallel studies in other areas of cell biology, such as membrane
trafficking and viral and cellular fusion events in nonneuronal systems.
Thus, despite the new realities and complexities that confront the study of
cytoplasmic signaling and transmitter release, it seems safe to predict that
these problems will be solved in the near future and that the romantic phase
of neuronal signaling, synaptic transmission, and synaptic plasticity will
reach closure in the first decades of the twenty-first century.
The great challenge for a reductionist approach in the subsequent decades
of the twenty-first century will be of two sorts: first in its application
to disease states, and second in its ability to contribute to the analysis of
brain systems important for cognition.
Neural Science 237
Molecular biology of disease
During the last two decades, we have made remarkable progress in analyzing
genes important for neurological disorders, especially monogenic diseases.
That this progress has been so dramatic encourages one to believe that within
the next decade the corpus of neurology may be transformed (for a review, see
Cowan et al. 1999). By contrast, progress in understanding the complex polygenic
diseases that characterize psychiatry has been noticeably slower.
The analysis of monogenic diseases dates to the beginning of the twentieth
century, but it accelerated markedly in 1989 when Louis Kunkel and his
associates first succeeded in cloning the gene for Duchenne’s muscular dystrophy
and found that the protein that it encodes, dystrophin, is homologous
to α-actinin and spectrin, two cytoskeletal proteins found on the inner surface
of the plasma membrane of muscle (Hoffman and Kunkel 1989; Hoffman
et al. 1987). Kunkel and his associates were able to show that in severe
forms of Duchenne’s dystrophy the dystrophic protein (dystrophin) is lacking
completely, whereas in a milder form, Becker dystrophy, functional protein
is present but in much reduced amounts. Kevin Campbell and his
colleagues extended this work importantly by showing that dystrophin is
only one component of a larger complex of glycoproteins (the dystroglycoprotein
complex) that links the cytoskeleton of the sarcoplasm to the extracellular
matrix (Straub and Campbell 1997).
A second major step in the analysis of monogenic diseases was taken in
1993 when James Gusella, Nancy Wexler, and their colleagues in the Huntington’s
Disease Collaborative Research Group isolated the gene responsible
for Huntington’s disease. In so doing, they discovered that the gene contains
an extended series of CAG repeats, thereby placing it together with a number
of other important neurological diseases in a new class of disorders: the
trinucleotide repeat diseases. These repeats were first encountered in the gene
responsible for the fragile X form of mental retardation (Kremer et al. 1991;
Verkerk et al. 1991). Subsequently, other hereditary disorders of the nervous
system were found to have similar repeats. Together, the trinucleotide repeat
disorders now constitute the largest group of dominantly transmitted neurological
diseases (for reviews see Paulson and Fischbeck 1996; Reddy and
Housman 1997; Ross 1997). Based on the nature of their repeats, the trinucleotide
repeat disorders can be divided into two groups: type I and type II
(Paulson and Fischbeck 1996).
In type I disorders, which include Huntington’s disease, the number of
CAG repeats usually does not exceed 90. The repeats lie within the coding
region of the gene, are translated as polyglutamine runs, and seem to cause
disease by a gain-of-function mechanism. The observation that the glutamine
repeats form β sheets consisting of six to eight residues per strand sug238
Psychiatry, Psychoanalysis, and the New Biology of Mind
gested to Max Perutz and his colleagues (1994) that the repeats could act as
a polar zipper that binds and traps other copies of either the same protein or
other proteins. This trapping might not only prevent the protein from functioning
normally but also could form large aggregates that may be toxic to
the cells. In the case of Huntington’s disease, Perutz postulated that the accumulation
of huntingtin in neurons might lead to the formation of toxic
protein aggregates, similar to those observed in Alzheimer’s disease or certain
prion disorders. Recent studies have indeed shown the existence of such
nuclear aggregates, although whether such aggregates reflect the cause or
the consequence of the disease remains an unsolved issue.
Type II repeat disorders, which include fragile X, have repeats found in either
the 5´ or 3´ untranslated regulatory regions of the gene that result in the
mRNA and protein not being expressed. In fragile X, for example, the FMR1
protein is not expressed. The wild-type protein contains RNA-binding motifs
(Warren and Ashley 1995) and in one severely retarded patient the mutation
was not in the regulatory region but in the coding region. Here a single
point mutation in one of the RNA-binding domains is sufficient to cause the
disease. These disorders are manifest by attenuated or absent expression of
the gene, and the disorder is not progressive but remains fixed from early development
Even in the case of these monogenic neurodegenerative disorders, however,
the problem of defining the molecular basis of the disease does not stop
with the identification of mutant genes. For several familial forms of neurological
diseases, notably Parkinson’s disease and amyotrophic lateral sclerosis
(ALS or Lou Gehrig’s disease), the identification of mutant protein
isoforms has not yet resulted in a clearer understanding of the cellular basis
of the disease. For example, our appreciation of the fact that gain-of-function
mutations in the superoxide dismutase 1 (SOD1) protein underlie certain
familial forms of ALS has not revealed the nature of the alteration in the
function of this protein. Similarly, the identification of mutated forms of synuclein
and Parkin proteins responsible for certain familial cases of Parkinson’s
disease has left unresolved the issue of how altered forms of these
proteins lead to the degeneration of mesencephalic dopaminergic neurons.
In addition, for these two disorders and for Huntington’s disease, the miscreant
proteins are widely expressed by virtually all neurons in the central nervous
system, yet in each disease quite distinct classes of neurons undergo
degeneration. The advent of more refined methods for translating the information
revealed through genomic sequencing to biochemical information
about the function of specific proteins in individual classes of neurons, the
so-called proteomic approaches, appears to offer considerable promise in resolving
these critical issues.
Of all the monogenic diseases, perhaps the most spectacular progress has
Neural Science 239
been made in elucidating the defects that underlie the hereditary myotonias,
periodic paralysis, and certain forms of epilepsy. These defects have now
been shown to reside in one or another voltage- or ligand-gated ion channels
of muscle. These disorders therefore are now referred to as the channelopathies—
disorders of ion channel function (for review, see Brown 1993; Cowan
et al. 1999; Ptácek 1997, 1998). As can be inferred from our earlier discussions,
the remarkable progress in understanding these diseases can be attributed
directly to the extensive knowledge about ion channel function that
was already available.
For example, hyperkalemic periodic paralysis and paramyotonia congenita,
two channelopathies due to ion channel disorders that result from
mutations in the α subunit of the Na+ channel, are caused by a number of
slightly different dominant mutations that make the Na+ channel hyperactive
by altering the inactivation mechanisms either by changing the voltage
dependency of Na+ activation or by slowing the coupling of activation and
inaction (for reviews, see Brown 1993; Ptácek et al. 1997). As was already
evident from earlier physiological studies, rapid and complete inactivation
of the Na+ channel is essential for normal physiological functioning of nerve
and muscle cells (Catterall 2000). These mutations do not occur randomly
but in three specific regions of the channel: the inactivation gate, the inactivation
gate receptor, and the voltage sensor regions that have been shown to
be functionally important by the earlier biophysical and molecular studies.
In contrast to these particular monogenic diseases, the identification of the
genetic basis of other degenerative neurological disorders has been slower.
Nevertheless, in some complex diseases such as Alzheimer’s disease, appreciable
progress has been made recently. This disease begins with a striking loss of
memory and is characterized by a substantial loss of neurons in the cerebral
cortex, the hippocampus, the amygdala, and the nucleus basalis (the major
source of cholinergic input to the cortex). On the cellular level, the disease is
distinguished by two lesions: 1) there is an extracellular deposition of neuritic
plaques; these are composed largely of β-amyloid (Aβ), a 42/43–amino acid
peptide; and 2) there is an intracellular deposition of neurofibrillary tangles;
these are formed by bundles of paired helical filaments made up of the microtubule-
associated protein tau. Three genes associated with familial Alzheimer’s
disease have been identified: 1) the gene encoding the β-amyloid
precursor protein (APP), 2) presenilin 1, and 3) presenilin 2.
The molecular genetic study of Alzheimer’s disease has also provided us
with the first insight into a gene that modifies the severity of a degenerative
disease. The various alleles of the apo E gene serve as a bridge between monogenic
disorders and the complexity we are likely to encounter in polygenic
disorders. As first shown by Alan Roses and his colleagues, one allele
of apolipoprotein E (apo E-4) is a significant risk factor for late-onset Alzhe240
Psychiatry, Psychoanalysis, and the New Biology of Mind
imer’s disease, acting as a dose-dependent modifier of the age of onset (Strittmatter
and Roses 1996).
The findings with apo E-4 stand as a beacon of hope for the prospect of
understanding the much more difficult areas of psychiatric disorders. Here
the general pace of progress has been disappointing for two reasons. First,
the diseases that characterize psychiatry, diseases such as schizophrenia, depression,
bipolar disorder, and anxiety states, tend to be complex, polygenic
disorders. Second, even prior to the advent of molecular genetics, neurology
had already succeeded in localizing the major neurological disorders to various
regions of the brain. By contrast, we know frustratingly little about the
anatomical substrata of most psychiatric diseases. A reliable neuropathology
of mental disorders is therefore severely needed.
Systems problems in the study of memory
and other cognitive states
As these arguments about anatomical substrata of psychiatric illnesses make
clear, neural science in the long run faces problems of understanding aspects
of biology of normal function and of disease, the complexity of which transcends
the individual cell and involves the computational power inherent in
large systems of cells unique to the brain.
For example, in the case of memory, we have here only considered the
cell and molecular mechanisms of memory storage, mechanisms that appear
to be shared, at least in part, by both declarative and nondeclarative memory.
But, at the moment, we know very little about the much more complex systems
problems of memory: how different regions of the hippocampus and
the medial temporal lobe—the subiculum, the entorhinal, parahippocampal,
and perirhinal cortices—participate in the storage of nondeclarative
memory and how information within any one of these regions is transferred
for ultimate consolidation in the neocortex. We also know nothing about the
nature of recall of declarative memory, a recall that requires conscious effort.
As these arguments and those of the next sections will make clear, the systems
problems of the brain will require more than the bottom-up approach
of molecular and developmental biology; they will also require the top-down
approaches of cognitive psychology, neurology, and psychiatry. Finally, it
will require a set of syntheses that bridge between the two.
The Assembly of Neuronal Circuits
The primary goal of studies in developmental neurobiology has been to clarify
the cellular and molecular mechanisms that endow neurons with the
ability to form precise and selective connections with their synaptic partners—
a selectivity that underlies the appropriate function of these circuits
Neural Science 241
in the mature brain. Attempts to explain how neuronal circuits are assembled
have focused on four sequential developmental steps. Loosely defined,
these are: the specification of distinct neuronal cell types; the directed outgrowth
of developing axons; the selection of appropriate synaptic partners;
and finally, the refinement of connections through the elimination of certain
neurons, axons, and synapses. In recent years, the study of these processes
has seen enormous progress (Cowan et al. 1997), and to some extent, each
step has emerged as an experimental discipline in its own right.
In this section of the review, we begin by describing some of the major
advances that have occurred in our understanding of the events that direct
the development of neuronal connections, focusing primarily on the cellular
and molecular discoveries of the past two decades. Despite remarkable
progress, however, a formidable gap still separates studies of neuronal circuitry
at the developmental and functional levels. Indeed, in the context of
this review it is reasonable to question whether efforts to unravel mechanisms
that control the development of neuronal connections have told us
much about the functions of the mature brain. And similarly, it is worth considering
whether developmental studies offer any prospect of providing such
insight in the foreseeable future. In discussing the progress of studies on the
development of the nervous system, we will attempt to indicate why such a
gap exists and to describe how recent technical advances in the ability to manipulate
gene expression in developing neurons may provide new experimental
strategies for studying the function of intricate circuits embedded in
the mature brain. In this way it should be possible to forge closer links between
studies of development and systems-oriented approaches to the study
of neural circuitry and function.
The Emergence of Current Views of the
Formation of Neuronal Connections
Current perspectives on the nature of the complex steps required for the formation
of neuronal circuits have their basis in many different experimental
disciplines (Cowan 1998). We begin by discussing, separately, some of the
conceptual advances in understanding how the diversity of neuronal cell
types is generated, how the survival of neurons is controlled, and how different
classes of neurons establish selective pathways and connections.
Inductive Signaling, Gene Expression, and the
Control of Neuronal Identity
The generation of neuronal diversity represents an extreme example of the
more general problem of how the fates of embryonic cells are specified. Ex242
Psychiatry, Psychoanalysis, and the New Biology of Mind
treme in the sense that the diversity of neuronal cell types, estimated to be
in the range of many hundreds (Stevens 1998), far exceeds that for other tissues
and organs. Nevertheless, as with other cell types, neural cell fate is now
known to be specified through the interplay of two major classes of factors.
The first class constitutes cell surface or secreted signaling molecules that,
typically, are provided by localized embryonic cell groups that function as
organizing centers. These secreted signals influence the pathway of differentiation
of neighboring cells by activating the expression of cell-intrinsic determinants.
In turn, these determinants direct the expression of downstream
effector genes, which define the later functional properties of neurons, in essence
their identity. Tracing the pathways that link the action of secreted factors
to the expression and function of cell-intrinsic determinants thus lies at
the core of attempts to discover how neuronal diversity is established.
The first contribution that had a profound and long-lasting influence on
future studies of neural cell fate specification was the organizer grafting experiment
of Hans Spemann and Hilde Mangold, performed in the early 1920s
(Spemann and Mangold 1924). Spemann and Mangold showed that naive ectodermal
cells could be directed to generate neural cells in response to signals
secreted by cells in a specialized region of the gastrula-stage embryo, termed
the organizer region. Transplanted organizer cells were shown to maintain
their normal mesodermal fates but were able to produce a dramatic change in
the fate of neighboring host cells, inducing the formation of a second body
axis that included a well-developed and duplicated nervous system.
Spemann and Mangold’s findings prompted an intense, protracted, and
initially unsuccessful search for the identity of relevant neural inducing factors.
The principles of inductive signaling revealed by the organizer experiment
were, however, extended to many other tissues, in part through the
studies of Clifford Grobstein, Norman Wessells, and their colleagues in the
1950s and 1960s (see Wessells 1977). These studies introduced the use of
in vitro assays to pinpoint sources of inductive signals, but again failed to reveal
the molecular nature of such signals.
Only within the past decade or so has any significant progress been made
in defining the identity of such inductive factors. One of the first breakthroughs
in assigning a molecular identity to a vertebrate embryonic inductive
activity came in the late 1980s through the study of the differentiation
of the mesoderm. An in vitro assay of mesodermal induction developed by
Peter Nieuwkoop (see Jones and Smith 1999; Nieuwkoop 1997) was used
by Jim Smith, Jonathan Cooke, and their colleagues to screen candidate factors
and to purify conditioned tissue culture media with inductive activity.
This search led eventually to the identification of members of the fibroblast
growth factor and transforming growth factor β (TGF-β) families as mesoderm-
inducing signals (Smith 1989).
Neural Science 243
Over the past decade, many assays of similar basic design have been used
to identify candidate inductive factors that direct the formation of neural tissue
and specify the identity of distinct neural cell types. The prevailing view
of the mechanism of neural induction currently centers on the ability of several
factors secreted from the organizer region to inhibit a signaling pathway
mediated by members of the TGF-β family of peptide growth factors (see
Harland and Gerhart 1997). The function of TGF-β proteins, when not constrained
by organizer-derived signals, appears to be to promote epidermal
fates at the expense of neural differentiation. The constraint on TGF-β–
related protein signaling appears to be achieved in part by proteins produced
by the organizer, such as noggin and chordin, that bind to and inhibit the
function of secreted TGF-β–like proteins. Other candidate neural inducers
may act instead by repressing the expression of TGF-β–like genes. However,
even now, the identity of physiologically relevant neural inducing factors
and the time at which neural differentiation is initiated remain matters of
Some of the molecules involved in the specification of neuronal subtype
identity, notably members of the TGF-β, fibroblast growth factor, and
Hedgehog gene families, have also been identified (Lumsden and Krumlauf
1996; Tanabe and Jessell 1996). These proteins have parallel functions in
the specification of cell fate in many nonneural tissues. Thus, the mechanisms
used to induce and pattern neuronal cell types appear to have been
co-opted from those employed at earlier developmental stages to control
the differentiation of other cells and tissues. Some of these inductive signals
appear to be able to specify multiple distinct cell types through actions at
different concentration thresholds—the concept of gradient morphogen
signaling (Gurdon et al. 1998; Wolpert 1969). In the nervous system, for
example, signaling by Sonic hedgehog at different concentration thresholds
appears sufficient to induce several distinct classes of neurons at specific
positions along the dorsoventral axis of the neural tube (Briscoe and Ericson
The realization that many different neuronal cell types can be generated
in response to the actions of a single inductive factor has placed added emphasis
on the idea that the specification of cell identity depends on distinct
profiles of gene expression in target cells. Such specificity in gene expression
may be achieved in part through differences in the initial signal transduction
pathways activated by a given inductive signal. But the major contribution
to specificity appears to be the selective expression of different target genes
in cell types with diverse developmental histories and thus different responses
to the same inductive factor.
The major class of proteins that possess cell-intrinsic functions in the determination
of neuronal fate are transcription factors: proteins with the ca244
Psychiatry, Psychoanalysis, and the New Biology of Mind
pacity to interact directly or indirectly with DNA and thus to regulate the
expression of downstream effector genes. The emergence of the central role
of transcription factors as determinants of neuronal identity has its origins
in studies of cell patterning in nonneural tissues and in particular in the genetic
analysis of pattern formation in the fruit fly Drosophila. The pioneering
studies of Edward Lewis (1985) on the genetic control of the Drosophila
body plan led to the identification of genes of the HOM-C complex, members
of which control tissue pattern in individual domains of the overall body
plan. Lewis further showed that the linear chromosomal arrangement of
HOM-C genes correlates with the domains of expression and function of
these genes during Drosophila development. Subsequently, Christine
Nüsslein-Vollhard and Eric Wieschaus (1980) performed a systematic series
of screens for embryonic patterning defects and identified an impressive array
of genes that control sequential steps in the construction of the early embryonic
body plan. The genes defined by these simple but informative
screens could be ordered into hierarchical groups, with members of each
gene group controlling embryonic pattern at a progressively finer level of
resolution (see St. Johnston and Nüsslein-Volhard 1992).
Advances in recombinant DNA methodology permitted the cloning and
structural characterization of the HOM-C genes and of the genes controlling
the embryonic body plan. The genes of the HOM-C complex were found to
encode transcription factors that share a 60-amino acid DNA-binding cassette,
termed the homeodomain (McGinnis et al. 1984; Scott and Weiner
1984). Many of the genes that control the embryonic body plan of Drosophila
were also found to encode homeodomain transcription factors and others
encoded members of other classes of DNA-binding proteins. The product of
many additional genetic screens for determinants of neuronal cell fate in
Drosophila and C. elegans led notably to the identification of basic helixloop-
helix proteins as key determinants of neurogenesis (Chan and Jan
1999). In the process, these screens reinforced the idea that cell-specific patterns
of transcription factor expression provide a primary mechanism for
generating neuronal diversity during animal development.
The cloning of Drosophila and C. elegans developmental control genes
was soon followed by the identification of structural counterparts of these
genes in vertebrate organisms, in the process revealing a remarkable and
somewhat unanticipated degree of evolutionary conservation in developmental
regulatory programs. The identification of over 30 different families
of vertebrate transcriptional factors, each typically comprising tens of individual
family members (see Bang and Goulding 1996), has provided a
critical molecular insight into the extent of neural cell diversity during vertebrate
development. Prominent among these are the homeodomain protein
counterparts of many Drosophila genes. Vertebrate homeodomain proteins
Neural Science 245
have now been implicated in the control of regional neural pattern, neural
identity, axon pathfinding, and the refinement of exuberant axonal projections.
The individual or combinatorial profiles of expression of transcription
factors may soon permit the distinction of hundreds of embryonic neuronal
Genetic studies in mice and zebra fish have demonstrated that a high
proportion of these genes have critical functions in establishing the identity
of the neural cells within which they are expressed. In many cases, the
classes of embryonic neurons defined on the basis of differential transcription
factor expressions have also been shown to be relevant to the later patterns
of connectivity of these neurons. Because of these advances, the
problem of defining the mechanisms of cell fate specification in the developing
nervous system can now largely be reduced to the issue of tracing the
pathway that links an early inductive signal to the profile of transcription
factor expression in a specific class of postmitotic neuron—a still daunting,
but no longer unthinkable, task.
Control of Neuronal Survival
The tradition of experimental embryology that led to the identification of inductive
signaling pathways has also had a profound impact on studies of a
specialized, if unwelcome, fate of developing cells: their death.
Many cells in the nervous system and indeed throughout the entire embryo
are normally eliminated by a process of cell death. The recognition of
this remarkable feature of development has its origins in embryological
studies of the influence of target cells on the control of the neuronal number.
In the 1930s and 1940s, Samuel Detwiler, Viktor Hamburger, and others
showed that the number of sensory neurons in the dorsal root ganglion of
amphibian embryos was increased by transplantation of an additional limb
bud and decreased by removing the limb target (Detwiler 1936). The targetdependent
regulation of neuronal number was initially thought to result
from a change in the proliferation and differentiation of neuronal progenitors.
A then-radical alternative view, proposed by Rita Levi-Montalcini and
Viktor Hamburger in the 1940s, suggested that the change in neuronal number
reflected instead an influence of the target on the survival of neurons
(Hamburger and Levi-Montalcini 1949). For example, about half of the motor
neurons generated in the chick spinal cord are destined to die during embryonic
development. The number that die can be increased by removing
the target and reduced by adding an additional limb (Hamburger 1975). The
phenomenon of neuronal overproduction and its compensation through cell
death is now known to occur in almost all neuronal populations within the
central and peripheral nervous systems (Oppenheim 1981).
246 Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 247
The findings of Levi-Montalcini and Hamburger led to the formulation
of the neurotrophic factor hypothesis: the idea that the survival of neurons depends
on essential nutrient or trophic factors that are supplied in limiting
amounts by cells in the environment of the developing neuron, often its target
cells (see Oppenheim 1981). This hypothesis prompted Levi-Montalcini
and Stanley Cohen to undertake the purification of a neurotrophic activity—
an ambitious quest, but one that led eventually to the identification of nerve
growth factor (NGF), the first peptide growth factor and a protein whose existence
dramatically supported the neurotrophic factor hypothesis (Hamburger
1993; Levi-Montalcini 1966) (Figure 6–10A). The isolation of NGF
was a milestone in the study of growth factors and, in turn, motivated
searches for additional neurotrophic factors. The efforts of Hans Thoenen,
Yves Barde, and others revealed that NGF is but the vanguard member of a
large array of secreted factors that possess the ability to promote the survival
of neurons (Reichardt and Fariñas 1997).
The best-studied class of neurotrophic factors, which includes NGF itself,
are the neurotrophins. Work by Mariano Barbacid, Luis Parada, Eric
Shooter, and others subsequently showed that neurotrophin signaling is mediated
by the interaction of these ligands with a class of membrane-spanning
tyrosine kinase receptors, the trk proteins (see Reichardt and Fariñas 1997)
(Figure 6–10B). Nerve growth factor interacts selectively with trkA, and
other neurotrophins interact with trkB and trkC. Other classes of proteins
that promote neuronal survival include members of the TGF-β family, the
FIGURE 6–10. Growth factors and their receptors (opposite
(A) The trophic actions of nerve growth factor on dorsal root ganglion neurons. Photomicrographs
of a dorsal root ganglion of a 7-day chick embryo that had been cultured
in medium supplemented with nerve growth factor for 24 hours. Silver
impregnation. The extensive outgrowth of neurites is not observed in the absence of
nerve growth factor.
(B) The actions of neurotrophins depend on interactions with trk tyrosine kinase receptors.
Neurotrophins interact with tyrosine kinase receptors of the trk class. The
diagram illustrates the interactions of members of the neurotrophin family with distinct
trk proteins. Strong interactions are depicted with solid arrows; weaker interactions
with broken arrows. In addition, all neurotrophins bind to a low-affinity
neurotrophin receptor p75NTR.
Abbreviations: NGF=nerve growth factor; NT=neurotrophin; BDNF=brain-derived
neurotrophic factor.
Source. (A) From studies of R. Levi-Montalcini; courtesy of the American Association
for the Advancement of Science. (B) From Kandel ER, Schwartz JH, Jessell T:
Principles of Neural Science, 4th Edition. New York, McGraw-Hill, 2000.
248 Psychiatry, Psychoanalysis, and the New Biology of Mind
interleukin 6–related cytokines, fibroblast growth factors, and hedgehogs
(Pettmann and Henderson 1998). Thus, classes of secreted proteins that
have inductive activities at early stages of development can also act later to
control neuronal survival. Neurotrophic factors were initially considered to
promote the survival of neural cells through their ability to stimulate cell
metabolism. Quite the contrary. Such factors are now appreciated to act predominantly
by suppressing a latent cell suicide program. When unrestrained
by neurotrophic factor signaling, this suicide pathway kills cells by apoptosis,
a process characterized by cell shrinkage, the condensation of chromatin,
and eventually cell disintegration (Jacobson et al. 1997; Pettmann and
Henderson 1998).
A key insight into the biochemical machinery driving this endogenous
cell death program emerged from genetic studies of cell death in C. elegans
by Robert Horvitz and his colleagues (Hengartner and Horvitz 1994;
Metzstein et al. 1998). Over a dozen cell death (ced) genes have now been
ordered in a pathway that controls cell death in C. elegans. Of these genes
two, ced-3 and ced-4, have pivotal roles. The function of both genes is required
for the death of all cells that are normally fated to die by apoptosis. A
third key gene, ced-9, antagonizes the activities of ced-3 and ced-4, thus protecting
cells from death. Remarkably, this death pathway is highly conserved
in vertebrate cells. The ced-3 gene encodes a protein closely related to members
of the vertebrate family of caspases, cysteine proteases that function as
cell death effectors by degrading target proteins essential for cell viability.
The ced-4 gene encodes a protein structurally related to another vertebrate
apoptosis-promoting factor, termed Apaf-1. The ced-9 gene encodes a protein
that is structurally and functionally related to the Bcl-2–like proteins,
some of which also act to protect vertebrate cells from apoptotic death. Apaflike
proteins appear to promote the processing and activation of caspases,
whereas some Bcl-2–like proteins interact with Apaf-1/ced-4 and in so doing,
inhibit the processing and activation of caspases.
These findings have revealed a core biochemical pathway that regulates
the survival of cells and is thought to serve as the intracellular target of neurotrophic
factors. The practical significance of this core cell death pathway
has not escaped attention. Pharmacological strategies to inhibit caspase activation
are now widely sought after in attempts to prevent the apoptotic
neuronal death that accompanies many neurodegenerative disorders.
Axonal Projections and the Formation of
Selective Connections
Attempts to unravel how selective neuronal connections are formed in the
developing brain have a somewhat different provenance. The electrophysioNeural
Science 249
logical studies of John Langley (1897), Charles Sherrington (1906), and others
at the turn of the twentieth century, as discussed earlier, had revealed the
exquisite selectivity with which mature neuronal circuits function and in the
process provided an early hint that their formation may also be a selective
process. In parallel, histological studies of the developing brain, applied
most decisively by Ramón y Cajal (1911/1955) but also by many others, provided
dramatic illustration of embryonic neurons captured in the process of
extending dendrites and axons, apparently in a highly stereotyped manner.
These pioneering anatomical descriptions provided circumstantial but persuasive
evidence that the assembly of neuronal connections is orchestrated
in a highly selective manner. By the middle of the twentieth century, many
elegant in vivo observations in simple vertebrate organisms had further
shown that developing axons extend in a highly reproducible fashion (see
Speidel 1933). But even these findings did not result in general acceptance
of the idea that the specificity evident in mature functional connections had
its basis in selective axonal growth and in selective synapse formation.
An alternative view, advanced most forcefully by Paul Weiss (1941) in
the 1930s and 1940s, and termed the resonance hypothesis, argued instead
that axonal growth and synapse formation were largely random events, with
little inherent predetermination. Advocates of the resonance view proposed
instead that the specificity of mature circuits emerges largely through the
elimination of functionally inappropriate connections, and only at a later developmental
stage. This extreme view, however, became gradually less tenable
in the light of experiments by Roger Sperry, notably on the formation of
topographic projections in the retinotectal system of lower vertebrates.
Sperry’s studies revealed a high degree of precision in the topographic order
of retinal axon projections to the tectum during normal development and
further established that this topographic specificity is maintained after experimental
rotation of the target tectal tissue—a condition in which the
maintenance of an anatomically appropriate connection results in a behaviorally
defective neuronal circuit (Sperry 1943; see Hunt and Cowan 1990)
(Figure 6–11). Over the subsequent two decades, the consolidation of these
early findings led Sperry (1963), in the 1960s, to formulate the chemoaffinity
hypothesis, a general statement to the effect that the most plausible explanation
for the selectivity apparent in the formation of developing
connections is a precise system of matching of chemical labels between preand
postsynaptic neuronal partners.
Sperry’s studies also emphasized the utility of combining embryological
manipulation and neuroanatomical tracing methods to probe the specificity
of neuronal connectivity. This tradition was extended in the 1970s by Lynn
Landmesser and her colleagues to demonstrate the specificity of motor axon
projections in vertebrate embryos (Lance-Jones and Landmesser 1981) and
250 Psychiatry, Psychoanalysis, and the New Biology of Mind
by Corey Goodman, Michael Bate, and their colleagues in analyses of the stereotyped
nature of axonal pathfinding in insect embryos (Bate 1976; Thomas
et al. 1984). Thus by the late 1970s, the cellular evidence for a high
degree of predetermination and selectivity in axonal growth and synapse formation
was substantial, although still not universally accepted (see Easter et
al. 1985).
In the 1980s and 1990s, attempts to clarify further the cellular mechanisms
of axonal growth and guidance focused on reducing the apparent
complexity inherent in the development of axonal projections to a few basic
modes of environmental signaling and growth cone response (Goodman and
Shatz 1993). As a first approximation, the multitude of cues thought to exist
FIGURE 6–11. Sperry’s demonstration of topographically specific
retinotectal projections.
Anatomical evidence for retinal axon regeneration to original sites of termination in
the optic tectum. Sperry’s studies showed the pattern of regenerated fibers in the
goldfish optic tract and tectum after removal of the anterior (left) or posterior (right)
half-retina. The optic nerve was cut at the time of retinal extirpation. The course and
termination of the regenerated axons was observed several weeks later, visualized by
silver staining. Regenerating axons terminate in appropriate regions despite the availability
of additional tectal tissue. M and L indicate medial and lateral optic tract bundles.
Source. Adapted from Attardi and Sperry 1963 as illustrated in Purves D, Lichtman
JW: Principles of Neural Development. Sunderland, MA, Sinauer, 1985.
Neural Science 251
in the environment of a growing axon was proposed to act in one of two
ways: 1) at long range, through the secretion of diffusible factors, or 2) at
short range, through cell surface-tethered or extracellular matrix-associated
factors. In addition, such long- and short-range cues were argued to act either
as attractants or local factors permissive for axonal growth or, in a complementary
manner, as repellents or factors that inhibit axon extension.
What remained unclear after this phase of conceptional reductionism and
simplification was the molecular basis of selective axon growth.
The Molecular Era of Axon Growth and Guidance
Today, there is no longer a paucity of molecules with convincing credentials
as regulators of axonal growth and guidance (see Tessier-Lavigne and Goodman
1996). This molecular cornucopia is the product of two main experimental
approaches: in vertebrate tissues, the biochemical purification of
proteins that promote cell adhesion and axonal growth; and in Drosophila
and C. elegans, the application of genetic screens to identify and characterize
mutations that perturb axonal projection patterns. Over the past decade,
these two complementary approaches have often supplied convergent information
and have resulted in the compilation of a rich catalog of molecules
with conserved functions in the control of axonal growth in insects, worms,
and vertebrates.
An early advance in the molecular characterization of proteins that control
axonal growth came with the biochemical dissection of two major adhesive
forces that bind neural cells, one calcium independent and the other
calcium dependent (Brackenbury et al. 1981). The design of assays to identify
neural adhesion molecules based on antibody-mediated perturbation of
cell adhesion by Gerald Edelman, Urs Rutishauser, and their colleagues led
to the purification of NCAM, a major calcium-independent homophilic cell
adhesion molecule (Rutishauser et al. 1982). The widespread expression of
NCAM initially argued against a role for this protein in specific aspects of
neuronal recognition. The discovery that NCAM is expressed in many different
molecular isoforms, however, preserves the possibility that it has
more specific functions in neural cell recognition and circuit assembly
(Edelman 1983). Although the precise contribution of NCAM to the growth
of axons and the formation of neuronal connections remains uncertain, its
isolation provided important credibility for the view that cell-adhesive interactions
in the nervous system can be dissected in molecular terms. In addition,
the realization that NCAM constitutes a divergent member of the
immunoglobulin (Ig) domain superfamily (Barthels et al. 1987) brought the
study of neural cell adhesion and recognition into the well-worked framework
of cell and antigen recognition in the immune system. Since the dis252
Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 253
covery of NCAM, over a hundred Ig domain-containing neural adhesion and
recognition proteins have been identified, although the function of most of
these proteins in vivo remains unclear (Brummendorf and Rathjen 1996).
In parallel, studies by Masatoshi Takeichi and his colleagues isolated the
major calcium-dependent adhesive force binding vertebrate cells, the cadherin
proteins (Takeichi 1990). Cadherins have been shown to have major
roles in the calcium-dependent adhesive interaction of virtually all cells in
the vertebrate embryo, and cadherins have also been identified in Drosophila
and C. elegans. The calcium dependence of cadherin function can be
mapped to a critical calcium-binding domain required for protein stability.
As we discuss below, cadherins, like Ig domain proteins, are now known to
represent a very large family.
A third general adhesive system characterized in the 1980s was that involved
in the interaction of cells with glycoproteins of the extracellular matrix.
At this time, biochemical studies by many groups had identified
collagens, fibronectins, and laminins as key adhesive glycoprotein components
of the extracellular matrix. The search for cellular receptors for these
structurally distinct glycoproteins converged with the identification of integrins,
a large family of heterodimeric integral membrane proteins (Hynes
1987; Ruoslahti 1996). Integrins have prominent roles in cell-matrix adhesion
within the nervous system and in virtually all other tissue types. Thus,
FIGURE 6–12. A role for ephrins and Eph kinases in the formation
of the retinotectal map (opposite page).
(A) Members of the Eph kinase class of tyrosine kinase receptors are distributed in
gradients in the retina, and some of their ligands, the ephrins, are distributed in gradients
in the optic tectum. These two molecular gradients have been proposed to regulate
retinotectal topography through the binding of ephrins to kinases and the
consequent inhibition of axon growth. The levels of ephrin A2 and ephrin A5 are
higher in the posterior tectum than in the anterior tectum, and thus may contribute
to the inhibition of extension of posterior retinal axons, which are rich in the kinase
eph A3.
(B) Diagram showing the consequences of ephrin A2 expression in portion of the
chick optic tectum that normally have low levels of this ligand. Posterior retinal axons
avoid sites of ephrin A2 overexpression and terminate in abnormal positions. In
contrast, anterior retinal axons, which normally grow into the ephrin-rich posterior
tectum, behave normally when they encounter excess ephrin A2.
(C) In mice lacking ephrin A5 function, some posterior retinal axons terminate in
inappropriate regions of the tectum.
Source. From the studies of O’Leary, Flanagan, Frisen, Barbacid, and others, as
summarized in Kandel ER, Schwartz JH, Jessell T: Principles of Neural Science, 4th
Edition. New York, McGraw-Hill, 2000.
254 Psychiatry, Psychoanalysis, and the New Biology of Mind
three main classes of neuronal surface membrane proteins—Ig domain proteins,
cadherins, and integrins—appear to provide neural cells with the major
adhesive systems necessary for the growth of axons, and these proteins
may also contribute to more selective forms of neuronal recognition.
Many additional proteins that are expressed more selectively and appear
to have selective roles in axonal growth have now been identified. For example,
genetic screens in C. elegans and biochemical assays of axon growth regulatory
factors in vertebrates collided with the characterization of netrins, a
small class of secreted proteins with cell context-dependent axonal attractant
and repellent activities (see Culotti and Merz 1998). A similar convergence
of biochemical and genetic assays led to the isolation of the
semaphorin/collapsin class of growth cone collapse-inducing factors
(Kolodkin 1998) and to the characterization of a slit signaling pathway that
appear to function both to repel axons and to promote axon branching
(Guthrie 1999). Independently, in vitro assays to examine the molecular basis
of the topographic mapping of retinotectal projections culminated in the
identification and functional characterization of ephrins: surface proteins
that function as ligands for receptor tyrosine kinases of the Eph class
(Drescher et al. 1997). Ephrin-Eph kinase signaling is now thought to have
a dominant role in the establishment of the molecular gradients used to form
projection maps in the retinotectal system and in other regions of the central
nervous system (Figure 6–12)—perhaps corresponding to some of the
matching chemical labels postulated earlier by Sperry.
With each of these discoveries, the veils that had previously shrouded
the molecular analysis of axon guidance have been progressively stripped
away. As a consequence, it is now realistic to begin to consider, at a molecular
level, how the guidance of axons is directed by dynamic sets of molecular
cues that either entice or deter the growth of axons at successive stages on
their path to a final target. Despite these indisputable advances, many aspects
of the logic of axon guidance remain unclear. With the multitude of
candidate cues now shown to possess repellent or attractant functions, we
still need to understand why individual sets of molecules are used in particular
cellular contexts. Are there unique and as yet unappreciated functions
provided by one but not another class of guidance cue? Or is there simply
molecular opportunism? That is, can similar steps in selective axon pathfinding
be achieved by any one of a large and structurally unrelated group of
guidance molecules?
One route to resolving such issues will be through the dissection of the
signal transduction pathways triggered in growth cones by activation of receptors
for guidance cues. Already, such studies have begun to lead to the
molecular classification of biochemical signaling pathways and their modulators
within the growth cone (Mueller 1999). They have also provided draNeural
Science 255
matic evidence in vitro that the ability of a growth cone to perceive an
extrinsic signal as attractant or repellent can be modified by changing the
ambient level of cyclic nucleotide activity. Further dissection of transduction
mechanisms in the growth cone may thus help to clarify the logic that underlies
the apparent selectivity of action of certain axonal growth and guidance
factors. Another critical but poorly resolved issue is that of determining
which guidance factors genuinely have instructive roles in directing axon
growth and which merely provide permissive signals that enable growth
cones to respond to other, more critical, signals.
The Selection and Refinement of Neuronal Connections
With the arrival of developing axons in the vicinity of their final position,
growth cones are required to select specific target cells with which to form
and maintain functional connections. Although this process is critical in establishing
the later functional properties of neural circuits, insight into the
molecular basis of neuronal target cell selection remains fragmentary. As discussed
above, one recurring issue has been the attempt to determine
whether the formation of selective connections is the product of genetically
determined factors that specify rules of connectivity in a precise manner, or
whether the initial pattern of connections can tolerate a degree of inaccuracy
that is subsequently resolved through the elimination of some connections
and the consolidation of others (Cowan et al. 1984; Shatz 1997). This latter
view then represents the reemergence, albeit in a more restricted and comprehensible
form, of the ideas originally articulated by Weiss in the 1940s.
A modern consensus view holds that both genetic predetermination and
use-dependent refinement of connections are important contributors to the
organization of mature circuits. The relative contribution of these two sets
of factors are, however, likely to vary considerably with the particular neural
circuit under study. One possibility is that circuits constructed early in evolution
or at early stages in the development of an organism, as for example
the spinal monosynaptic stretch reflex circuit, are established in a predominantly
activity-independent manner (Frank and Wenner 1993). In contrast,
the more sophisticated cortical circuits associated with the processing of
cognitive information, which emerge later in evolution and development,
may require functional validation for the establishment of final patterns of
connectivity (Shatz 1997).
The pioneering studies of David Hubel and Torsten Wiesel in the 1960s
provided the first evidence for a role for visually driven neural activity in the
functional organization of the primary visual cortex (Hubel and Wiesel
1998). Hubel and Wiesel deprived one eye of vision for several weeks during
an early critical period of postnatal life. After this procedure, they observed
256 Psychiatry, Psychoanalysis, and the New Biology of Mind
that most neurons in layer four of the primary visual cortex could be activated
only by input from the eye that had remained open, thus revealing a
marked shift in the pattern of ocular dominance columns in the cortex. At
an anatomical level, the terminal arbors of the axons of lateral geniculate
neurons supplied by the intact eye were found by Simon LeVay, Michael
Stryker, and their colleagues to be considerably more extensive than those
supplied by the deprived eye (Antonini and Stryker 1993a, 1993b; Hubel et
al. 1977). Many subsequent studies have confirmed the essential role of activity
in the formation of visual connections and have shown further that the
temporal pattern of activity provided by the two eyes is an important parameter
in the establishment of ocular dominance columns (Shatz 1997). Under
conditions in which visual input is provided to both eyes in a synchronous
manner, the formation of ocular dominance columns is again perturbed
(Stryker and Harris 1986). Additional studies have shown that the level of
activity in postsynaptic cortical neurons is necessary for ocular dominance
column formation (Hata and Stryker 1994). Collectively, these findings have
begun to focus attention on the possible mechanisms by which the state of
activity of postsynaptic cortical neurons could influence the pattern of arborization
of presynaptic afferent fibers as they enter the cortex.
One advance in addressing this problem came with the proposal that the
activation of the NMDA subclass of glutamate receptors on postsynaptic
neurons might be involved in the normal segregation of afferent input to visual
centers (Hofer and Constantine-Paton 1994). An extension of this idea
is that the NMDA receptor–mediated activation of cortical neurons results
in the release of an activity-dependent retrograde signal that influences the
growth and maintenance of presynaptic branches and nerve terminals. Several
candidate mediators of such a retrograde signal have now been advanced,
including nitric oxide and certain peptide growth factors. Much
attention has also been directed at testing the possibility that the activitydependent
release of neurotrophins by cortical neurons is a critical step in
the establishment of eye-specific projections into the visual cortex. Some
support for this idea has been provided with the demonstration by Carla
Shatz and colleagues that local infusion of the neurotrophins NT4 or BDNF
into the developing cortex prevents the segregation of ocular dominance columns
(Cabelli et al. 1995). Similar developmental defects are observed if the
ligand-binding domains of neurotrophin receptors are introduced into the
cortex, presumably the consequence of sequestering endogenous neurotrophins
(Cabelli et al. 1997). Thus, an attractive if still speculative idea is that
neurotrophic factors—classes of proteins identified initially on the basis of
their critical roles in promoting the survival of neurons—have later and
more subtle roles in shaping neuronal connections in the mammalian CNS.
Although the critical role of activity in the formation of neuronal circuits in
Neural Science 257
the visual system and in many other regions of the CNS is well established, the
precise nature of its contribution is less well defined. Information encoded by
patterns of activity could be sufficient to direct certain connections. It remains
possible, however, that for many neuronal circuits, a basal but unpatterned
level of activity is all that is required. In this view, activity may simply permit
neurons to respond to other signals that have more direct roles in the control
of selective connections or may permit the maintenance of connections formed
at earlier stages and through separate mechanisms. Evidence supportive of this
latter view has come from studies by Michael Stryker and his colleagues on the
role of visually driven activity in the formation of orientation and ocular dominance
columns in the developing visual cortex (Crair et al. 1998). Neural
activity may therefore exert its influence in large part by consolidating connections
that have been established earlier through mechanisms which have their
basis in molecular recognition between afferent neurons and their cortical target
cells (see Crowley and Katz 1999; Weliky and Katz 1999).
Defining the relative contributions of sensory-evoked activity and genetically
determined factors remains difficult, first because the molecular basis
of target recognition in any circuit is still unknown and second because the
pathways by which activity modifies connectivity are poorly understood.
Progress in resolving these issues will therefore require additional insight
into the molecules that control synaptic specificity. One anticipated feature
of molecules that contribute to the selection of neural connections is that of
molecular diversity (Serafini 1999). Several classes of proteins that exhibit
inordinate molecular variation have recently been identified and, not surprisingly,
have been implicated in the formation of selective connections.
The cadherins as discussed above represent one class of cell surface recognition
protein that exists in large numbers. Diversity in cadherin structure
can be enhanced dramatically through a process in which one of a chromosomally
arrayed cluster of variable cadherin domain gene sequences is appended
to a nearby constant region sequence (Wu and Maniatis 1999). The
molecular mechanism used to assemble such modularly constructed cadherin
proteins remains unclear, but the number of these variable domains is
high, bringing the total number of predicted cadherins to well over 100. The
vast majority of cadherins are known to be expressed by neural cells and
studies of the patterns of expression of the classical cadherins have revealed
a striking segregation of individual cadherins within functionally interconnected
regions of the brain (Takeichi et al. 1997). In addition, cadherins are
concentrated at apposing pre- and postsynaptic membranes at central synapses
(Shapiro and Colman 1999). Although intriguing, the link between selective
cadherin expression and the specificity of synaptic connections
remains to be demonstrated functionally.
A second class of proteins with the potential for considerable structural
258 Psychiatry, Psychoanalysis, and the New Biology of Mind
variation is the neurexins. Neurexins are surface proteins identified originally
by virtue of their interaction with the neurotoxin α-latrotoxin (Missler
and Südhof 1998; Rudenko et al. 1999). Analysis of the potential for alternative
splicing of the neurexin genes suggests, in principle, that ~1,000 protein
isoforms can be generated and at least some of these potential isoforms are
known to be expressed by central neurons. In addition, a class of neurexin
receptors termed neuroligins has been identified (Song et al. 1999). Again,
though, a functional role for neurexin-neuroligin interactions in the formation
of synapses remains to be established.
A third highly diverse class of neuronal surface proteins are the seven-pass
odorant receptors expressed on primary sensory neurons in the olfactory epithelium.
Several major classes of odorant or pheromone receptors have now
been identified in vertebrates, and in total this class of receptors is thought to
be encoded by over 1,000 distinct genes (Axel 1995; Buck and Axel 1991).
This genetic diversity is likely to underlie the remarkable discriminatory capacity
of the mammalian olfactory sensory system. The creative manipulation
of odorant receptor gene regulatory sequences to map the central projections
of olfactory sensory axons through reporter gene expression in transgenic
mice has also revealed a precise anatomical convergence of sensory axons
linked by common receptor gene expression to individual target glomeruli in
the olfactory bulb (Mombaerts et al. 1996). This finding poses the additional
question of the mechanisms directing sensory axon targeting to individual
glomeruli. Strikingly, manipulation of the pattern of expression of individual
odorant receptor genes in transgenic mice results in a predictable change in
the central projection pattern of olfactory sensory axons (Wang et al. 1998).
An intriguing implication of these findings is that olfactory sensory receptors
function not only in peripheral odor discrimination but also in axon targeting,
potentially providing a direct link between the sensory receptive properties of
a neuron and its central pattern of connectivity.
Determining whether each or any of these classes of proteins have roles
in selective synapse formation in the developing central nervous system
CNS) is an important goal in itself and may also provide the entry point for
a more rigorous examination of the relationship between neuronal activity,
gene expression, and synaptic connectivity.
The events that initiate the formation of selective contacts between preand
postsynaptic partners are, however, unlikely to provide sufficient information
to establish the functional properties of synapses necessary for effective
neuronal communication. A separate set of molecules and mechanisms
appears to promote the maturation of early neuron–target contacts into
specialized synaptic structures. Current views of this aspect of neuronal development
derive largely from studies of one peripheral synapse, the neuromuscular
junction (Sanes and Lichtman 1999). These studies have their
Neural Science 259
origins in many classical physiological studies of synaptic transmission at
the neuromuscular junction. In particular, the ability to measure dynamic
changes in the pattern of expression of ACh receptors on the surface of muscle
fibers as they become innervated (Fischbach et al. 1978) provided many
early insights into the cellular mechanisms by which the motor axon organizes
the elaborate program of postsynaptic differentiation necessary for efficient
synaptic transmission. By the 1980s, powerful in vivo and in vitro
assays to examine synaptic organization under conditions of muscle denervation
and reinnervation had been developed, and these assays facilitated
biochemical efforts to purify neuronally derived factors with synaptic organizing
capacities (McMahan 1990; Sanes and Lichtman 1999).
These efforts culminated in the identification of two major pre- to
postsynaptic signaling pathways that appear to coordinate many aspects of
the synaptic machinery in the postsynaptic muscle membrane. Signals mediated
by agrin, a nerve- and muscle-derived proteoglycan, through its tyrosine
kinase receptor MuSK have an essential role in the clustering of ACh
receptors and also of other synaptically localized proteins at postsynaptic
sites located in precise register with the presynaptic zones specialized for
transmitter release (see Kleiman and Reichardt 1996; McMahan 1990). A
second set of nerve- and muscle-derived factors, the neuregulins which signal
through ErbB class tyrosine kinase receptors, appears instead to control
the local synthesis of ACh receptor genes in muscle cells (see Sandrock et al.
1997), and perhaps also to direct the local insertion of newly synthesized receptors
at synaptic sites.
These dramatic molecular successes have provided the foundations of a
comprehensive understanding of the steps involved in the formation and organization
of nerve-muscle synapses. The extent to which the principles that
have emerged from the study of this synapse peripherally extend also to the
organization of central synapses remains uncertain. There has, however,
been considerable progress in recent years in defining the structural components
of the presynaptic release apparatus at central synapses (Bock and
Scheller 1999) and the proteins that concentrate postsynaptic receptors
(Sheng and Pak 1999). From the information now emerging, it seems likely
that the identity of molecular signals that orchestrate the maturation of central
synapses will soon be known, and in the process we will come to recognize
principles of central synaptic organization similar to those that operate
at the neuromuscular junction.
A Future for Studies of Neural Development
Despite the dramatic advances of the two past decades, several important but
unresolved issues cloud our view of the assembly of synaptic connections.
260 Psychiatry, Psychoanalysis, and the New Biology of Mind
These problems will need to be addressed before any satisfying understanding
of neural circuit assembly can be claimed.
One issue stems from the pursuit of mechanisms of neuronal cell fate determination
and of the control of axonal pathfinding and connectivity as
largely separate disciplines. With the many available details of cell fate specification
and of the regulation of axonal growth and guidance, it is still not clear
if and how the transcriptional codes that control neuronal identity intersect
with the expression of the effector molecules that direct axonal connectivity.
For example, in only a few cases have relevant genetic targets of the transcription
factors that control early steps in neuronal identity been identified. Indeed,
a superficial survey of patterns of expression of transcription factors and
axonal receptors for guidance cues reveals little obvious coincidence at the cellular
level. Thus, the extent to which the regulated expression of genes that encode
receptors for axon guidance cues depends on the sets of determinant
factors implicated in earlier aspects of neuronal subtype identity remains unclear.
Defining the full complement of transcription factors that specify the
identity of an individual neuronal subtype and the molecular sequence of cellcell
interactions that guide the axon of the same neuron to its target is one obvious
but laborious route to resolving this issue.
Similarly, the relationship between transcription factor expression and
other later aspects of neuronal phenotype—for example neurotransmitter
synthesis and chemosensitivity—also remains unclear. In a few instances,
cell-specific transcription factors have been linked to the expression of genes
that control neurotransmitter synthesis (see Goridis and Brunet 1999).
Nevertheless, the general logic linking transcriptional identity and the
expression of the neuronal traits that confer specialized synaptic signaling
properties and connectivity remains obscure.
Assuming, as seems likely, that these issues can be solved in a relatively
rapid fashion, what does the future hold for studies of neural development?
Clearly, there will be interesting variation in the strategies used to establish
selective connections in different regions of the developing brain and in different
circuits. The documentation of these variations will provide a richer
and more profound appreciation of the core principles of neuronal circuit assembly.
But the reiteration of a few basic themes in different brain regions
can sustain excitement in the field only briefly, and in any event will not provide
an obvious intellectual bridge between studies of development and of
the function of mature neuronal circuits.
Application of neural development to the
study of neurological disease
One future area in which studies of neural development are likely to have
significant impact is in the application of fundamental information on the
Neural Science 261
specification of cell fate and the guidance of axons to problems posed by
neurodegenerative diseases and traumatic injury to the nervous system.
As discussed above, we are beginning to obtain a rather detailed outline
of the relationship between inductive signaling and the expression of cellspecific
transcription factors that define cell fate. In some cases, details of
these pathways have progressed to the point that certain transcription factors
expressed by single classes of CNS neurons have been shown to be
sufficient to direct neuronal subtype fate in a manner that is largely independent
of the prior developmental history of the progenitor cell (Tanabe et al.
1998). If this is the case for the few classes of neurons in which inductive
signaling pathways have been particularly well studied, it seems likely that
similar dedicated determinant factors will exist for many other classes of
neurons in the CNS. The identification of such factors may be of significance
in the context of the many ongoing attempts to identify neural progenitor
cells and then to drive them along specific pathways of neurogenesis (Doetsch
et al. 1999; Johansson et al. 1999; Morrison et al. 1999; Panchision et al.
1998). One outcome of such developmental studies may therefore be to rationalize
strategies for reintroduction of fate-restricted neural progenitor
cells into the CNS in vivo. In principle, these advances could offer the potential
of more efficient cell replacement therapies in a wide variety of neurological
degenerative disorders.
Similarly, the wealth of information on molecules that promote or inhibit
axonal growth is likely to be of relevance for studies of axonal regeneration and
repair. The pioneering studies of Albert Aguayo and colleagues of the regenerative
capacity of central neurons in a cellular environment composed of peripheral
rather than CNS nerve cells revealed the potential of central neurons to
regenerate (see Goldberg and Barres 2000; Richardson et al. 1997). These studies
prompted the search for molecules expressed by cells of the mature CNS
that inhibit the growth of axons (see Tatagiba et al. 1997) and for molecules expressed
in early development that have the capacity to promote the growth of
axons of CNS neurons (Tessier-Lavigne and Goodman 1996). The progress in
identification of axon growth–promoting and inhibitory factors may therefore
eventually permit rational changes to be made in the environment through
which regenerating axons in the mature CNS are required to project. Of equal
promise are studies to clarify the signal transduction pathways by which axons
respond to these environmental cues. The elucidation of these pathways may
permit a more general manipulation of axonal responses—for example rendering
axons insensitive to broad classes of inhibitory factors, or supersensitive to
many distinct axonal growth–promoting factors. It may also be worth considering
whether there is a common molecular basis for the marked differences in
the regenerative capacity of different vertebrate species evident in studies of
both nerve and limb regeneration (see, for example, Brockes 1997).
262 Psychiatry, Psychoanalysis, and the New Biology of Mind
Establishing a link between the development
and function of neuronal circuits
An additional, and potentially a more far-reaching, contribution of neural
development may emerge by taking advantage of the compendium of information
now available on cell-specific gene expression in developing neurons
and of the ease of genetic manipulation in mammals, notably the mouse.
With these methods in hand, it may be possible to modify the function of
highly restricted classes of neurons in the adult animal and to assay resultant
changes in the function of specific neuronal circuits.
One initial limitation in the application of information about neuronal
subtype–specific gene expression during development is that the majority of
such genes are transiently expressed. Thus, the normal temporal profile of
gene expression does not permit direct tracing of the relationship between embryonic
neuronal subtype identity and the physiological properties of the
same neuronal subsets in the adult. This problem can now be overcome
through the use of genetically based lineage tracing methods. For example,
genes encoding yeast or bacterially derived recombinase enzymes can be introduced
into specific genetic loci by targeted recombination (Dymecki 1996;
Schwenk et al. 1998), to generate mouse strains which can then be crossed
with other genetically modified mice in which recombinase-driven DNA rearrangement
results in the irreversible activation of reporter gene expression at
all subsequent stages in the life of a neuron (Lee et al. 2000; Zinyk et al. 1998).
This relatively simple methodology offers the immediate promise of providing
a direct link between subsets of neurons defined at embryonic stages and the
location, and functional identity of these neurons within the mature CNS.
With the compilation of such lineage information, variants of this same
basic genetic strategy can be used to modify the function of neuronal subsets
at predefined times. One drastic method for eliminating neuronal function
involves the activation of toxins in a neuron-specific manner, under precise
temporal control (see, for example, Grieshammer et al. 1998; Watanabe et
al. 1998), thus permitting the physical ablation of predefined populations of
CNS neurons with a specificity unattainable by conventional lesioning
methods. More subtly, specific populations of neurons could, in principle, be
activated or inactivated reversibly in the adult animal through temporally
regulated expression of ion channels that change the threshold for neuronal
excitability (Johns et al. 1999). In addition, the development of transgenic
mice methods for anterograde or retrograde transynaptic transport of foreign
marker proteins (Coen et al. 1997; Yoshihara et al. 1999) may be helpful in
providing novel information on neuronal connectivity in the CNS that cannot
easily be extracted by other anatomical tracing methods.
In this way, the increasingly detailed molecular information that derives
Neural Science 263
from attempts to examine the principles of neural circuit assembly during
development should have clear application to the major problems of systems
neuroscience discussed in the following sections of this review. At present,
the routine application of these genetic methodologies is feasible only in the
mouse, and thus the issue of linking studies obtained in lower mammals
with information obtained in primates and ideally in man still needs to be
addressed. Nevertheless, with advances in the resolution of functional imaging
methods that are outlined later in this review, and in the application of
these methods to small mammals, the link between studies in mice and primates
can be strengthened. When this is achieved, the information that
emerges from studies of the development of neural circuits may assume a
more prominent place in the repertoire of experimental strategies that aim
to decipher how such circuits function in the adult brain.
Neural Systems: From Neurons
to Perception
The individual neurons that make up the brain work together in specialized
groups, or systems, each of which serves a distinct function. Systems neuroscience
is the study of these neural systems, which include those involved in
vision, memory, and language. Neural systems possess a number of common
properties, not the least of which is the fact that they all process higher-order
information about an organism’s environment and biological needs. In humans,
this information often gains access to consciousness. Systems neuroscience
thus places great emphasis on uncovering the neural structures and
events associated with the steps in an information-processing hierarchy.
How is information encoded (sensation), how is it interpreted to confer
meaning (perception), how is it stored or modified (learning and memory),
how is it used to predict the future state of the environment and the consequences
of action (decision making/emotion), and how is it used to guide
behavior (motor control) and to communicate (language)? The twentieth
century has seen remarkable progress in understanding these processes.
This ascendance of modern systems neuroscience is attributable, in part, to
the convergence of five key subdisciplines, each of which contributed major
technical or conceptual advances.
Neuropsychology: Localization of the
Biological Source of Mental Function
The first question one might ask about an information-processing device
concerns its gross structure and the relationship between structural ele264
Psychiatry, Psychoanalysis, and the New Biology of Mind
ments and their functions. The simplest approach to this question—and the
approach that has best withstood the test of time—is to observe the behavioral
or psychological consequences of localized lesions of brain tissue. The
modern discipline of neuropsychology was founded on this approach and
draws both from human clinical case studies—often provided during the
early decades of the twentieth century by brain injuries sustained in battle—
and from experimental studies of the effects of targeted destruction of brain
tissue in animals. Through these means the functions of specific brain regions,
such as those involved in sensation, perception, memory, and language,
have been inferred.
Neuroanatomy: Patterns of Connectivity
Identify Information Processing Stages
The discipline of neuroanatomy, which blossomed at the turn of the century
following the adoption of the neuron doctrine and which has benefited from
many subsequent technical advances, has revealed much about the fine
structure of the brain’s components and the manner in which they are connected
to one another. As we have seen, one of the earliest and most influential
technical developments was the discovery by Camillo Golgi of a
method for selective staining of individual neurons, which permitted their
visualization by light microscopy. By such methods, it became possible to
use differences in the morphology of cells in different brain regions as markers
for functional diversity. This procedure, known as cytoarchitectonics,
was promoted vigorously in the early decades of the twentieth century by
the anatomists Korbinian Brodmann and Oscar and Cecile Vogt. Brodmann’s
cytoarchitectonic map of the human cerebral cortex, which was published in
1909 and charted the positions of some 50 distinct cortical zones, has served
as a guidebook for generations of scientists and clinicians, and as a catalyst
for innumerable studies of cortical functional organization.
Arguably the most important outcome of the means to label neurons,
however, was the ability it provided to trace connections between different
brain regions. To this end, cell labeling techniques have undergone enormous
refinement over the past three decades. Small quantities of fluorescent
or radioactive substances, for example, can now be injected with precision
into one brain region and subsequently detected in other regions, which provides
evidence for connectivity. The products of anatomical tract tracing are
wiring diagrams of major brain systems, which are continuously evolving in
their precision and completeness, and have been indispensable to the analysis
of information flow through the brain and for understanding the hierarchy
of processing stages.
Neural Science 265
Neurophysiology: Uncovering Cellular
Representations of the World
Adoption of the neuron doctrine and recognition of the electrical nature of
nervous tissue paved the way to an understanding of the information represented
by neurons via their electrical properties. Techniques for amplification
and recording of small electrical potentials were developed in the 1920s
by Edgar Adrian. This new technology enabled neurobiologists to relate a
neuronal signal directly to a specific event, such as the presentation of a sensory
stimulus, and became a cornerstone of systems neuroscience. By the
1930s, electrophysiological methods were sufficiently refined to enable recordings
to be made from individual neurons. Sensory processing and motor
control emerged as natural targets for study. The great successes of singleneuron
electrophysiology are most evident from the work of Vernon Mountcastle
in the somatosensory system, and David Hubel and Torsten Wiesel in
the visual cortex, whose investigations, beginning in the late 1950s, profoundly
shaped our understanding of the relationship between neuronal and
sensory events.
Psychophysics: The Objective Study of Behavior
Historically, quantitation of behavior has been the province of experimental
psychology, which emerged in the nineteenth century from deep-rooted
philosophical traditions to become a distinct scientific discipline and a key
component of modern systems neuroscience. Among the most notable steps
in this emergence was the development by the German physicist and philosopher
Gustav Fechner of a systematic scientific methodology for assessing
the relationship between behavior and internal states. Fechner’s Elements of
Psychophysics, published in 1860, founded an “exact science of the functional
relationship. ..between body and mind,” based on the assumption
that the relationship between brain and perception could be measured experimentally
as the relationship between a stimulus and the sensation it
gives rise to (Fechner 1860/1966). In practice, Fechner’s psychophysics is
applied by varying a sensory stimulus along some physical dimension—
such as the intensity or wavelength of light—and obtaining reports from an
observer regarding the sensations experienced. In this manner, one can identify
the function that relates the physical dimension of the stimulus to an internal
sensory dimension, and from that relationship infer the rules by
which the sensory information is processed.
Throughout the twentieth century, the tools of psychophysics have been
extremely useful in identifying the information-processing strategies of sensory,
perceptual, and motor systems of the brain. Beginning with the work
of Mountcastle in the 1960s (Mountcastle et al. 1969), psychophysics has
266 Psychiatry, Psychoanalysis, and the New Biology of Mind
frequently been paired directly with electrophysiological methods to extraordinary
effect in identifying the neuronal events that give rise to specific
sensory and perceptual processes.
Computation: Divining the Mechanisms of
Information Processing
Large neural systems such as those involved in vision, combine and analyze
incoming signals to “interpret” their causes and generate appropriate outputs.
The logical steps in these neuronal mechanisms have become accessible
to quantitative and theoretical treatment. The goal has been to extract
generic computational principles that can account for existing data and have
predictive value. Some of the earliest work along these lines was directed at
sensory and motor processing and was founded on engineering techniques
and principles designed for the study of simple linear systems. One of the
most successful examples of this approach is Georg von Bekesy’s (1960) investigation
of the cochlea and its relation to the frequency encoding of
sound. von Bekesy began by investigating the patterns of vibration of the
various components of the inner ear, and the relationship of these patterns
to the characteristics of sound waves. From these observations he concluded
that this system analyzes sound by a linear frequency decomposition—that
is, the mechanical properties of the cochlea allow specific frequency components
of sound to be independently isolated and detected by the sensory epithelium.
Considerable gains have also been made using similar theoretical
approaches to understand early stages of visual processing and the control
of movements of the eyes.
Many levels of processing in neural systems deviate from linear forms of
computation. The search for alternative computational principles, which
was fueled in part by the rise of cognitive science in the 1980s and an unprecedented
richness of physiological and anatomical data, has led to a
number of novel and sophisticated theoretical approaches, such as those embodied
by neural networks (Rumelhart et al. 1987). These networks operate
on the biologically plausible principle that information can be represented
in a distributed fashion across a large population of “units,” or modeled neurons.
Moreover, this information may be combined in many different ways
to yield complex cellular representations, simply by changing the strength
of “synaptic” connections between modeled neurons.
Vision as a Model System
Collectively, these five areas of neuroscience—neuropsychology, neuroanatomy,
neurophysiology, psychophysics, and computation—constitute an exNeural
Science 267
perimental arsenal, which has already revealed in outline the structure,
operational mechanisms, and functions of large neural systems, such as
those involved in vision, memory, and language. Although the range of successes
is broad, and many general principles of system organization and
function have been discovered, the visual system has emerged as the model
for experimental investigation and is consequently the area in which we
have the greatest understanding.
Setting the stage: early explorations of
visual perception and brain
Two early developments presaged the extraordinary progress in understanding
visual function that is now a legacy of the twentieth century. The first of
these occurred within the field of experimental psychology. Hermann von
Helmholtz (1860/1924) and Wilhelm Wundt (1902), two of Fechner’s nineteenth-
century contemporaries, attempted to identify how different visual
stimuli lead to different subjective experiences. Their method was initially
observational and introspective, but later they exploited the objective methodology
of psychophysics. The lasting outcome of these efforts was a quantitative
appreciation of the elements of visual experience—color, brightness,
motion, distance—and an initial set of ideas about how they might be represented
by the brain. A second early development occurred within the field
of neuropsychology. With mounting experimental evidence for localization
of function within specific brain regions, Hermann Munk (1881) and Edward
Schafer (1888) each used the method of focal ablation of brain tissue
at the end of the nineteenth century to identify brain regions that serve visual
function. They found that the occipital lobe of the cerebral cortex plays
an essential role in the processing of visual information.
The Golden Era of Single-Neuron Electrophysiology
Perhaps the single greatest technical advance in vision science was the application
of the electrophysiological methods that had emerged in the late
1920s and 1930s. In a pioneering series of studies begun in the early 1930s,
Keefer Hartline recorded from single cells in the eye of the horseshoe crab
(Limulus) and examined the relationship between the properties of the incoming
sensory stimulus—which in this case happened to be a small spot of
light projected onto the eye—and the neuronal response (i.e., the frequency
of action potentials). Through these (Hartline and Graham 1932) and subsequent
experiments of a similar nature that involved recordings from single
axons in the frog optic nerve (Hartline 1938), Hartline made two important
discoveries. First, he found that individual neurons respond to light only
within a well-defined region of visual space, which Hartline termed the vi268
Psychiatry, Psychoanalysis, and the New Biology of Mind
sual receptive field. Operationally defined, the receptive field is the portion of
the sensory epithelium (the sheet of photoreceptors, in the case of vision)
that when stimulated elicits a change in the frequency of action potentials
for a given neuron. In anatomical terms, the receptive field describes all of
the receptor and subsequent cells that converge upon and influence the firing
pattern of the neuron under study. The concept of the receptive field has
proven to be an extremely useful and general concept in systems neuroscience.
Hartline’s second major discovery was that the visual responses to light
were dependent upon contrast. Specifically, the amplitude of the neuronal
response to a light in one region of visual space was greatest when there was
no light in an adjacent region of space. Thus, rather than simply conveying
the presence or absence of light, neurons in the visual system communicate
information about the spatial structure or pattern of the incoming stimulus.
Because these observations paralleled the well-known perceptual enhancement
of brightness at contrast boundaries—exploited for centuries by artists
wishing to enhance the range of light intensities perceived in their paintings
(Leonardo da Vinci 1956)—they were seized upon as a potential physiological
substrate for the perceptual experience of brightness.
The tradition of single-neuron studies of visual processing continued
through the 1940s and 1950s with a shift of emphasis to mammals (affording
closer ties to human visual perception and made possible by advances in electrophysiological
techniques). In 1953, Stephen Kuffler, a student of Eccles and
Katz, examined the behavior of neurons in the cat retina. Kuffler focused on
the retinal ganglion cells, the output cells of the retina, which carry visual information
from the photoreceptors through the optic nerve to the lateral geniculate
nucleus of the thalamus and other central processing regions.
Following Hartline’s model, Kuffler described the response characteristics of
ganglion cells in terms of their receptive field properties. He discovered that
the receptive fields of retinal ganglion cells were round in shape and had distinct
concentric excitatory and inhibitory zones, which made them maximally
sensitive to spatial contrast. On the basis of the architecture of their receptive
field properties, Kuffler divided these cells into two groups. One class of cells
had a central excitatory zone and a surrounding inhibitory region (“oncenter”
cells), whereas the other class of cells had an inhibitory central region
and an excitatory surround region (“off-center” cells).
Beyond the retina: visual contours are detected
by neurons in primary visual cortex
Anatomical tracing experiments conducted over the past several decades
have shown that the outputs of the retina terminate in several distinct brain
Neural Science 269
regions. One of the largest projections extends from the retina to the lateral
geniculate nucleus (LGN) of the thalamus and continues on via the geniculostriate
pathway to primary visual cortex. Otherwise known as striate cortex
or area V1, this latter visual processing stage lies on the occipital pole of the
cerebral cortex, and is known from the early neuropsychological studies of
Munk, Schafer, Gordon Holmes (1927), and others to be critical for normal
visual function.
The electrophysiological approach pioneered by Kuffler in studies of
mammalian retinal ganglion cells was carried to these higher processing
stages by two of Kuffler’s young colleagues, David Hubel and Torsten Wiesel.
In the late 1950s, Hubel and Wiesel began to examine the response properties
of neurons in the cat and monkey lateral geniculate nucleus. These neurons
were found to possess center-surround receptive field properties not
unlike those of retinal ganglion cells. By contrast, Hubel and Wiesel (1959)
found that the response properties of cells in the primary visual cortex of
both cats and monkeys were very much more complicated. Cortical cells
could not be effectively stimulated by the simple spots of light that proved
so effective in the retina and in the lateral geniculate nucleus. To be effective,
a stimulus had to have linear properties; the best stimuli were lines, bars,
rectangles, or squares (Figure 6–16). Hubel and Wiesel divided the cortical
cells into simple and complex (for a review, see Hubel and Wiesel 1977). We
will use simple cells to illustrate in greater detail the types of stimulus selectivities
observed (Figure 6–13).
A typical receptive field for a simple cell in primary visual cortex might
have a central rectangular excitatory area with its long axis running from
twelve to six o’clock, flanked on each side by similarly shaped inhibitory areas.
For this type of cortical cell, the most effective excitatory stimulus is a
bar of light with a specific axis of orientation—in this case, from twelve
o’clock to six o’clock—projected on the central excitatory area of the receptive
field. Since this rectangular zone is framed by two rectangular inhibitory
areas, the most effective stimulus for inhibition is one that stimulated one or
both of the two flanking inhibitory zones. A horizontal or oblique bar of
light would stimulate both excitatory and inhibitory areas and would therefore
be relatively ineffective. Thus, a stimulus that is highly effective if projected
vertically onto a given area of retina so as to be on target for the
excitatory zone would become ineffective if held horizontally or obliquely.
Other cells had similar receptive field shapes but different axes of orientation
(vertical or oblique). For example, the most effective stimulus for a cell with
an oblique field would be a bar of light running from ten o’clock to four
o’clock or from two o’clock to eight o’clock (Figure 6–13).
The most interesting feature of the simple cortical cells is that they are
much more particular in their stimulus requirement than the retinal gan270
Psychiatry, Psychoanalysis, and the New Biology of Mind
FIGURE 6–13. Neuronal orientation selectivity, as first observed
by Hubel and Wiesel in primary visual cortex (area V1) of a rhesus
The receptive field of the cell that was recorded is indicated by broken rectangles in
the left column. The visual stimulus that was viewed by the monkey consisted of a
bar of light that was moved back and forth through the receptive field of the cell in
each of seven different orientations (rows A–G). The different directions of motion
used for each orientation are indicated by the small arrows. Recorded traces of cellular
activity are shown at right, in which the horizontal axis represents time (2 s/
trace) and each vertical line represents an action potential. This neuron responded
most strongly to a bar of light oriented along the diagonal (stimulus D), particularly
when the bar was moved through the receptive field from lower left to upper right.
Neurons bearing such properties are common in the visual cortex, and their discovery
revolutionized views on the neuronal bases of visual perception.
Source. From Hubel DH, Wiesel TN: “Receptive Fields and Functional Architecture
of Monkey Straite Cortex.” The Journal of Physiology 195:215–243, 1968.
Neural Science 271
glion cells or those in the lateral geniculate nucleus in requiring a proper
axis of orientation. For a stimulus to be effective for a retinal ganglion or a
geniculate cell, it only has to have the proper shape—in general, circular—
and the proper retinal position so as to activate appropriate receptors in the
retina. Simple cortical cells not only have to represent all retinal positions
and several shapes (lines, bars, rectangles), but also for each shape they have
to represent all axes of orientation. These findings provided an initial insight
as to why the visual cortex (or any cortex) needs so many cells for its normal
functions. Cells are required to represent every retinal area in all axes of orientation
so as to abstract the information presented to the cortex. Hubel and
Wiesel suggested that the simplest explanation for the response properties
of a cortical cell with a simple receptive field was that they received innervation
from a set of geniculate cells that had appropriate on-center and offcenter
properties and appropriate retinal positions.
Another feature distinguishes cells in primary visual cortex from those
in the lateral geniculate nucleus. Neurons of the lateral geniculate nucleus
only respond to stimulation of one or the other eye. In primary visual cortex,
one begins to find cells that are activated by stimulation of either eye (Figure
6–14). These cells provide a neural substrate for the integration of information
from the two eyes. Binocular properties of this type are essential to the
use of stereoscopic cues for depth vision in animals, such as primates with
frontally placed eyes.
Beyond V1: specialized functions of higher cortical visual areas
Neuropsychological studies carried out over much of the past century have
shown that deficits in visual function follow from damage anywhere within
a vast expanse—over 30% in humans—of the cerebral cortex. Moreover, the
type of deficit depends upon the site of damage: temporal lobe lesions cause
impairments in object recognition (termed “agnosias” by Sigmund Freud)
(for a review, see Farah 1995), whereas parietal lobe lesions interfere with
use of visual cues for spatially directed actions (for a review, see Mesulam
1999). These early findings suggested that the complex cellular properties
discovered in area V1 by Hubel and Wiesel were only the tip of the iceberg.
Motivated by this prospect, in the 1970s there was a dramatic increase in
electrophysiological and anatomical studies designed to explore the organization
of the extrastriate visual cortex, which lies beyond area V1. Two
groups—Semir Zeki, and John Allman and John Kaas—noted that not only
were extrastriate neurons vastly heterogeneous in their response properties
but that the extrastriate visual cortex could be neatly subdivided into a large
number of distinct modules on the basis of these properties (for a review, see
Van Essen 1985). At present, the visual cortex of monkeys is thought to be
272 Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 273
composed of over 30 such modules (Figure 6–16).
These efforts to reveal order in the heterogeneity of visual cortex were a
natural extension of the nineteenth-century concept of localization of function.
They were, moreover, reinforced by the computational view, advanced
by the theorist David Marr (1982), that large system operations (such as seeing)
can be subdivided and assigned to task-specific modules. Although little
is yet known of the specific tasks “assigned” to the vast majority of
extrastriate visual areas, there are some noteworthy exceptions. Of particular
interest is the middle temporal (MT) visual area, which appears to be specialized
for motion.
Area MT lies near the junction of the occipital, parietal, and temporal
lobes and is known to possess a high proportion of neurons that represent
the trajectory of a moving visual stimulus, suggesting an important role in
visual motion processing (for a review, see Albright 1993). This idea has
been supported by three related findings by William Newsome and his colleagues
that imply a causal link between the activity of MT neurons and perceived
motion. In the first experiment, Newsome and his colleagues found
that focal destruction of area MT in monkeys results in motion blindness,
demonstrating that MT is necessary for motion perception (Newsome and
Paré 1988) (Figure 6–15). In a second experiment, Newsome and Anthony
Movshon obtained psychophysical measurements of a monkey’s ability to
discriminate direction of motion, simultaneously with electrophysiological
measurements of the motion sensitivity of MT neurons in the monkey’s
brain. They found that the sensitivity of individual neurons correlated extremely
well with performance on the behavioral task, demonstrating that
FIGURE 6–14. Anatomical representation of ocular dominance
columns in primate visual cortex (opposite page).
(A) The right eye of a normal adult was injected with the radiolabeled proline and
fucose. This dark-field autoradiograph obtained after 10 days shows a tangential section
of area 17 of the right hemisphere. Radioactivity can be seen in the form of white
stripes, which correspond to thalamic axon terminals in layer 4 that relay input from
the injected eye. The alternating dark stripes depict the position to geniculate afferents
from the uninjected eye.
(B) Reconstruction of ocular dominance columns in area 17 of the right hemisphere
showing the intricate organization of the map.
Source. (A) From Hubel DH, Wiesel TN, LeVay S: “Plasticity of Ocular Dominance
Columns in the Monkey Striate Cortex.” Philosophical Transactions of the Royal Society
of London. Series B: Biological Sciences 278:377–409, 1977. (B) From LeVay S,
Wiesel TN, Hubel, DH: “The Development of Ocular Dominance Columns in Normal
and Visualy Deprived Monkeys.” The Journal of Comparative Neurology 191:1–
51, 1980.
274 Psychiatry, Psychoanalysis, and the New Biology of Mind
the direction information encoded by neurons of MT is sufficient to account
for the monkey’s judgment of motion (Newsome et al. 1989). Newsome and
colleagues reasoned that if this logic were correct, then it should be possible
to alter the monkey’s perception of motion by artificially modifying the firing
rate of MT neurons. In a third experiment, these investigators did exactly
that. Small electrical currents were used to stimulate clusters of neurons sensitive
to a common direction of motion. Remarkably, doing so was found to
bias the monkey’s judgment toward that direction of motion (Salzman et al.
1990). Electrical stimulation of this sort thus has the effect of adding a fixed
motion signal to the signal received by MT from the retina. Not only do these
results strongly support the hypothesized role of area MT in motion processing,
but also they imply that perceptual decisions can be based on the activity
of relatively small populations of neurons.
Detection of behaviorally significant visual features
The initial discovery that visual neurons integrate complex spatial information
led to speculation about the role of such neurons in detection of behaviorally
significant visual features. Experiments conducted in the 1950s by
Horace Barlow (1953) and by Jerome Lettvin and Humberto Maturana reinforced
this view with the finding that ganglion cells in the frog retina respond
optimally to patterns of light that resemble the silhouette and flight
of an insect (Lettvin et al. 1959), which is, of course, the primary food source
for a frog. This finding led naturally to the concept that single visual cells
were feature detectors. According to this view, receptive fields may operate as
highly specialized templates for detection of significant features (“trigger
features”) in the visual image. The concept was expanded upon in the 1960s
by the Polish psychologist Jerzy Konorski (1967), who proposed the existence
of “gnostic units”—cellular representations of visual features, such as
faces, that convey highly meaningful information to the observer. Horace
Barlow (1972) expressed similar views in his “cardinal cell” hypothesis. The
possibility that cells specialized for detection of faces might exist at higher
levels of processing was a logical extension of the findings of Hubel and Wiesel
(1977), which documented increasingly abstract cellular representations
as one ascended the hierarchy of processing stages. In addition, the feature
detector/cardinal cell hypothesis followed naturally from several decades of
neuropsychological research demonstrating that damage to the inferior temporal
lobe of the cerebral cortex compromised a human or monkey’s ability
to identify complex objects (Farah 1995; Gross 1973). Most importantly,
damage to a small subregion of this cortex in humans results in an inability
to recognize faces, a syndrome called prosopagnosia. Patients with prosopagnosia
can identify a face as a face, its parts, and even specific emotions exNeural
Science 275
FIGURE 6–15. The involvement of cortical area MT in the perception
of visual motion.
A monkey with an MT lesion and a human patient with damage to extrastriate visual
cortex have similar deficits in motion perception.
(A) Displays used to study the perception of motion. In the display on the left, there
is no correlation between the directions of movement of several dots, and thus no net
motion in the display. In the display on the right, all the dots move in the same direction
(100% correlation). An intermediate case is in the center; 50% of the dots move
in the same direction while the other 50% move in random directions (and are perceived
as noise added to the signal).
(B) The perception of visual motion by a monkey before and after a lesion of MT
(left). The performance of a human subject with bilateral brain damage is compared
to two normal subjects (right). The ordinate of the graph shows the percent correlation
in the directions of all moving dots (as in part [A]) required for the monkey to
select the common direction. The abscissa indicates the size of the displacement of
the dot and thus the degree of apparent motion. Note the general similarity between
the performance of the human and that of the monkey and the devastation to this
performance after the cortical lesions (Baker et al. 1991; Pare 1998).
Source. From experiments of Newsome and others as illustrated in Kandel ER,
Schwartz JH, Jessell T: Principles of Neural Science, 4th Edition. New York, McGraw-
Hill, 2000.
276 Psychiatry, Psychoanalysis, and the New Biology of Mind
Neural Science 277
pressed by the face, but they are unable to identify a particular face as
belonging to a specific person (Farah 1995).
Strong electrophysiological evidence in support of the feature detector/
cardinal cell hypothesis came from the work of Charles Gross beginning in
the late 1960s. Using the same methods established by Kuffler, Hubel and
Wiesel, and others, Gross discovered cells in the inferior temporal cortex of
monkeys that respond to specific types of complex stimuli such as hands
and faces (Gross et al. 1969). For cells that respond to a hand, individual fingers
are particularly critical. For cells that respond to faces, the frontal view
of the face is the most effective stimulus for some cells, while for others it is
the side view. Whereas some neurons respond preferentially to faces, others
respond to facial expressions. It seems likely that such cells contribute directly
to the perceptually meaningful experience of face recognition.
FIGURE 6–16. Visual information is processed by divergent cortical
pathways (opposite page).
One of the most important discoveries of the past century in the field of sensory biology
is the multiplicity of areas in the cerebral cortex that are involved in visual perception
and visually guided behavior. These areas are thought to be functionally
specialized and hierarchically organized into two parallel processing streams.
(A) The image is a lateral view of the rhesus monkey brain, illustrating these two major
pathways, both originating from V1, the striate cortex. There is a dorsal (“where”)
cortical stream, which takes a dorsal route to the parietal cortex, and a ventral
(“what”) cortical stream, which takes a ventral route to the temporal cortex.
(B) The image illustrates the primary visual cortex (area V1) located on the occipital
pole (left); the “extrastriate” cortical visual areas extend anteriorally (rightward) and
are labeled by their commonly used abbreviations (see below). Indicated borders of
visual areas (dashed lines) are approximate. Some sulci have been partially opened
(shaded regions) to show visual areas that lie buried within these sulci.
(C) The image illustrates some of the anatomical connections known to exist from
the retina through visual cortex. Except where indicated by arrows, anatomical connections
are known to be reciprocal.
Abbreviations: FST=fundus superior temporal, LGN=lateral geniculate nucleus of
the thalamus, M=magnocellular subdivisions, MST=medial superior temporal,
MT=middle temporal, P1 and P2=parvocellular subdivisions, PP=posterior parietal
cortex, RGC=retinal ganglion cell layer, STP=superior temporal polysensory
VIP=ventral intraparietal.
Source. From Albright TD: “Cortical Processing of Visual Motion.” Reviews of Oculomotor
Research 5:177–201, 1993. Used with permission of Elsevier.
278 Psychiatry, Psychoanalysis, and the New Biology of Mind
General Principles of Visual System
Organization and Function
We have here covered only a few highlights of twentieth-century research on
the visual system. The complete legacy has led to a number of general principles
of visual system organization and function to which we now turn.
The visual system is hierarchically organized
A consistent feature of the visual system is the presence of multiple hierarchically
organized processing stages (Figure 6–16), through which information
is represented in increasingly complex and abstract forms (for a review,
see Van Essen 1985). As first suggested by Hubel and Wiesel (1962), properties
present at each stage result, in part, from selective convergence of
information from the previous stage. The hierarchy begins with the photoreceptors,
which detect the presence of a spot of light shining upon a particular
part of the retina. Each retinal ganglion cell—the output cells of the
retina—surveys the activity of retinal bipolar cells, which in turn survey activity
in a group of receptors. The product of this convergence of information
in the retina is a simple abstraction of light intensity, namely a
representation of luminance contrast. A neuron in the lateral geniculate nucleus
surveys a group of retinal ganglion cells, and activity in the geniculate
cell also signals luminance contrast. A simple cell in primary visual cortex
surveys a population of geniculate neurons and the firing of that simple cell
reflects a still higher level of abstraction: the presence of an oriented contour
in the visual image. At successively higher stages of processing, information
is combined to form representations of even greater complexity, such that,
for example, individual neurons within the pathway for visual pattern processing
encode complex behaviorally significant objects, such as faces.
How far does this hierarchy go? Is there a group of cells that observes the
hierarchies of simple cells and makes one aware of the total pattern? And if
so, is there a still higher group in the hierarchy that looks at combinations
of complex patterns as these enter our awareness? These are important questions
for the future of systems neuroscience, which we address below in our
discussions of visual feature “binding” and consciousness.
The visual system is organized in parallel processing streams
In addition to a hierarchy of processing stages, the visual system is organized
in parallel streams (Figure 6–16). Incoming information of different types is
channeled through these different streams, such that the output of each
stream serves a unique function. This type of channeling occurs on several
scales and at different hierarchical levels (for a review, see Van Essen 1985).
Neural Science 279
One of the most pronounced examples of channeling occurs within the
projection from retina to the lateral geniculate nucleus of the thalamus, and
beyond. Different types of retinal ganglion cells project selectively to three
different laminar subdivisions of the lateral geniculate nucleus, known as
parvocellular, magnocellular, and koniocellular laminae. Each of these subdivisions
is known to convey a unique spectrum of retinal image information
and to maintain that information in a largely segregated form at least as
far into the system as primary visual cortex.
Beyond V1, the ascending anatomical projections give rise to two visual
pathways, each of which is organized hierarchically (Figure 6–16). One pathway
extends dorsally to terminate within the parietal lobe and includes area
MT and visual areas of the posterior parietal cortex (Figure 6–19). A second
pathway extends ventrally to terminate within the temporal lobe and includes
areas V4 and inferior temporal cortex (for a review, see Felleman and Van Essen
1991). On the basis of a large body of neuropsychological data, Leslie Ungerleider
and Mortimer Mishkin concluded in the early 1980s that these two
cortical pathways serve different functions (Ungerleider and Mishkin 1982).
Accordingly, cortical areas of the dorsal pathway are concerned with “where”
an object is in visual space. These areas represent motion, distance, and the
spatial relations between surfaces in the visual world, and provide a crucial
source of information for initiating and guiding movements. By contrast, areas
of the ventral pathway are concerned with “what” an object is. These areas
represent information about form and the properties of visual surfaces, such
as color or texture, and play important roles in object recognition (for a review,
see A.D. Milner and Goodale 1995). Electrophysiological studies of the response
properties of neurons in areas of the proposed “where” and “what”
pathways have provided support for this functional dichotomy (Figure 6–17).
The “what” versus “where” distinction continues at still higher levels,
where visual information is stored in memory for later retrieval. The dorsal
stream projects to a subdivision of frontal cortex that is known to be critical
for visual spatial memory. The ventral stream projects to a different subdivision
of frontal cortex, which serves object recognition memory.
Many visual processing stages are topographically organized
Near the turn of the century, Hermann Munk (1881) suggested, in part on
the basis of observed effects of cortical lesions, that the visual cortex may
contain a spatial map of the retinal surface. This hypothesis was confirmed
in the 1910s by Gordon Holmes (1927), who discovered a precise relationship
between the site of damage in human visual cortex and the location in
visual space where visual sensitivity was lost. Even better evidence came in
1941 from the work of Wade Marshall and Samuel Talbot, who used gross
280 Psychiatry, Psychoanalysis, and the New Biology of Mind
electrophysiological techniques to demonstrate an orderly topographic representation
of visual space across the cortical surface, such that neurons with
spatially adjacent receptive fields lie adjacent to one another in the brain
(Talbot and Marshall 1941). It has since been discovered that neuronal maps
of visual space are characteristic of many visual processing stages (for a review,
see Van Essen 1985). Such maps may facilitate computations based on
comparisons of visual information at adjacent regions of visual space. These
maps are commonly distorted relative to the visual field, such that the numbers
of neurons representing the center of gaze, which is particularly important
for visual object recognition, greatly exceed those representing the
visual periphery. These variations in “magnification factor” are thought to
underlie variations in the observer’s resolving power and sensitivity.
The visual cortex is organized in vertical columns
The existence of a column-like anatomical substructure in the cerebral cortex
has been known since the beginning of the twentieth century, following
FIGURE 6–17. The influence of local sensory context on visual
Each of the three images displayed here contains a horizontal dark gray rectangle. Although
the rectangles are physically identical, the surrounding features (the contexts)
differ in the three images. As a result, the rectangle is attributed perceptually
to different environmental causes in the three instances: In the image shown in (A),
the rectangle appears to result from the overlap of two surfaces, one of which is transparent
(e.g., a piece of tinted glass). In the image shown in (B), the rectangle appears
to result from a variation in surface reflectance (e.g., a stripe painted across a large
flat canvas). In the image shown in (C), the rectangle appears to result from partial
shading of a three-dimensional surface (i.e., variation in the angle of the surface with
respect to the source of illumination). These markedly different perceptual interpretations
argue for the existence of different neuronal representations of the rectangle
in each of the three instances. These representations can only be identified in neurophysiological
experiments if the appropriate contextual cues are used for visual stimulation.
See the text for details.
Source. Courtesy of T. Albright and colleagues.
Neural Science 281
the work of Ramón y Cajal, Constantin von Economo, and Rafael Lorente de
Nó. Although Lorente de Nó (1938) first suggested that this characteristic
structure might have some functional significance, it was the physiologist
Vernon Mountcastle who developed the concept fully and proposed in the
1950s that this may be a general principle of cortical organization. Using single-
cell electrophysiological techniques, Mountcastle (1957) obtained the
first evidence in support of this proposal through his investigations of the
primate somatosensory system.
The best-studied example of columnar organization, however, is that discovered
in the 1960s by Hubel and Wiesel. These investigators found that primary
visual cortex is arranged in a series of narrow vertical columns, about
100–200 μm in width, running from the surface of the cortex to the white matter
(Hubel and Wiesel 1962, 1968). In a given column, cells have similar
receptive field positions and generally similar receptive field properties, including
preferred orientation. Consistent with Mountcastle’s hypothesis, the columns
seem to serve as elementary units of cortical organization designed both
to bring cells together so that they can be appropriately interconnected and to
generate from their interconnections the properties needed for cells with
higher-order receptive fields. Additional evidence for functional columns, and
for the validity of Mountcastle’s proposal, has come from studies of higher visual
areas. In the early 1980s, Thomas Albright and colleagues (1984) identified
a system of functional columns in area MT. Interestingly, the spatial scale
of this columnar system, which represents direction of stimulus motion, is virtually
identical to that for orientation in primary visual cortex—consistent
with the expression of a general organizational principle. Columnar systems
have also been observed in inferior temporal cortex (Fujita et al. 1992).
Why are columns a preferred form of cortical architecture? Mountcastle’s
original proposal assumed the need for adequate “coverage,” such that, for
example, the machinery for detecting all contour orientations is available for
all parts of the visual field represented in the cortex. There are also computational
advantages (Schwartz 1980) afforded by representing similar features
adjacent to one another—such as the ability readily to compare and
contrast similar orientations. Finally, it may be that columnar structure is derived
simply from developmental constraints, such that it is easier and more
economical to wire together a cortex that has similar properties in close
proximity (Goodhill 1997; Miller 1994; Swindale 1980).
The visual system is modifiable by experience
during early postnatal development
The mammalian brain develops through a complex multistaged process that
extends from embryogenesis through early postnatal life. The end product
282 Psychiatry, Psychoanalysis, and the New Biology of Mind
of this developmental sequence is a set of patterned anatomical connections
that give rise to the mature system properties of the brain. Once the principal
features of mature visual system organization and function became known
through the work of Hubel and Wiesel in the 1960s, it was natural to question
whether those features reflect a genetically predetermined plan that is
implemented during development, or whether they are influenced by the
amount and type of visual stimulation that occurs before the developmental
process is complete.
These questions have been addressed through experiments in which
1) the properties of the system are assessed at birth or shortly thereafter
(precluding the possibility of any significant contribution of experience), or
2) animals are subjected to abnormal visual experience during postnatal development.
The earliest experiments of the former type were conducted by
Hubel and Wiesel (1963), who reported the visual sensitivities of neurons in
the primary visual cortex of newborn kittens to be similar in most respects
to those of mature animals. Although these findings suggested a large degree
of genetic control, Wiesel and Hubel (1963, 1965) also discovered that abnormal
visual experience, such as extended closing of the eyelids or induction
of strabismus (both achieved surgically), dramatically altered the visual
sensitivities of cortical neurons, provided that the intervention occurred
during a “critical period,” which extended for several weeks postnatally.
Other studies have demonstrated close relationships between such critical
periods and their presumed causes and effects—that is, the formation of appropriate
anatomical connections (for a review, see L.C. Katz and Shatz
1996) and the development of visual perceptual abilities (for a review, see
Teller 1997).
The general view that has emerged from all of these experiments is that
the newborn visual system possesses a considerable degree of order, but that
visual experience is essential during critical periods to maintain that order
and to fine-tune it to achieve optimal performance in adulthood. Hubel and
Wiesel (1965) summed up the implications of this view with characteristic
prescience and breadth: “All of this makes one wonder whether more subtle
types of deprivation may not likewise exert their ill effects through the deterioration
of complex central pathways that either were not used or else were
used inappropriately.” The degree to which this may be true throughout the
life of an organism is an issue we address below in the context of perceptual
A Future for the Study of Neural Systems
Despite unprecedented progress, our understanding of visual system organization
and function is far from complete. On the contrary, developments and
Neural Science 283
discoveries of the past century have raised many new and often unanticipated
questions regarding the visual system and other large brain systems.
Here we focus on a few of the bigger issues at stake, with some predictions
about where this field of research may be headed in the new millennium.
How do sensory representations lead to perception?
The physiological studies of Hubel and Wiesel and many others over the past
50 years have revealed much about how basic features of the visual image,
such as oriented lines and patches of color, are detected and represented by
cortical neurons. But how do these cellular representations account for our
perceptual experience of the world? The underlying assumption has been
that perception of complex scenes would result from the collective activities
of neurons whose properties we so far have considered and were characterized
under reduced stimulus conditions. It is, however, increasingly apparent
that this assumption is flawed because it posits that individual neuronal
representations of sensory features are independent of one another, and the
field of sensory physiology is consequently at a turning point in its evolution.
One can fully appreciate the problem and begin to chart a new course for
the future by tracing the origins of current views on the cellular bases of perception.
One popular nineteenth-century view was known as “elementism.”
According to this influential doctrine, any percept can be explained as a collection
of independent internal states (sensations) elicited by individual sensory
elements, such as brightness, color, and distance. The undeniable
appeal of elementism rests on the power of reductionism, whereby it should
be possible in principle to dissect out the elemental causes of perceptual experience—
a red patch here, a yellow contour there, some motion in the
center, etc.—much in the way one might dismantle a pump. There were,
nonetheless, many early critics of this view, including the physicist Ernst
Mach (1886/1924), and its foundations crumbled at the turn of the century
with the emergence of a Gestalt theory of visual function, which maintained
that the perceptual whole is indeed far greater than the sum of the sensory
This Gestalt perspective was promoted vigorously by the psychologists
Max Wertheimer (1924/1950), Wolfgang Köhler (1929), and Kurt Koffka
(1935), and its legitimacy was most effectively communicated by simple and
compelling visual demonstrations. Through such demonstrations, it could
easily be seen that the percept elicited by one stimulus element (e.g., brightness)
is heavily dependent upon other stimulus attributes (e.g., threedimensional
form) in the same image. A key feature of Gestalt theory is thus
contextual interaction: the perceptual interpretation of each visual image fea284
Psychiatry, Psychoanalysis, and the New Biology of Mind
ture is a function of the context offered by other features in the image. Why
contextual interaction occurs is in itself an interesting question. The answer
can be found in the fact that visual images are only ambiguously related to
the visual scenes that give rise to them (i.e., there are, in principle, an infinite
number of scenes that can lead to the same visual image on the retina). The
context in which an image feature appears provides a rich source of information
that can be used to resolve its ambiguity—to, in other words, assist the
viewer in identifying the “meaning” of the feature, as defined by the content
of the scene that led to its appearance.
Although the holistic and eminently functional perspective of the Gestalt
tradition bears great validity, the tradition has long lacked momentum, owing
in part to its failure to develop mechanistic or neuronal foundations, and
it has had surprisingly little influence throughout much of the twentieth
century. Indeed, with the rise of single-cell studies of the mammalian visual
system, we have witnessed an unwitting return to the principles of elementism,
largely as a matter of investigative convenience. And therein lies the
problem. When Hubel and Wiesel stimulated cortical neurons with single
oriented lines, they purchased the power to reduce the response to a simple
code for oriented image features, which has been of enormous benefit, but
at the cost of a lack of generality. From the orientation tuning of a V1 neuron
obtained by such means, one learns how the cell encodes the physical properties
of the retinal stimulus. If, however, the meaning of the stimulus—that
is, its environmental cause, the thing that is perceived—is only revealed by
context, as shown by the Gestalt theorists, then it is frankly impossible to
learn what role the cell plays in perception using this experimental approach.
In the search for an alternative approach to carry us into a new millennium,
it is useful to return to the principles of Gestalt theory and to develop
an operational distinction between candidate neuronal substrates for sensation
and for perception. Accordingly, candidate neuronal substrates for sensation—
which have been the primary subjects of study over the past
50 years—encode the physical properties of the proximal stimulus (the visual
image), such as orientation or direction of motion. Perceptual representations,
on the other hand, reflect the world (the visual scene) that likely
gave rise to the sensory stimulus. Contextual manipulations make it possible
to dissociate local sensory properties from perception, and thereby offer a
means to identify neuronal responses that are correlated with perception
rather than the proximal sensory stimulus. Francis Crick and Christof Koch
(1998) have recently equated perceptual representations of the sort defined
here with “neural correlates of consciousness,” owing to their belief that perceptual
awareness is a legitimate operational definition of consciousness. We
take this issue up below in the section on vision and consciousness.
Neural Science 285
There are many ways in which this research strategy for studying candidate
neuronal substrates for perception can be applied. These fall into two
complementary categories. First, one can investigate whether neuronal responses
to identical receptive field stimuli covary with the different percepts—
determined by different contexts—that those stimuli elicit. The set
of stimuli illustrated in Figure 6–17 are of this class, and a valid experimental
goal would be to identify neuronal responses that vary with the percept
elicited, even though the receptive field stimulus (the gray rectangle in Figure
6–17) remains physically unchanged. Second, one can investigate the
neuronal responses to different receptive field stimuli—“sensory synonyms”—
that elicit the same percept, owing to context. Both of these situations,
which are prevalent in normal experience, afford opportunities to
experimentally decouple sensation and perception.
The first of these two approaches—whether neuronal responses covary
with different contexts—was used by Thomas Albright, Gene Stoner, and
colleagues to understand the role of cortical visual area MT in motion perception
(for a review, see Albright and Stoner 1995). Moving objects in a typical
visual scene commonly generate a complex array of moving visual image
features. One objective of the visual system is to integrate these moving features
to recover the coherent motions of the objects that gave rise to them.
That integration process is heavily and necessarily context dependent, such
that, for example, the object motions that are perceived from two identical
collections of image motions can vary greatly as a function of the context in
which the motions appear. A simple example of this phenomenon can be
found in the appearance of two overlapping stripes that move in different directions.
This type of visual stimulus, which is known as a “plaid pattern,”
is a simple laboratory incarnation of a common real-world occurrence,
namely two objects moving past one another.
Albright and Stoner proposed that if contextual cues present in the image
indicated that the two stripes lay at different distances from the observer,
then the two stripes would appear to move in two different directions. On
the other hand, if the very same motions were viewed in the presence of contextual
cues that indicated no such depth ordering, then the two striped
components of the plaid would appear to move coherently in one direction.
There are a number of different contextual cues that can be used to elicit a
percept in which the stripes are ordered in depth. In their initial experiment,
Stoner, Albright, and V.S. Ramachandran used luminance cues for perceptual
transparency to achieve this goal. Simply by adjusting the intensities of
light coming from different regions of the plaid pattern, it was possible to
make the plaids appear as either a single surface or two overlapping surfaces
(Stoner et al. 1990). Moreover, as predicted, these contextual manipulations
dramatically altered perceived motion, even though the image motion was
286 Psychiatry, Psychoanalysis, and the New Biology of Mind
unchanged. Once this dissociation between image and perceived motion was
discovered, the dissociation became a useful tool to investigate whether the
responses of cortical motion-sensitive neurons encode image motion or perceived
In a second experiment, Stoner and Albright (1992) pursued this idea
and found that when directionally selective neurons in cortical visual area
MT were presented with identical image motions that were perceived differently
as a function of context, the responses of many neurons covaried with
the motion that was perceived. These findings support the view that an important
step in the visual processing hierarchy is the use of context to construct
cellular representations of visual scene attributes—the stuff of
perceptual experience—out of cellular representations of visual image features,
such as oriented contours.
The second approach to decoupling sensation and perception is the reciprocal
of the first and is based upon the phenomenon of perceptual constancy,
in which multiple sensory stimuli give rise to the same percept, owing to appropriate
contextual cues. Perceptual constancies reflect efforts by the visual
system to recover behaviorally significant attributes of the visual scene, in
the face of variation along behaviorally irrelevant sensory dimensions. Size
constancy—the invariance of perceived size of an object across different retinal
sizes—and brightness/color constancy—the invariance of perceived reflectance
or color of a surface in the face of illumination changes—are classic
examples. Generally speaking, the physiological approach advocated is one
in which neuronal responses are evaluated to determine whether they covary
with the changing receptive field stimulus, or whether they exhibit an
invariance that mirrors the percept. Physiologists have only begun to employ
this approach. Several studies, for example, have recently explored a
phenomenon termed “form-cue invariance,” in which a percept of motion of
shape is invariant across different “form cues,” such as luminance, chrominance,
or texture, that enable the stimulus to be seen. In one such study, Albright
(1992) discovered a population of motion-sensitive neurons in area
MT that appear to encode the direction of motion of a stimulus independently
of the fact that the form cue—an aspect of the retinal stimulus that is,
in principle, irrelevant to motion detection—is varied. To paraphrase from
Horace Barlow’s (1972) neuron doctrine for perceptual psychology, the main
function of such cells appears not to be the encoding of specific characteristics
of retinal illumination, “but to continue responding invariantly to the
same external patterns”—that is, to the meaningful attributes of the input.
From the outset, the physiological approach to the operations of the visual
system has seen itself as being in the service of perceptual psychology.
If the “exploration of psychological territory” is to continue, however, physiologists
must advance beyond the acontextual approach that has been the
Neural Science 287
standard of twentieth-century research in this field. New experimental approaches
in which contextual influences are exploited as tools for the study
of neural substrates of perception are thus likely to be an important feature
of future research in this area.
Binding it all together
As we have seen, the representational strategy that the visual system has
adopted is one in which the properties of the incoming signal are distributed
across many neurons, such that each neuron only conveys a small piece of
the larger picture. At the level of the retina, for example, the information represented
by single cells is limited to a small circular region of space. At the
level of primary visual cortex, cells integrate information from earlier stages
in order to convey information about contour orientation, but they remain
highly specialized. At still higher levels, visual information is further combined
and abstracted to yield even more complex properties and greater specialization
of function, as evidenced by the multiplicity of extrastriate visual
areas. In view of this strategy, one cannot help but wonder how all of the specialized
representations are bound together to render a neuronal signal that
conforms to the complex patterns that we perceive. How is it, for example,
that the cells representing the edges, the varying orientations, the colors and
textures, and the different distances associated with the tree outside my corner
window, are linked together to produce my percept of that tree? How are
other properties of the same visual image, such as the attributes of a different
but nearby tree, “segmented” and bound together as a separate entity from
the first? Even more puzzling is the fact that only some of these complex patterns
enter my awareness at any point in time. How are those patterns selected,
and how are objects that we are simultaneously aware of linked
together? What role does visual attention play in this process? Is there a distinguishing
feature of the collections of neurons that happen to represent
objects or collections of objects that we have become conscious of? Is there,
as contemplated by Sherrington (1941) a half century ago, one “pontifical
cell” that represents the final outcome of this integration process? The representational
problem addressed by these questions has become known as
the “binding problem.” In a more general form, the problem has preoccupied
philosophers and cognitive scientists for decades, and it now stands
among the most formidable challenges in modern neuroscience (see, for example,
the October 1999 issue of Neuron).
At its most basic level, the binding problem is simply that of representing
conjunctions of attributes. There are, in principle, two mechanistic strategies
that could accomplish this task, one based on neural space and the other
based on time. On the one hand, attributes that must be conjoined—such as
288 Psychiatry, Psychoanalysis, and the New Biology of Mind
the color and the direction of motion of an object—could be represented in
that form by selective convergence of information onto single neurons,
yielding a neuron, for example, that selectively encodes rightward-moving
red objects. The appeal of such a strategy is that it follows naturally from
what is already known of the hierarchical properties of the visual system.
The neuronal representation of an oriented contour is, after all, nothing
more than a product of selective convergence of information from the previous
stage. Moreover, long-standing evidence indicates that neurons well up
in the hierarchy encode very complex conjunctions of visual attributes, such
as those associated with faces (Gross et al. 1969). The problem with this
form of binding, however, is one of generality: the variety of unique perceivable
conjunctions of visual attributes vastly exceeds the number of available
neurons. So while this may be a strategy that the visual system has adopted
for early levels of integration and for highly specialized and vital functions
like facial recognition, it is simply untenable as a general mechanism for
The alternative strategy is one in which visual attributes are bound in
time, rather than by static spatial convergence. The obvious advantage of a
dynamic binding mechanism is that, unlike the static design, it places no serious
combinatorial limits on the pieces of information that can be conjoined.
A form of this mechanism was suggested as early as 1949 by the
psychologist Donald Hebb (1949), who hypothesized the existence of “cell
assemblies.” Each such assembly was conceived as a collection of neurons
that are dynamically associated with one another as needed to link the features
they independently represent. A key feature of this concept is the ability
of each cell to hold membership in multiple overlapping assemblies—
such that, for example, a cell that represents upward motion may be assembled
with a cell representing the color red on one occasion but assembled
with a cell representing the color green on a different occasion. This view of
binding was subsequently elaborated upon by Horace Barlow (1972), who
noted it to be a particularly efficient form of representation because perceptions
commonly “overlap with one another, sharing parts which continue
unchanged from one moment to another.”
The trick, of course, is identifying a dynamic binding code that can be
used to transiently link cells into an assembly. One idea, which was implicit
in Hebb’s original proposal, developed significantly in the early 1980s by the
theorist Christoph von der Malsburg (1981), and which has subsequently
drawn a great deal of attention (see, for example, reviews in the October
1999 issue of Neuron), is that temporal synchrony of neuronal firing patterns
may underlie binding. As suggested in 1989 by Charles Gray, Wolf Singer,
and colleagues, “synchrony of oscillatory responses in spatially separate regions
of the cortex may be used to establish a transient relationship between
Neural Science 289
common but spatially distributed features of a pattern” (Gray et al. 1989).
This solution is in effect a dynamic switchboard that binds collections of
complex features “on demand” via synchronized firing of the neurons that
represent the individual features. Gray and Singer presented provocative
data in support of this hypothesis. They found that the temporal spiking patterns
of pairs of simultaneously recorded neurons in visual cortex were
likely to be correlated if the separate visual stimuli that elicited those patterns
appeared (to human observers) to be part of a common object. Other
studies, however, have failed to find support for this synchrony hypothesis
(e.g., Lamme and Spekreijse 1998; for a review, see Shadlen and Movshon
1999) and the matter remains unsettled.
If we accept the concept of dynamic cell assemblies, and the related proposal
for temporal binding by synchronous firing (if only for the sake of argument),
we face many critical questions. How, for example, are the
transient patterns of synchrony “read out”? Does synchronous firing lead to
transient synaptic facilitation of converging inputs onto a multipurpose
pontifical cell (or, perhaps more appropriately, given the democratic and
ephemeral nature of the hypothesized convergence, a “presidential cell”)?
Or is the perceptual binding simply implicit in the activity of the synchronized
neurons, which constitute flexible cell assemblies for specific percepts?
What elicits synchrony to begin with? Is there a top-down supervisory module
that identifies attributes that experience tells us are likely to be parts of
the same object? Or is the process bottom-up, using a variety of “image segmentation
cues” to parse out attributes that belong to the same versus different
objects? And what happens when—as is often the case—there are
multiple objects perceived simultaneously? Is spike timing sufficiently precise
to allow multiple synchrony events to occur simultaneously?
While we await the answers to these and other questions, it nevertheless
appears likely that visual integration rests upon a combination of static and
dynamic binding mechanisms. Indeed, except for a few clever hypotheses
and controversial details, our understanding of these processes has advanced
little beyond the view advocated by Barlow in the early 1970s (as a
counterpoint to Sherrington’s pontifical cell metaphor), according to which
a series of “cardinal cells” reside at the top of static convergence hierarchies,
but “among the many cardinals only a few speak at once” in the form of dynamic
cell assemblies (Barlow 1972). But research now moves rapidly on
these fronts and, with vast improvements in technology for monitoring the
firing patterns of many neurons simultaneously, the existence and operations
of cell assemblies and their role in binding should come into sharper
focus in the coming years.
290 Psychiatry, Psychoanalysis, and the New Biology of Mind
Vision and consciousness
Interestingly, the binding problem and its proposed solution by a dynamic
code have also been coupled with the phenomena of visual awareness and
consciousness. In the case of visual awareness, the argument is quite natural
(if not tautological), as there are good reasons to believe that the perceptual
binding of visual attributes is tantamount to their reaching the perceiver’s
awareness. Indeed, Singer and colleagues have argued that “appropriate synchronization
among cortical neurons may be one of the necessary conditions
for the . ..awareness of sensory stimuli” (Engel et al. 1999). Developing the
concept of neuronal “metarepresentations,” Singer (1998) has furthermore
suggested that this code may underlie all of the complex patterns that enter
our awareness at any time. The extension of this line of argument to consciousness
depends, of course, on how one defines the term. Although there
is a long history of confusion about what consciousness actually means, and
a plethora of colloquial uses of the term, Francis Crick and Christof Koch
(1998) have recently attempted to facilitate scientific progress by arguing for
a specific and limited definition that is relevant to vision. (We consider the
issue more broadly in “Consciousness: A Challenge for the Next Century”
below). Roughly speaking, that definition can be equated with perceptual
awareness; it is “enriched” by visual attention, and may fill a window of
time, which Gerald Edelman (1983) has termed “the remembered present”
(see also James 1890). If we accept that operational and not unreasonable
definition, then dynamic representations of the sort proposed by Singer and
others may be relevant to consciousness. But this is slick and unstable terrain,
newly trodden by neuroscience and lacking the guideposts of established
experimental paradigms (deflecting, to paraphrase Bertrand Russell,
nearly all but fools and Nobel laureates). It is, nonetheless, one of the most
compelling issues facing the future of neuroscience, and is certain to be a focal
point of research in the next century.
What are the local cellular mechanisms of vision?
As we have seen, physiological studies have revealed much about the types
of visual information carried by neurons at different processing stages. In
parallel, anatomical studies have told us a great deal about the gross pattern
of connections within and between different stages of processing. Until recently,
however, comparatively little has been known about the local circuits
that confer neuronal properties and mediate the computations required for
perception. This knowledge is an essential starting point for understanding
how neurons integrate and store different sources of visual information, as
well as alter their sensitivity to compensate for environmental and behavioral
Neural Science 291
Recent progress in this area has been fueled by new technologies that allow
finer-resolution tracing of anatomical connections, in conjunction with
methods that allow assessment of the contributions of these connections in
their functional state. For example, optical imaging of neuronal activity,
combined with cell labeling, is enabling us to determine the relationships
between functional architecture and cortical circuits (Malach et al. 1993).
Investigation of correlated firing patterns between pairs of neurons, in conjunction
with precise measures of receptive field properties, has provided an
approach complementary to anatomical tracing of local circuitry (for a review,
see Usrey and Reid 1999).
Another promising technique of this sort, known as photostimulation,
was applied recently by Edward Callaway and Lawrence Katz. This technique,
which enables one to assess the pattern and strength of synaptic connections
between neurons in local cortical circuits, exploits a form of the
excitatory neurotransmitter glutamate that is inactive (“caged”) until illuminated
(L.C. Katz and Dalva 1994). Callaway and colleagues have revealed
that the pattern of functional connections between different laminae in primary
visual cortex provides far more opportunities for cross talk between
different visual processing streams than was evident from traditional anatomical
studies (Callaway 1998).
Another approach to understanding the relationship between circuitry
and function might involve deactivation of individual circuit components,
such as specific classes of cells, and assessment of the ensuing loss of function.
At a gross level, this approach is recognizable as the lesion method that
has been used in neuropsychology for over a century to reveal the functions
of large neuronal systems. But is it realistic to expect that these methods can
be extended to identify the fine details of circuit organization and function?
The fact that circuit components are both anatomically and functionally intermingled—
particularly in the cerebral cortex—would seem to preclude
this possibility. A resolution, however, can be found in new molecular techniques
that enable one to manipulate gene expression and to exploit the genetic
distinctiveness of cells that serve different functions.
These new techniques incorporate three key features: 1) the ability to introduce
novel genes into neurons, the expressed products of which will alter
neuronal function, 2) the ability to regulate expression of these transgenes
in a time-dependent manner, and 3) the ability to regulate expression of
these transgenes selectively in specific classes of cells. The first of these techniques
has been standard fare for some time now in the form of germline
transgenic manipulations in mice (for a review, see Picciotto 1999). The
same end point is now possible in other species, including primates, using
viral vector transfection (Takahashi et al. 1999). In principle, as discussed
earlier, it should be possible by these means to introduce novel genes that
292 Psychiatry, Psychoanalysis, and the New Biology of Mind
block neuronal cell firing when expressed, which would effectively remove
affected cells from the circuit. Recent evidence suggests that overexpression
of K+ channels may be an effective means to transiently inhibit conduction
of action potentials or their propagation into dendrites (Johns et al. 1999).
The second technique is also becoming routine using one of a number of inducible
systems that promote gene expression only in the presence of exogenous
factors, which can be delivered by the experimenter (e.g., Nó et al.
1996). This temporal control permits before and after measures of the contributions
of affected cells. The third technique—cell-specific expression—
is absolutely critical, of course, if these tools are to provide any greater resolution
than standard cell-ablation techniques. As discussed earlier in this
review, this technique taps into gene “promoters” that are known to regulate
expression of specific genes only in specific cell types. By replacing the genes
that are normally regulated with novel genes, one can restrict expression of
these transgenes to cells that recognize the promoter. Mark Mayford, Eric
Kandel, and colleagues have demonstrated the feasibility of these three basic
techniques using germline transgenic manipulations in mice to explore the
functions of hippocampal neurons in relation to memory storage (Mayford
et al. 1996).
Related techniques might also be used to facilitate anatomical analysis of
local circuits. For example, instead of introducing and expressing a gene that
disrupts cell firing, one might simply transfect neurons with genes that encode
visible proteins, such as GFP. The end result in this case would be selective
labeling of a specific class of cells, which could be used, for example, to identify
those cells in a brain slice preparation for physiological recording, or simply
for analysis of cell morphology and connections using light microscopy.
Recent experiments document the feasibility of this general approach using
germline transgenic manipulations in mice (Yoshihara et al. 1999).
There are many technical details that will need to be worked out—not the
least of which is the identification of additional cell-specific promoters perhaps
through the strategies outlined in our earlier discussion of the assembly
of neural connections—before these fantasy experiments become a practical
means to investigate the organization and function of local circuits in the primate
visual system. Nonetheless, the potential benefit afforded by this unprecedented
merger of molecular tools and systems approaches to brain function
is clearly enormous. They are certain to become a staple of future experiments
aimed at understanding the entire realm of brain systems.
How do cellular representations change with visual experience?
As we have seen, a central tenet of modern neuroscience is that stages in
brain development correspond to specific stages in the development of perNeural
Science 293
ceptual abilities. These stages are known as critical periods, and they are
characterized by an extraordinary degree of neuronal and perceptual plasticity.
Only recently has it been recognized that the plasticity of the visual system
is not restricted to these critical periods early in development but is
modifiable throughout the adult life of the organism. The forms of this adult
plasticity are many and varied, but all can be viewed as recalibration of incoming
signals to compensate for changes in the environment, the fidelity
of signal detection (such as that associated with normal aging or trauma to
the sensory periphery), or behavioral goals.
One of the most striking and revealing forms of adult neuronal plasticity
is that associated with perceptual learning, which is an improvement with
practice in the ability to discriminate sensory attributes. In humans, these
learning phenomena are ubiquitous in everyday life and generally selfevident,
and they have been a subject of scientific investigation for decades
(for a review, see Karni and Bertini 1997). Consider, for example, the copy
editor who over time becomes particularly sensitive to graceless word pairings,
or the assembly line worker who can instantly recognize the miswired
transistor. Until recently, however, little effort had been made to investigate
their neuronal bases. Indeed, the critical period concept had become so
widespread and deeply rooted that there seemed little ground for believing
that visual representations might be modifiable throughout life.
Thus, it was well before the modern neuroscience community was prepared
to embrace the concept of plastic representations in mature animals
that Michael Merzenich began to address the degree to which sensory maps
could change in response to a variety of manipulations (for a review, see
Buonomano and Merzenich 1998). This work began to have a broad impact
in the mid-1980s with the demonstration of marked and systematic reorganization
of somatosensory cortex in response to a change in the peripheral
sensory field (e.g., selective deafferentation). Even more exciting and provocative
was the subsequent demonstration that cortical maps reorganize in
response to selective use of components of a sensory modality, in a manner
that mirrors perceptual learning.
Following in the footsteps of Merzenich, Charles Gilbert has recently begun
to investigate the relationship between adult visual perceptual learning
and the receptive field properties of cortical neurons. In one set of experiments,
Gilbert and colleagues have found that increases in perceptual sensitivity
fail to generalize to spatial locations or stimulus configurations beyond
those in the set of training stimuli (Crist et al. 1997). This high degree of
spatial specificity suggests that the underlying neuronal changes may occur
at a very early stage of processing—perhaps V1—where the spatial resolving
power of cortical neurons is greatest. Other behavioral observations support
this conclusion (e.g., Karni and Sagi 1993). Using a complementary ap294
Psychiatry, Psychoanalysis, and the New Biology of Mind
proach, Gilbert and others have demonstrated plasticity of cortical representations
more directly (Chino et al. 1992; Gilbert and Wiesel 1992). In this
case, the receptive field properties of V1 neurons were found to change following
localized interruption of retinal input (caused by small retinal lesions).
Similar to previous findings from the somatosensory system, these
changes took the form of shifts in the spatial profile of receptive fields, such
that cells normally responding to light in the area covered by the retinal lesion
become sensitive to stimulation of adjacent regions of visual space.
These changes began to occur in a matter of minutes following deafferentation.
Although in this case, unlike perceptual learning, the plasticity was not
induced by repeated exposure to a sensory stimulus but was rather a response
to a marked loss of stimulation, both can be viewed as forms of renormalization
and the underlying cellular mechanisms may be similar.
This is among the most exciting areas of systems neuroscience today,
bridging as it does the topics of sensory processing and learning. Early contributions
to this field have been particularly inspiring and influential because
the prevailing wisdom held that sensory representations were largely
immutable following critical periods of developmental reorganization. As we
have seen, recent observations prove that this is not the case. On the contrary,
representational changes occur throughout life as part of a normalization
process to compensate for damage or deterioration of the sensory
periphery or to meet novel behavioral and perceptual requirements. But
these findings naturally raise many new questions that will occupy neuroscientists
for years to come. Little is yet known, for example, of the specific
neuronal events that give rise to plasticity of the adult visual system, although
such processes are sure to include changes in synaptic efficacy,
changes in neuronal cell structure, and possibly neurogenesis. Future research
is also likely to address the following questions: How are these experiencedependent
changes in visual processing mediated? Evidence indicates that
higher cortical areas, such as regions of the frontal cortex, contain neurons
that represent attributes of memorized stimuli. What role, if any, do these
mnemonic representations play in the formation of experience-dependent
changes in visual cortex? What are the control signals that initiate such
changes? Does representational plasticity occur at all stages of visual processing?
Does it occur in the retina? Do such changes constitute the neuronal
repository of long-term visual memories?
Beyond vision: exploring links with other brain systems
It is a pedagogical convenience to treat the visual system—as we have done
here—as functionally independent and separable from other brain systems.
The fact of the matter is that vision is but one cog in the wheel and is in many
Neural Science 295
ways integrated with other major systems, including those responsible for
memory, emotion, and motor control. Although we know less—much less,
in some cases—about these other systems, it is now clear that the areas of
interface between vision and those systems that serve storage, evaluation,
and action are among the most important targets for future research in systems
neuroscience. Here we consider one of these areas of interface: that associated
with motor control.
Visual guidance of behavior: from retina to muscle
A major function of the visual system is to provide sensory input to guide
actions, such as moving through the environment. Visual and motor control
systems have in common the fact that they both represent space. But the relevant
frames of reference—retinal space, in the case of vision, and ultimately
muscle space, in the case of action—are radically different. How then does
light falling on a particular location on my retina lead to a reaching arm
movement (or an eye movement, or a leg movement, etc.) to the source of
the light? The problem becomes even more puzzling if we consider that exactly
the same arm movement will be executed regardless of what direction
I am looking, implying that vastly different retinal signals can lead to the
same motor output. In principle, there are a number of different means by
which this coordinate transformation could be accomplished.
Perhaps in large part because of the compelling subjective sense that
space is stable regardless of the orientations of our sense organs and our
muscles, it has often been proposed that the brain contains a unified representation
of space. This unified map might represent space in a three-dimensional
“world-centered” frame of reference, as opposed to the more specific coordinates
of the sensory (retina) and effector (muscle) organs. According to
this view, we have an internal neuronal map of the spatial locations of all of
the items on my desk in front of us. That map remains coherent and unchanged
regardless of which way we are looking—or, for that matter, which
way the entire body is oriented. The advantage of such a system is that it provides
a generic source of spatial information that can be used to guide all
movements, which is independent of the state of the sense organs or muscles.
The disadvantage is that, because of its independence from sense organs
and muscles, a generic reference frame is extremely difficult to compute.
Numerous studies conducted over the past few decades have evaluated
the hypothesis that space is transformed from sensation to action via a unified
reference frame. Neuropsychologists have examined the effects of damage
to brain regions that lie between early visual processing and motor
control—specifically the parietal and premotor areas of the cerebral cortex.
The typical consequence of such damage is “neglect,” in which subjects ig296
Psychiatry, Psychoanalysis, and the New Biology of Mind
nore stimuli that appear in certain regions of visual space (for a review, see
Mesulam 1999) (Figure 6–19). (Neglect is distinguished from blindness by
the fact that a neglect patient can clearly see a neglected stimulus if his or
her attention is drawn to it.) If there were a single unified map of space, one
would expect that neglect would be manifested in the same part of the spatial
map—always to the right side of the observer, for example. On the contrary,
results indicate that neglect can be present in any of a number of
different spatial reference frames (retinal coordinates, body part coordinates,
object-based coordinates, world-based coordinates), suggesting that there
are multiple spatial maps, which may serve specialized functions (for a review,
see Colby and Goldberg 1999).
Neurophysiological data also support the hypothesis that there are multiple
types of spatial maps used to transform information from sensory to
motor coordinates. This issue has been explored extensively by Richard
Andersen and colleagues, who recorded from neurons in the parietal cortex
in search of a representation of visual space in head-centered coordinates
(which would, in principle, be useful for directing movements of the eyes).
Andersen discovered instead that these neurons possess “gain fields,” by
which the amplitude of the response to a visual stimulus is modulated systematically
by the direction of gaze (Andersen et al. 1985). Because these
neuronal responses take eye position into account, it is in principle possible
to deduce the spatial location of the visual stimulus from the activity of a
population of such neurons, regardless of where the eyes are looking (for a
review, see Andersen et al. 1993). This information can then be used to guide
movements to the stimulus.
Recent physiological experiments by two groups—Carl Olson and Sonya
Gettner, and Michael Graziano and Charles Gross—provide fascinating
evidence for more explicit but highly specialized spatial maps that could mediate
visual-motor control. Olson and Gettner (1995) recorded from individual
neurons in premotor cortex of monkeys and found that the neurons
responded if an eye movement was made to a particular part of an object, regardless
of the spatial location of the object. These neurons thus appear to
represent space in an object-based coordinate frame. Graziano and Gross
studied single neurons in the premotor cortex that possess both visual and
tactile receptive fields. The visual receptive field of each neuron was found
to be linked to the spatial location of the tactile receptive field, such that, for
example, a neuron that was activated by tactile stimulation of the arm was
also activated by a visual stimulus in the vicinity of the arm (Graziano et al.
1994). Remarkably, if the arm moved to a new location, the visual receptive
field moved along with it. The visual receptive fields of these neurons thus
appear to have been transformed from a retinal frame of reference to a reference
frame centered on the position of the arm. Graziano and Gross propose
Neural Science 297
that this arm-centered reference frame may be well suited for orchestrating
movements of the arm to stimuli that are near the arm. More generally, they
speculate that visual-motor transformations of many types may rely upon
specialized body-part centered maps of space, rather than upon a single unified
spatial map. Although both of these physiological findings suggest
promising new approaches to the study of sensorimotor coordinate transformations,
they leave many questions of a mechanistic nature unanswered.
Perhaps the most nagging question raised by the Graziano and Gross study
concerns the apparently profound spatial mobility of the visual receptive
field, which dances across retinal space with every movement of the arm.
How are retinal signals dynamically rerouted, as it were, to continuously update
the visual receptive field of the premotor neuron, using information
about arm position as a guide? This and other related questions will be an
important focus of research in years to come.
Consciousness: A Challenge for the Next Century
Perhaps the greatest unresolved problem in visual perception, in memory,
and, indeed, in all of biology resides in the analysis of consciousness. This is
a particularly difficult problem, in part because there is no widespread agreement
on exactly what constitutes a successful solution. There is agreement
nevertheless that a successful solution will require, at a minimum, insight
into two major issues that lie at the heart of the study of consciousness: 1)
awareness of the sensory world and 2) volition, the voluntary control of
thoughts and feelings.
In this section, we consider awareness by focusing on two of its components:
attentional orienting to sensory signals in the presence and in the absence
of stimuli (imagery). We shall then go on to consider volition by
focusing on the self-regulation of thoughts, feelings, and actions. These two
problems are at once relevant to consciousness, yet tractable, and therefore
serve to illustrate how consciousness can be dissected biologically.
As with other problems in biology, there are both reductionist and holistic
approaches to these components of consciousness. A reductionist approach
would view these aspects of consciousness from a genetic, synaptic,
and cellular level. However, in the case of consciousness, it is hard to imagine
any solution that would not also require an understanding of the large
neural networks that underlie cognition, actions, and emotion. In our view,
the appropriate direction in seeking a solution to the problems of consciousness
lies in successfully linking understanding at all of these levels, from
genes to behavior. In this section, we try to illustrate this integrative approach
in relation to orienting to sensory stimuli, imagery, and self-regulation. Finally,
we examine how far these scientific approaches will take us in under298
Psychiatry, Psychoanalysis, and the New Biology of Mind
standing the most subjective aspects of consciousness.
Rigorous top-down approaches to consciousness have been limited by the
lack of good methods for resolving the activity of populations of cells. The use
of neuroimaging methods during the last decade has made it possible to observe
the activity of large numbers of neurons in human subjects while they are studied
for their awareness of the sensory world and for their voluntary control of
thoughts and feelings (Posner and Raichle 1994, 1998). Cognitive studies using
these imaging methods, such as positron emission tomography (PET) and functional
magnetic resonance imaging (fMRI), are based upon changes in blood
flow and blood oxygenation that occur in localized regions of the brain when
neurons increase their activity (Raichle 1998; Rosen et al. 1998). These methods
have now been applied with some success to the study of attentional orienting,
visual imagery, and regulation of cognitive and emotional states. In each of these
domains, the individual functional components have proven to be surprisingly
well localized; however, each of the major functions of consciousness—such as
attentional orienting to sensory stimuli and volition—involves not one but several
functional components. As a result, each function of consciousness appears
to involve several networks and these are distributed across a variety of brain areas.
Fortunately, the enormous complexity of the problems has been made
somewhat more tractable by use of appropriate animal models. In the best case,
as with studies of the visual system, it has proven possible to relate neural activity
studied at the cellular level in nonhuman primates to the activity of large neural
networks studied in the same brain areas but now in human subjects using
brain imaging (Tootell et al. 1998). While the results are not definitive, they
show that specific aspects of consciousness can even now be analyzed on the cellular
level with methods currently available to neuroscience.
Orienting of Attention to Sensory Stimuli
Origins of the modern study of sensory attention
The modern study of attention can be traced to 1958 and the publication by
Donald Broadbent of a monograph entitled Perception and Communication
(Broadbent 1958). Broadbent proposed that when we focus attention on one
object to the exclusion of other surrounding objects, the focus of selective
attention requires a filter that holds back messages from unattended channels.
According to Broadbent’s view, attention is a high-level skill that is so
developed in some people as to allow them to perform remarkable feats such
as simultaneous translation. This skill allows even untrained subjects to
have a role in selecting their environment by attending only to certain stimuli
while shutting out others. Although there have been challenges to this
view in the four decades that have passed since it was proposed, even BroadNeural
Science 299
bent’s strongest critics have embraced his general approach. In the next
section we begin to explore the neuronal implementation of the type of selective
attention studied by Broadbent.
All visual areas, including primary visual cortex (V1),
can be biased by a shift of attention
To obtain an idea of how brain areas become involved in selection of a stimulus,
consider the task of looking for a file on your computer desktop. If the
desktop is cluttered with files, you will have to search for the one you want.
Such a search may be accompanied by eye movements, but if the objects are
close, the search may involve covert shifts of attention without eye movements.
Such visual search tasks involve the coordinated action of the two
large-scale brain networks. One network, the ventral visual pathway, which
we discussed in the previous section, is concerned with objects and with
form recognition, required for obtaining the identity of each file. The second
network—located in the posterior parietal cortex of the dorsal visual pathway—
is related to the act of shifting attention to the locations where the file
might be found. Early studies of the dorsal pathway were conducted by
Michael Goldberg and Robert Wurtz (1972). They found that cells in the
posterior parietal cortex of alert monkeys responded differentially to identical
stimuli depending on whether or not the monkey was attending to the
stimulus (Figure 6–18). When the monkey attended, the firing of the cell
was much more intense than when the monkey ignored the stimulus. These
results provided the first data on the cellular level that neurons in the parietal
cortex are correlated with attention to the location of visual objects.
With the advent of neuroimaging, it became possible to see the distributed
network of brain areas involved in attention in human subjects. This network
includes the frontal eye fields, the superior colliculus, and the posterior
parietal lobe, all of which are also involved in eye movements (Corbetta
Studies conducted by Robert Desimone and colleagues have addressed
the role of the ventral visual pathway (particularly areas V4 and IT) in attentional
control. A typical paradigm involves first establishing the stimulus selectively
for a cortical neuron. Suppose, for example, that the neuron under
study responded well to a red bar of light and poorly to a green bar when
these stimuli were individually placed in the neuronal receptive field. At this
point, both stimuli—the red and green bars—would be placed in the receptive
field of the cell at the same time. If the animal was instructed to attend
to the “good” stimulus (red bar) then the neuron responded well. If, however,
the animal was instructed to attend to the “poor” stimulus (green bar)
then the response was correspondingly poor—despite the fact that the retinal
300 Psychiatry, Psychoanalysis, and the New Biology of Mind
FIGURE 6–18. The influence of attention on the response of cortical
Neurons in the posterior parietal cortex of a monkey respond more effectively to a
stimulus when the animal is attentive to the stimulus.
(A) A spot of light elicits only a few action potentials in a cell when the animal’s gaze
is fixed away from the stimulus.
(B) The same cell’s activity is enhanced when the animal takes visual notice of the
stimulus through a saccadic eye movement.
(C) The cell’s activity is enhanced further when the monkey touches the spot, even
without eye movement.
Source. From Wurtz and Goldberg 1989, as illustrated in Kandel ER, Schwartz JH,
Jessell T: Principles of Neural Science, 4th Edition. New York, McGraw-Hill, 2000.
Neural Science 301
stimulus was the same in both cases. Desimone and colleagues interpreted
these results as evidence that the receptive field shrinks to conform to the
attended stimulus, thereby excluding the unattended stimulus and implementing
a filtering mechanism of the sort proposed by Broadbent (see Desimone
and Duncan 1995 for a review).
The exact visual area that will be biased in the manner revealed by these
physiological studies appears to depend upon the task required of the subject
(Desimone and Duncan 1995; Kastner et al. 1999; Posner and Gilbert
1999). Imaging studies have shown that if people are asked to attend to target
motion, activity is increased in a brain area in the dorsal visual pathway
sensitive to movement (area MT). Quite different visual areas become active
for attention to other stimulus dimensions such as color or orientation (Corbetta
et al. 1991). When attention is shifted to a new location, the neural activity
of cells in the ventral, object recognition network is increased even
before any target is presented at that location (Kastner et al. 1999).
There is a fundamental distinction between
focal and ambient attention
Of course, there is a sense in which, without even trying to attend, you are
conscious of all the objects on your desktop. However, when careful tests are
made that involve making changes in a complex visual scene, these tests reveal
that when attention is focused on one object, other objects within the
scene, even large and important ones, can be altered without the subject being
aware that a change has taken place (Rensink et al. 1997). Thus, while
attention can be summoned efficiently to a novel event, there is surprisingly
little awareness of changes that occur at loci that are not attended. It, therefore,
is useful to distinguish between focal attention, which allows reporting
of details of the scene, and ambient attention, which forms our general awareness
of the scene around us. While both are aspects of consciousness, their
underlying neurobiological mechanisms may be quite different. It seems
likely that ambient attention may depend primarily upon posterior brain areas.
By contrast, focal attention, which is often switched between objects
based on instructions, may depend on more anterior areas related to voluntary
control of action.
There may be only a small number of networks concerned
with attention, and these can be distinguished on the
cellular and even on the molecular level
It seems likely that the neuronal computation that occurs in most cortical
areas can be influenced by attention. Indeed, there is a surprisingly large
number of sites in the brain where attentional influences can be demon302
Psychiatry, Psychoanalysis, and the New Biology of Mind
strated. However, the source of those effects is thought to emanate from a
small number of networks that perform different functions.
For example, a novel visual stimulus serves both to alert the organism
and to orient attention to the location of the stimulus. This distinction can
be demonstrated by using separate cues for alerting and orienting. An alerting
cue provides the monkey with information about when a target will occur,
but not about where that target will be located. An orienting cue
provides the monkey with specific information about where the target will
be located and thus allows the subject to move attention to the cued location.
Two separate brain networks, both located in the parietal lobe, but using
distinctly different chemical transmitter systems, are involved in
changing the level of alertness and in switching attention toward the stimulus.
Thus, the influence of alerting cues is reduced by drugs that block norepinephrine
activity, but these drugs do not influence orienting. By contrast,
drugs that block cholinergic activity influence orienting to the cue, but do
not diminish the alerting effect (Marrocco and Davidson 1998). These pharmacological
studies illustrate how one can separate a simple act of attention
to a novel event into two distinct components, and pinpoint both the anatomical
systems and the modulatory synaptic mechanisms involved.
Findings in patients confirm the importance of neural systems for alerting
and orienting to our normal awareness of the world around us. Strokes
that interfere with the blood supply to the posterior parietal cortex on the
right side produce an inability to orient attention to the left side, the side opposite
the lesion. Patients who suffer from these right-sided lesions will
show striking deficits in body image and in their perception of spatial relations.
Although their somatic sensations are intact, these patients may ignore
(neglect) the spatial aspects of all sensory input from the left side of
their body as well as of external space, and they will ignore the left half of
any visual object with which they are confronted. For example, patients with
neglect syndrome will exhibit a severe disturbance in their ability to copy
drawn figures. This deficit can be so severe that the patient may draw a
flower with only petals on the right side of the plant. When asked to copy a
clock, the patient may ignore the digits on the left and try to fill in all the
digits on the right, or draw them down the side, running off the clock face
(Figure 6–19). These patients also may ignore the left half of the body and
fail to dress, undress, and wash the affected side.
Less dramatic but similar difficulties in orienting accompany loss of parietal
neurons due to degenerative disorders such as Alzheimer’s dementia
(Parasuraman and Greenwood 1998). In these cases, stimuli going directly
to the lesioned area, that would normally produce orienting of attention,
may no longer do so and consequently the person may be completely unaware
of these stimuli.
Neural Science 303
FIGURE 6–19. The contribution of the posterior parietal cortex
to visual attention?
The three drawings on the right were made from the models on the left by patients
with unilateral visual neglect following lesion of the right posterior parietal cortex.
Source. From Bloom and Lazerson 1988, as illustrated in Kandel ER, Schwartz JH,
Jessell T: Principles of Neural Science, 4th Edition. New York, McGraw-Hill, 2000.
304 Psychiatry, Psychoanalysis, and the New Biology of Mind
Many researchers agree with Francis Crick’s view that sensory awareness
probably is the most tractable area for a rigorous understanding of consciousness
at a mechanistic level. As a result, much research has recently
been focused on the study of orienting to sensory stimuli. The discovery that
attention can influence activity within primary visual cortex (area 17) has
allowed investigators to explore attentional effects within a visual area
whose other cellular and physiological features and anatomical characteristics
are extremely well characterized (Posner and Gilbert 1999). The work
in area 17 therefore provides an opportunity to specify exactly what cellular
structures and functions can be modified by the act of attending.
There also now are methods for exploring the consequences of attention
on the natural life of the organism. One direction of current research involves
the study of the maturation of orienting mechanisms in the human
infant. These studies illustrate that the ability to orient attention to visual
stimuli undergoes a substantial maturation during the first year of life. For
example, an infant 2–4 months of age has great difficulty in disengaging
from a strong visual attractor. If the attractor is a checkerboard, the difficulty
in disengaging may cause the infant to become distressed; if the attractor is
the eyes of the mother, the lack of engagement may contribute to the development
of parent-infant bonding. As with age, the parietal mechanisms involved
in orienting of attention mature; they appear to allow the infant to
disengage from strong attractors. Thus, by 4 months infants can begin to
move their eyes in anticipation of the occurrence of a visual stimulus. The
ability to show anticipatory eye movements demonstrates an influence of
learning on orientation, on where infants attend.
The maturation of the ability to orient to visual stimuli has dramatic consequences
for an infant’s response to novelty and their ability to know where
to look. At 4 months and older, a parent can regulate negative affect by use
of distraction, by orienting the infant to a novel stimulus (Posner and Rothbart
1998). Much of the development of orienting skills must relate to the
maturation of specific pathways between neural areas. The advent of new
forms of neuroimaging may allow one to follow the time course of the maturation
of the pathways required for the development of orienting and thus
give us a new means of exploring the mechanisms of these developmental
changes in infants and children (Conturo et al. 1999).
Imagery differs from perception in efficiency of coding
Visual images are an excellent example of a purely mental event, and as such,
they are a promising entryway for the neurobiological study of consciousNeural
Science 305
ness. Images seem to have a sensory character even though no sensory stimulus
is presented. In the early part of the twentieth century, visual images
could only be studied by the methods of verbal report and by the systematic
collection of surveys (Galton 1907). Because behaviorists thought that the
subjective nature of the image would never allow a scientific approach, the
study of imagery was largely abandoned. In the period following World
War II, imagery again became a focus of study in modern cognitive psychology,
and objective experimental methods for probing the characteristics of
these mental events were successfully developed (see Kosslyn 1980 for a review).
It is now possible to study brain mechanisms
of imagery in humans and animals
It is now possible to design objective tests of imagery. If a test subject is given
the name of a letter (e.g., R) and an angle of orientation (75 degrees) most
people can construct a mental image of the R at the correct angle. If people
are then shown visual probes of R that may be a real or mirror-image R, they
can quickly construct a mental image of the R when the probe letter is at the
same angle as the image. In fact, they are faster to respond to probes that
match the angle of the imaged R than to an upright R (Cooper and Shepard
But if one tests a subject by asking them to develop a visual image of a
somewhat lengthy word, such as “pumpkin,” most people believe they can
do it until required to perform an objective task of this accomplishment like
being asked to spell the word backward. The difficulty encountered when
asking a subject to spell backward a word like “pumpkin” led Weber and
Harnish (1974) to question whether there really was an image of the word
“pumpkin.” They then proceeded to compare the performance of subjects
when a word stimulus like “pumpkin” was physically present (perceptual
condition) and when it had to be created as an image (imagery condition).
With short three- or five-letter words, the subjects show the same reaction
time for an imagery condition as they show for the perceptual condition. But
when the word has more than five letters imagery is much slower than perception
(Weber and Harnish 1974). Thus, imagery can indeed produce a remarkably
efficient representation, but it is also rather limited to only about
three to five separate items. By contrast, most people can hold in memory
about seven to eight separate letter names.
It also takes longer to develop a visual image than to provide a name. If
you are asked to image each lowercase letter of the alphabet in turn, it will
take you 10–20 seconds to go through the alphabet, much longer than if you
were to name the letters silently. Creating mental images is thus a complex
306 Psychiatry, Psychoanalysis, and the New Biology of Mind
task that consists of many mental operations. Accordingly, many parts of the
brain are involved (Kosslyn 1994).
The ability to study the brain while people construct and inspect visual
images has greatly enhanced the field of mental imagery (Kosslyn 1994).
These studies have revealed that most of the visual areas that are involved in
pattern recognition and in orienting of attention also are recruited during visual
imagination. The overlap between areas recruited for the perception of
real as opposed to imagined visual objects is substantial. It is clear that visual
imagination uses the same apparatus of visual perception and the same systems
of visual attention as would be involved if the stimulus were actually
presented to the sense organ.
This finding also is supported by clinical evidence. Studies of patients
with lesions of the right posterior parietal cortex that produce visual neglect
in viewing real objects also disrupt the experience of imaging the left side of
a visual image (Bisiach and Luzzatti 1978). This defect in imagery was first
described in a fascinating study by Bisiach and Luzzatti of a group of patients
in Milan, all of whom had injury to the right parietal lobe. As the patients
were sitting in the hospital’s examining room, they were asked to imagine
that they were standing in the city’s main square, the Piazza del Duomo, facing
the cathedral, and to describe from memory the key buildings around the
square. These subjects identified from memory all the buildings on the right
side of the square (ipsilateral to the lesion) but could not recall the buildings
on the left, even though these buildings were thoroughly familiar to them.
The patients were next asked to imagine that they were standing on the
opposite side of the square, on the steps of the cathedral, so that right and
left were reversed. In this imagined position, the patients were now able to
name the buildings they previously had been unable to identify but failed to
identify or name the buildings they had previously listed. The patients now
described what they previously neglected, and neglected what they had previously
described, suggesting that they retained in memory full knowledge
of the space. What these patients lost was access to memories associated with
the side of the body opposite the lesion, no matter which way the patient
imagined himself facing.
Study of patients with lesions in the right posterior parietal cortex has
yielded three insights. First, these lesions commonly lead to a disturbance in
orienting of attention. Second, patients who neglect the left side of their
body after damage to the right parietal lobe show a disturbance not only in
cortical representation of their own body but also in representation of external
space. Specifically, they neglect visual stimuli on that side of the body.
Third, patients with neglect syndrome also lack access to memories against
which perceptions on the neglected side can be compared. Thus, these patients
neglect not only real external objects but also objects in memory.
Neural Science 307
Finally, the lesions not only lead to disorders in perceptive-spatial relationships;
they also commonly lead to a disturbance in directed attention (Figure
It has even been possible to develop an animal model of mental rotation
of the sort we described previously (Georgopoulos et al. 1989). Monkeys
were taught to move a lever in a direction 90 degrees from a target light. Recording
from cells in the motor system shows that immediately after the
light comes on, the set of active cells has an equivalent vector that would
move the limb directly to the light. However, over a 0.25-second interval, the
population of cells changes to compute a vector which is at the proper 90-
degree angle. Imaging studies of mental rotation in humans have shown that
motor areas of the brain as well as the parietal lobe become involved when
rotation involves one’s own hand (Parsons and Fox 1998).
Future studies of imagery can probe the influence
of learning and individual differences
Traditionally, imagery has been defined by subjective report, but that is no
longer necessary. It is now possible to know when the visual system has been
activated from the top down, even if the person is unaware that any form of
imagery has been used. For example, when reading a vivid description, people
often create a visual representation of the scene and individual words
may automatically evoke images related to their meaning. Some people are
aware of these representations, but others deny having any subjective visual
experience. By appropriate imaging studies, we can now determine if the
representations differ between people according to their reports or whether
the representations are the same, but only some reach threshold for a verbal
report of awareness.
Normal people can report whether they have created an image, but in
some cases, imagery can be pathological, as in the voices heard by paranoid
schizophrenics which are attributed to outside forces, or in the hallucinations
of drug states. Lesions of the frontal midline can result in attributing
control of one’s own hand to alien forces (G. Goldberg 1985). It seems very
likely that high-level attentional networks involving this frontal midline area
are a source of knowledge that the information has been internally generated.
These networks are normally involved in voluntary control of actions.
The pathological belief of alien control of one’s thoughts that is found in
some schizophrenics may arise from abnormalities in the regulation of these
networks. In the next section, we examine the operation of these frontal networks
as a means of aiding our understanding of the conscious control of
308 Psychiatry, Psychoanalysis, and the New Biology of Mind
FIGURE 6–20. Localization of alerting and emotional function.
(A) The localization of alerting and orienting functions in the parietal lobe. A diagram
of the lateral surface of the human brain indicating the relation of alerting and
orienting mechanisms in the right parietal lobe to the vigilance area in the right frontal
(B) The localization of cortical regions involved in cognitive and emotional states. A
diagram of the medial part of the human brain indicating the dorsal anterior cingulate,
which appears to be involved in the monitoring and/or regulation of cognitive
activity, and the more ventral portions of the cingulate, which appear to be related to
the regulation of emotion.
Source. (A) Courtesy of M. Posner. (B) Adapted from Bush et al. 1998.
Neural Science 309
Executive Control Includes Volition as a Major Component
Areas of the frontal midline appear to be important
in voluntary control of cognition and emotion
Normal people have a strong subjective feeling of their intentions. They
have a clear sense that they have voluntary control of their own behavior.
These subjective feelings of intentions and voluntary control can be freely
verbalized. Indeed, asking people about their goals or intentions is probably
the single most predictive indicator of their behavior during problem solving
(Newell and Simon 1972). The importance of intention is also illustrated in
patients with frontal lesions (Duncan 1986) or in patients suffering from
with mental disorders (Frith and Dolan 1998), who show disruption in either
their voluntary control over behavior or their subjective feelings of control.
What are the neural mechanisms of voluntary control?
Norman and Shallice (1986) have argued that an executive attention system
is necessary for situations in which routine or automatic processes are
inadequate. These nonroutine, nonautomatic executive functions include
selection among conflicting inputs, resolution of conflict among responses,
and monitoring and correcting errors.
Priming produced by automatic activation without
attention can facilitate reaction time to the item
when presented consciously
The existence of executive control was made more concrete in the 1970s and
1980s when cognitive studies first succeeded in separating conscious control
of mental events from automatic activation of the same events (Posner
1978). This separation used priming of a target word by a prime word related
in meaning to the target. The method is simple. Subjects are given a task to
perform on a target word. For example, they may be asked to classify
whether the target word is a meaningful word or not, or to categorize it as
representing a living object or not. On some trials, prior to the presentation
of the target word, the subject is given a prime word, a word that is flashed
briefly on the screen without further comment. The prime may be related to
the target (e.g., prime “toy” and target “doll”) or unrelated (prime “toy” and
target “stop”). Although the prime provides no direct information on how to
respond to the target, nonetheless, related primes were found to speed up
and unrelated ones to slow down reaction time in comparison to a neutral
warning signal that merely tells the person that a target will be coming.
People do not have to attend to the prime or even be aware that it has
happened to get the priming effect. Priming still takes place—if the prime
word is followed immediately by a visual masking noise (a random visual in310
Psychiatry, Psychoanalysis, and the New Biology of Mind
put) so that subjects are unaware of the identity of the prime word. However,
the effects of priming were somewhat different from trials in which the
prime word had been carefully attended. Consider a condition when people
are given a series of targets that either involve trees or body parts. If ambiguous
prime words such as “palm” are followed by a mask so they cannot be
reported, they serve to improve the performance on subsequent targets related
to both meanings of the word (e.g., “tree” and “hand”). However, when
the prime “palm” is presented in the context of trees and unmasked, only the
meaning related to the category of the previous trials (e.g., “tree”) is primed
(Marcel 1983).
James Neely (1977) studied the conscious use of primes by his subjects.
In one condition a word from one category (e.g., “animal”) was presented as
a prime word, and subjects were instructed to associate the category “animal”
with the category “building.” Target words in the category “building”
(e.g., “window”) had faster reaction times than targets in a category unrelated
category (e.g., “tin”). The subject had voluntarily activated the instructed
category. If a specific animal target (e.g., “dog”) was presented after
a very short interval so that subjects did not have enough time to switch
from the prime category “animal” to the instructed category “building,” fast
reaction times were made to the word “dog.” However, if “dog” was delayed
until after subjects had a chance to execute the switch to the instructed category,
the target “dog” would have a slow reaction time since subjects were
now attending to the wrong category. Within a second, Neeley was able to
trace the conscious effort involved in switching categories, by its influence
on the reaction time to probes.
There is more than one form of priming
These findings give a reality to the difference between the voluntary, conscious
control of mental events and the same event when driven unconsciously
and automatically by input. Priming can be produced in both ways.
First, priming can be produced by automatic activation of a pathway without
attention, facilitating reaction times for related items. This form of priming
can be totally subconscious. Imaging studies have shown that automatic
priming of this sort is produced by a reduction of blood flow within the brain
area related to processing the target. For semantic priming, this reduction
would be within areas of the brain related to the meaning of the word (Demb
et al. 1995). It is as if the prime had tuned the neuron pool so that attention
of fewer neurons is required to process the target. Possibly, as a consequence
of this reduced overall activity, a primed target, although classified rapidly,
is often poorly remembered in a later recall or recognition test.
A second form of priming is produced by directing attention to semantic
Neural Science 311
information. This form of attention appears not to depend on the brain area
related to the processing of the word but depends upon different frontal networks,
and this information is available to the conscious awareness of the
person. Within a second, subjects can voluntarily choose an associated category,
and the consequence of that selection is faster processing of related
targets and retarded processing of unrelated targets. When a category is attended,
items within the category are facilitated in reaction time, but items
in other categories are retarded over what they would have been if no priming
had taken place. Imaging studies have suggested that attention to a computation
increases blood flow within the attended area. Thus, priming may
be produced by different brain mechanisms that have quite different consequences
for performance. It remains for future studies to tell us how these
two priming mechanisms produce what seem like opposite changes at the
neural level.
The dorsal anterior cingulate cortex
is essential for executive control
Suppose you are asked how many objects you see between the brackets:
[two, two, two]. You may at first want to say “two,” even though the correct
answer is “three.” This is because there is a conflict in your mind between
the meaning of the word as read and the specified task of saying how many
words are present. This is one form of the Stroop effect. The most frequent
form of Stroop effect occurs when a subject has to name the color of ink in
which a word is written when the color of the word conflicts with the word
name (e.g., when the word “red” is written in blue ink). These conflicting
tasks involve focal attention to the critical element of the task when that element
must be selected in competition with a more dominant element. Imaging
studies of the Stroop effect produced by conflict between elements
tend to find very strong activity in the dorsal anterior cingulate gyrus (see
Figure 6–20B) often in concert with areas of the basal ganglia and lateral
frontal cortex (Bush et al. 1998). For this reason, dorsal anterior cingulate
gyrus has been thought to be involved in some aspect of focal or executive
attention (Carter et al. 1999).
As is the case with humans, rhesus monkeys trained to associate digits
with a quantity show conflict between deciding which of two displays has
the greatest number of objects when there is an incompatible relation between
the two (e.g., when the larger number of objects is made up of the
smaller digit). The monkeys made many more errors on incompatible trials
than do humans, despite many hundreds of trials at the task (Washburn
1994). It is as though the monkeys have somewhat less capacity for avoiding
interference, despite very extensive training.
312 Psychiatry, Psychoanalysis, and the New Biology of Mind
In humans, activity in the anterior cingulate gyrus generally is related to
the degree of practice or automation of the task. Perhaps the best example is
a task where subjects are required to ascribe a use for each noun in a list
(e.g., “hammer”→“pound”). There is a conflict between saying the word
name aloud and the required task of generating the use of the word. There
was strong activation of anterior cingulate when the list was first presented,
but with practice on a single list, activity in the cingulate disappears and instead
there is activity in the anterior insula, a portion of cortex that lies buried
beneath Broca’s area (Raichle et al. 1994). Both the anterior insula and
Broca’s area are closely related to the automatic task of reading the word
aloud. Imaging studies have identified two different pathways for producing
the use of a word. One pathway is involved when conscious thought is
needed to generate a word. This pathway involves the cingulate in conjunction
with left lateral areas of the cortex and the right cerebellum. Another,
more automatic, pathway is involved when the words are well practiced so
that the feeling of conscious search disappears. Now the activity in the cingulate
(as well as in the lateral cortex and cerebellum) disappears, and instead
one finds activity in Broca’s area and in the anterior insula, the
structures that are usually involved in the automatic tasks of reading words
out loud.
The ventral and dorsal anterior cingulate are concerned
with emotion and cognition, respectively
We often think of sensory orienting and memory retrieval as related to focal
attention. However, another source of information that frequently engages
our attention is emotion. When emotional words are presented in the same
conflict tasks described above, a more ventral area of the anterior cingulate
becomes active (see Figure 6–20B). In some neuroimaging experiments, the
cognitive and emotional areas of the cingulate seem to be mutually inhibitory
(Drevets and Raichle 1998). Thus, when strong emotions are involved
in the task, the dorsal area is less active than at rest and cognitive conflict
tasks tend to reduce activity in the more ventral area of the cingulate.
If the dorsal area of the cingulate is involved in selecting dimensions of
a stimulus when there is conflict among competing dimensions, a reasonable
idea might be that the dorsal area serves a similar selecting function for emotional
conflict. Indeed, we have already discussed the idea that orienting of
attention in infancy serves as one means by which caregivers seek to distract
their infants from the expression of distress. The control of distress is an important
concern of early childhood, and caregivers have the task of first regulating
emotions in their infant and later teaching the child to regulate its
own emotions. Perhaps, areas of the brain that regulate emotion in infancy
Neural Science 313
have acquired the ability to perform the same functions in response to cognitive
challenges. If this idea is correct, children who are well advanced in
emotional regulation should be at a specific advantage in regulating cognitive
We know that children differ in their ability to regulate their emotions.
This can be elicited from caregivers when they are asked specific questions
about the child’s ability to control distress, orient attention and be sensitive
to pleasures. The dimension of individual variation in regulation has been
called effortful control. Studies of 6- to 7-year-olds have found that effortful
control can be defined in terms of scales measuring attentional focusing, inhibitory
control, low intensity pleasure, and perceptual sensitivity (Rothbart
et al. 2000). Effortful control is consistently negatively related to a negative
affect in keeping with the notion that attentional skill may help attenuate
negative affect. Effortful control also is correlated with the performance of
2-to 4-year-old children in Stroop tasks that require them to handle conflict
(Posner and Rothbart 1998). Effortful control is related both to empathy and
to the acquisition of conscience, of a sense of moral behavior. Kochanska
(1995) has found that individual differences in effortful control have important
implications both for the inhibition of antisocial behavior and for the
acquisition of prosocial behavior. Children who can effectively employ attention
to regulate behavior are better able to inhibit prepotent responses
(e.g., striking out, stealing) and are better at taking into consideration the
effect of their actions on others.
Empathy and a sense of moral behavior or conscience are at the heart of
child socialization. The link between the attentional network of the frontal
lobe and conscience might make it possible to at least imagine how aspects
of morality might be studied on the neuronal level discussed below in relation
to disorders.
Disorders that recruit the cingulate cortex suggest
a connection between cognition and emotion
Attention deficit disorder is defined by a set of cognitive and emotional
symptoms. This disorder is usually diagnosed in children but often remains
present into adulthood. Neuroimaging of adults who suffer from attention
deficit disorder has been carried out under circumstances that require them
to do a numerical version of the Stroop effect. Here they are asked to respond
to the number of items present. When that number is sometimes in conflict
with the quantity indicated by the word (e.g., three copies of the word
“two”), adults with attention deficit disorder performed on conflict trials
only slightly less efficiently than normals. But unlike the normal controls,
adults with attention deficits show no activation of the anterior cingulate.
314 Psychiatry, Psychoanalysis, and the New Biology of Mind
Instead, they show greater activity on incompatible trials in the anterior insula
(Bush et al. 1999). As was suggested in the study of word association
discussed above, the insula represents a more automatic pathway than the
anterior cingulate, thus allowing for less effortful control over the task.
Another disorder that produces a disruption of voluntary control as well
as other emotional and cognitive problems is schizophrenia. Benes (1998)
has reported subtle abnormalities of the anterior cingulate in postmortem
analyses of schizophrenic brains. She argues that the problem with the anterior
cingulate, in the brains of schizophrenics, may be a shift in dopamine
regulation from pyramidal to nonpyramidal cells. She has also argued that
these changes in the cingulate are related to circuitry involving the amygdala
and hippocampus. The schizophrenia studies provide a lead at the cellular
level of the possible disregulation of the anterior cingulate in a second abnormality
noted for its attentional deficits.
Both schizophrenia and attention deficit disorder have a genetic basis.
Studies of attention deficit disorder families have shown that some of them
possess a particular allele of the dopamine 4 receptor (LaHoste et al. 1996;
Smalley et al. 1998). These studies provide some potential cellular and genetic
links between attentional abnormalities found in various pathologies.
A Future for the Study of Consciousness
We have outlined that the problem of consciousness can be considered to
consist of two subproblems: awareness and volition. Studies of orienting and
of imagery are concerned with the first, and self-regulation with the second.
As future research penetrates the organization of attention networks in the
frontal cortex, the two functions could prove to be linked. Studies of complex
scenes show that presentation of a stimulus does not lead automatically
to awareness of even the most central aspects of the scene (Rensink et al.
1997). Even though subjects report that they are aware of a whole scene,
they only become aware of a change when their attention is drawn to a
change in the scene. We have suggested that posterior brain areas involved
in orienting to sensory stimuli may be closely related to ambient awareness
and that focal attention might be more associated with the anterior cingulate
and other frontal areas related to voluntary control. Perhaps only as we understand
the neural basis of the distinction between focal attention to limited
aspects of the external world and a more general ambient awareness of the
general scene will we be able to understand the parts of the brain related to
consciousness of sensory events (Iwasaki 1993).
The exact functions that the anterior cingulate plays in higher-level attention
are not yet clear. We have learned that even very simple acts of attention
such as orienting to sensory stimuli involve a network of brain areas
Neural Science 315
that carry out specific functions. During conflict tasks, activation of the anterior
cingulate is usually accompanied by activity in lateral frontal cortical
areas and in the basal ganglia. It is an important goal to find out what each
of these areas contributes.
The functions of attention also relate to issues other than those usually
discussed under the term “consciousness”. Recently, the contribution of biology
to understanding the acquisition of high-level skills such as those
learned in schools has been extensively debated (e.g., Bruer 1999). In some
areas, such as the neural networks involved in reading and arithmetic,
progress has been extensive (Dehaene 1997; Posner et al. 1999). Both skills
required attention networks related to those discussed in this section. A better
understanding of these mechanisms might help in realizing the goal of a
neuroscience-based approach to aspects of education.
What is the future for neural science in the next millennium? We have seen
remarkable and rapid progress in understanding neuronal and synaptic signaling.
These advances now invite a structural approach to visualize the
static and dynamic structures of ion channels, receptors, and the molecular
machinery for signal transduction postsynaptically and for vesicle transport,
fusion, and exocytosis presynaptically. We also have made some progress in
the analysis of the elementary synaptic mechanisms that contribute to memory
storage. These studies have revealed that the different memory systems
of brain seem to use similar synaptic mechanisms for the storage of both declarative
and nondeclarative knowledge. Similarly, we now have achieved an
understanding, at least in broad outline, of the development of the nervous
system. Specific inducers, morphogens, attractants, and repellents of process
outgrowth and synapse organizing molecules have now been defined, providing
a molecular reality to concepts that previously were shrouded in mystery.
Progress in these several areas has in turn made possible a molecularbased
neurology, a neurology that will, one hopes, finally be able to address
the degenerative diseases of the brain that have for so long eluded our best
scientific efforts. In time, advances in these areas may also yield insight into
and perhaps solutions for some of the most debilitating diseases confronting
medical science—the psychiatric and neurological illnesses of schizophrenia,
depression, and Alzheimer’s disease. Implicit in this prediction is the expectation
that in the future molecular biology will be able to contribute to
the system problems of cognitive neural science much as it has recently contributed
to signaling, plasticity, and development.
The advances in the cellular understanding of the organization of the so316
Psychiatry, Psychoanalysis, and the New Biology of Mind
matosensory and visual system by Mountcastle, Hubel and Wiesel, and their
followers have helped turn our interest to perception and in the broader
sense to cognitive psychology. In turn, contact between cognitive psychology
and neural science has given us a new approach to the classical problems
of mental function, including attention and consciousness. In both early
sensory processes and higher cognitive perceptual and motor systems, we
find evidence for the localization of components within a broadly distributed
network carrying out complex functions. Indeed, one of the early insights
into consciousness is that it shares the properties of other cognitive
systems in that like vision and action, it can be dissected into components:
attention, imagery, and volition. Each component consists of a set of subcomponents
that can be localized within a larger, distributed neural system.
Having pointed to that similarity, we must nevertheless acknowledge that of
all fields in neural science, in fact of all the fields in all of science, the problems
of perception, action, memory, attention, and consciousness provide us
with the greatest evidence for our lack of understanding as well as the greatest
Even if one agreed that the scientific agenda outlined here may be adequate
to handle the issues of awareness and volition, there is another aspect
of consciousness that needs to be confronted and that is the nature of subjectivity.
The subjective aspect of consciousness is seen by philosophers of
mind such as Searle (1993, 1998) and Nagel (1993) as its defining characteristic
and the aspect that poses the greatest scientific challenge. Searle and
Nagel argue that each of us experiences a world of private and unique experiences
and that these seem much more real to us than the experiences of
others. We experience our own ideas, moods, and sensations—our successes
and disappointments, joys and pains—directly, whereas we can only appreciate
other people’s ideas, moods, and sensations. Are the purple you see and
the jasmine you smell identical to the purple that we see and the jasmine that
we smell? The fact that conscious experience is uniquely personal and intensely
subjective raises the question whether it is ever possible to determine
objectively some common characteristics of experience. We cannot, the argument
goes, use those same senses to arrive at an objective understanding
of experience.
Clearly, we should be prepared for the possibility that there are aspects
of consciousness that will not be solved by the approaches discussed in this
review. Some might believe that all that is scientific about the study of life is
illuminated at all levels from the molecular to the behavioral by what we
know about DNA. But others might believe that there are issues about what
it means to be a living being that are really not explained by the most detailed
account of DNA. Many issues of awareness and voluntary control are
likely to be explained at all levels, from genes to behavior. This might conNeural
Science 317
stitute a theory of consciousness in much the same way as DNA serves as the
basis for any scientific analysis of what constitutes life. Nonetheless, it is at
present hard to imagine how the progress discussed above, even if it continues
and intensifies, will solve all the issues of the subjective nature of our
experience. We leave it to the readers of Cell and Neuron in the next millennium
to determine how much insight about human consciousness will result
from the type of work we have discussed here.
We thank G. Gasic, E. Marcus, and L. Pond for helpful comments on the
manuscript, Harriet Ayers, Millie Pellan, and Kathy MacArthur for assistance
in preparing the text, and Ira Schieren, Sarah Mack, and Charles Lam
for help in preparation of the figures. Many of the illustrations used in this
article derive from Principles of Neural Science, with permission from
McGraw-Hill. T.D.A., E.R.K., and T.M.J. are Investigators of the Howard
Hughes Medical Institute.
Abel T, Nguyen PV, Barad M, et al: Genetic demonstration of a role for PKA in the
late phase of LTP and in hippocampus-based long-term memory. Cell 88:615–
626, 1997
Adrian ED: The analysis of the nervous system. Sherrington Memorial Lecture. Proc
R Soc Med 50:991–998, 1957
Albright TD: Form-cue invariant motion processing in primate visual cortex. Science
255:1141–1143, 1992
Albright TD: Cortical processing of visual motion. Rev Oculomot Res 5:177–201,
Albright TD, Stoner GR: Visual motion perception. Proc Natl Acad Sci USA 92:2433–
2440, 1995
Albright TD, Desimone R, Gross CG: Columnar organization of directionally selective
cells in visual area MT of the macaque. J Neurophysiol 51:16–31, 1984
Andersen RA, Essick GK, Siegel RM: Encoding of spatial location by posterior parietal
neurons. Science 230:456–458, 1985
Andersen RA, Snyder LH, Li CS, et al: Coordinate transformations in the representation
of spatial information. Curr Opin Neurobiol 3:171–176, 1993
Antonini A, Stryker M: Development of individual geniculocortical arbors in cat striate
cortex and effects of binocular impulse blockade. J Neurosci 13:3549–3573,
Antonini A, Stryker M: Rapid remodeling of axonal arbors in the visual cortex. Science
260:1819–1821, 1993b
318 Psychiatry, Psychoanalysis, and the New Biology of Mind
Armstrong CM, Hille B: Voltage gated ion channels and electrical excitability. Neuron
20:371–380, 1998
Attardi DG, Sperry RW: Preferential selection of central pathways by regenerating optic
fibers. Exp Neurol 7:46–64, 1963
Axel R: The molecular logic of smell. Sci Am 273:154–159, 1995
Bang AG, Goulding MD: Regulation of vertebrate neural cell fate by transcription factors.
Curr Opin Neurobiol 6:25–32, 1996
Barlow HB: Summation and inhibition in the frog’s retina. J Physiol 119:69–88, 1953
Barlow HB: Single units and sensation: a neuron doctrine for perceptual psychology?
Perception 1:371–394, 1972
Barthels D, Santoni MJ, Wille W, et al: Isolation and nucleotide sequence of mouse
NCAM cDNA that codes for a Mr, 79,000 polypeptide without a membranespanning
region. EMBO J 6:907–914, 1987
Bartsch D, Ghirardi M, Skehel PA, et al: Aplysia CREB-2 represses long-term facilitation:
relief of repression converts transient facilitation into long-term functional
and structural change. Cell 83:979–992, 1995
Bartsch D, Casadio A, Karl KA, et al: CREB-1 encodes a nuclear activator, a repressor,
and a cytoplasmic modulator that form a regulatory unit critical for long-term
facilitation. Cell 95:211–223, 1998
Bate CM: Pioneer neurones in an insect embryo. Nature 260:54–56, 1976
Benes FM: Model generation and testing to probe neural circuitry in the cingulate
cortex of postmortem schizophrenic brains. Schizophr Bulletin 24:219–230,
Bennett MVL: A comparison of electrically and chemically mediated transmission, in
Structure and Function of Synapses. Edited by Pappas GD and Purpura DP. New
York, Raven, 1972, pp. 221–256
Bennett MVL: Seeing is relieving: electrical synapses between visualized neurons. Nat
Neurosci 3:7–9, 2000
Bisiach E, Luzzatti C: Unilateral neglect of representational space. Cortex 14:129–
133, 1978
Bliss TVP, Lømo T: Long-lasting potentiation of synaptic transmission in the dentate
area of the anaesthetized rabbit following stimulation of the perforant path.
J Physiol 232:331–356, 1973
Bloom F, Lazerson A: Brain, Mind, and Behavior, 2nd Edition. New York, WH Freeman,
Bock JB, Scheller RH: SNARE proteins mediate lipid bilayer fusion. Proc Natl Acad
Sci USA 96:12227–12229, 1999
Boistel J, Fatt P: Membrane permeability change during inhibitory transmitter action
in crustacean muscle. J Physiol 4:176–191, 1958
Bourtchouladze RA, Franguelli B, Bendy J, et al: Deficient long-term memory in mice
with a targeted mutation of the cAMP responsive element binding protein. Cell
79:59–68, 1994
Boyd IA, Martin AR: The end-plate potential in mammalian muscle. J Physiol
132:74–91, 1956
Neural Science 319
Brackenbury R, Rutishauser U, Edelman GM: Distinct calcium-independent and calcium-
dependent adhesion systems of chicken embryo cells. Proc Natl Acad Sci
USA 78:387–391, 1981
Briscoe J, Ericson J: The specification of neuronal identity by graded Sonic Hedgehog
signalling. Semin Cell Dev Biol 10:353–362, 1999
Broadbent DE: Perception and Communication. London, Pergamon, 1958
Brock LG, Coombs JS, Eccles JC: The recording of potentials from motoneurones
with an intracellular electrode. J Physiol 117:431–460, 1952
Brockes JP: Amphibian limb regeneration: rebuilding a complex structure. Science
276:81–87, 1997
Brown RH Jr: Ion channel mutations in periodic paralysis and related myotonic diseases.
Ann NY Acad Sci 707:305–316, 1993
Bruer JT: The Myth of the First Three Years: A New Understanding of Early Brain Development
and Lifelong Learning. New York, Free Press, 1999
Brummendorf T, Rathjen FG: Structure/function relationships of axon-associated adhesion
receptors of the immunoglobulin superfamily. Curr Opin Neurobiol
6:584–593, 1996
Buck L, Axel R: A novel multigene family may encode odorant receptors: a molecular
basis for odor recognition. Cell 65:175–187, 1991
Buonomano DV, Merzenich MM: Cortical plasticity: from synapses to maps. Annu
Rev Neurosci 21:149–186, 1998
Burns BD: The Mammalian Cerebral Cortex (Monographs of the Physiological Society,
No 5). London, Edward Arnold [Publishers] Ltd., 1958
Bush G, Whalen PJ, Rosen BR, et al: The counting Stroop: an interference task specialized
for functional neuroimaging—validation study with functional MRI.
Hum Brain Map 6:270–282, 1998
Bush G, Frazier JA, Rauch SL, et al: Anterior cingulate cortex dysfunction in attention-
deficit/hyperactivity disorder revealed by fMRI and the counting Stroop.
Biol Psychiatry 45:1542–1552, 1999
Byrne JH, Kandel ER: Presynaptic facilitation revisited: state and time dependence.
J Neurosci 16:425–435, 1996
Cabelli RJ, Hohn A, Shatz CJ: Inhibition of ocular dominance column formation by
infusion of NT-4/5 or BDNF. Science 267:1662–1666, 1995
Cabelli RJ, Shelton DL, Segal RA, et al: Blockade of endogenous ligands of trkB inhibits
formation of ocular dominance columns. Neuron 19:63–76, 1997
Callaway EM: Local circuits in primary visual cortex of the macaque monkey. Annu
Rev Neurosci 21:47–74, 1998
Carew TJ, Hawkins RD, Kandel ER: Differential classical conditioning of a defensive
withdrawal reflex in Aplysia californica. Science 219:397–400, 1983
Carter CS, Botvinick MM, Cohen JD: The contribution of the anterior cingulate to
executive processes in cognition. Rev Neurosci 10:49–57, 1999
Casadio A, Martin KC, Giustetto M, et al: A transient neuron-wide form of CREBmediated
long-term facilitation can be stabilized at specific synapses by local
protein synthesis. Cell 99:221–237, 1999
320 Psychiatry, Psychoanalysis, and the New Biology of Mind
Castellucci V, Pinsker H, Kupfermann I, et al: Neuronal mechanisms of habituation
and dishabituation of the gill-withdrawal reflex in Aplysia. Science 167:1745–
1748, 1970
Catterall WA: Structure and function of voltage-sensitive ion channels. Science
242:50–61, 1988
Catterall WA: From ionic currents to molecular mechanisms: the structure and function
of voltage-gated sodium channels. Neuron 26:13–25, 2000
Chan YM, Jan YN: Conservation of neurogenic genes and mechanisms. Curr Opin
Neurobiol 9:582–588, 1999
Changeux J-P, Galzi JL, Devillers-Thiery A, et al: The functional architecture of the
acetylcholine nicotinic receptor explored by affinity labeling and site directed
mutagenesis. Q Rev Biophys 25:395–432, 1992
Chen G-Q, Cui C, Mayer ML, et al: Functional characterization of a potassium-selective
prokaryotic glutamate receptor. Nature 402:817–821, 1999
Chino YM, Kaas JH, Smith ELD 3rd, et al: Rapid reorganization of cortical maps in
adult cats following restricted deafferentation in retina. Vision Res 32:789–796,
Coen L, Osta R, Maury M, et al: Construction of hybrid proteins that migrate retrogradely
and transynaptically into the central nervous system. Proc Natl Acad Sci
USA 94:9400–9405, 1997
Colby CL, Goldberg ME: Space and attention in parietal cortex. Annu Rev Neurosci
22:319–349, 1999
Cole KS, Curtis HJ: Electrical impedance measurements in the squid giant axon during
activity. J Gen Physiol 22:649–670, 1939
Collingridge GL, Bliss TV: Memories of NMDA receptors and LTP. Trends Neurosci
18:54–56, 1995
Colquhoun P, Sakmann B: From muscle endplate to brain synapses: a short history
of synapses and agonist-activated ion channels. Neuron 20:381–387, 1998
Conturo TE, Nicolas FL, Cull TS, et al: Tracking fiber pathways in the living human
brain. Proc Natl Acad Sci USA 96:10422–10427, 1999
Cooper LA, Shepard RN: Chronometric studies of the rotation of visual images, in
Visual Information Processing. Edited by Chase WG. New York, Academic
Press, 1973, pp 75–176
Corbetta M: Frontoparietal cortical networks for directing attention and the eye to
visual locations: identical, independent, or overlapping neural systems? Proc
Natl Acad Sci USA 95:831–838, 1998
Corbetta M, Miezin FM, Dobmeyer S, et al: Selective and divided attention during visual
discriminations of shape, color, and speed: functional anatomy by positron
emission tomography. J Neurosci 11:2383–2402, 1991
Cowan WM: The emergence of modern neuroanatomy and developmental neurobiology.
Neuron 20:413–426, 1998
Cowan WM, Kandel ER: A brief history of synapses and synaptic transmission, in
The Synapse. Baltimore, MD, Johns Hopkins University Press, 2000
Cowan WM, Fawcett JW, O’Leary DD, et al: Regressive events in neurogenesis. Science
225:1258–1265, 1984
Neural Science 321
Cowan WM, Jessell TM, Zipursky SL: Molecular and Cellular Approaches to Neural
Development. New York, Oxford University Press, 1997
Cowan WM, Harter DH, Kandel ER: The emergence of modern neuroscience: some
implications for neurology and psychiatry. Annu Rev Neurosci 23:343–391,
Crair MC, Gillespie DC, Stryker MP: The role of visual experience in the development
of columns in cat visual cortex. Science 279:566–570, 1998
Crick F, Koch C: Consciousness and neuroscience. Cereb Cortex 2:97–107, 1998
Crist RE, Kapadia MK, Westheimer G, et al: Perceptual learning of spatial localization:
specificity for orientation, position, and context. J Neurophysiol 78:2889–
2894, 1997
Crowley JC, Katz LC: Development of ocular dominance columns in the absence of
retinal input. Nat Neurosci 2:1125–1130, 1999
Culotti JG, Merz DC: DCC and netrins. Curr Opin Cell Biol 10:609–613, 1998
Dehaene S: The Number Sense. New York, Oxford University Press, 1997
del Castillo J, Katz B: Quantal components of the endplate potential. J Physiol
124:560–573, 1954
Demb JB, Desmond JE, Wagner AD, et al: Semantic encoding and retrieval in the left
inferior prefrontal cortex: a functional MRI study of task difficulty and process
specificity. J Neurosci 15:5870–5878, 1995
Desimone R, Duncan J: Neural mechanisms of selective visual attention. Annu Rev
Neurosci 18:193–222, 1995
Detwiler SR: Neuroembryology: An Experimental Study. New York, Macmillan, 1936
Doetsch F, Caille I, Lim DA, et al: Subventricular zone astrocytes are neural stem cells
in the adult mammalian brain. Cell 97:703–716, 1999
Doyle DA, Morais Cabral J, Pfuetzner RA, et al: The structure of the potassium channel:
molecular basis of K+ conduction and selectivity. Science 280:69–77, 1998
Drescher U, Bonhoeffer F, Muller BK: The Eph family in retinal axon guidance. Curr
Opin Neurobiol 7:175–180, 1997
Drevets WC, Raichle ME: Reciprocal suppression of regional blood flow during emotional
versus higher cognitive processes: implications for interactions between
emotion and cognition. Cogn Emotion 12:353–385, 1998
Dubnau J, Tully T: Gene discovery in Drosophila: new insights for learning and memory.
Annu Rev Neurosci 21:407–444, 1998
Dudel J, Kuffler SW: Presynaptic inhibition at the crayfish neuromuscular junction.
J Physiol 155:543–562, 1961
Duncan J: Disorganization of behavior after frontal lobe damage. Cognitive Neuropsychology
3:271–290, 1986
Dymecki SM: Flp recombinase promotes site-specific DNA recombination in embryonic
stem cells and transgenic mice. Proc Natl Acad Sci USA 93:6191–6196,
Easter SS Jr, Purves D, Rakic P, et al: The changing view of neural specificity. Science
230:507–511, 1985
Eccles JC: The Neurophysiological Basis of Mind: The Principles of Neurophysiology.
Oxford: Clarendon Press, 1953
322 Psychiatry, Psychoanalysis, and the New Biology of Mind
Eccles JC: The Physiology of Synapses. New York, Academic Press, 1964
Edelman GM: Cell adhesion molecules. Science 219:450–457, 1983
Engel AK, Fries P, Konig P, et al: Temporal binding, binocular rivalry, and consciousness.
Conscious Cogn 8:128–151, 1999
Farah MJ: Visual Agnosia. Cambridge, MA, MIT Press, 1995
Fatt P, Katz B: An analysis of the end-plate potential recorded with an intra-cellular
electrode. J Physiol 115:320–370, 1951
Fatt P, Katz B: Spontaneous subthreshold activity at motor nerve endings. J Physiol
117:109–126, 1952
Fechner G: Elements of Psychophysics, Vol 1 (1860). Translated by Adler HE. New
York: Holt, Rinehart and Winston, 1966
Felleman DJ, Van Essen DC: Distributed hierarchical processing in the primate cerebral
cortex. Cereb Cortex 1:1–47, 1991
Fernandez-Chacon R, Südhof TC: Genetics of synaptic vesicle function: toward the
complete functional anatomy of an organelle. Annu Rev Physiol 61:753–776,
Fischbach GD, Frank E, Jessell TM, et al: Accumulation of acetylcholine receptors
and acetylcholinesterase at newly formed nerve-muscle synapses. Pharmacol
Rev 30:411–428, 1978
Forbes A: The interpretation of spinal reflexes in terms of present knowledge of nerve
conduction. Physiol Rev 2:361–414, 1922
Frank E, Wenner P: Environmental specification of neuronal connectivity. Neuron
10:779–785, 1993
Frey U, Morris RG: Synaptic tagging and long-term potentiation. Nature 385:533–
536, 1997
Frith C, Dolan RJ: Images of psychopathology. Curr Opin Neurobiol 8:259–262,
Fujita I, Tanaka K, Ito M, et al: Columns for visual features of objects in monkey inferotemporal
cortex. Nature 360:343–346, 1992
Fuortes MGF, Frank K, Becker MC: Steps in the production of motoneuron spikes.
J Gen Physiol 40:735–752, 1957
Furshpan EJ, Potter DD: Mechanism of nerve impulse transmission at a crayfish synapse.
Nature 180:342–343, 1957
Galton F: Inquiry Into Human Faculty and Its Development. London, JM Dent and
Sons, 1907
Galvani L: Commentary on the Effect of Electricity on Muscular Motion (1791).
Translated by Green RM. Cambridge, MA, Licht, 1953
Georgopoulos AP, Lurioto JT, Petrides M, et al: Mental rotation of the neuronal population
vector. Science 243:234–236, 1989
Gilbert CD, Wiesel TN: Receptive field dynamics in adult primary visual cortex. Nature
356:150–152, 1992
Goldberg G: Supplementary motor area structure and function: review and hypothesis.
Behav Brain Sci 8:567–615, 1985
Goldberg JL, Barres AB: Nogo in nerve regeneration. Nature 403:369–370, 2000
Neural Science 323
Goldberg ME, Wurtz RH: Activity of superior colliculus in behaving monkey, II: effect
of attention on neuronal responses. J Neurophysiol 35:560–574, 1972
Goodhill GJ: Stimulating issues in cortical map development. Trends Neurosci
20:375–376, 1997
Goodman CS, Shatz CJ: Developmental mechanisms that generate precise patterns of
neuronal connectivity. Cell 72:77–98, 1993
Goridis C, Brunet JF: Transcriptional control of neurotransmitter phenotype. Curr
Opin Neurobiol 9:47–53, 1999
Gray CM, Konig P, Engel AK, et al: Synchronization of oscillatory responses in visual
cortex: a plausible mechanism for scene segmentation, in Proceedings of the
Conference on Synergetics of the Brain, 1989
Graziano MSA, Yap GS, Gross CG: Coding of visual space by premotor neurons. Science
266:1054–1057, 1994
Green T, Heinemann SI, Gusella JF: Molecular neurobiology and genetics: investigation
of neural function and dysfunction. Neuron 20:427–444, 1998
Grieshammer U, Lewandoski M, Prevette D, et al: Muscle-specific cell ablation conditional
upon Cre-mediated DNA recombination in transgenic mice leads to
massive spinal and cranial motoneuron loss. Dev Biol 197:234–247, 1998
Gross CG: Visual functions of inferotemporal cortex, in Handbook of Sensory Physiology,
Vol 7. Edited by Autrum H, Jung R, Loewenstein WR, et al. Berlin, Springer-
Verlag, 1973, pp 451–482
Gross CG, Bender DB, Rocha-Miranda CE: Visual receptive fields of neurons in inferotemporal
cortex of the monkey. Science 166:1303–1306, 1969
Gulbis JM, Mann S, MacKinnon R: Structure of a voltage-dependent K+ channel
β subunit. Cell 97:943–952, 1999
Gurdon JB, Dyson S, St. Johnston D: Cells’ perception of position in a concentration
gradient. Cell 95:159–162, 1998
Guthrie S: Axon guidance: starting and stopping with slit. Curr Biol 9:R432–R435,
Hamburger V: Cell death in the development of the lateral motor column of the chick
embryo. J Comp Neurol 160:535–546, 1975
Hamburger V: The history of the discovery of the nerve growth factor. J Neurobiol
24:893–897, 1993
Hamburger V, Levi-Montalcini R: Proliferation differentiation and degeneration in
the spinal ganglia of the chick embryo under normal and experimental conditions.
J Exp Zool 111:457–501, 1949
Han EB, Stevens CI: Of mice and memory. Learn Mem 6:539–541, 1999
Hanson PJ, Roth R, Morisaki H, et al: Structure and conformational changes in NSF
and its membrane receptor complexes visualized by quick-freeze/deep-etch
electron microscopy. Cell 90:523–535, 1997
Harland R, Gerhart J: Formation and function of Spemann’s organizer. Annu Rev Cell
Dev Biol 13:611–667, 1997
Hartline HK: The response of single optic nerve fibers of the vertebrate eye to illumination
of the retina. Am J Physiol 121:400–415, 1938
324 Psychiatry, Psychoanalysis, and the New Biology of Mind
Hartline HK, Graham CH: Nerve impulses from single receptors in the eye. J Cell
Comp Physiol 1:277–295, 1932
Hata Y, Stryker M: Control of thalamocortical afferent rearrangement by postsynaptic
activity in developing visual cortex. Science 265:1732–1735, 1994
Hebb DO: The Organization of Behavior: A Neuropsychological Theory. New York,
Wiley, 1949
Hengartner MO, Horvitz HR: Programmed cell death in Caenorhabditis elegans. Curr
Opin Genet Dev 4:581–586, 1994
Hertting G, Axelrod J: Fate of tritiated noradrenaline at sympathetic nerve endings.
Nature 192:172–173, 1961
Heuser JE: Synaptic vesicle exocytosis revealed in quick-frozen frog neuromuscular
junctions treated with 4-amino-pyridine and given a single electric shock, in
Approaches to the Cell Biology of Neurons. Edited by Cowan WM, Ferrendelli
JA. Washington, DC, Society for Neuroscience, 1977, pp 215–239
Hille B, Armstrong CM, MacKinnon R: Ion channels: from ideas to reality. Nat Med
5:iii–vii, 1999
Hodgkin AL: Evidence for electrical transmission in nerve, parts I and II. J Physiol
90:183–232, 1937
Hodgkin AL, Huxley AF: Action potentials recorded from inside a nerve fiber. Nature
144:710–711, 1939
Hodgkin AL, Katz B: The effect of sodium ions on the electrical activity of the giant
axon of the squid. J Physiol 108:37–77, 1949
Hodgkin AL, Huxley AF, Katz B: Measurement of current voltage relations in the
membrane of the giant axon of Loligo. J Physiol 116:424–448, 1952
Hofer M, Constantine-Paton M: Regulation of N-methyl-D-aspartate (NMDA) receptor
function during the rearrangement of developing neuronal connections.
Prog Brain Res 102:277–285, 1994
Hoffman EP, Kunkel LM: Dystrophin abnormalities in Duchenne/Becker muscular
dystrophy. Neuron 2:1019–1029, 1989
Hoffman EP, Brown RH Jr, Kunkel LM: Dystrophin: the protein product of the Duchenne
muscular dystrophy locus. Cell 51:919–928, 1987
Hokfelt T: Neuropeptides in perspective: the last ten years. Neuron 7:867–879, 1991
Holmes G: Disorders of sensation produced by cortical lesions. Brain 50:413–427,
Hubel DH, Wiesel TN: Receptive fields of single neurones in the cat’s striate cortex.
J Physiol 148:574–591, 1959
Hubel DH, Wiesel TN: Receptive fields, binocular interaction and functional architecture
in the cat’s visual cortex. J Physiol 160:106–154, 1962
Hubel DH, Wiesel TN: Receptive fields of cells in striate cortex of very young, visually
inexperienced kittens. J Neurophysiol 26:994–1002, 1963
Hubel DH, Wiesel TN: Binocular interaction in striate cortex of kittens reared with
artificial squint. J Neurophysiol 28:1041–1059, 1965
Hubel DH, Wiesel TN: Receptive fields and functional architecture of monkey straite
cortex. J Physiol 195:215–243, 1968
Neural Science 325
Hubel DH, Wiesel TN: Ferrier lecture: functional architecture of macaque monkey
visual cortex. Proc R Soc Lond B Biol Sci 198:1–59, 1977
Hubel DH, Wiesel TN: Early exploration of the visual cortex. Neuron 20, 401–412,
Hubel DH, Wiesel TN, LeVay S: Plasticity of ocular dominance columns in the monkey
striate cortex. Philos Trans R Soc Lond Biol 278:377–409, 1977
Hunt RK, Cowan WM: The chemoaffinity hypothesis: an appreciation of Roger W.
Sperry’s contributions to developmental biology, in Brain Circuits and Functions
of the Mind. Edited by Trevarthen C. New York, Cambridge University Press,
1990, pp 19–74
Huntington’s Disease Collaborative Research Group: A novel gene containing a trinucleotide
repeat that is expanded and unstable on Huntington’s disease chromosomes.
Cell 72:971–983, 1993
Hynes RO: Integrins: a family of cell surface receptors. Cell 48:549–554, 1987
Impey S, Mark M, Villacres EC, et al: Induction of CRE-mediated gene expression by
stimuli that generate long-lasting LTP in area CA1 of the hippocampus. Neuron
16:973–982, 1996
Impey S, Smith DM, Obrietan K, et al: Stimulation of cAMP response element (CRE)-
mediated transcription during contextual learning. Nat Neurosci 1:595–601,
Impey S, Obrietan K, Storm DR: Making new connections: role of ERK/MAP kinase
signaling in neuronal plasticity. Neuron 23:11–14, 1999
Isom LL, De Jongh KS, Catterall WA: Auxiliary subunits of voltage-gated ion channels.
Neuron 12:1183–1194, 1994
Iversen LL: The Uptake and Storage of Noradrenaline in Sympathetic Nerves. London,
Cambridge University Press, 1967
Iwasaki S: Spatial attention and two modes of visual consciousness. Cognition
49:211–233, 1993
Jacobson MD, Weil M, Raff MC: Programmed cell death in animal development. Cell
88:347–354, 1997
James W: Principles of Psychology. New York, Henry Holt, 1890
Johansson CB, Momma S, Clarke DL, et al: Identification of a neural stem cell in the
adult mammalian central nervous system. Cell 96:25–34, 1999
Johns DC, Marx R, Mains RE, et al: Inducible genetic suppression of neuronal excitability.
J Neurosci 19:1691–1697, 1999
Jones CM, Smith JC: Mesoderm induction assays. Methods Mol Biol 97:341–350,
Kandel ER, Spencer WA: Cellular neurophysiological approaches in the study of
learning. Physiol Rev 48:65–134, 1968
Kandel ER, Schwartz JH, Jessell T: Principles of Neural Science, 4th Edition. New
York, McGraw-Hill, 2000
Karlin A, Akabas MH: Toward a structural basis for the function of nicotinic acetylcholine
receptors and their cousins. Neuron 15:1231–1244, 1995
Karni A, Bertini G: Learning perceptual skills: behavioral probes into adult cortical
plasticity. Curr Opin Neurobiol 7:530–535, 1997
326 Psychiatry, Psychoanalysis, and the New Biology of Mind
Karni A, Sagi D: The time course of learning a visual skill. Nature 365:250–252, 1993
Kastner S, Pinsk MA, De Weerd P, et al: Increased activity in human visual cortex during
directed attention in the absence of visual stimulation. Neuron 22:751–761,
Katz B: The release of neural transmitter substances, in The Sherrington Lectures, 10.
Springfield, IL, Charles C. Thomas, 1969
Katz LC, Dalva MB: Scanning laser photostimulation: a new approach for analyzing
brain circuits. J Neurosci Methods 54:205–218, 1994
Katz LC, Shatz CJ: Synaptic activity and the construction of cortical circuits. Science
274:1133–1138, 1996
Kleiman RJ, Reichardt LF: Testing the agrin hypothesis. Cell 85:461–464, 1996
Kochanska G: Children’s temperament, mothers’ discipline, and security of attachment:
multiple pathways to emerging internalization. Child Dev 66:597–615,
Koffka K: Principles of Gestalt Psychology. New York, Harcourt Brace, 1935
Köhler W: Gestalt Psychology. London, Bell and Sons, 1929
Konorski J: Integrative Activity of the Brain. Chicago, IL, University of Chicago
Press, 1967
Kosslyn SM: Image and Mind. Cambridge, MA, Harvard University Press, 1980
Kosslyn SM: Image and Brain. Cambridge, MA, MIT Press, 1994
Kremer EJ, Pritchard M, Lynch M, et al: Mapping of DNA instability at the fragile X
to a trinucleotide repeat sequence P(CCG)n. Science 252:1711–1714, 1991
Kuffler SW: Discharge patterns and functional organization of mammalian retina.
J Neurophysiol 16:37–68, 1953
LaHoste GJ, Swanson JM, Wigal SB, et al: Dopamine D4 receptor gene polymorphism
is associated with attention deficit hyperactivity disorder. Mol Psychiatry 1:121–
124, 1996
Lamme VA, Spekreijse H: Neuronal synchrony does not represent texture segregation.
Nature 396:362–366, 1998
Lance-Jones C, Landmesser L: Pathway selection by chick lumbosacral motoneurons
during normal development. Proc R Soc Lond B Biol Sci 214:1–18, 1981
Langley JN: On the regeneration of pre-ganglionic and post-ganglionic visceral nerve
fibers. J Physiol 22:215–230, 1897
Langley JN: On nerve endings and on special excitable substances in cells. Proc R Soc
Lond B Biol Sci 78:170–194, 1906
Lee K, Dietrich P, Jessell TM: Genetic ablation reveals the essential role of the roof
plate in dorsal interneuron specification. Nature 403:734–740, 2000
Leonardo da Vinci: Treatise on Painting (Codes Urbinus Latinus 1270). Princeton,
NJ, Princeton University Press, 1956
Lettvin JY, Maturana HR, McCulloch WS, et al: What the frog’s eye tells the frog’s
brain. Proc Instit Radio Engineers 47:1940–1951, 1959
LeVay S, Wiesel TN, Hubel DH: The development of ocular dominance columns in
normal and visually deprived monkeys. J Comp Neurol 191:1–51, 1980
Levi-Montalcini R: The nerve growth factor: its mode of action on sensory and sympathetic
nerve cells. Harvey Lect 60:217–259, 1966
Neural Science 327
Lewis EB: Regulation of the genes of the bithorax complex in Drosophila. Cold
Spring Harb Symp Quant Biol 50:155–164, 1985
Liley AW: The quantal components of the mammalian endplate potential. J Physiol
133:571–587, 1956
Lorente de Nó R: Analysis of the activity of the chains of internuncial neurons. J Neurophysiol
1:207–244, 1938
Lumsden A, Krumlauf R: Patterning the vertebrate neuraxis. Science 274:1109–
1115, 1996
Mach E: Contributions to the Analysis of Sensations (1886). Translated by Southall
JPC. Optical Society of America, 1924
Malach R, Amir Y, Harel M, et al: Relationship between intrinsic connections and
functional architecture revealed by optical imaging and in vivo targeted biocytin
injections in primate striate cortex. Proc Natl Acad Sci USA 90:10469–10473,
Marcel AJ: Conscious and unconscious perception: experiments on visual masking
and word recognition. Cognit Psychol 15:197–237, 1983
Marr D: Vision: A Computational Investigation Into the Human Representation and
Processing of Visual Information. San Francisco, CA, WH Freeman, 1982
Marrocco RT, Davidson MC: Neurochemistry of attention, in The Attentive Brain.
Edited by Parasuraman R. Cambridge, MA, MIT Press, 1998
Martin KC, Casadio A, Zhu H, et al: Synapse-specific transcription-dependent longterm
facilitation of the sensory to motor neuron connection in Aplysia: a function
for local protein synthesis in memory storage. Cell 91:927–938, 1998
Mayford M, Kandel ER: Genetic approaches to memory storage. Trends Genet
15:463–470, 1999
Mayford M, Bach ME, Huang Y-Y, et al: Control of memory formation through regulated
expression of a CaMKII transgene. Science 274:1678–1683, 1996
McGinnis W, Levine MS, Hafen E, et al: A conserved DNA sequence in homoeotic
genes of the Drosophila Antennapedia and bithorax complexes. Nature 308:
428–433, 1984
McMahan UJ: The agrin hypothesis. Cold Spring Harb Symp Quant Biol 55:407–418,
Mesulam MM: Spatial attention and neglect: parietal, frontal and cingulate contributions
to the mental representation and attentional targeting of salient extrapersonal
events. Philos Trans R Soc Lond B Biol Sci 354:1325–1346, 1999
Metzstein MM, Stanfield GM, Horvitz HR: Genetics of programmed cell death in
C. elegans: past, present and future. Trends Genet 14:410–416, 1998
Miller KD: A model for the development of simple cell receptive fields and the ordered
arrangement of orientation columns through activity-dependent competition
between ON- and OFF-center inputs. J Neurosci 14:409–441, 1994
Milner AD, Goodale MA: The Visual Brain in Action. New York, Oxford University
Press, 1995
Milner B: Memory disturbance after bilateral hippocampal lesions, in Cognitive Processes
and the Brain. Edited by Milner PM, Glickman SE. Princeton, NJ, Van
Nostrand, 1965
328 Psychiatry, Psychoanalysis, and the New Biology of Mind
Milner B, Squire LR, Kandel ER: Cognitive neuroscience and the study of memory.
Neuron 20:445–468, 1998
Missler M, Südhof TC: Neurexins: three genes and 1,001 products. Trends Genet
4:20–26, 1998
Mombaerts P, Wang F, Dulac C, et al: Visualizing an olfactory sensory map. Cell
4:675–686, 1996
Montarolo PG, Goelet P, Castellucci VF, et al: A critical period for macromolecular
synthesis in long-term heterosynaptic facilitation in Aplysia. Science 234:1249–
1254, 1986
Morrison SJ, White PM, Zock C, et al: Prospective identification, isolation by flow
cytometry, and in vivo self-renewal of multipotent mammalian neural crest stem
cells. Cell 96:737–749, 1999
Mountcastle VB: Modality and topographic properties of single neurons of cat’s somatic
sensory cortex. J Neurophysiol 20:408–434, 1957
Mountcastle VB, Talbot WH, Sakata H, et al: Cortical neuronal mechanisms in fluttervibration
studied in unanesthetized monkeys. Neuronal periodicity and frequency
discrimination. J Neurophysiol 32:452–484, 1969
Mueller BK: Growth cone guidance: first steps towards a deeper understanding.
Annu Rev Neurosci 22:351–388, 1999
Munk H: Über die Funktionen der Grosshirnrinde, 3te Mitteilung. Berlin, A Hirschwald,
1881, pp 28–53
Nagel T: What is the mind-brain problem? in Experimental and Theoretical Studies
of Consciousness (CIBA Foundation Symposium, 174). New York, Wiley Interscience/
CIBA Foundation, 1993, pp 1–13
Neher E, Sakmann B: Single channel membrane of denervated frog muscle fibers. Nature
260:799–802, 1976
Neher E, Sakmann B: Single channel membrane of denervated frog muscle fibers. Nature
260:799–802, 1976
Neely JH: Semantic priming and retrieval from lexical memory: roles of inhibitionless
spreading activation and limited-capacity attention. J Exp Psychol 106:226–254
Newell A, Simon HA: Human Problem Solving. Engelwood Cliffs, NJ, Prentice-Hall,
Newsome WT, Paré EB: A selective impairment of motion perception following lesions
of the middle temporal visual area (MT). J Neurosci 8:2201–2211, 1988
Newsome WT, Britten KH, Movshon JA: Neuronal correlates of a perceptual decision.
Nature 341:52–54, 1989
Nickel W, Weber T, McNew JA, et al: Content mixing and membrane integrity during
membrane fusion driven by pairing of isolated v-SNAREs and t-SNAREs. Proc
Natl Acad Sci USA 96:12571–12576, 1999
Nieuwkoop PD: Short historical survey of pattern formation in the endo-mesoderm
and the neural anlage in the vertebrates: the role of vertical and planar inductive
actions. Cell Mol Life Sci 53:305–318, 1997
Nó D, Yao TP, Evans RM: Ecdysone-inducible gene expression in mammalian cells
and transgenic mice. Proc Natl Acad Sci USA 93:3346–3351, 1996
Neural Science 329
Norman DA, Shallice T: Attention to action: willed and automatic control of behavior,
in Consciousness and Self-Regulation. Edited by Davidson RJ, Schwartz GE,
Shapiro D. New York, Plenum, 1986, pp 1–18
Numa S: A molecular view of neurotransmitter regions and ion channels. Harvey
Lect 63:121–165, 1989
Nüsslein-Volhard C, Wieschaus E: Mutations affecting segment number and polarity
in Drosophila. Nature 287:795–801, 1980
Olson CR, Gettner SN: Object-centered direction selectivity in the macaque supplementary
eye field. Science 269:985–988, 1995
Oppenheim RW: Neuronal cell death and some related regressive phenomena during
neurogenesis: a selective historical review and progress report, in Studies in Developmental
Neurobiology: Essays in Honor of Viktor Hamburger. Edited by
Cowan WM. New York, Oxford University Press, 1981
Panchision D, Hazel T, McKay R: Plasticity and stem cells in the vertebrate nervous
system. Curr Opin Cell Biol 10:727–733, 1998
Parasuraman R, Greenwood PM: Selective attention in aging and dementia, in The
Attentive Brain. Edited by Parasuraman R. Cambridge, MA, MIT Press, 1998
Parlati F, Weber T, McNew JA, et al: Rapid and efficient fusion of phospholipid vesicles
by the α-helical core of a SNARE complex in the absence of an N-terminal
regulatory domain. Proc Natl Acad Sci USA 96:12565–12570, 1999
Parsons LM, Fox PT: The neural basis of implicit movements used in recognizing
hand shape. Cognitive Neuropsychology 15:583–615, 1998
Paulson HL, Fischbeck KH: Trinucleotide repeats in neurogenetic disorders. Annu
Rev Neurosci 19:79–107, 1996
Perutz MF, Johnson T, Suzuki M, et al: Glutamine repeats as polar zippers: their possible
role in inherited neurodegenerative diseases. Proc Natl Acad Sci USA
91:5355–5358, 1994
Pettmann B, Henderson CE: Neuronal cell death. Neuron 20:633–647, 1998
Picciotto MR: Knock-out mouse models used to study neurobiological systems. Crit
Rev Neurobiol 13:103–149, 1999
Posner MI: Chronometric Explorations of Mind. Hillsdale, NJ, Erlbaum, 1978
Posner MI, Gilbert CD: Attention and primary visual cortex. Proc Natl Acad Sci USA
96:2585–2587, 1999
Posner MI, Raichle ME: Images of Mind. New York, Scientific American Books, 1994
Posner MI, Raichle ME: The neuroimaging of human brain function. Proc Natl Acad
Sci USA 95:763–764, 1998
Posner MI, Rothbart MK: Attention, self-regulation and consciousness. Philos Trans
R Soc Lond B Biol Sci 353:1915–1927, 1998
Posner MI, Abdullaev YG, McCandliss BD, et al: Neuroanatomy, circuitry and plasticity
of word reading. Neuroreport 10:R12–R23, 1999
Ptácek LJ: Channelopathies: ion channel disorders of muscle as a paradigm for paroxysmal
disorders of the nervous system. Neuromuscul Disord 7:250–255, 1997
Ptácek LJ: The familial periodic paralyses and nondystrophic myotonias. Am J Med
104:58–70, 1998
330 Psychiatry, Psychoanalysis, and the New Biology of Mind
Purves D, Lichtman JW: Principles of Neural Development. Sunderland, MA,
Sinauer, 1985
Raichle ME: Behind the scenes of functional brain imaging: a historical and physiological
perspective. Proc Natl Acad Sci USA 95:765–772, 1998
Raichle ME, Fiez JA, Videen TO, et al: Practice-related changes in the human brain:
functional anatomy during non-motor learning. Cereb Cortex 4:8–26, 1994
Ramón y Cajal S: The Croonian Lecture: la fine structure des centres nerveux. Proc
R Soc Biol Lond B Sci 55:444–467, 1894
Ramón y Cajal S: The structure and connexions of neurons (1906), in Nobel Lectures:
Physiology or Medicine (1901–1921). Amsterdam, Elsevier, 1967, pp
Ramón y Cajal S: Histologie du système nerveux de l’homme et des vertébrés, Vols 1
and 2 (1911). A Maloine, Paris. Reprinted by Consejo Superior de Investigaciones
Cientificas, Inst. Ramón y Cajal, Madrid, 1955
Reddy PS, Housman DE: The complex pathology of trinucleotide repeats. Curr Opin
Cell Biol 9:364–372, 1997
Reichardt LF, Fariñas I: Neurotrophic factors and their receptors, in Molecular and
Cellular Approaches to Neural Development. Edited by Cowan WM, Jessell TM,
Zipursky SL. New York, Oxford University Press, 1997, pp 220–263
Rensink RA, O’Regan JK, Clark JJ: To see or not to see: the need for attention to perceive
changes in scenes. Psychol Sci 8:368–373, 1997
Richardson PM, McGuinness UM, Aguayo AJ: Axons from CNS neurons regenerate
into PNS grafts. Nature 284:264–265, 1997
Rosen BR, Buckner RL, Dale AM: Event related functional MRI: past, present, and future.
Proc Natl Acad Sci USA 95:773–780, 1998
Ross CA: Intranuclear neuronal inclusions: a common pathogenic mechanism for
glutamine-repeat neurogenerative diseases? Neuron 19:1147–1150, 1997
Rothbart MK, Ahadi SA, Evans DW: Temperament and personality: origins and outcomes.
J Per Soc Psychol 78:122–135, 2000
Rudenko G, Nguyen T, Chelliah Y, et al: The structure of the ligand-binding domain
of neurexin Iβ: regulation of LNS domain function by alternative splicing. Cell
99:93–101, 1999
Rumelhart DE, McClelland JL, Group PR: Parallel Distributed Processing. Cambridge,
MA, MIT Press, 1987
Ruoslahti E: RGD and other recognition sequences for integrins. Annu Rev Cell Dev
Biol 12:697–715, 1996
Rutishauser U, Hoffman S, Edelman GM: Binding properties of a cell adhesion molecule
from neural tissue. Proc Natl Acad Sci USA 79:685–689, 1982
Ryalls J, Lecours AR: Broca’s first two cases: from bumps on the head to cortical convolutions,
in Classic Cases in Neuropsychology. Edited by Code C, Wallesch
C W, Joanelle Y, et al. New York, Psychology Press, 1996
Salzman CD, Britten KH, Newsome WT: Microstimulation of visual area MT influences
perceived direction of motion. Invest Ophthalmol Vis Sci 31:238, 1990
Neural Science 331
Sandrock AW Jr, Dryer SE, Rosen KM, et al: Maintenance of acetylcholine receptor
number by neuregulins at the neuromuscular junction in vivo. Science 276:
599–603, 1997
Sanes JR, Lichtman JW: Development of the vertebrate neuromuscular junction.
Annu Rev Neurosci 22:389–442, 1999
Schafer EA: Experiments on special sense localisations in the cortex cerebri of the
monkey. Brain 10:362–380, 1888
Schiller F: Paul Broca: Explorer of the Brain. New York, Oxford University Press,
Schwartz EL: Computational anatomy and functional architecture of striate cortex:
spatial mapping approach to perceptual coding. Vision Res 20:645–669, 1980
Schwenk F, Kuhn R, Angrand PO, et al: Temporally and spatially regulated somatic
mutagenesis in mice. Nucleic Acids Res 26:1427–1432, 1998
Scott MP, Weiner AJ: Structural relationships among genes that control development:
sequence homology between the Antennapedia, Ultrabithorax, and fushi tarazu
loci of Drosophila. Proc Natl Acad Sci USA 81:4115–4119, 1984
Searle JR: The problem of consciousness, in Experimental and Theoretical Studies of
Consciousness (CIBA Foundation Symposium, 174). New York, Wiley Interscience/
CIBA Foundation, 1993, pp 61–80
Searle JR: How to study consciousness scientifically, in Towards an Understanding of
Integrative Brain Function. Edited by Fuxe K, Grillner S, Hokfelt T, et al. Amsterdam,
Elsevier, 1998, pp 379–387
Serafini T: Finding a partner in a crowd: neuronal diversity and synaptogenesis. Cell
98:133–136, 1999
Shadlen MN, Movshon JA: Synchrony unbound: a critical evaluation of the temporal
binding hypothesis. Neuron 24:67–77, 1999
Shapiro L, Colman DR: The diversity of cadherins and implications for a synaptic adhesive
code in the CNS. Neuron 23:427–430, 1999
Shatz CJ: Neurotrophins and visual system plasticity, in Molecular and Cellular Approaches
to Neural Development. Edited by Cowan WM, Jessell TM, Zipursky
SL. New York, Oxford University Press, 1997
Sheng M, Pak DT: Glutamate receptor anchoring proteins and the molecular organization
of excitatory synapses. Ann NY Acad Sci 868:483–493, 1999
Shepherd GM: Foundations of the Neuron Doctrine. New York, Oxford University
Press, 1991
Sherrington CS: The Central Nervous System, Vol 3, in A Textbook of Physiology, 7th
Edition. Edited by Foster M. London, Macmillan, 1897
Sherrington CS: The Integrative Action of the Nervous System, 2nd Edition. New Haven,
NJ, Yale University Press, 1906
Sherrington CS: Flexor-reflex of the limb, crossed extension reflex, and reflex stepping
and standing (cat and dog). J Physiol 40:28–116, 1910
Sherrington CS: Inhibition as a Coordinative Factor. Nobel Lecture. Stockholm, PA
Norstedt, 1932
Sherrington CS: Man on His Nature. New York, Macmillan, 1941
332 Psychiatry, Psychoanalysis, and the New Biology of Mind
Silva AJ, Kogan JH, Frankland PW, et al: CREB and memory. Annu Rev Neurosci
21:127–148, 1998
Singer W: Consciousness and the structure of neuronal representations. Philos Trans
R Soc Lond B Biol Sci 353:1829–1840, 1998
Smalley SL, Bailey JN, Palmer CG, et al: Evidence that the D4 receptor is a susceptibility
gene in attention deficit hyperactivity disorder. Mol Psychiatry 3:427–430,
Smith JC: Induction and early amphibian development. Curr Opin Cell Biol 1:1061–
1070, 1989
Söllner T, Whiteheart SW, Brunner M, et al: SNAP receptors implicated in vesicle targeting
and fusion. Nature 362:318–324, 1993
Song JY, Ichtchenko K, Südhof TC, et al: Neuroligin 1 is a postsynaptic cell-adhesion
molecule of excitatory synapses. Proc Natl Acad Sci USA 96:1100–1105, 1999
Speidel CC: Studies of living nerves. II. Activities of ameboid growth cones, sheath
cells, and myelin segments, as revealed by prolonged observation of individual
nerve fibers in frog tadpoles. Am J Anat 52:1–75, 1933
Spemann H, Mangold H: Ü Induktion von Embryonalanlagen durch Implantation
artfremder Organisatoren. Arch Mikrosk Anat Entwicklungsmech 100:599–
638, 1924
Sperry RW: Effect of 180° rotation of the retinal field on visuomotor coordination. J
Exp Zool 92:263–279, 1943
Sperry RW: Chemoaffinity in the orderly growth of nerve fiber patterns and connections.
Proc Natl Acad Sci USA 50:703–710, 1963
Squire LR, Kandel ER: Memory: From Mind to Molecules. New York, Scientific
American Books, 1999
Squire LR, Zola-Morgan S: The medial temporal lobe memory system. Science
253:1380–1386, 1991
Stevens CF: Making a submicroscopic hole in one. Nature 349:657–658, 1991
Stevens CF: Neuronal diversity: too many cell types for comfort? Curr Biol 8:R708–
R710, 1998
St. Johnston D, Nüsslein-Volhard C: The origin of pattern and polarity in the Drosophila
embryo. Cell 68:201–219, 1992
Stoner GR, Albright TD: Neural correlates of perceptual motion coherence. Nature
358:412–414, 1992
Stoner GR, Albright TD, Ramachandran VS: Transparency and coherence in human
motion perception. Nature 344:153–155, 1990
Straub V, Campbell KP: Muscular dystrophies and the dystrophin-glycoprotein complex.
Curr Opin Neurol 10:168–175, 1997
Strittmatter WJ, Roses AD: Apolipoprotein E and Alzheimer’s disease. Annu Rev
Neurosci 19:53–77, 1996
Stryker MP, Harris W: Binocular impulse blockade prevents the formation of ocular
dominance columns in cat visual cortex. J Neurosci 6:2117–2133, 1986
Sutton RB, Fasshauer D, Jahn R, et al: Crystal structure of a SNARE complex involved
in synaptic exocytosis at 2.4 Å resolution. Nature 395:347–353, 1998
Neural Science 333
Swindale NV: A model for the formation of ocular dominance stripes. Proc R Soc
Lond B Biol Sci 208:243–264, 1980
Takahashi M, Miyoshi H, Verma IM, et al: Rescue from photoreceptor degeneration
in the rd mouse by human immunodeficiency virus vector-mediated gene transfer.
J Virol 73:7812–7816, 1999
Takeichi M: Cadherins: a molecular family important in selective cell-cell adhesion.
Annu Rev Biochem 59:237–252, 1990
Takeichi M, Uemura T, Iwai Y, et al: Cadherins in brain patterning and neural network
formation. Cold Spring Harb Symp Quant Biol 62:505–510, 1997
Takeuchi A, Takeuchi N: On the permeability of the end-plate membrane during the
action of transmitter. J Physiol (Paris) 154:52–67, 1960
Talbot SA, Marshall WH: Physiological studies on neural mechanisms of visual localization
and discrimination. Am J Ophthalmol 24:1255–1264, 1941
Tanabe Y, Jessell TM: Diversity and pattern in the developing spinal cord. Science
274:1115–1123, 1996
Tanabe Y, William C, Jessell TM: Specification of motor neuron identity by the MNR2
homeodomain protein. Cell 95:67–80, 1998
Tang YP, Shimizu E, Duber R, et al: Genetic enhancement of learning and memory in
mice. Nature 401:63–69, 1999
Tatagiba M, Brosamle C, Schwab ME: Regeneration of injured axons in the adult
mammalian central nervous system. Neurosurgery 40:541–547, 1997
Teller DY: First glances: the vision of infants. The Friedenwald Lecture. Invest Ophthalmol
Vis Sci 38:2183–2203, 1997
Tempel BL, Papazian PM, Schwartz TL, et al: Sequence of probable potassium channels
at Shaker locus of Drosophila. Science 237:770–775, 1987
Tessier-Lavigne M, Goodman CS: The molecular biology of axon guidance. Science
274:1123–1133, 1996
Thomas JB, Bastiani MJ, Bate M, et al: From grasshopper to Drosophila: a common
plan for neuronal development. Nature 310:203–207, 1984
Tootell RBH, Nouchine K, Hadjikhani WV, et al: Functional analysis of primaryvisual
cortex in humans. Proc Natl Acad Sci USA 95:811–817, 1998
Tsien JZ, Heurta PT, Tonegawa S: The essential role of hippocampal CA1 NMDA receptor-
dependent synaptic plasticity in spatial memory. Cell 87:1327–1338,
Ungerleider LG, Mishkin M: Two cortical visual systems, in Analysis of Visual Behavior.
Edited by Ingle DJ, Goodale MA, Mansfield RJW. Cambridge, MA, MIT
Press, 1982, pp 549–586
Usrey WM, Reid RC: Synchronous activity in the visual system. Annu Rev Physiol
61:435–456, 1999
Van Essen DC: Functional organization of primate visual cortex, in Cerebral Cortex,
Vol 3. Edited by Peters A, Jones EG. New York, Plenum, 1985, pp 259–327
Verkerk AJ, Pieretti M, Sutcliffe JS, et al: Identification of a gene (FMR-1) containing
a CGG repeat coincident with a breakpoint cluster region exhibiting length variations
in fragile X syndrome. Cell 65:905–914, 1991
334 Psychiatry, Psychoanalysis, and the New Biology of Mind
von Bekesy G: Experiments in Hearing. Edited and translated by Wever EG. New
York, McGraw-Hill, 1960
von der Malsburg C: The correlation theory of brain function. MPI Biophysical
Chemistry, Internal Report 81–2. MPI Biophysical Chemistry, 1981
von Helmholtz H: Treatise on Psychological Optics, Vol 2 (1860).Translated by
Southall JPC. [Rochester, NY], Optical Society of America, 1924
Wang F, Nemes A, Mendelsohn M, et al: Odorant receptors govern the formation of
a precise topographic map. Cell 93:47–60, 1998
Warren ST, Ashley CT Jr: Triplet repeat expansion mutations: the example of fragile
X syndrome. Annu Rev Neurosci 18:77–99, 1995
Washburn DA: Stroop-like effects for monkeys and humans: processing speed or
strength of association? Psychol Sci 5:375–379, 1994
Watanabe D, Inokawa H, Hashimoto K, et al: Ablation of cerebellar Golgi cells disrupts
synaptic integration involving GABA inhibition and NMDA receptor activation
in motor coordination. Cell 95:17–27, 1998
Watkins JC, Evans RH: Excitatory amino acid transmitters. Annu Rev Pharmacol
Toxicol 21:165–204, 1981
Weber RJ, Harnish R: Visual imagery for words: the Hebb test. J Exp Psychol
102:409–414, 1974
Weber T, Zemelman BV, McNew JA, et al: SNAREpins: minimal machinery for membrane
fusion. Cell 92:759–772, 1998
Weiss P: Self-differentiation of the basic patterns of coordination. Comparative Psychology
Monographs 17:1–96, 1941
Weliky M, Katz LC: Correlational structure of spontaneous neuronal activity in the
developing lateral geniculate nucleus in vivo. Science 285:599–604, 1999
Wertheimer M: Gestalt Theory (1924). New York, Humanities Press, 1950
Wessells NK: Tissue Interactions and Development. Menlo Park, CA, Benjamin/
Cummings, 1977
Wiesel TN, Hubel DH: Single-cell responses in striate cortex of kittens deprived of
vision in one eye. J Neurophysiol 26:1003–1017, 1963
Wiesel TN, Hubel DH: Comparison of the effects of unilateral and bilateral eye closure
on cortical unit responses in kittens. J Neurophysiol 28:1029–1040, 1965
Wolpert L: Positional information and the spatial pattern of cellular differentiation.
J Theor Biol 25:1–47, 1969
Wu Q, Maniatis T: A striking organization of a large family of human neural cadherin-
like cell adhesion genes. Cell 97:779–790, 1999
Wundt WM-G: Principles of Physiological Psychology. Translated from the 5th German
Edition. New York, Macmillan, 1902
Wurtz RH, Goldberg ME (eds): The Neurobiology of Saccadic Eye Movements (Reviews
of Oculomotor Research, Vol 3). Amsterdam, Elsevier, 1989
Yin JC, Tully T: CREB and the formation of long-term memory. Curr Opin Neurobiol
6:264–268, 1996
Yin JCP, Del Vecchio M, Zhou H, et al: CREB as a memory modulator: induced expression
of a dCREB-2 activator isoform enhances long-term memory in Drosophila.
Cell 81:107–115, 1995
Neural Science 335
Yoshihara Y, Mizuno T, Nakahira M, et al: A genetic approach to visualization of multisynaptic
neural pathways using plant lectin transgene. Neuron 22:33–41, 1999
Zamanillo D, Sprengel R, Hvalby O, et al: Importance of AMPA receptors for hippocampal
LTP but not for spatial learning. Science 284:1805–1811, 1999
Zinyk DL, Mercer EH, Harris E, et al: Fate mapping of the mouse midbrain-hindbrain
constriction using a site-specific recombination system. Curr Biol 8:665–668,
This page intentionally left blank
Charles F. Zorumski, M.D.
The ability of an organism to modify its behavior based on experience is arguably
the most important and fascinating property of the nervous system.
For serious students of behavior, it is a given that the neural changes underlying
learning and memory involve alterations in the strength of connections
between neurons. It is important to remember that this was not the case even
50 years ago. In the late 1950s, when Eric Kandel embarked on a scientific
journey that radically changed how we now think about complex behaviors,
there were several plausible and competing hypotheses under consideration,
none of which had particularly strong experimental support. Using what he
called a “radically reductionist approach” based on studying relatively simple
behaviors in the giant marine snail Aplysia californica, Kandel systematically
changed the field. By focusing on a simpler organism with a simple
nervous system, Kandel and colleagues were able to map in detail the neural
pathways underlying specific reflex behaviors in the snail to determine ways
in which these behaviors could be modified by experience in a laboratory
setting. In a series of dramatic and definitive electrophysiological studies,
Kandel demonstrated that the changes underlying several forms of learning
result from persistent changes in the function of specific synapses used to
338 Psychiatry, Psychoanalysis, and the New Biology of Mind
generate the behaviors of interest. Furthermore, learning in Aplysia, like
learning in more complex mammalian systems, has several components
with differing degrees of long-term stability. Over several decades of study,
the Kandel group dissected the physiological, biochemical, and molecular
details of learning, including the identification of specific changes in synapses,
ion channels, second messenger systems, protein synthesis, and gene
Of great importance, the spectacular insights gained from work in Aplysia
did not end with the invertebrate system. Rather, Kandel and colleagues
extended these insights into mouse models, where they coupled the power
of synaptic physiology and sophisticated behavioral analyses with the ability
to manipulate gene expression to examine specific molecules that contribute
to mammalian declarative memory. While the molecular and biophysical
mechanisms underlying synaptic plasticity in the mouse do not always exactly
mimic the events in Aplysia, the overall conservation of guiding principles
and similarity in processes, including some of the key molecular
components, is staggering and a bit humbling for those who see mammals,
particularly humans, as so much more complex than mollusks.
The following essay, “The Molecular Biology of Memory Storage: A Dialogue
Between Genes and Synapses,” was delivered on December 8, 2000, as
Kandel’s Nobel lecture, and highlights the paths that Kandel and his colleagues
pursued in developing what are now textbook insights into learning
and memory. These studies unequivocally set the stage for the next generation
of studies.
What have we learned from this work? As Kandel summarizes, there are
at least four major sets of insights. First, the mechanisms underlying memory
reside in changes in synaptic transmission and are conserved across species.
These changes can be both pre- and postsynaptic, and, importantly, the
same sets of synapses can be altered in different and completely opposite
ways by different types of learning. Second, learning affects neuronal excitability
as well as synaptic transmission. Thus, the firing properties of neurons
are also subject to modification and contribute to synaptic and network
changes. Third, synaptic modifications underlying learning can have different
temporal components, contributing specifically to short- and long-term
forms of memory; overlapping, but not necessarily identical, cellular processes
contribute to the different time courses. Fourth, changes underlying
memory not only affect synaptic function but also involve structural changes
in synaptic contacts. While short-term information storage involves covalent
modifications of already existing proteins, long-term storage requires
changes in gene expression and protein synthesis.
These four points are profound insights into the workings of nervous
systems across species. It is already clear that these principles have a much
The Molecular Biology of Memory Storage 339
broader impact than the conditioned reflexes in a mollusk or declarative
memory in a rodent. The insights we are now gaining into the properties of
emotional and motivational systems, including the actions of therapeutic
and abused psychoactive drugs, reflect many of the points outlined above.
While the specific molecular details may differ in different systems, the principles
developed by the Kandel group serve as a guiding light for future
In closing, there is one additional concept that this work teaches us. This
concerns the importance of working in simple but definable systems for
gaining knowledge of more complex processes. The brilliance of Kandel’s
foresight in choosing to work with Aplysia cannot be overemphasized.
While studies of the cellular basis of learning and memory remain a work in
progress, it is safe to say that we would not have the insights into mammalian
learning and the scientific directions that we have today had it not been
for Kandel’s determined decision to employ a “radically reductionist approach.”
This is a lesson that may trump all others as the field of psychiatry
anticipates its scientific future.
This page intentionally left blank
C H A P T E R 7
A Dialogue Between Genes and Synapses
Eric R. Kandel, M.D.
One of the most remarkable aspects of an animal’s behavior is the ability to
modify that behavior by learning, an ability that reaches its highest form in
human beings. For me, learning and memory have proven to be endlessly
fascinating mental processes because they address one of the fundamental
features of human activity: our ability to acquire new ideas from experience
and to retain these ideas over time in memory. Moreover, unlike other mental
processes such as thought, language, and consciousness, learning seemed
from the outset to be readily accessible to cellular and molecular analysis.
This article was originally published in Science, Volume 294, Number 5544, pp.
Howard Hughes Medical Institute, Center for Neurobiology and Behavior, College
of Physicians and Surgeons of Columbia University, New York State Psychiatric
Institute, 1051 Riverside Drive, New York, NY 10032, USA.
This essay is adapted from the author’s address to the Nobel Foundation, December
342 Psychiatry, Psychoanalysis, and the New Biology of Mind
I, therefore, have been curious to know, What changes in the brain when we
learn? And, once something is learned, how is that information retained in
the brain? I have tried to address these questions through a reductionist approach
that would allow me to investigate elementary forms of learning and
memory at a cellular molecular level—as specific molecular activities within
identified nerve cells.
I first became interested in the study of memory in 1950 as a result of my
readings in psychoanalysis while still an undergraduate at Harvard College.
Later, during medical training, I began to find the psychoanalytic approach
limiting because it tended to treat the brain, the organ that generates behavior,
as a black box. In the mid-1950s, while still in medical school, I began
to appreciate that during my lifetime the black box of the brain would be
opened and that the problems of memory storage, once the exclusive domain
of psychologists and psychoanalysts, could be investigated with the
methods of modern biology. As a result, my interest in memory shifted from
a psychoanalytic to a biological approach. As a postdoctoral fellow at the National
Institutes of Health (NIH) in Bethesda from 1957 to 1960, I focused
on learning more about the biology of the brain and became interested in
knowing how learning produces changes in the neural networks of the
My purpose in translating questions about the psychology of learning
into the empirical language of biology was not to replace the logic of psychology
or psychoanalysis with the logic of cellular molecular biology, but
to try to join these two disciplines and to contribute to a new synthesis that
would combine the mentalistic psychology of memory storage with the biology
of neuronal signaling. I hoped further that the biological analysis of
memory might carry with it an extra bonus, that the study of memory storage
might reveal new aspects of neuronal signaling. Indeed, this has proven
A Radical Reductionist Strategy to
Learning and Memory
At first thought, someone interested in learning and memory might be
tempted to tackle the problem in its most complex and interesting form.
This was the approach that Alden Spencer and I took when we joined forces
at NIH in 1958 to study the cellular properties of the hippocampus, the part
of the mammalian brain thought to be most directly involved in aspects of
complex memory (Kandel and Spencer 1961). We initially asked, rather naively:
Are the electrophysiological properties of the pyramidal cells of the
hippocampus, which were thought to be the key hippocampal cells involved
in memory storage, fundamentally different from other neurons in the
The Molecular Biology of Memory Storage 343
brain? With study, it became clear to us that all nerve cells, including the pyramidal
cells of the hippocampus, have similar signaling properties. Therefore,
the intrinsic signaling properties of neurons would themselves not give
us key insights into memory storage (Kandel and Spencer 1968). The
unique functions of the hippocampus had to arise not so much from the intrinsic
properties of pyramidal neurons but from the pattern of functional interconnections
of these cells, and how those interconnections are affected by
learning. To tackle that problem, we needed to know how sensory information
about a learning task reaches the hippocampus and how information
processed by the hippocampus influences behavioral output. This was a formidable
challenge, since the hippocampus has a large number of neurons
and an immense number of interconnections. It seemed unlikely that we
would be able to work out in any reasonable period of time how the neural
networks, in which the hippocampus was embedded, participate in behavior
and how those networks are affected by learning.
To bring the power of modern biology to bear on the study of learning,
it seemed necessary to take a very different approach—a radically reductionist
approach. We needed to study not the most complex but the simplest instances
of memory storage, and to study them in animals that were most
tractable experimentally. Such a reductionist approach was hardly new in
twentieth-century biology. One need only think of the use of Drosophila in
genetics, of bacteria and bacteriophages in molecular biology, and of the
squid giant axon in the study of the conduction of nerve impulses. Nevertheless,
when it came to the study of behavior, many investigators were reluctant
to use a reductionist strategy. In the 1950s and 1960s, many
biologists and most psychologists believed that learning was the one area of
biology in which the use of simple animal models, particularly invertebrate
ones, was least likely to succeed. They argued that only higher animals exhibit
interesting forms of learning and that these forms require neuronal organizations
and neuronal mechanisms qualitatively different from those
found in simple animals.
It was my belief, however, that concerns about the use of a simple experimental
system to study learning were misplaced. If elementary forms of
learning are common to all animals with an evolved nervous system, there
must be conserved features in the mechanisms of learning at the cell and
molecular level that can be studied effectively even in simple invertebrate
A Simple Learned Behavior in an Invertebrate
After an extensive search for a suitable experimental animal, I settled on the
giant marine snail Aplysia (Figure 7–1A) because it offers three important
344 Psychiatry, Psychoanalysis, and the New Biology of Mind
The Molecular Biology of Memory Storage 345
advantages: its nervous system is made up of a small number of nerve cells;
many of these are gigantic; and (as became evident to me later) many are
uniquely identifiable (Frazier et al. 1967; Kandel 1976). Whereas the mammalian
brain has a trillion central nerve cells, Aplysia has only 20,000, and
the simplest behaviors that can be modified by learning may directly involve
less than 100 central nerve cells. In addition to being few in number, these
cells are the largest nerve cells in the animal kingdom, reaching up to
1,000 μm in diameter, large enough to be seen with the naked eye. One can
record from these large cells for many hours without any difficulty, and the
same cell can be returned to and recorded from over a period of days. The
cells can easily be dissected out for biochemical studies, so that from a single
cell one can obtain sufficient mRNA to make a cDNA library. Finally, these
identified cells can readily be injected with labeled compounds, antibodies,
or genetic constructs, procedures that opened up the molecular study of signal
transduction within individual nerve cells.
Irving Kupfermann and I soon delineated a very simple defensive reflex:
the withdrawal of the gill upon stimulation of the siphon, an action that is
like the quick withdrawal of a hand from a hot object. When a weak tactile
stimulus is applied to the siphon, both the siphon and gill are withdrawn
into the mantle cavity for protection under the mantle shelf (Pinsker et al.
1970) (Figure 7–1A). Kupfermann, Harold Pinsker, and later Tom Carew,
Robert Hawkins, and I found that this simple reflex could be modified by
three different forms of learning: habituation, sensitization, and classical
conditioning (Carew et al. 1972; Pinsker et al. 1970, 1973). As we examined
FIGURE 7–1. A simple learned behavior (opposite page).
(A) A dorsal view of Aplysia showing the gill, the animal’s respiratory organ. A light
touch to the siphon with a fine probe causes the siphon to contract and the gill to
withdraw. Here, the mantle shelf is retracted for a better view of the gill. Sensitization
of the gill-withdrawal reflex, by applying a noxious stimulus to another part of the
body, such as the tail, enhances the withdrawal reflex of both the siphon and the gill.
(B) Spaced repetition converts short-term memory into long-term memory in Aplysia.
Before sensitization training, a weak touch to the siphon causes only a weak, brief
siphon- and gill- withdrawal reflex. Following a single noxious, sensitizing shock to
the tail, that same weak touch produces a much larger siphon and gill reflex withdrawal
response, an enhancement that lasts about 1 hour. More tail shocks increase
the size and duration of the response.
Source. Modified from Frost WN, Castellucci VF, Hawkins RD, et al: “Monosynaptic
Connections Made by the Sensory Neurons of the Gill- and Siphon-Withdrawal Reflex
in Aplysia Participate in the Storage of Long-Term Memory for Sensitization.”
Proceedings of the National Academy of Sciences of the United States of America
82:8266–8269, 1985.
346 Psychiatry, Psychoanalysis, and the New Biology of Mind
these three forms of learning, we were struck by the resemblance each had
to corresponding forms of learning in higher vertebrates and humans. As
with vertebrate learning, memory storage for each type of learning in Aplysia
has two phases: a transient memory that lasts minutes and an enduring
memory that lasts days. Conversion of short-term to long-term memory
storage requires spaced repetition—practice makes perfect, even in snails
(Figure 7–1B) (Carew et al. 1972; Frost et al. 1985; Pinsker et al. 1973).
We focused initially on one type of learning. Sensitization is a form of
learned fear in which a person or an experimental animal learns to respond
strongly to an otherwise neutral stimulus (Frost et al. 1985; Pinsker et al.
1970, 1973). For example, if a person is suddenly exposed to an aversive
stimulus, such as a gunshot going off nearby, that person will be sensitized
by the unexpected noise. As a result, that person will be frightened and will
now startle to an otherwise innocuous stimulus like a tap on the shoulder.
Similarly, on receiving an aversive shock to a part of the body such as the tail,
an Aplysia recognizes the stimulus as aversive and learns to enhance its defensive
reflex responses to a variety of subsequent stimuli applied to the siphon,
even innocuous stimuli (Castellucci et al. 1989) (Figure 7–1A). The
animal remembers the shock, and the duration of this memory is a function
of the number of repetitions of the noxious experience (Figure 7–1B). A single
shock gives rise to a memory lasting only minutes; this short-term memory
does not require the synthesis of new protein. In contrast, four or five
spaced shocks to the tail give rise to a memory lasting several days; this longterm
memory does require the synthesis of new protein. Further training,
four brief trainings a day for 4 days, gives rise to an even more enduring
memory lasting weeks, which also requires new protein synthesis. Thus, just
as in complex learning in mammals (Ebbinghaus 1885/1963; Flexner and
Flexner 1966), the long-term memory for sensitization differs from the
short-term memory in requiring the synthesis of new proteins. This was our
first clear evidence for the conservation of biochemical mechanisms between
Aplysia and vertebrates.
Kupfermann, Castellucci, Carew, Hawkins, John Byrne, and I worked
out significant components of the neural circuit gill-withdrawal reflex (Figure
7–2). The circuit is located in the abdominal ganglion and has 24 central
mechanoreceptor sensory neurons that innervate the siphon skin and make
direct monosynaptic connections with 6 gill motor cells (Byrne et al. 1974,
1978a; Castellucci et al. 1970) (Figure 7–2C). The sensory neurons also
made indirect connections with the motor cells through small groups of excitatory
and inhibitory interneurons (Hawkins et al. 1981a,b). In addition to
being identifiable, individual cells also have surprisingly large effects on behavior
(Figure 7–2B) (Byrne et al. 1978a,b; Kandel 1976). As we examined
the neural circuit of this reflex, we were struck by its invariance. In every anThe
Molecular Biology of Memory Storage 347
imal we examined, each cell connected only to certain target cells and not to
others (Figure 7–2C). This also was true in the neural circuitry of other behaviors
in Aplysia including inking, control of the circulation, and locomotion
(Kandel 1976, 1979). This raised a key question in the cell-biological
study of learning: How can learning occur in a neural circuit that is so precisely
In 1894, Santiago Ramón y Cajal (1894) proposed a theory of memory
storage according to which memory is stored in the growth of new connections.
This prescient idea was neglected in good part for half a century as students
of learning fought over newer competing ideas. First, Karl Lashley,
Wolfgang Köhler, and a number of Gestalt psychologists proposed that
learning leads to changes in electric fields or chemical gradients, which they
postulated surround neuronal populations and are produced by the aggregate
activity of cells recruited by the learning process. Second, Alexander
Forbes and Lorente de Nó proposed that memory is stored dynamically by a
self-reexciting chain of neurons. Donald Hebb later championed this idea as
a mechanism for short-term memory. Finally, Holger Hyden proposed that
learning led to changes in the base composition of DNA or RNA. Even
though there was much discussion about the merits of each of these ideas,
there was no direct evidence to support any of them (Kandel 1968).
Kupfermann, Castellucci, Carew, Hawkins, and I addressed these alternative
ideas directly by confronting the question of how learning can occur
in a circuit with fixed neuronal elements. To address this question, we examined
the neural circuit of the gill-withdrawal reflex while the animal underwent
sensitization, classical conditioning, or habituation. Our studies
provided clear evidence for the idea proposed by Ramón y Cajal, that learning
results from changes in the strength of the synaptic connections between
precisely interconnected cells (Castellucci et al. 1970; Kupfermann et al.
1970). Thus, while the organism’s developmental program assures that the
connections between cells are invariant, it does not specify their precise
strength. Rather, experience alters the strength and effectiveness of these
preexisting chemical connections. Seen in the perspective of these three
forms of learning, synaptic plasticity emerged as a fundamental mechanism
for information storage by the nervous system, a mechanism that is built
into the very molecular architecture of chemical synapses (Milner et al.
Molecular Biology of Short- and
Long-Term Memory Storage
What are the molecular mechanisms whereby short-term memory is established,
and how is it converted to long-term memory? Initially, we focused
348 Psychiatry, Psychoanalysis, and the New Biology of Mind
The Molecular Biology of Memory Storage 349
on short-term sensitization. In collaboration with James H. Schwartz, we
found that the synaptic changes, like short-term behavior, were expressed
even when protein synthesis was inhibited. This finding first suggested to us
that short-term synaptic plasticity might be mediated by a second-messenger
system such as cyclic AMP (cAMP) (Schwartz et al. 1971). Following up
on this idea, Schwartz, Howard Cedar, and I found in 1972 that stimulation
of the modulatory pathways recruited during heterosynaptic facilitation led
to an increase in cAMP in the abdominal ganglion (Cedar et al. 1972). Cedar
and Schwartz (1972) found that the neurotransmitter candidates serotonin
FIGURE 7–2. The neural circuit of the Aplysia gill-withdrawal reflex
(opposite page).
(A) In this dorsal view of the abdominal ganglion, the six identified motor cells to the
gill are brown and the seven sensory neurons are blue. A sensory neuron that synapses
on gill motor neuron L7 is stimulated electrically with an intracellular electrode
and a microelectrode in the motor neuron records the synaptic potential produced by
the action potential in the sensory neuron [see middle trace in (B)]. The sensory neuron
carries the input from the siphon skin; the motor neuron makes direct connections
onto the gill.
(B) Individual cells make significant contributions to the reflex. Stimulating a single
motor neuron (traces on the left) produces a detectable change in the gill and stimulating
a single sensory neuron produces a large synaptic potential in the motor neuron
(traces in the middle). Repeated stimulation of a single sensory neuron increases
the frequency of firing in the motor neuron, leading to a visible reflex contraction of
the gill (traces on the right). A single tactile stimulus to the skin normally activates
6–8 of the 24 sensory neurons, causing each to fire 1–2 action potentials. The repetitive
firing of 10 action potentials in a single sensory neuron, designed to simulate
the firing of the total population (trace on the right) simulates the reflex behavior reasonably
(C) Diagram of the circuit of the gill-withdrawal reflex. The siphon is innervated by
24 sensory neurons that connect directly with the 6 motor neurons. The sensory neurons
also connect to populations of excitatory and inhibitory interneurons that in
turn connect with the motor neurons. Stimulating the tail activates three classes of
modulatory interneurons (serotonergic neurons, neurons that release the small cardioactive
peptide, and the L29 cells) that act on the terminals of the sensory neurons
as well as on those of the excitatory interneurons. The serotonergic modulatory action
is the most important; blocking the action of these cells blocks the effects of sensitizing
Source. From Glanzman DL, Mackey SL, Hawkins RD, et al: “Depletion of Serotonin
in the Nervous System of Aplysia Reduces the Behavioral Enhancement of Gill
Withdrawal as Well as the Heterosynaptic Facilitation Produced by Tail Shock.” The
Journal of Neuroscience 9:4200–4213, 1989.
350 Psychiatry, Psychoanalysis, and the New Biology of Mind
and dopamine could simulate this action of electrical stimulation and increase
levels of cAMP. Later, Hawkins, Castellucci, David Glanzman, and I
delineated the modulatory system activated by a sensitizing stimulus to the
tail (Glanzman et al. 1989; Hawkins et al. 1981b; Mackey et al. 1989), and
confirmed that it contains serotonergic interneurons.
We next found that serotonin acts on specific receptors in the presynaptic
terminals of the sensory neuron to enhance transmitter release. In 1976,
Marcello Brunelli, Castellucci, and I injected cAMP directly into the presynaptic
cells and found that it too produced presynaptic facilitation (Brunelli
et al. 1976; Kandel et al. 1976). This provided the most compelling evidence
then available that cAMP is involved in controlling synaptic strength and
gave us our first insight into one molecular mechanism of short-term memory—
the regulation of transmitter release (Figure 7–3).
How does cAMP enhance transmitter release? Serotonin, or injected
cAMP, leads to increased excitability and a broadening of the action potential
by reducing specific K+ currents, allowing greater Ca2+ influx into the presynaptic
terminal with each action potential (Klein and Kandel 1980). The
greater Ca2+ influx could contribute to the enhanced transmitter release.
Following the lead of Paul Greengard, who had proposed that cAMP produces
its action in the brain through the cAMP-dependent protein kinase
(PKA), Marc Klein and I suggested that cAMP may cause phosphorylation
of this K+ channel by activating PKA (Klein and Kandel 1980). In collaborative
experiments with Paul Greengard in 1980, Castellucci, Schwartz, and
I found that the active catalytic subunit of PKA by itself produced broadening
of the action potential and enhancement of glutamate release (Castellucci
et al. 1980). Conversely, the specific peptide inhibitor of PKA (PKI)
blocked the actions of serotonin. These findings provided direct evidence for
the role of PKA in short-term presynaptic facilitation (Byrne and Kandel
1995; Castellucci et al. 1982).
In an elegant series of experiments, Steven Siegelbaum, Joseph Camardo,
and Michael Shuster identified a novel K+ channel, the S-type K+ channel,
and showed that it too could be modulated by cAMP (Siegelbaum et al.
1982) and that PKA could act on the S-type K+ channel directly (Shuster et
al. 1985). Later, Byrne showed that serotonin also modulates a delayedrectifier
K+ (Byrne and Kandel 1995). The S-type channel mediated the increase
in excitability with a minor contribution to broadening, whereas the
delayed-rectifier K+ channel contributed little to excitability but had a major
role in spike broadening. Finally, Hochner, Klein, and I—and independently,
Jack Byrne and his colleagues—showed that, in addition to spike broadening,
serotonin also enhanced release by an as-yet-unspecified action on the
release machinery. Thus, serotonin leads to an increase in presynaptic cAMP,
which activates PKA and leads to synaptic strengthening through enhanced
The Molecular Biology of Memory Storage 351
FIGURE 7–3. Effects of short- and long-term sensitization on the
monosynaptic component of the gill-withdrawal reflex of Aplysia.
In short-term sensitization (lasting minutes to hours), a single tail shock causes a
transient release of serotonin that leads to covalent modification of preexisting proteins.
The serotonin acts on a transmembrane serotonin receptor to activate the enzyme
adenylyl cyclase (AC), which converts ATP to the second-messenger cAMP. In
turn, cAMP recruits the cAMP-dependent protein kinase A (PKA) by binding to the
regulatory subunits (spindles), causing them to dissociate from and free the catalytic
subunits (ovals). These subunits can then phosphorylate substrates (channels and
exocytosis machinery) in the presynaptic terminals, leading to enhanced transmitter
availability and release. In long-term sensitization, repeated stimulation causes the
level of cAMP to rise and persist for several minutes. The catalytic subunits can then
translocate to the nucleus, and recruit the mitogen-activated protein kinase (MAPK).
In the nucleus, PKA and MAPK phosphorylate and activate the cAMP response element-
binding (CREB) protein and remove the repressive action of CREB-2, an inhibitor
of CREB-1. CREB-1 in turn activates several immediate-response genes,
including a ubiquitin hydrolase necessary for regulated proteolysis of the regulatory
subunit of PKA. Cleavage of the (inhibitory) regulatory subunit results in persistent
activity of PKA, leading to persistent phosphorylation of the substrate proteins of
PKA. A second immediate-response gene activated by CREB-1 is C/EBP, which acts
both as a homodimer and as a heterodimer with activating factor (AF) to activate
downstream genes [including elongation factor 1α (EF1α)] that lead to the growth
of new synaptic connections.
352 Psychiatry, Psychoanalysis, and the New Biology of Mind
transmitter release produced by a combination of mechanisms (Figure 7–3)
(Byrne and Kandel 1995).
CREB-1 Mediated Transcription
By substituting puffs of serotonin (the transmitter released by tail shocks),
for the tail shocks themselves, Samuel Schacher, Pier Giorgio Montarolo,
Philip Goelet, and I modeled sensitization in a culture dish consisting of a
single sensory cell making synaptic connections with a single motor cell
(Montarolo et al. 1986). We were able to induce both short- and long-term
facilitation and found, as with the intact animal, that the long-term process
differed from the short-term process in requiring the synthesis of new proteins.
We used this cell culture to ask: What genes are activated to convert the
short-term to the long-term process, and what genes are essential for the
maintenance of the long-term process? We found that five spaced puffs of
serotonin (simulating five spaced shocks to the tail) activate PKA, which in
turn recruits the mitogen-activated protein kinase (MAPK). Both translocate
to the nucleus, where they activate a transcriptional cascade beginning with
the transcription factor CREB-1, the cAMP response element-binding protein–
1, so called because it binds to a cAMP response element (CRE) in the
promoters of target genes (Figure 7–3). The first clue to the importance of
CREB in long-term memory was provided in 1990 by Pramod Dash and Binyamin
Hochner (Dash et al. 1990). They injected, into the nucleus of a sensory
neuron in culture, oligonucleotides carrying the CRE DNA element,
thereby titrating out CREB. This treatment blocked long-term but not shortterm
facilitation (Figure 7–3). Later, Dusan Bartsch cloned Aplysia CREB-1a
(ApCREB-1a) and showed that injection of the phosphorylated form of this
transcription factor by itself could initiate the long-term memory process.
Downstream from ApCREB (Bartsch et al. 1998), Cristina Alberini and Bartsch
found two additional positive transcription regulators, the CAAT box
enhancer binding protein (ApC/EBP) and activation factor (Ap/AF) (Alberini
et al. 1994; Bartsch et al. 2000). CREB-1 activates this set of immediate
response genes, which in turn act on downstream genes, to give rise to
the growth of new synaptic connections (Bacskai et al. 1993; Bailey and Kandel
1993; Bailey et al. 1992; Dash et al. 1990; Glanzman et al. 1990; Kaange
et al. 1993; Martin et al. 1997; Schacher et al. 1988) (Figure 7–3). As first
shown by Craig Bailey and Mary Chen, long-term memory endures by virtue
of the growth of new synaptic connections, a structural change that parallels
the duration of the behavioral memory (Bailey and Chen 1988, 1989; Bailey
and Kandel 1993; Bailey et al. 1992). As the memory fades, the connections
retract over time. A typical sensory neuron in the intact Aplysia has about
The Molecular Biology of Memory Storage 353
1,200 synaptic varicosities. Following long-term sensitization, the number
more than doubles to about 2,600; with time the number returns to about
Inhibitory Constraints
In 1995, Bartsch found that positive regulators are only half the story—there
are also inhibitory constraints on memory. Long-term synaptic facilitation
requires not only activation of memory-enhancer genes but also inactivation
of memory-suppressor genes (Figure 7–3). One of these, the transcription
factor ApCREB-2, can repress ApCREB-1a mediated transcription; relieving
this repression lowers the threshold for the long-term process.
Thus, during long-term memory storage, a tightly controlled cascade of
gene activation is switched on, with memory-suppressor genes providing a
threshold or checkpoint for memory storage, presumably to ensure that only
salient features are learned. Memory suppressors may allow for the modulation
of memory storage by emotional stimuli, as occurs in “flashbulb memories,”
memories of emotionally charged events that are recalled in detail, as
if a complete picture had been instantly and powerfully etched in the brain.
Synapse Specificity of Long-Term Facilitation
The finding of a transcriptional cascade explained why long-term memory
requires new protein synthesis immediately after training, but it posed a new
cell-biological problem. A single neuron makes hundreds of contacts on
many different target cells. Short-term synaptic changes are synapsespecific.
Since long-lasting synaptic changes require transcription and thus
the nucleus, is long-term memory storage a cell-wide process, or are there
cellbiological mechanisms that maintain the synapse specificity of long-term
To examine these questions, Kelsey Martin cultured one Aplysia sensory
cell with a bifurcating axon with two motor neurons, forming two widely
separated synapses (Figure 7–4A). In this culture system, a single puff of serotonin
applied to one synapse produces transient facilitation at that synapse
only, as expected (Casadio et al. 1999; Martin et al. 1998). Five puffs of serotonin
applied to one branch produces long-lasting facilitation (72 hours),
also restricted to the stimulated synapse (Figure 7–4B). This long-lasting
synapse-specific facilitation requires CREB and also leads to structural
changes. Thus, despite recruitment of nuclear processes, long-term changes
in synaptic function and structure are confined only to those synapses stimulated
by serotonin.
How does this come about? Martin, Andrea Casadio, Bailey, and I found
354 Psychiatry, Psychoanalysis, and the New Biology of Mind
The Molecular Biology of Memory Storage 355
that five puffs of serotonin send a signal to the nucleus to activate CREB-1,
which then appears to send proteins to all terminals; however, only those
terminals that have been marked by serotonin can use these proteins productively
for synaptic growth. Indeed, one puff of serotonin to the previously
unstimulated synapse is sufficient to mark that synapse so that it can
capture a reduced form of the long-term facilitation induced at the other site
by five puffs of serotonin (Figure 7–4B).
These results gave us a new and surprising insight into short-term facilitation.
The stimulus that produces the short-term process has two functions
(Figure 7–4C). When acting alone, it provides a selective, synapse-specific enhancement
of synaptic strength, which contributes to short-term memory,
lasting minutes. When acting in conjunction with the activation of CREB initiated
by a long-term process in either that synapse or in any other synapse on
the same neuron, the stimulus locally marks those synapses at which it occurs.
The marked synapse can then utilize the proteins activated by CREB for synaptic
growth to produce a persistent change in synaptic strength. Thus, the
logic for the long-term process involves a long-range integration that is differ-
FIGURE 7–4. A single sensory neuron connects to many target
cells (opposite page).
The requirement of a transcriptional mechanism for long-term memory raises the
question, What is the unit of long-term information storage? Is it a single synapse, as
with short-term facilitation, or the entire neuron? Is there a mechanism for restricting
synaptic facilitation to some synaptic connections?
(A) This photomicrograph shows a culture system developed to examine the action
of two independent branches of a single in Aplysia sensory neuron (the small neuron
in the middle) on two different motor neurons (large neurons). Serotonin can be selectively
applied to one and not the other of the two branches. The flow of the serotonin
can be monitored with the dye, fast green.
(B) Long-term facilitation is synapse-specific and can be captured at another branch
by the stimulus that initiates the short-term process. Five puffs of serotonin applied
at the initiation site (cell A) produce a synapse-specific facilitation shown in (B). This
synapse-specific facilitation is not evident at the synapse of cell B unless that synapse
is itself primed with a single puff of serotonin.
(C) Two effects of short-term facilitation: short-term memory storage when acting by
itself and marking of the specific synapse to which it is applied for subsequent capture
of the proteins necessary for long-term facilitation and growth when applied in
conjunction with five pulses to another set of terminals.
Source. (A) and (B) From Martin KC, Casadio A, Zhu H, et al: “Synapse-Specific
Transcription-Dependent Long-Term Facilitation of the Sensory to Motor Neuron
Connection in Aplysia: A Function for Local Protein Synthesis in Memory Storage.”
Cell 91:927–938, 1998. Used with permission of Elsevier.
356 Psychiatry, Psychoanalysis, and the New Biology of Mind
ent from the short-term process. In the long term, the function of a synapse is
not only determined by the history of usage of that synapse. It is also determined
by the state of the transcriptional machinery in the nucleus.
How does one puff of serotonin mark a synapse for long-term change?
For structural changes to persist, local protein synthesis is required (Casadio
et al. 1999). Oswald Steward’s important work in the early 1980s had shown
that dendrites contain ribosomes, and that specific mRNAs are transported
to the dendrites and translated there (Steward 1997). Our experiments
showed that one function of these locally translated mRNAs was to stabilize
the synapse-specific long-term functional and structural changes.
Neurotransmitter Regulation of Local Protein Synthesis
These studies thus revealed a new, fourth type of synaptic action mediated by
neurotransmitter signaling (Figure 7–5). Three of these four have emerged, at
least in part, from the study of learning and memory. First, in 1951, Katz and
Fatt opened up the modern study of chemical transmission with their discovery
of ionotropic receptors that regulate ion flux through transmitter-gated ion
channels to produce fast synaptic actions, lasting milliseconds (Fatt and Katz
1951). Second, in the 1970s, metabotropic receptors were found to activate
second-messenger pathways, such as the cAMP-PKA pathway, to produce
slow synaptic activity lasting minutes (Greengard 1976). As we have seen in
Aplysia, this slow synaptic action can regulate transmitter release, thereby contributing
to short-term memory for sensitization. Third, an even more persistent
synaptic action, lasting days, results from repeated action of a modulatory
transmitter such as serotonin. With repeated applications of serotonin, second-
messenger kinases translocate to the nucleus, where they activate a cascade
of gene induction leading to the growth of new synaptic connections.
This of course raises the problem of synapse specificity that we have considered
above. Our experiments, in the bifurcated culture system, revealed a
novel fourth action of neurotransmitters, the marking of the synapse and the
regulation of local protein synthesis, which contributes to the establishment
of synapse-specific long-term facilitation.
Explicit Memory
I have so far considered only the simplest cases of memory storage—those
involving reflexes—a form called implicit or procedural memory. Implicit
memory is memory for perceptual and motor skills and is expressed through
performance, without conscious recall of past episodes. In contrast, the
memories we hold near and dear are called explicit (or declarative) memories.
These memories require conscious recall and are concerned with peoThe
Molecular Biology of Memory Storage 357
FIGURE 7–5. A dialog between genes and synapses.
Four consequences of the action of neurotransmitters. 1) Transmitter activation of
a ligand-gated ion channel leads to a rapid synaptic action lasting milliseconds.
2) Transmitter activation of a seven-transmembrane receptor and a second-messenger
kinase leads to a more enduring synaptic action lasting minutes. 3) Repeated
transmitter activation of a seven transmembrane receptor leads to the translocation
of the kinase to the nucleus and to activation of transcription, producing a persistent
synaptic action. 4) Transmitter activation of local protein synthesis to stabilize the
synapse-specific facilitation.
358 Psychiatry, Psychoanalysis, and the New Biology of Mind
The Molecular Biology of Memory Storage 359
ple, places, objects, and events. Explicit memory involves a specialized
anatomical system in the medial temporal lobe, and a structure deep to it,
the hippocampus (Bacskai et al. 1993; Castellucci et al. 1978; Milner et al.
1998) (Figure 7–6A). How is explicit memory stored? Louis Flexner, Bernard
Agranoff, Sam Barondes, and Larry Squire had shown that explicit
memory, like implicit memory, has a short-term phase that does not require
protein synthesis and a long-term phase that does (Bacskai et al. 1993). Are
these two components of memory storage also represented at the cellular
level? What rules govern explicit memory storage?
A decade ago, when I reached my sixtieth birthday, I gathered up my
courage and returned to the hippocampus. Mario Capecchi and Oliver
Smithies, by achieving targeted gene ablation in mouse embryonic stem
cells, provided a superb genetic system for relating individual genes to synaptic
plasticity, on the one hand, and to complex explicit memory storage on
the other. Mice have a medial temporal lobe system, including a hippocampus,
that resembles that of humans, and they use their hippocampus much
as we do to store memory of places and objects (Figure 7–6A).
FIGURE 7–6. Long-term potentiation (LTP) in the hippocampus
(opposite page).
(A) Three major pathways, each of which gives rise to LTP. The perforant pathway
from the subiculum forms excitatory connections with the granule cells of the dentate
gyrus. The mossy fiber pathway, formed by the axons of the granule cells of the
dentate gyrus, connects the granule cells with the pyramidal cells in area CA3 of the
hippocampus. The Schaffer collateral pathway connects the pyramidal cells of the
CA3 region with the pyramidal cells in the CA1 region of the hippocampus.
(B) The early and late phases of LTP in the Schaffer collateral pathway. A single train
of stimuli for 1 second at 100 Hz elicits an early LTP, and four trains at 10-minute intervals
elicit the late phase of LTP. The early LTP lasts about 2 hours, the late LTP
more than 24 hours.
(C) A model for the late phase of LTP in the Schaffer collateral pathway. A single train
of action potentials initiates early LTP by activating NMDA receptors, Ca2+ influx
into the postsynaptic cell, and the activation of a set of second messengers. With repeated
trains of action potentials (illustrated here), the Ca2+ influx also recruits an
adenylyl cyclase (AC), which activates the cAMP-dependent protein kinase. The kinase
is transported to the nucleus, where it phosphorylates CREB. CREB in turn activates
targets (C/EBPβ, tPA, BDNF) that are thought to lead to structural changes.
Mutations in mice that block PKA or CREB reduce or eliminate the late phase of LTP.
The adenylyl cyclase can also be modulated by dopamine signals and perhaps other
modulatory inputs. In addition, there are constraints (bold lines) that inhibit L-LTP
and memory storage. Removal of these constraints lowers the threshold for L-LTP
and enhances memory storage.
360 Psychiatry, Psychoanalysis, and the New Biology of Mind
Although we still do not know much about how information is transformed
as it gets into and out of the hippocampus, it is well established that
the hippocampus contains a cellular representation of extrapersonal space—
a cognitive map of space—and lesions of the hippocampus interfere with
spatial tasks (Grant et al. 1992). Moreover, in 1972, Terje Lømo and Tim
Bliss discovered that the perforant path, a major pathway within the hippocampus,
exhibits activity-dependent plasticity, a change now called longterm
potentiation (LTP) (Figure 7–6B). In the CA1 region of the hippocampus,
LTP is induced postsynaptically by activation of an NMDA receptor to
glutamate. In the late 1980s, Richard Morris found that blocking the NMDA
receptor pharmacologically not only interfered with LTP but also blocked
memory storage (Bliss and Lømo 1973; Morris et al. 1986).
This earlier work on LTP in hippocampal slices had focused on the response
to one or two trains of electrical stimuli. But in Aplysia we had found
that long-term memory emerges most effectively with repeated stimuli (Figure
7–1B). So when Uwe Frey, Yan-You Huang, Peter Nguyen, and I turned
to the hippocampus, we examined whether LTP changed with repeated
stimulation (Frey et al. 1993; Nguyen et al. 1994; Nicoll and Malenka 1999)
and found that hippocampal LTP has phases, much like facilitation in Aplysia.
The early phase of LTP, produced by a single train of stimuli, lasts only
1–3 hours and does not require new protein synthesis (Nguyen et al. 1994);
it involves covalent modifications of preexisting proteins that lead to the
strengthening of preexisting connections, similar in principle to short-term
facilitation in Aplysia. By contrast, repeated trains of electrical stimuli produce
a late phase of LTP, which has properties quite different from early LTP
and similar to long-term facilitation in Aplysia (Figure 7–6B). The late phase
of LTP persists for at least a day and requires both translation and transcription.
The late phase of LTP, like long-term storage of implicit memory, requires
PKA, MAPK, and CREB and appears to lead to the growth of new
synaptic connections (Bolshakov et al. 1997; Bourtchouladze et al. 1994;
Engert and Bonhoeffer 1999; Frey et al. 1993; Impey et al. 1998; Ma et al.
1999; Muller 1997; Nguyen et al. 1994; Nicoll and Malenka 1999; Yin and
Tully 1996) (Figure 7–6C).
The Late Phase of LTP and Explicit Memory
To explore further the specific role of PKA and late LTP in memory storage,
Ted Abel, Mark Barad, Rusiko Bourtchouladze, Peter Nguyen, and I generated
transgenic mice that express R(AB), a mutant form of the regulatory
subunit of PKA that inhibits enzyme activity (Abel et al. 1997). In these
R(AB) transgenic mice, the reduction in hippocampal PKA activity was paralleled
by a significant decrease in late LTP, while basal synaptic transmisThe
Molecular Biology of Memory Storage 361
sion and early LTP remained unchanged. Most interesting, this deficit in the
late phase of LTP was paralleled by behavioral deficits in hippocampusdependent
long-term memory for extrapersonal space, whereas learning,
and short-term memory, are unimpaired (Figure 7–7A and B). Thus, in the
storage of explicit memory of extrapersonal space in the mammalian hippocampus,
PKA plays a critical role in the transformation of short-term memory
into long-term memory, much as it does in the storage of implicit memory
in Aplysia and Drosophila.
Using the R(AB) mice we could now ask, Why do animals with compromised
PKA signaling have difficulty with space (Abel et al. 1997)? We were
influenced by the classic studies of John O’Keefe and John Dostrovsky, who
in 1971 discovered that the pyramidal cells of the hippocampus—the cells
one examines artificially by electrically stimulating the Schaffer collateral
pathway while studying LTP—are “place cells;” they actually encode extrapersonal
space in the animal (O’Keefe and Nadel 1978). A given pyramidal
cell will fire only when the head of the mouse is in a certain part of an enclosed
space—the cell’s place field. When placed in a new environment,
within minutes an animal develops an internal representation of the space
(by the coordinated firing of a population of place cells), which is normally
stable for days. The same cell will have the same firing field each time the
animal is reintroduced to that environment. When now placed in a second
environment, a new map is formed—again in minutes—in part from some
of the cells that made up the map of the first environment and in part from
pyramidal cells that had been silent previously (O’Keefe and Nadel 1978).
It struck me that the formation of a new map resembled a learning process.
The map develops with time as the animal familiarizes itself with the
space, and once learned, the map of space is retained for days and weeks. To
first test whether the molecular pathways underlying the late phase of LTP
were important for the long-term stabilization of this map, Cliff Kentros,
Robert Muller, Hawkins, and I simply blocked LTP pharmacologically with
an NMDA receptor antagonist (Kentros et al. 1998). When placed in a new
environment, the animals with blocked NMDA receptors formed a good spatial
map that was still stable 1 hour later. However, by 24 hours, most pyramidal
cells no longer retained the representation of the field they had
initially. This suggested that activation of NMDA receptors—perhaps a step
in modifying the strength of the synapse—is required for the long-term stabilization
of a place cell map, a result consistent with the role for the late
phase of LTP in the stabilization of a place cell map.
We next asked whether a selective deficit that affects only the late phase
of LTP causes a selective abnormality in the long-term stability of place cells.
Since only the late phase of LTP requires PKA, Alex Rotenberg, Muller, Abel,
Hawkins, and I returned to the R(AB) transgenic mice with diminished PKA
362 Psychiatry, Psychoanalysis, and the New Biology of Mind
FIGURE 7–7. Contextual learning and the stability of place cells.
(A) The protocol for context conditioning consists of exposure to the context followed
by a tone and then a shock. The animals are then tested 1 hour and 24 hours
after training.
(B1) Mutant mice that express the R(AB) gene in the hippocampus, blocking the action
of PKA, have a selective defect for long-term contextual memory. Mice that express
R(AB) were conditioned to freeze to the context. After becoming familiar with
the context, the mice heard a sound and received a shock through the electrified grid
in the floor. As a result, the animals learned to associate the context of the space with
shock and to freeze when placed in the box at a future time. These mice had good
short-term memory at 1 hour for freezing to context, but at 24 hours they no longer
froze to context, indicating a defect in a form of long-term explicit (declarative)
memory that requires the hippocampus. (B2) Wild-type mice exposed to anisomycin,
an inhibitor of protein synthesis, during training show a similar defect for longterm
memory when tested 24 hours after conditioning.
(C) Place cell stability for R(AB) and wild-type mice. R(AB) mice with a defect in
PKA and late LTP form place fields that are stable at 1 hour. These fields are not stable
at 24 hours.
Source. (A) and (B) From Abel T, Nguyen PV, Barad M, et al: “Genetic Demonstration
of a Role for PKA in the Late Phase of LTP and in Hippocampus-Based Long-
Term Memory.” Cell 88:615–626, 1997. (C) From Rotenberg A, Abel T, Hawkins RD,
et al: “Parallel Instabilities of Long-Term Potentiation, Place Cells, and Learning
Caused by Decreased Protein Kinase A Activity.” The Journal of Neuroscience
20:8096–8102, 2000; and Agnihotri N, Hawkins RD, Kandel ER, et al: “Protein Synthesis
Inhibition Selectively Abolishes Long-Term Stability of Hippocampal Place
Cell Maps.” Abstracts—Society for Neuroscience 27:316.14, 2001. Used with permission
of Elsevier.
The Molecular Biology of Memory Storage 363
activity and a diminished form of late LTP (Rotenberg et al. 2000). If reduced
activity of PKA affected the stability of place cells, R(AB) mice should
be able to form a stable map of space in a novel environment, as in normal
animals, that is stable for at least 1 hour. However, the cell field should be
unstable when recorded 24 hours later. This is precisely what we found (Figure
7–7C). The fact that long-term instability in the spatial map and the deficit
in long-term memory paralleled the deficit in the late phase of LTP
suggested that PKA-mediated gene activation and the synthesis of new protein
might be essential for the stabilization of the spatial map. Naveen Agnihotri,
Kentros, Hawkins, and I tested this idea and found that inhibiting
protein synthesis indeed destabilized the place fields in the long term much
as does inhibiting PKA (Kentros et al. 2001).
In the course of this work, Kentros and Agnihotri found, remarkably,
that, as is the case with explict memories in humans, a key feature in the stabilization
of PKA and protein synthesis–dependent phase of memory is attention.
When a mouse does not attend to the space it walks through, the
map forms but is unstable after 3–6 hours. When the mouse is forced to attend
to the space, however, the map invariably is stable for days!
Inhibitory Constraints on Explicit Memory
Recently we (Malleret et al. 2001) and others (Blitzer et al. 1998) have found
that the threshold for hippocampal synaptic plasticity and memory storage is
determined by the balance between protein phosphorylation governed by
PKA and dephosphorylation (Malleret et al. 2001; Mansuy et al. 1998). To determine
whether the endogenous Ca2+-sensitive phosphatase calcineurin acts
as a constraint on this balance, we inhibited calcineurin and examined the effects
on synaptic plasticity and memory storage. Isabelle Mansuy, Gael
Malleret, Danny Winder, Tim Bliss, and I found that a transient reduction of
calcineurin activity resulted in facilitation of LTP both in vitro and in vivo
(Malleret et al. 2001). This facilitation persisted for several days in the intact
animal and was accompanied by enhanced learning and strengthening of
short- and long-term memory on several spatial and nonspatial tasks requiring
the hippocampus. These results, together with previous findings by Winder
and Mansuy showing that overexpression of calcineurin impairs PKA-dependent
components of LTP and memory (Mansuy et al. 1998; Winder et al.
1998), demonstrate that endogenous calcineurin can act as a negative regulator
of synaptic plasticity, learning, and memory (Figure 7–6C).
An Overall View
Our studies of the storage component of memory, the molecular mechanism
whereby information is stored, have led to two general conclusions.
364 Psychiatry, Psychoanalysis, and the New Biology of Mind
First, our research suggests that the cellular and molecular strategies
used in Aplysia for storing short- and long-term memory are conserved in
mammals and that the same molecular strategies are employed in both implicit
and explicit memory storage. With both implicit and explicit memory
there are stages in memory that are encoded as changes in synaptic strength
and that correlate with the behavioral phases of short- and long-term memory.
The short-term synaptic changes involve covalent modification of
preexisting proteins, leading to modification of preexisting synaptic connections,
whereas the long-term synaptic changes involve activation of gene expression,
new protein synthesis, and the formation of new connections.
Whereas short-term memory storage for implicit and explicit memory requires
different signaling, long-term storage of both implicit and explicit
memory uses as a core signaling pathway PKA, MAPK, and CREB-1. At least
in the mouse, additional components are likely recruited. In both implicit
and explicit memory the switch from short-term to long-term memory is
regulated by inhibitory constraints.
Second, the study of learning has revealed new features of synaptic transmission
and new cell-biological functions of synaptic signaling. For example,
different forms of learning recruit different modulatory transmitters,
which then act in one of three ways: 1) They activate second-messenger kinases
that are transported to the nucleus, where they initiate processes required
for neuronal growth and long-term memory; 2) they mark the
specific synapses for capture of the long-term process and regulate local protein
synthesis for stabilization; and 3) they mediate, in ways we are just beginning
to understand, attentional processes required for memory formation
and recall.
Most important, the study of long-term memory has made us aware of
the extensive dialog between the synapse and the nucleus, and the nucleus
and the synapse (Figure 7–5). In the long-term process, the response of a
synapse is not determined simply by its own history of activity (as in shortterm
plasticity) but also by the history of transcriptional activation in the
I started this essay by pointing out that 40 years ago, at the beginning of
my career, I thought that a reductionist approach based on the use of a simple
experimental system such as Aplysia might allow us to address fundamental
questions in learning and memory. That was a leap of faith for which I have
been rewarded beyond my fondest hopes. Still, the complexity of explicit
memory is formidable, and we have only begun to explore it. We as yet know
little about the molecular mechanisms that initiate or stabilize the synaptic
growth associated with long-term memory. What signaling molecules lead to
the cytoskeletal rearrangements during synaptic remodeling? How do they relate
to the molecules that control synapse formation during development?
The Molecular Biology of Memory Storage 365
In addition, we have here only considered the molecular mechanisms of
memory storage. The more difficult part of memory—especially explicit
memory—is a systems problem. We still need to seek answers to a family of
important questions. How do different regions of the hippocampus and the
medial temporal lobe—the subiculum, the entorhinal, parahippocampal,
and perirhinal cortices—interact in the storage of explicit memory? How is
information in any of these regions transferred for ultimate consolidation in
the neocortex? We do not, for example, understand why the initial storage
of long-term memory requires the hippocampus, whereas the hippocampus
is not required once a memory has been stored for weeks or months (Milner
et al. 1998; Squire and Zola-Morgan 1991). What critical information does
the hippocampus convey to the neocortex? We also know very little about
the nature of recall of explicit (declarative) memory, a recall that requires
conscious effort. These systems problems will require more than the bottoms-
up approach of molecular biology. They will also require the top-down
approaches of cognitive psychology, neurology, and psychiatry. Ultimately,
we will need syntheses that bridge the two approaches.
Despite these complexities, these and other questions in the biology of
learning no doubt will be vigorously addressed in the near future. For the
biology of the mind has now captured the imagination of the scientific community
of the twenty-first century, much as the biology of the gene fascinated
the scientists of the twentieth century. As the biological study of the
mind assumes the central position within biology and medicine, we have every
reason to expect that a succession of brain scientists will be called to
Stockholm and honored for their own leaps of faith (Kentros et al. 2001).
I have had the privilege to work with and to learn from many gifted students,
fellows, and collaborators, and I have tried throughout this lecture to acknowledge
their contributions. My science has benefited enormously from
the interactive environment created by the Center for Neurobiology and Behavior
at the College of Physicians and Surgeons of Columbia University. It
would be hard to find a more ideal environment in which to mature as a scientist.
Specifically, I have benefited greatly from my long-standing friendship
with R. Axel, C. Bailey, J. Dodd, R. Hawkins, J. Koester, T. Jessell, J.H.
Schwartz, S. Siegelbaum, and G. Fischbach, the current dean of the College
of Physicians and Surgeons. I am further grateful to J. Koester for his excellent
leadership of the Center for Neurobiology and Behavior, and to D.
Hirsh, S. Silverstein, and J. Oldham, chairs of the three departments to
which I belong. Finally, I am indebted to H. Pardes who, until recently,
served as dean of the College of Physicians and Surgeons. My research has
366 Psychiatry, Psychoanalysis, and the New Biology of Mind
been generously supported by the Howard Hughes Medical Institute, the
NIH, the Mathers Foundation, FRAXA, and the Lieber Trust. I am particularly
indebted to the Howard Hughes Medical Institute and its leadership, D.
Fredrickson, G. Cahill, P. Chopin, M. Cowan, D. Harter, and more recently
T. Cech and G. Rubin, whose farsighted vision has encouraged Hughes investigators
to take a long-term perspective so as to be able to tackle challenging
problems. Research on learning and memory certainly meets both of
these criteria!
Abel T, Nguyen PV, Barad M, et al: Genetic demonstration of a role for PKA in the
late phase of LTP and in hippocampus-based long-term memory. Cell 88:615–
626, 1997
Agnihotri N, Hawkins RD, Kandel ER, et al: Protein synthesis inhibition selectively
abolishes long-term stability of hippocampal place cell maps. Abstr Soc Neurosci
27:316.14, 2001
Alberini C, Ghirardi M, Metz R, et al: C/EBP is an immediate-early gene required for
the consolidation of long-term facilitation in Aplysia. Cell 76:1099–1114, 1994
Bacskai BJ, Hochner B, Mahaut-Smith M, et al: Spatially resolved dynamics of cAMP
and protein kinase A subunits in Aplysia sensory neurons. Science 260:222–
226, 1993
Bailey CH, Chen M: Long-term memory in Aplysia modulates the total number of
varicosities of single identified sensory neurons. Proc Natl Acad Sci USA
85:2373–2377, 1988
Bailey CH, Chen M: Time course of structural changes at identified sensory neuron
synapses during long-term sensitization in Aplysia. J Neurosci 9:1774–1780,
Bailey CH, Kandel ER: Structural changes accompanying memory storage. Annu Rev
Physiol 55:397–426, 1993
Bailey CH, Montarolo P, Chen M, et al: Inhibitors of protein and RNA synthesis block
structural changes that accompany long-term heterosynaptic plasticity in Aplysia.
Neuron 9:749–758, 1992
Bartsch D, Ghirardi M, Skehel PA, et al: Aplysia CREB2 represses long-term facilitation:
relief of repression converts transient facilitation into long-term functional
and structural change. Cell 83:979–992, 1995
Bartsch D, Casadio A, Karl KA, et al: CREB1 encodes a nuclear activator, a repressor,
and a cytoplasmic modulator that form a regulatory unit critical for long-term
facilitation. Cell 95:211–223, 1998
Bartsch D, Ghirardi M, Casadio A, et al: Enhancement of memory-related long-term
facilitation by ApAP, a novel transcription factor that acts downstream from
both CREB1 and CREB2. Cell 103:595–608, 2000
The Molecular Biology of Memory Storage 367
Bliss TV, Lømo T: Long-lasting potentiation of synaptic transmission in the denate
area of the anaesthetized rabbit following stimulation of the perforant path.
J Physiol 232:331–356, 1973
Blitzer RD, Connor JH, Brown GP, et al: Gating of CaMKII by cAMP-regulated protein
phosphatase activity during LTP. Science 280:1940–1942, 1998
Bolshakov VY, Golan H, Kandel ER, et al: Recruitment of new sites of synaptic transmission
during the cAMP-dependent late phase of LTP at CA3-CA1 synapses in
the hippocampus. Neuron 19:635–651, 1997
Bourtchouladze R, Franguelli B, Blendy J, et al: Deficient long-term memory in mice
with a targeted mutation of the cAMP-responsive element-binding protein. Cell
79:59–68, 1994
Brunelli M, Castellucci V, Kandel ER: Synaptic facilitation and behavioral sensitization
in Aplysia: possible role of serotonin and cyclic AMP. Science 194:1178–
1181, 1976
Byrne JH, Kandel ER: Presynaptic facilitation revisited: state and time dependence.
J Neurosci 16:425–435, 1995
Byrne J, Castellucci V, Kandel ER: Receptive fields and response properties of mechanoreceptor
neurons innervating skin and mantle shelf of Aplysia. J Neurophysiol
37:1041–1064, 1974
Byrne JH, Castellucci VF, Carew TJ, et al: Stimulus-response relations and stability of
mechanoreceptor and motor neurons mediating defensive gill-withdrawal reflex
in Aplysia. J Neurophysiol 41:402–417, 1978a
Byrne JH, Castellucci VF, Kandel ER: Contribution of individual mechanoreceptor
sensory neurons to defensive gill-withdrawal reflex in Aplysia. J Neurophysiol
41:418–431, 1978b
Carew TJ, Pinsker HM, Kandel ER: Long-term habituation of a defensive withdrawal
reflex in Aplysia. Science 175:451–454, 1972
Casadio A, Martin KC, Giustetto M, et al: A transient neuron-wide form of CREBmediated
long-term facilitation can be stabilized at specific synapses by local
protein synthesis. Cell 99:221–237, 1999
Castellucci VF, Pinsker H, Kupfermann I, et al: Neuronal mechanisms of habituation
and dishabituation of the gill-withdrawal reflex in Aplysia. Science 167:1745–
1748, 1970
Castellucci VF, Carew TJ, Kandel ER: Cellular analysis of long-term habituation of
the gill-withdrawal reflex in Aplysia. Science 202:1306–1308, 1978
Castellucci VF, Kandel ER, Schwartz JH, et al: Intracellular injection of the catalytic
subunit of cyclic AMP–dependent protein kinase simulates facilitation of transmitter
release underlying behavioral sensitization in Aplysia. Proc Natl Acad Sci
USA 77:7492–7496, 1980
Castellucci VF, Nairn A, Greengard P, et al: Inhibitor of adenosine 3´:5´-monophosphate-
dependent protein kinase blocks presynaptic facilitation in Aplysia.
J Neurosci 2:1673–1681, 1982
Castellucci VF, Blumenfeld H, Goelet P, et al: Inhibitor of protein synthesis blocks
long-term behavioral sensitization in the isolated gill-withdrawal reflex of
Aplysia. J Neurobiol 20:1–9, 1989
368 Psychiatry, Psychoanalysis, and the New Biology of Mind
Cedar H, Schwartz JH: Cyclic adenosine monophosphate in the nervous system of
Aplysia californica, II: effect of serotonin and dopamine. J Gen Physiol 60:570–
587, 1972
Cedar H, Kandel ER, Schwartz JH: Cyclic adenosine monophosphate in the nervous
system of Aplysia californica, I: increased synthesis in response to synaptic stimulation.
J Gen Physiol 60:558–569, 1972
Dash PK, Hochner B, Kandel ER: Injection of cAMP-responsive element into the nucleus
of Aplysia sensory neurons blocks long-term facilitation. Nature 345:718–
721, 1990
Ebbinghaus H: Memory: A Contribution to Experimental Psychology (1885). New
York, Dover, 1963
Engert F, Bonhoeffer T: Dendritic spine changes associated with hippocampal longterm
synaptic plasticity. Nature 399:66–70, 1999
Fatt P, Katz B: An analysis of the end-plate potential recorded with an intracellular
electrode. J Physiol 115:320–370, 1951
Flexner LB, Flexner JB: Effect of acetoxycycloheximide and of an acetoxycycloheximide-
puromycin mixture on cerebral protein synthesis and memory in mice.
Proc Natl Acad Sci USA 55:369–374, 1966
Frazier WT, Kandel ER, Kupfermann I, et al: Morphological and functional properties
of identified neurons in the abdominal ganglion of Aplysia californica.
J Neurophysiol 30:1288–1351, 1967
Frey U, Huang Y-Y, Kandel ER: Effects of cAMP simulate a late stage of LTP in hippocampal
CA1 neurons. Science 260:1661–1664, 1993
Frost WN, Castellucci VF, Hawkins RD, et al: Monosynaptic connections made by the
sensory neurons of the gill- and siphon-withdrawal reflex in Aplysia participate
in the storage of long-term memory for sensitization. Proc Natl Acad Sci USA
82:8266–8269, 1985
Glanzman DL, Mackey SL, Hawkins RD, et al: Depletion of serotonin in the nervous
system of Aplysia reduces the behavioral enhancement of gill withdrawal as well
as the heterosynaptic facilitation produced by tail shock. J Neurosci 9:4200–
4213, 1989
Glanzman DL, Kandel ER, Schacher S: Target-dependent structural changes accompanying
long-term synaptic facilitation in Aplysia neurons. Science 249:799–
802, 1990
Grant SGN, O’Dell TJ, Karl KA, et al: Impaired long-term potentiation, spatial learning,
and hippocampal development in fyn mutant mice. Science 258:1903–
1910, 1992
Greengard P: Possible role for cyclic nucleotides and phosphorylated membrane proteins
in postsynaptic actions of neurotransmitters. Nature 260:101–108, 1976
Hawkins RD, Castellucci VF, Kandel ER: Interneurons involved in mediation and
modulation of the gill-withdrawal reflex in Aplysia, I: identification and characterization.
J Neurophysiol 45:304–314, 1981a
The Molecular Biology of Memory Storage 369
Hawkins RD, Castellucci VF, Kandel ER: Interneurons involved in mediation and
modulation of the gill-withdrawal reflex in Aplysia, II: identified neurons produce
hetero-synaptic facilitation contributing to behavioral sensitization. J Neurophysiol
45:315–326, 1981b
Impey S, Obrietan K, Wong ST, et al: Cross talk between ERK and PKA is required
for Ca2+ stimulation of CEB-dependent transcription and ERK nuclear translocation.
Neuron 21:869–883, 1998
Kaang B-K, Kandel ER, Grant SGN: Activation of cAMP-responsive genes by stimuli
that produce long-term facilitation in Aplysia sensory neurons. Neuron 10:427–
435, 1993
Kandel ER: Cellular Basis of Behavior: An Introduction to Behavioral Biology. San
Francisco, CA, Freeman, 1976
Kandel ER: The Behavioral Biology of Aplysia: A Contribution to the Comparative
Study of Opisthobranch Molluscs. San Francisco, CA, Freeman, 1979
Kandel ER, Spencer WA: Electrophysiological properties of an archicortical neuron.
Ann NY Acad Sci 94:570–603, 1961
Kandel ER, Spencer WA: Cellular neurophysiological approaches in the study of
learning. Physiol Rev 48:65–134, 1968
Kandel ER, Brunelli M, Byrne J, et al: A common presynaptic locus for the synaptic
changes underlying short-term habituation and sensitization of the gillwithdrawal
reflex in Aplysia. Cold Spring Harb Symp Quant Biol 40:465–482,
Kentros C, Hargreaves E, Hawkins RD, et al: Abolition of long-term stability of new
hippocampal place cell maps by NMDA receptor blockage. Science 280:2121–
2126, 1998
Kentros C, Agnihotri N, Hawkins R, et al: Stabilization of the hippocampal representation
of space in mice requires attention. Abstr Soc Neurosci 27:316.15, 2001
Klein M, Kandel ER: Mechanism of calcium current modulation underlying presynaptic
facilitation and behavioral sensitization in Aplysia. Proc Natl Acad Sci Abstr
77:6912–6916, 1980
Kupfermann I, Castellucci V, Pinsker H, et al: Neuronal correlates of habituation and
dishabituation of the gill-withdrawal reflex in Aplysia. Science 167:1743–1745,
Ma L, Zablow L, Kandel ER, et al: Cyclic AMP induces functional presynaptic boutons
in hippocampal CA3-CA1 neuronal cultures. Nat Neurosci 2:24–30, 1999
Mackey SL, Kandel ER, Hawkins RD: Identified serotonergic neurons LCB1 and
RCB1 in the cerebral ganglia of Aplysia produce presynaptic facilitation of siphon
sensory neurons. J Neurosci 9:4227–4235, 1989
Malleret G, Haditsch U, Genoux D, et al: Inducible and reversible enhancement of
learning, memory, and long-term potentiation by genetic inhibition of calcineurin.
Cell 104:675–686, 2001
Mansuy IM, Mayford M, Jacob B, et al: Restricted and regulated overexpression reveals
calcineurin as a key component in the transition from short-term to longterm
memory. Cell 92:39–49, 1998
370 Psychiatry, Psychoanalysis, and the New Biology of Mind
Martin KC, Michael D, Rose JC, et al: MAP kinase translocates into the nucleus of the
presynaptic cell and is required for long-term facilitation in Aplysia. Neuron
18:899–912, 1997
Martin KC, Casadio A, Zhu H, et al: Synapse-specific transcription-dependent longterm
facilitation of the sensory to motor neuron connection in Aplysia: a function
for local protein synthesis in memory storage. Cell 91:927–938, 1998
Milner B, Squire LR, Kandel ER: Cognitive neuroscience and the study of memory.
Neuron 20:445–468, 1998
Montarolo PG, Goelet P, Castellucci VF, et al: A critical period for macromolecular
synthesis in long-term heterosynaptic facilitation in Aplysia. Science 234:1249–
1254, 1986
Morris RGM, Anderson E, Lynch GS, et al: Selective impairment of learning and
blockade of long-term potentiation by an N-methyl-D-aspartate receptor antagonist,
AP5. Nature 319:774–776, 1986
Muller D: Ultrastructural plasticity of excitatory synapses. Rev Neurosci 8:77–93,
Nguyen PV, Abel T, Kandel ER: Requirement of a critical period of transcription for
induction of a late phase of LTP. Science 265:1104–1107, 1994
Nicoll RA, Malenka RC: Expression mechanisms underlying NMDA receptor-dependent
long-term potentiation. Ann NY Acad Sci 868:515–525, 1999
O’Keefe J, Nadel L: The Hippocampus as a Cognitive Map. Oxford, Clarendon, 1978
Pinsker HM, Kupfermann I, Castellucci V, et al: Habituation and dishabituation of
the gill-withdrawal reflex in Aplysia. Science 167:1740–1742, 1970
Pinsker HM, Hening WA, Carew TJ, et al: Long-term sensitization of a defenseive
withdrawal reflex in Aplysia. Science 182:1039–1042, 1973
Ramón y Cajal S: The Croonian Lecture: la fine structure des centres nerveux. Proc
R Soc London B Bio Sci 55:444–468, 1894
Rotenberg A, Abel T, Hawkins RD, et al: Parallel instabilities of long-term potentiation,
place cells, and learning caused by decreased protein kinase A activity.
J Neurosci 20:8096–8102, 2000
Schacher S, Castellucci VF, Kandel ER: cAMP evokes long-term facilitation in Aplysia
sensory neurons that requires new protein synthesis. Science 240:1667–1669,
Schwartz JH, Castellucci VF, Kandel ER: Functioning of identified neurons and synapses
in abdominal ganglion of Aplysia in absence of protein synthesis. J Neurophysiol
34:939–953, 1971
Shuster MJ, Camardo JS, Siegelbaum SA, et al: Cyclic AMP–dependent protein kinase
closes the serotonin-sensitive K+ channels of Aplysia sensory neurones in cellfree
membrane patches. Nature 313:392–395, 1985
Siegelbaum S, Camardo JS, Kandel ER: Serotonin and cAMP close single K+ channels
in Aplysia sensory neurones. Nature 299:413–417, 1982
Squire LR, Zola-Morgan S: The medial temporal lobe memory system. Science
253:1380–1386, 1991
Steward O: mRNA localization in neurons: a multipurpose mechanism? Neuron
18:9–12, 1997
The Molecular Biology of Memory Storage 371
Winder DG, Mansuy IM, Osman M, et al: Genetic and pharmacological evidence for
a novel, intermediate phase of long-term potentiation (I-LTP) suppressed by
calcineurin. Cell 92:25–37, 1998
Yin J, Tully T: CREB and the formation of long-term memory. Curr Opin Neurobiol
6:264–268, 1996s
This page intentionally left blank
John M. Oldham, M.D.
Kandel’s essay “Genes, Brains, and Self-Understanding: Biology’s Aspirations
for a New Humanism” is eloquent, integrative, and visionary. With characteristic
prescience, Kandel outlines the rapidly changing face and breathtaking
potential of the science and practice of medicine. Indeed, it takes only a
moment of reflection to envy the graduating medical students at Columbia
University College of Physicians and Surgeons as they embarked upon their
careers, launched with this inspirational, scientifically derived prophecy.
The potential for the new knowledge of the human genome to move us
from a focus on populations at risk to the specific genetic vulnerabilities of
an individual is exciting and increasingly real, paving the way for renewed
and individualized emphasis on protective mechanisms and prevention.
However, how effective this new individualized information will be remains
unclear, and herein lies a fundamental challenge that, with all of our new
knowledge, we must not overlook. We already know from studies of clinical
populations, for example, that a healthy diet and regular exercise are protective
factors for individuals at risk for coronary artery disease, or that careful
adherence to antihypertensives and minimization of stress are protective
against dangerous hypertensive episodes, or that major depressive disorder
374 Psychiatry, Psychoanalysis, and the New Biology of Mind
is a serious medical condition that should be treated quickly (which is usually
quite effective) when it occurs. And we know that countless numbers of
people who repeatedly receive this type of advice do not follow it. Human
behavior is, to understate it, complicated.
Kandel ranks at the very top of “the family of deep problems that confront
the study of the mind,” the need to understand “the biology of consciousness,”
which is so clear and urgent that surely there is no controversy
here. But I would extend this challenge to emphasize the need to understand
the biology of the unconscious—to continue our efforts to understand what
motivates human behavior at all levels. We already know a great deal about
certain behaviors that are self-injurious, such as the molecular neurobiology
of addiction, but even so, addiction is a complex mix of biology and behavior.
We have every reason to expect that advances in our knowledge of genetics
will help us sort out those individuals at highest genetic risk for a
given type of addictive behavior—who may need our greatest attention—
from those who, with help, may have better odds to leave the addiction behind.
But what about patterns of self-injurious behavior that may go unrecognized
by the individual in question? For example, gambling to the brink
of bankruptcy while being the sole support of a loving family; being driven
by extreme, narcissistic personal ambition while leaving an unnoticed trail
of used and discarded co-workers behind; making repeated bad choices in
relationships without knowing why, leading to escalating frustration and bewilderment,
or even, for those at risk, the emergence of severe depression.
Kandel’s three laws emphasize pleasures and obligations that are wonderfully
relevant to his listening audience of graduating students. There are,
however, many future visitors to the consulting room who have obligations
and burdens, yet little or no access to pleasure or relief—those illustrated
above who create and perpetuate their own unhappiness, and a large universe
of others who may be at or near the poverty level, without education,
or ill equipped to keep their balance in tough parts of the world. It may well
be here, in the realm of negligible resources that the pleasures of a high-fat
meal trump the self-discipline of a healthy diet, especially if living longer
looks like prolonged misery.
So while science moves at warp speed and takes medicine to a new humanism
that is individualized, we must not forget to look for ways to help
each individual’s world become a better one. Otherwise, that stress-filled environment
doubles right back to attack the health of the individual—the
very health that we’re trying hard to sustain and improve.
C H A P T E R 8
Biology’s Aspirations for a New Humanism
Eric R. Kandel, M.D.
Revolution in Genomics
Members of the graduating class of the year 2001, relatives and friends of the
graduates, Dean Gerald Fischbach, colleagues, ladies and gentlemen: I am
extremely pleased to be asked to participate in the commencement today because
it gives me the opportunity of celebrating the academic achievements
of this college and of this class. I owe this college an enormous personal and
scientific debt! I have found this medical school to be the very best place in
the world to do scientific research and I have benefited greatly from the interactive
and supportive environment engendered by the faculty and the students
of this school. In addition, throughout my 27 years on this faculty, I
have always enjoyed teaching medical students, including, of course, the
privilege of teaching the distinguished class we celebrate today. In fact, Principles
of Neuroscience, the textbook that your class has come to know and
love—which is now universally acknowledged to be the heaviest and most
expensive book of its kind—is based on the neural science course our fac-
This paper is a slightly modified version of the graduation address given on May 16,
2001, at the Columbia University College of Physicians and Surgeons.
376 Psychiatry, Psychoanalysis, and the New Biology of Mind
ulty teaches here at this college, a course for which I was privileged to serve
as first course director.
So when I am asked to what do I aspire after receiving the Nobel Prize in
the year 2000, my answer is clear: to be selected, by the graduating class of
the year 2001, to give the convocation address at the College of Physicians
and Surgeons of Columbia University! What more meaningful and satisfying
recognition can one ever imagine?
For no celebration is more satisfying for this college or more inspiring to
the intellectual community throughout the world than an academic commencement.
For each commencement celebrates the entry into academic
ranks of another class of scholars. Since the task of a great university is not
to simply replicate its own image in scholarship but to create a new knowledge,
it is implicit in the charge to a faculty to develop scholars who are better
than we are, more knowledgeable, more thoughtful, more moral, finer
human beings.
Given that we think you are all of these things, what is there left to tell
you as you now progress from being our students to being our peers? What
are you likely to confront as you move into the next stage of your life? And,
in turn, what can we expect of you in that confrontation? Let me put these
questions, and your past 4 years in medical school, into a bit of historical
The years you have spent in medical school—the remarkable 4 years that
spanned the transition from the twentieth century to the twenty-first century—
have produced both the elucidation of the human genome and an increased
understanding of the biology of the human brain. We have every
reason to expect that the revolution in genomics and in brain science will
radically change the way we practice medicine. And it will do so in two ways.
First, medicine will be transformed from a population-based to an individual-
based medical science; it will become more focused on the individual
and his or her predisposition to health and disease. Second, we will, for the
first time, have a meaningful and nuanced biology of human mental processes
and human mental disorders. If we are fortunate, your generation will
help join these two intellectual streams—that of the human genome and that
of brain science—to realize biology’s aspiration for a new humanism, a humanism
based in part on insights into our biology. If we are successful in advancing
this new humanistic agenda, the genomic revolution and the new
insights into the biological nature of mind will not only enhance medical
care but will also change fundamentally the way we view ourselves and one
The influence of biology on the way informed people think about each
other and about the world in which they live is, of course, not new. In modern
times, this influence first became evident in 1859 with Darwin’s insight
Genes, Brains, and Self-Understanding 377
into the evolution of species. Darwin first argued that human beings and
other animals evolved gradually from animal ancestors quite unlike themselves.
He also emphasized the even more daring idea, that the driving force
for evolutionary change stems not from the heavens, not from a conscious
purpose, but from natural selection, a completely mechanistic, sorting process
based on hereditary variations.
This radical idea split the bond between religion and biology, a bond
based on the idea that an important function of biology was to explain divine
purpose—to account for the overall design of nature. Indeed, natural selection
even caused difficulty for nonbelievers because it was vague as a scientific
To understand hereditary variations, scientists first needed to know: how
is information about biological structure passed from one generation to another?
This question was answered only in the first decades of the twentieth
century. We owe first to Gregor Mendel and then to Thomas Hunt Morgan
(of our own Columbia University) the remarkable discovery that hereditary
information necessary to specify the construction of the organism is passed
from one generation to the next by means of discrete biological structures
we now call genes. Forty years later, first Avery, McCarthy, and McCloud and
subsequently Watson and Crick gave us the seminal insight that the genes
of all living organisms are embodied in the physics and chemistry of a single
large molecule, DNA. Nature, in all its beauty and variety, results from variations
in the sequence of bases in DNA.
In the 1960s and 1970s, our understanding of genes was further enhanced
by the cracking of the genetic code, the three-letter alphabet
whereby the sequence of bases in DNA is translated into the amino acids of a
protein. This breakthrough was followed by DNA sequencing, which allowed
us to read directly the nucleotide sequences that form the instructions
of each gene. Creative application of these and other molecular insights
made possible genetic engineering and more recently the sequencing of the
human genome.
The current generation of physicians will be the first to reap the benefits
of the human genome and use its insights not only to provide better care to
patients—better diagnoses, better treatment—but, also, I would hope, more
individualized care, more individually tailored diagnoses, and more individualized
treatment. Indeed, one would hope that this generation will move us
away from the impersonality of managed health care into a new, biologically
inspired personalized medicine.
What reason do we have to believe that this will come to pass? What will
we learn from the genome that might orient us more to see the patient as a
person rather than as a disease state? The genome of course provides us with
a periodic table of life. It contains the complete list and structure of all genes.
378 Psychiatry, Psychoanalysis, and the New Biology of Mind
But it provides us not simply with an average-expectable genome. It provides
each of us with our own unique genome. In time, our genome will be a part of
our private medical record. As a result, we in academic medicine will collectively
have a catalog of all the human genetic variations that account for all
the heritable differences between individuals.
We now know that any two individuals share an amazing 99.9% DNA sequence
similarity. This means that all the heritable differences among individuals
of a species can be attributed to a mere 0.1% of the sequence. Most
differences between the genomes of any two individuals take the form of
very small changes, where one single base is substituted for another in the
sequence of nucleotides that form a gene. These changes are called single
base changes or single nucleotide polymorphisms.
We already know of about 3 million such polymorphisms, and more will
be identified with time. They are spread throughout the genome and at least
93% of all genes contain at least one such polymorphism. Thus, for the first
time, we will have for every gene all the polymorphic sequence variations
that exist. Many of these will prove unimportant, but some of them will be
fundamental to understanding disease.
These common, polymorphic variations differ fundamentally from the
rare mutations that lead invariably to inherited disease, and that have been
the focus of medical genetics up to now. The common polymorphisms that
we now will have full access to for the first time do not cause disease per se;
rather, they influence the expression of disease; they predict our predisposition
to, and our protection from, disease in all of its manifestations.
To give but one example, there are rare genetic mutations on chromosome
21 that invariably cause an early-onset form of Alzheimer’s disease in
the rare person who carries the mutation. By contrast, there is a fairly common
polymorphism that does not produce Alzheimer’s disease directly. But
the 17% of the population that carry this single base change polymorphism
have a 10-times greater risk of developing a late onset form of Alzheimer’s
disease than those individuals who do not carry this polymorphism. Other
genetic polymorphisms similarly predispose people to various forms of diabetes,
hypertension, cancer, and mental disorders. Indeed, every disease to
which we are prone—including our response to infection, to the consequences
of aging, and even our very longevity itself—will be shown to be influenced
by polymorphisms in our genes. As a corollary, the polymorphisms
also will help reveal that complex diseases such as hypertension, depression,
and Alzheimer’s disease are likely not to be unitary but to be made up of a
number of different, intricately related subtypes, each requiring its own distinctive
medical management.
What will knowledge of these predispositions and subtypes mean for the
practice of clinical medicine? This knowledge will serve to decrease the
Genes, Brains, and Self-Understanding 379
uncertainty in the management of disease. It is likely that clinical DNA
testing—the search for genetic polymorphisms in ourselves and in our patients—
will reveal our individual risk for all major diseases and therefore allow
us to intervene prophylactically in these diseases through diet, surgery,
exercise, or drugs, years before the disease becomes manifest. Indeed, genetic
polymorphisms will be found to underlie the way our patients respond
to these interventions, so that DNA testing will also allow us to predict individual
responses to drugs and to determine the degree to which individuals
are susceptible to particular side effects. This will allow the pharmaceutical
industry to develop new targets and new tools to sharpen the specificity of
the drugs they deliver to meet the needs of the individual patient.
This knowledge of the biological uniqueness of our patients will alter all aspects
of medicine. Currently, newborn babies are only screened for treatable
genetic diseases, such as phenylketonuria. Perhaps in the not too distant future,
children at high risk for coronary artery disease, Alzheimer’s, or multiple
sclerosis will be identified and treated to prevent changes occurring later
in life. For middle-aged and older people, you will be able to determine the
risk profiles for numerous late-onset diseases; ideally, people at risk will
know of their risk before the appearance of symptoms, so that their disease
might, at least, be partially prevented through medical intervention.
The Biological Basis of Uniqueness
This new emphasis on the biological basis of uniqueness, encouraged by the
human genome, brings me to my second point. Our uniqueness is reflected,
in its highest form, in the uniqueness of our mind, a uniqueness that
emerges from the uniqueness of our brain. Now that we understand natural
selection and the molecular basis of heredity, it has become clear that the last
great mystery that confronts biology is the nature of the human mind. This
is the ultimate challenge, not just for biology but for all of science. It is for
this reason that many of us believe that the biology of the mind will be for
the twenty-first century what the biology of the gene was for the twentieth
The biology of mind represents the final step in the philosophical progression
that began in 1859 with Darwin’s insights into evolution of bodily
form. Here, with the biology of mind, we are confronted with the even more
radical and profound realization that the mental processes of humans also
have evolved from animal ancestors and that the mind is not ethereal but can
be explained in terms of nerve cells and their interconnections.
One reason that people have difficulty altering their view of the mind is
that the science of the brain, like all experimental science, is at once mechanistic
in thought and reductionist in method. We have become comfortable
380 Psychiatry, Psychoanalysis, and the New Biology of Mind
with the knowledge that the heart is not the seat of emotions but a muscular
organ that pumps blood through the circulation. Yet some of us still find it
difficult to accept that what we call mind is a set of functions carried out by
the brain, a computational organ made marvelously powerful not by its mystery
but by its complexity, by the enormous number, variety, and interactions
of its building blocks, its nerve cells. We find it difficult to accept that every
mental process, from our most public action to our most private thought, is
a reflection of biological processes in the brain.
With modern imaging and cell-biological studies of brain, we are now
beginning to understand aspects of both our public actions and our private
thoughts: we are beginning to understand how we perceive, act, feel, learn,
and remember. And the insights we so far have obtained are truly remarkable!
For example, these studies show that the brain does not simply
perceive the external world by replicating it, like a three-dimensional photograph.
Rather, the brain reconstructs reality only after first analyzing it
into component parts. In scanning a visual scene, for example, the brain analyzes
the form of objects separately from their movement, and both separately
from the color of the objects, all before reconstituting the full image
again, according to the brain’s own rules. Thus, the belief that our perceptions
are precise and direct is an illusion. We re-create in our brain the external
world in which we live.
We now appreciate that simply to see—merely to look out into the world
to recognize a face or to enjoy a landscape—entails an amazing computational
achievement on the part of the brain that no current computer can
even begin to approach. All of our perceptions and actions—seeing, hearing,
smelling, touching, or reaching for a glass of water—are analytic triumphs.
In addition to creating our perceptions and actions, our brain provides
us with a sense of awareness, it creates for us a historical record, a consciousness
not only of ourselves but of the world around us. Within the family of
deep problems that confront the study of mind, the biology of consciousness
must surely rank at the very top.
The brain can achieve consciousness of self, and can perform remarkable
computational feats because its many components, its nerve cells, are wired
together in very precise ways. Equally remarkable, we now know that the
connections between cells are not fixed but can be altered by experience, by
learning. The ability of experience to change connections in our brain means
that the brain of each person in this audience is slightly different from the
brain of every other person in this audience because of distinctive differences
in our life history. Even identical twins, with identical genomes, will
have slightly different brains because they will invariably have been exposed
to somewhat different life histories.
Genes, Brains, and Self-Understanding 381
The Individuality of Mental Life
It is very likely that during your careers, brain imaging will succeed in resolving
these unique differences of our brain. We will then have, for the first time,
a biological foundation for the individuality of our mental life. If that is so, we
will have a powerful new way of diagnosing behavioral disorders and evaluating
the outcome of treatment including the outcome of psychotherapy.
Seen in this light, the biology of mind represents not only a scientific and
clinical goal of great promise but one of the ultimate aspirations of humanistic
scholarship. It is part of the continuous attempt of each generation of
scholars to understand human thought and human action in new terms.
Personalized Medicine
Your generation—the first postgenomic generation—will have adequate information
from both the human genome and from brain sciences to explore,
more meaningfully than ever before, the genetic contribution to mental processes.
Indeed, we already know that not only psychiatric disorders but almost
all long-standing patterns of behavior—from wearing bow ties to being
socially gregarious—show moderate to high degrees of heritability. The human
genome will thus not only aid in revolutionizing psychiatry and neurology,
but it also will allow us a better understanding of normal behavior—
of how you and I function.
For example, the analysis of genetic polymorphisms may at last uncover
how genetic factors interact with the environment to encourage our various
intellectual capabilities, our mathematical and musical talents and perhaps
even our differing capabilities for creativity, for empathy, and for selfunderstanding.
Whatever the details, we can expect that the genome will reveal
new links between genetics and environment that our society will eventually
have to confront.
As these and other questions are addressed, biology and medicine will
help transform our society as they transform our understanding of the individuals
in society. You will therefore be creating a world in which it is imperative
for each individual to have sufficient understanding of this new
knowledge so that we, as a society, can apply it wisely.
But like all knowledge, biological knowledge is a double-edged sword. It
can be used for ill as well as for good, for private profit as well as for public
benefit. In the hands of the misinformed or the malevolent, natural selection
was distorted into social Darwinism, genetics was corrupted into eugenics.
Brain sciences have also been, and can again be, misused for social control
and manipulation.
This brings me to one final point. We are entering a world that is being
382 Psychiatry, Psychoanalysis, and the New Biology of Mind
changed because of advances in science and in technology and by the social
ramifications of these advances. It will be our obligation to reach out to understand
these advances, to evaluate them, to encourage some and restrict others.
By extension, beyond our own education, we will need to assume the
leadership roles for which you have been trained to ensure the scientific literacy
of the general public, especially the scientific literacy of the patients that you
will be treating.
Kandel’s Laws
Let me then conclude my comments about medicine’s aspirations for a new
humanism by enunciating three principles that I now, in my seniority, invoke
with some frequency. These principles, which I believe reflect some of
my best thinking, are of such importance that I have come to refer to them,
in the modesty and privacy of my own study, as Kandel’s three laws.
Kandel’s first law states that belonging to a university community is one
of the deepest intellectual pleasures of one’s life. Universities are the institutions
that make society great. People from all over the world come to the
United States to study in our universities because the rest of the world sees
the American university as our most extraordinary national product. I will
go further and say that I fully believe there is nothing more important for our
society, and indeed for the world at large, than the two great missions of the
university: to produce new ideas and to train young people to assume responsible
roles in their society.
Belonging to a university assures you that you will be a scholar in perpetuity—
one of the great sensual pleasures of life. Kandel’s first law, you will
appreciate, is not original. I want to remind you that the first medical school
convocation in the American colonies was held at the College of Physicians
and Surgeons, when this college conferred the first M.D. degree in the Americas,
an honorary M.D., for his services to this college, to Samuel Bard, our
first professor of medicine. In his commencement address on May 16,
1769—232 years ago to the very day—Samuel Bard said:
Do not therefore imagine, that from this Time your Studies are to cease; so
far from it; you are to be considered as but just entering upon them; and unless
your whole Lives, are one continued Series of Applications and Improvement,
you will fall short of your Duty…In a Profession then, like that you
have embraced, where the Object is of so great Importance as the Life of a
Man; you are accountable even for the Errors of Ignorance, unless you have
embraced every Opportunity of obtaining Knowledge.
Kandel’s second law is that within the university, teaching is a particularly
rewarding activity. There is no better way to assure yourself that you
Genes, Brains, and Self-Understanding 383
understand an issue than to try to explain it to others. Teaching will guarantee
that you understand the major scientific issues of your time. It will also
give you a perspective on how your thinking and your work fit in with the
rest of medicine.
Kandel’s third law is that patient care is beyond question our most important
responsibility. That is why we are here. Never let patient care take a
secondary role to any other activity in your professional life. Patient welfare
is the ultimate goal of biological science and it is the engine that drives the
whole scientific enterprise. Here, I again want to recall for you Samuel Bard’s
comments of 1769:
In your Behavior to the Sick, remember always that your Patient is the Object
of the tenderest Affection to some one, or perhaps to many about him; it is
therefore your Duty, not only to endeavour to preserve his Life; but to avoid
wounding the Sensibility of a tender Parent, a distressed Wife, or an affectionate
Child. Let your Carriage be humane and attentive, be interested in his
Welfare, and shew your Apprehension of his Danger.
As I hope these three laws make clear, you should leave here confident
that the best days of medical care and the best days of your lives are ahead of
you. As a result of the training you have received at the College of Physicians
and Surgeons of Columbia University, we are confident that you will be able
to influence, through your knowledge and your actions, the emergence of a
new humanism, a humanism made more rational by a deeper respect for the
genome and a greater understanding of the human mind. You are entering
an exciting time in your lives and in the history of medicine, a time that will
afford you the opportunity to benefit your patients, your university, and your
society in novel, important, and humanizing ways. So enjoy the future, and
do it justice.
This page intentionally left blank
Psychotherapy and the
Single Synapse Revisited
Where is psychiatry heading? What areas of psychiatry will benefit most
from biology in the years ahead?
Perhaps the most important and most anticipated advances will come
from the delineation of the genes that render people vulnerable to various
mental illnesses and the characterization of those genes in experimental animals—
in worms, flies, and mice. A second priority is the development of a
new neuropathology of mental illness, one based on knowledge of how specific
molecules in specific regions of the brain make people vulnerable to
specific types of mental illness. A third priority is higher-resolution brain imaging
technologies that will enable us to see anatomical changes in the
brains of mental patients before and after treatment.
Advances along these lines will put us in a position to pinpoint the experiences
that act on genetic and anatomical predispositions to disease. Understanding
mental illness in genetic, anatomical, and experiential terms is
likely to open up new therapeutic approaches. In addition to better drugs,
we may expect better psychotherapies and better ways of selecting therapies
that are effective for specific types of patients. This last theme has interested
me for some time. In fact, the first essay in this volume, “Psychotherapy and
the Single Synapse,” makes the point that the effects of psychotherapy must
ultimately be explained empirically on the level of individual neurons and
their synapses, just as the effects of drugs are.
In view of our progress in the biological understanding of mental disorders,
we can now ask, Is the attempt to evaluate psychotherapy in biological
terms still a profitable endeavor? In the last three decades, we have devel386
Psychiatry, Psychoanalysis, and the New Biology of Mind
oped drugs that are effective in the treatment of a variety of psychiatric disorders:
obsessive-compulsive disorder, anxiety disorders, posttraumatic
stress disorder, depression, bipolar disorders, and the positive symptoms of
schizophrenia. Yet our experience has also made it clear that drugs alone are
often not sufficient treatment. Some patients do better when psychotherapy
is combined with drugs, while other patients do reasonably well with psychotherapy
In her book An Unquiet Mind, Kay Jamison describes the benefits of both
modes of treatment. Lithium prevented her disastrous highs, kept her out of
the hospital, saved her life by preventing her from committing suicide, and
made psychotherapy possible. “But, ineffably,” she writes, “psychotherapy
heals. It makes some sense of the confusion, reins in the terrifying thoughts
and feelings, returns some control and hope and possibility of learning from
it all. Pills cannot, do not, ease one back into reality” (Jamison 1996, p. 89).
Psychotherapy has not only contributed to the treatment of mental illness
it has provided us with a tool for examining the workings of the mind
by peeling back superficial layers of action and revealing the deeper motives.
Until quite recently, there were few independent, compelling ways to test
psychodynamic ideas or to evaluate the relative efficacy of one therapeutic
approach over another. However, neuroimaging may give us just that—a
method of revealing both mental dynamics and the workings of the living
brain. Had imaging been available in 1894, when Freud wrote “On a Scientific
Psychology,” he might well have directed psychoanalysis along very different
lines, keeping it in close relationship with biology, as he outlined in
that essay. In this sense, combining brain imaging with psychotherapy represents
top-down investigation of the mind, which continues the scientific
program that Freud originally conceived for psychoanalysis.
Indeed, one might say that clinical imaging is of even greater importance
to psychiatry than it is to neurology. One obstacle to understanding mental
illness has been the limitations posed by animal models. Most mental illnesses
affect functions that appear to be uniquely human—that is, language,
abstract thought, and complex social interactions. As a result, we cannot as
yet model a number of critical features of mental illness; these can only be
studied successfully in people. Psychotherapy presumably works by creating
an environment in which people learn to change. If those changes are maintained
over time, it is reasonable to conclude that psychotherapy leads to
structural changes in the brain, just as other forms of learning do. Indeed,
we can already image people’s brains before and after therapy and thus see
the consequences of psychotherapeutic intervention in certain disorders.
Preliminary imaging studies have found that obsessive-compulsive disorder
is associated with an increase in metabolism in the caudate nucleus.
Schwartz and his colleagues at the University of California in Los Angeles
Afterword 387
have described how the increased metabolism can be reversed by a form of
psychotherapy called exposure therapy as well as by selective serotonin uptake
inhibitors (Schwartz et al. 1996). Moreover, imaging studies of depression
commonly show a decrease in basal activity in the dorsolateral region
of the prefrontal cortex and an increase in activity in the ventrolateral region.
Psychotherapy and drugs reverse aspects of these two abnormalities, but
here the different modes of treatment affect distinctly anatomical loci in the
brain ones (for a review, see Etkin et al. 2005).
Thus, we may now be able to describe with some rigor the metabolic
changes in the brain that result from drug therapy and those that result from
psychotherapy. Indeed, since a variety of psychotherapies are now in use, it
may be possible to distinguish among them on the basis of the changes they
produce in the brain. It may be that all effective psychotherapies work
through a common set of anatomical mechanisms. Alternatively, they may
achieve their goals through distinctly different processes. Effective psychotherapies
may even have adverse side effects, as drugs do. Describing psychotherapies
in terms of empirical evidence could help maximize the safety
and effectiveness of these important treatments, much as has been done for
drugs. Empirical studies would also help predict the outcome of particular
types of therapeutic interventions and would direct patients to the therapies
most appropriate for them.
It is becoming increasingly apparent that a biological approach to psychiatry
will enable us to reach a deeper understanding of human behavior.
For instance, a number of experimental approaches can be used today to distinguish
conscious from unconscious mental processes. These approaches
are not limited to the implicit unconscious; they can also explore the dynamic
and the preconscious unconscious. One way of doing this is to compare
the brain activation patterns generated by unconscious and conscious
perceptual states (as with the perception of fear) and to identify the regions
of the brain that are recruited by each state.
Finally, biology can sharpen psychiatry’s dual contribution to modern
medicine: its ability to develop effective drug treatments based on neuroscience
and its ability to listen to and learn from patients. We need to combine
these two treatment modalities in ways that are at once objective and
effective. If we are successful in this undertaking, we will join radical reductionism,
which drives biology, with the humanistic goal of understanding
the human mind, which drives psychiatry.
Thus, a half century after I left psychoanalysis because it was unconcerned
with biology, science has progressed to the point where we now have
a rudimentary biology of the mind. The goal for the next decade is twofold:
first, we need to determine how specific combinations of genes give rise to
altered brain anatomy that results in mental illness by increasing vulnerabil388
Psychiatry, Psychoanalysis, and the New Biology of Mind
ity to specific social and environmental experiences. Second, we need to see
how drugs and psychotherapy can complement one another in the treatment
of mental disorders. In taking on this twofold task, biology will be addressing
some of the issues that first attracted so many of my generation to psychiatry
and psychoanalysis. Moreover, with this agenda the new biology of
mind will assume its ascribed role as the natural bridge between the humanities,
which is concerned with the nature of human existence, and the natural
sciences, which are concerned with the nature of the physical world.
Indeed, in the next half, century great universities will be judged by how
successfully they make that bridge and how much they contribute to our
understanding of the human mind. One would hope that psychiatry and
psychoanalysis would be central contributors to this historic effort to understand
the mind in biological terms. If it does succeed in exercising this role,
psychiatry and psychoanalysis will again capture the best and brightest of
the next generations.
My laboratory is supported by the Howard Hughes Medical Institute, the
Fred Kavli Institute for Brain Science, the Lieber Center for Schizophrenia
Research, the Mathers Foundation, and the National Institute of Mental
Health. I have benefited greatly from comments on the introduction and on
the afterword from Amit Etkin, Marianne Goldberger, Christopher Kellendonk,
Henry Numberg, Jonathan Polan, and Eleanor Simpson. I am also
grateful to Howard Beckman and Blair Burns Potter for their editorial advice.
Finally, I thank Dr. Robert Hales, editor-in-chief, and Julia Bozzolo, my book
editor at American Psychiatric Publishing, Inc., for their thoughtful help in
putting this book into its final form.
Etkin A, Pittenger CJ, Kandel ER: Biology in the service of psychotherapy, in The
American Psychiatric Publishing Textbook of Personality Disorders. Edited by
Oldham JM, Skodol AE, Bender DS. Washington, DC, American Psychiatric
Publishing, Inc. 2005, pp 669–682
Jamison K: An Unquiet Mind. New York, Vintage Books, 1996
Schwartz JM, Stoessel PW, Baxter LR Jr, et al: Systematic changes in cerebral glucose
metabolic rate after successful behavior modification treatment of obsessivecompulsive
disorder. Arch Gen Psychiatry 53:109–113, 1996
Page numbers printed in boldface type refer to tables or figures.
Abel, Ted, 360, 361
Academic psychiatry, 33–38, 53
Acetylcholine (ACh), 163, 167, 177,
216, 217, 228, 229
binding sites for, 169
quantal release from presynaptic
terminal, 224
Acetylcholine receptor (AChR),
168–172, 180, 217
assembly of, 170–171
in invertebrates, 172
molecular biology of, 168–169
muscarinic, 171–172
on muscle fibers, 259
myasthenia gravis and
autoantibodies to, 170
nicotinic, 163, 170, 171, 174,
opening and closing of, 168
structure of, 169–170, 216
subunits of, 168–169
monoclonal antibodies to, 170
Acetylcholinesterase (AChE), 168, 177,
ACh. See Acetylcholine
AChE (acetylcholinesterase), 168, 177,
AChR. See Acetylcholine receptor
ACTH (adrenocorticotropic hormone),
83, 178, 179
Action potentials, 134–136, 137, 149,
167, 175, 189, 205, 210, 211–215,
Activation factor (AF), 351,
Addictive behavior, 374
Adenosine triphosphate (ATP), 219
Adenylate cyclase, 136, 150, 189,
Adhesion molecules, neural, 183–184,
Adrenocorticotropic hormone (ACTH),
83, 178, 179
Adrian, Edgar, 211, 212, 265
AF (activation factor), 351, 352
Agnihotri, N., 363
Agnosias, 271
Agoraphobia, 109, 111
Agranoff, Bernard, 359
Agrin, 259
Aguayo, Albert, 261
Alberini, Cristina, 352
Albright, Thomas, 203–317
Aldrich, Richard, 213
Allen, L. S., 90
Allman, John, 271
ALS (amyotrophic lateral sclerosis),
Alzheimer’s disease, 2, 30, 47, 315
apolipoprotein E-4 in, 239–240
attention in, 302
genetic polymorphisms and risk of,
γ-Aminobutyric acid (GABA) receptors,
216, 217, 219
Amnesia. See Memory loss
390 Psychiatry, Psychoanalysis, and the New Biology of Mind
atrophy of, in Alzheimer’s disease,
in fear conditioning, 78, 79
β-Amyloid precursor protein (APP), in
Alzheimer’s disease, 239
Amyotrophic lateral sclerosis (ALS),
Anderson, C. R., 168
Anderson, Richard, 296
Antibodies, monoclonal, 164,
to acetylcholine receptor subunits,
to block neurite outgrowth, 185
detection of molecular
heterogeneity with, 192–193
Antibody diversity, 188–189
Anticipatory anxiety. See Signal
(anticipatory) anxiety
Antidepressants, tricyclic, 107
for panic attacks, 120
Antidisciplines, 7–8
Antipsychotics, 2, 107
acquired, 119–120
actual (automatic), 119
adaptive, 110
animal models of, 109–110,
114–116, 121–124
Aplysia californica, 128–148
from birth trauma, 112
castration, 111
chronic, 109–110, 115, 119, 120,
altered gene expression in,
139–146, 142, 144
in Aplysia, 128, 129
as long-term sensitization,
molecular explanation for,
134–138, 137
molecular model of, 146–147
morphological correlates of,
138–139, 139, 140
shared molecular components
with anticipatory anxiety,
147–148, 149
vs. signal anxiety, 123–124
clinical syndromes of, 119–121
dyspnea and, 111
Freud’s concept of, 109, 111,
119–121, 130–131, 132
insecure attachment and, 80
learned, 114, 120, 122, 122–123
molecular genetic model for
maintenance of, 146–147
neurotic, 79
panic attacks, 109, 111, 112, 120
pathological, 110, 111
pharmacological treatment of, 120
separation, 81–82, 111 (See also
Separation response)
in rodents, 82, 83, 84
signal (anticipatory), 78–79,
109–110, 115, 118, 120–124
animal models of, 109–110,
114–116, 121–124, 128
in Aplysia, 128, 129
behaviorists’ vs. Freud’s views of,
131, 132
in biological adaptation, 123
vs. chronic anxiety, 123–124
learned, 120, 122, 122–123
molecular model for, 148–150
pharmacological treatment of,
shared molecular components
with chronic anxiety,
147–148, 149
as stimulus response, 120–121
stimulus substitution and, 130
stranger, 111
subjective nature of, 128
underlying mechanisms of,
throughout phylogeny, 124–
Anxiety disorders, 109–112, 114–116
behavioral treatments of, 3
clinical nosology of, 109
Index 391
heritability of, 111
neuroimaging in, 48
pharmacotherapy for, 61, 386
ApC/EBP (CAAT box enhancer binding
protein), 352
Aplysia californica, 1, 17–18, 18, 49,
110, 114–116, 124, 128–148
acetylcholine receptors in, 172
advantages of studies in, 343–345
anticipatory and chronic anxiety in,
conditioning of anticipatory
anxiety, 131–132, 133
experimental protocols for
conditioning and
sensitization, 128–130, 129,
molecular explanation for
chronic anxiety, 135–138,
molecular model for anticipatory
anxiety, 148–150
molecular model for
maintenance of anxiety,
morphological correlates of
chronic anxiety, 138–139,
139, 140
pattern of effects created by,
presynaptic facilitation and, 134,
responses of conditioned and
sensitized animals after
training, 129–130, 131
sensitization and, 133–134, 134,
344, 345–346
shared molecular components
of, 147–148, 149
gill-withdrawal reflex in
neural circuitry of, 346–347,
sensitization of, 129, 129–130,
133–134, 134, 140, 344,
molecular mechanisms of,
350, 351
learning and memory in, 195–196,
201–202, 230, 231, 232,
337–339, 343–347, 344
long-term potentiation in
hippocampus, 360
molecular biology of memory
storage, 347–353, 351
CREB-1 mediated
transcription, 352–353
inhibitory constraints, 353
phases of memory storage, 346
synapse specificity of long-term
facilitation, 353–356, 354
number of nerve cells in, 345
peptides in, 179
potassium channels in, 173
Apolipoprotein E-4, in Alzheimer’s
disease, 239–240
Apoptosis, 248
APP (β-amyloid precursor protein), in
Alzheimer’s disease, 239
Arachidonic acid metabolites, 223
Armstrong, Clay, 213, 221
Aromatic L-amino acid decarboxylase,
ATP (adenosine triphosphate), 219
Attachment, 61, 67
anaclitic, 111
Bowlby’s theory of, 81–82
secure vs. insecure, 80
separation response and, 81–83
Attention, 29, 297, 298–304, 316
in Alzheimer’s disease, 302
brain networks concerned with,
301–304, 303
in explicit memory storage, 363
focal vs. ambient, 301, 314
in neglect syndrome, 302, 303
origins of modern study of,
selective, 298–299
visual areas biased by shift of,
299–301, 300
392 Psychiatry, Psychoanalysis, and the New Biology of Mind
Attention deficit disorder, 313–314
Autism, 10, 30, 81
“Average expectable environment,” 8,
Axelrod, Julius, 108, 228
Axons, 206–207
growth and guidance of, 182–183,
chemoaffinity hypothesis of, 249
molecular era of, 251–255, 252
resonance hypothesis of, 249
regeneration of, 261
Bach, Johann Sebastian, 91
Bachrach, H. M., 98
Bacteriorhodopsin, 170–171, 194, 234
Baddeley, Alan, 86–87
Bailey, Craig, 138, 147, 352, 353
Barad, Mark, 360
Barbacid, Mariano, 247
Bard, Samuel, 382, 383
Barde, Yves, 247
Barlow, Horace, 274, 286, 288
Barondes, Sam, 359
Bartsch, Dusan, 352, 353
Bate, Michael, 250
Becker dystrophy, 237
Behavior, 27
addictive, 374
biology and, 118–119
definition of, 126
effect of brain lesions on, 40
gene effects on, 39, 44–46
modification of gene expression by,
39, 46–48
objective study of, 265–266
self-injurious, 374
sexual, 89
twin studies of, 44
visual guidance of, 295–297
voluntary control of, 309
Behavior therapy, 3
for obsessive-compulsive disorder,
Behaviorism, 55, 118, 126
Bennett, Michael, 218
Benzer, Seymour, 163
Benzodiazepines, 107
for anxiety, 120
Betz, Heinrich, 218
Biological basis of uniqueness,
Biological pluralism, 28
Bipolar disorder, 30, 96–97
inherited susceptibility to, 48
pharmacotherapy for, 38, 386
Bisexuality, 87
Bisiach, E., 306
Bleuler, Eugen, 34
Bliss, Tim, 232, 360, 363
β-Blockers, for anticipatory anxiety,
Boring, E. G., 66
Boston Process of Change Study
Group, 72, 93
Bourtchouladze, Rusiko, 360
Bowlby, John, 81, 111
Brain. See also specific brain structures
in Alzheimer’s disease, 239
assembly of neuronal circuits of,
180–186, 240–263
Broca’s regions of, 54–55, 204
Brodmann’s cytoarchitectonic map
of, 264
cell labeling techniques in study of,
cortical-striatal-thalamic system in
obsessive-compulsive disorder,
defining cell types in, 190–192
dorsal anterior cingulate cortex of
in cognition, 312
in executive control,
dynamic models of neural activity
in, 28–29
gene expression in, 28, 187
language region of, 55, 204
macromolecular complexity in,
Index 393
maintenance of learned alterations
in gene expression by
structural alterations in, 49–51,
memory regions of, 56, 70, 71, 83,
mental processes reflecting
functions of, 39, 40–43, 380
mRNA processing in, 187, 188
of musicians, 91–92, 92
networks concerned with attention,
301–304, 303
psychiatric disorders and
anatomical abnormalities of,
21–23, 47
psychotherapy-induced structural
changes in, 3, 60, 91–93, 387
sexual dimorphisms in, 89–90
somatosensory cortex of, 49–51, 51
ventral anterior cingulate cortex of,
in emotion, 312–313
Brain imaging, 28, 29, 39, 48, 61, 200,
205, 298, 299
in anxiety disorders, 48
in attention deficit disorder, 313
in depression, 48, 387
importance to psychiatry, 380, 381,
to monitor progress of
psychotherapy, 52
in obsessive-compulsive disorder,
48, 386–387
in schizophrenia, 48
of unconscious mental processes,
Bremner, J. D., 84
Brenner, C., 74
Brenner, Sidney, 163
Broadbent, Donald, 298–299, 301
Broca, Paul, 54–55, 204
Brodmann, Korbinian, 264
Brunelli, Marcello, 350
Buie, Dan, 36
Burns, B. Deslisle, 230
Byrne, John, 346, 350
C. (Caenorhabditis) elegans
axonal growth in, 251, 254
cadherins in, 253
cell death in, 248
cell lineage of, 181–182
neurogenesis in, 244
CAAT box enhancer binding protein
(ApC/EBP), 352
Cadherins, 253, 254, 257
Calcineurin, effects on memory
storage, 358, 363
Calcitonin, 188
Calcium channels, 158–159, 172–174
in anticipatory anxiety, 149–150
cAMP effects on, 350, 351
in chronic anxiety, 135
in habituation and sensitization, 18,
19, 21
ion selectivity of, 172
membrane topology of, 216, 219
in quantal neurotransmitter release
from presynaptic terminals,
224, 228
voltage-gated, 167, 175
cloning of, 218
Callaway, Edward, 291
Camardo, Joseph, 136, 350
cAMP (cyclic AMP), 19, 134, 136, 137,
144, 146–147, 149, 173, 223, 231,
349, 350, 351, 356, 357
cAMP response element-binding
protein-1 (CREB-1), 231–232,
351, 352–353, 357, 358, 360, 364
cAMP response element-binding
protein-2 (CREB-2), 231–232,
351, 353, 358
Campbell, Kevin, 237
Capecchi, Mario, 359
Carew, Thomas, 128, 345, 347
Carlsson, Arvid, 108
Casadio, Andrea, 353
Caspases, 248
Castellucci, V. F., 136, 346, 347, 350
Castration anxiety, 111
Cataracts, congenital, 11
394 Psychiatry, Psychoanalysis, and the New Biology of Mind
Catecholamines, 82
synthesis of, 176
Catterall, William, 218
Caudate nucleus, in obsessivecompulsive
disorder, 52
Causality, psychological, 78–79
Cedar, Howard, 349
Cell death, 245, 248
Cell lineage, 181–182
Cell theory, 206
Cerebral blood flow, 298, 302
priming and, 310, 311
cGMP (cyclic GMP), 223
Chandler, Knox, 213
Changeux, Jean-Pierre, 163, 217
Channelopathies, 239
Chemotransduction, 195
Chen, M., 138, 147, 352
Child abuse or neglect, 79
Child development
A. Freud’s studies of, 80
“average expectable environment”
and, 8, 80
critical period for, 80, 81
deprivation and, 1, 8–12
early experience and predisposition
to psychopathology, 79–86
emotional regulation and, 313
mother–infant interaction and,
psychoanalytic studies of, 64, 65
psychosexual, 87–88
social, 313
Chloride channels, 167, 172
Choline acetyltransferase, 177
Cingulate cortex
anterior, 314–315
dorsal, in cognition and
executive control,
ventral, in emotion, 312–313
in attention deficit disorder, 313–
in fear response, 79
in schizophrenia, 314
Classical conditioning, 74–75,
122–123, 125–126, 195–196.
See also Learning
in Aplysia, 129–130, 129–130
appetitive, 78, 123
aversive, 110, 122, 123
conditioned and unconditioned
stimuli in, 121–122
contiguity and contingency in,
delay conditioning, 76, 77
psychic determinism of, 75–78
strength of, 75
trace conditioning, 76–77, 77, 87
Clozapine, 2
Clyman, Robert, 72
Codons, 141
Cognition, 27
dorsal anterior cingulate cortex in,
emotion and, 313–314
Cognitive neuroscience, 55, 200
of memory, 55–56
psychoanalysis and, 64–65
Cognitive psychology, 38, 55, 117–118,
127, 204, 205
psychoanalysis and, 64–65
Cognitive science, 29, 30, 38
Cohen, Stanley, 247
Cole, Kenneth, 212, 213
Collingridge, Graham, 232
appetitive, 78, 123
associative, 74, 75, 110, 125–126
aversive, 110, 122, 123
in Aplysia, 129, 129–130
classical, 74–75, 122–123, 125–126,
defensive, 78
delay, 76, 77
fear, 78–79, 115–116
operant, 35
trace, 76–77, 77, 87
Conflicts, unconscious, 71
Congenital adrenal hyperplasia, 89
Index 395
Consciousness, 29, 69, 297–315, 316,
awareness of sensory world and,
imagery, 297, 304–308
orienting of attention to sensory
stimuli, 297, 298–304
(See also Sensory attention)
future study of, 314–315
reductionist and holistic approaches
to, 297–298
subject aspect of, 316–317
vision and, 290
volition and, 297, 309–314
Cooke, Jonathan, 242
Cooper, Arnold M., 59–62, 95, 97–98
Coronary artery disease, 373
Corticotropin-releasing factor (CRF),
in depression, 84–86
early life experience and expression
of gene for, 83
CREB-1 (cAMP response elementbinding
protein-1), 231–232, 351,
352–353, 357, 358, 360, 364
CREB-2 (cAMP response elementbinding
protein-2), 231–232, 351,
353, 358
CRF (corticotropin-releasing factor), 179
in depression, 84–86
early life experience and expression
of gene for, 83
Crick, Francis, 162, 163, 284, 290,
304, 377
Cultural evolution, 43
Curtis, Howard, 212
Cushing’s syndrome, 84
Cyclic AMP (cAMP), 19, 134, 136,
137, 144, 146–147, 149, 173, 223,
231, 349, 350, 351, 356, 357
Cyclic GMP (cGMP), 223
Cytoarchitectonics, 264
Dahl, Hartvig, 66
Dale, Henry, 212
Darwin, Charles, 44, 115, 120, 125,
196, 376–377, 379
Darwinism, social, 41, 381
Dash, Pramod, 352
Day, M., 36
DBH (dopamine β-hydroxylase),
De Camilli, Pietro, 226
Deafferentation, 293–294
Defensive conditioning, 78
Defensive mechanisms, 71
Deiters, Otto, 206, 207
del Castillo, Jose, 224
Delay conditioning, 76, 77
Delbrück, Max, 162
Dementia, 21, 47. See also Alzheimer’s
Dendrites, 206–207, 356
Dennis, W., 10
Depolarization, 167
Depression, 34, 110, 315, 373–374
anaclitic, 9
early experiences and, 79, 84–86
as functional disorder, 21, 47
genetics of, 38, 141
hypothalamic-pituitary-adrenal axis
in, 84
inherited vulnerability to, 79, 141
neuroimaging in, 48, 387
pharmacotherapy for, 38, 61, 386
Deprivation in infancy, 1, 8–12
Harlow’s monkey studies of, 10,
social responsiveness and, 9
Spitz’s studies of, 9–10, 80
visual, 1, 11–12, 13, 14, 16, 282
Descartes, René, 199, 201
Desimone, Robert, 299–301
Detwiler, Samuel, 245
Dexamethasone suppression, in
depression, 84
Diacylglycerol, 223
Diagnostic and Statistical Manual of
Mental Disorders, 200
Dickinson, A., 124, 125
396 Psychiatry, Psychoanalysis, and the New Biology of Mind
DNA, 164, 316–317, 377. See also
clinical testing of, 379
complementary (cDNA), 165
mutations of, 42, 46, 141, 142, 378
recombinant, 164, 166, 244
sequencing of, 28, 164, 377
structure of, 141, 162
template function of, 42, 44
L-Dopa, 176
Dopamine, 28, 350
synthesis of, 176
Dopamine β-hydroxylase (DBH), 176
Dostrovsky, John, 361
Dreams, 73–74
aggressive, 127
biology of, 87–91
Freud’s concept of, 87
libidinal, 111, 127
motivational state and, 87
unconscious, 71
Drosophila, 158, 162, 164, 170, 193,
axonal projection patterns in, 251
cadherins in, 253
genes controlling embryonic body
plan of, 244
learning and memory in, 195–196,
nervous system development in,
P element of, 174
shaker mutant of, 174
Duchenne’s muscular dystrophy, 237
Dynamic polarization, principle of, 207
Dyspnea, 111, 112
Dystroglycoprotein complex, 237
Dystrophin, 237
E. (Escherichia) coli, 162, 163, 164
Early life experiences
attachment and separation
response, 81–83
depression and, 79, 84–86
deprivation in infancy, 1, 8–12
mother–infant interaction, 80
predisposition to psychopathology
and, 79–86
Eccles, John C., 209, 217
Eckstein-Schlossmann, E., 9
Edelman, Gerald, 251, 290
Edelson, Marshall, 67–68
EF1α (elongation factor 1α), 351
Ego, 71, 72, 87
Ego psychology, 59
Eichenbaum, H., 77
Eisenberg, Leon, 111
Eissler, Kurt, 66
Elementism, 283
Ellis, Havelock, 87
Elongation factor 1α (EF1α), 351
children's regulation of, 313
cognition and, 313–314
effortful control of, 313
encoding of, 29
ventral anterior cingulate cortex
and, 312–313
Empathy, 36–37, 61
Ephrins, 252, 254
Epilepsy, 239
Epinephrine, 176
Erikson, Eric, 68
Estrogen, 89
Ethology, 7
Eugenics, 41, 381
Ewalt, Jack, 37
Executive functions, 87, 309
dorsal anterior cingulate cortex and,
priming and, 309–311
volition and, 309
Exposure therapy, 387
Extracellular matrix, in nervous system
development, 185
Fambrough, Douglas, 163
Fatt, P., 167, 168, 215, 217, 224, 356
Fear, 114. See also Anxiety
Index 397
as actual anxiety, 119
learned, 110, 114
phobic, 111
of recurrence of panic attack, 109
Fear conditioning, 78–79, 115–116
Fechner, Gustav, 265
Feldberg, Wilhelm, 212
Fibroblast growth factor, 242, 243, 248
Fight-flight response, 79, 115, 121
Fishbach, Gerald, 375
Flexner, Abraham, 98
Flexner, Louis, 359
Flexner Report, 98
Fluoxetine, for obsessive-compulsive
disorder, 52
fMRI (functional magnetic resonance
imaging), 3, 48, 298
Forbes, Alexander, 347
Form-cue invariance, 286
Fragile X syndrome, 237, 238
Frank, Karl, 209
Frazier, S. H., 47
Free association, 65, 74–75
Freud, Anna, 80, 81
Freud, Sigmund, 35, 36, 54, 59, 67, 68,
69, 95, 111, 115, 116, 271
“Beyond the Pleasure Principle,” 63
concept of anxiety, 109, 111,
signal anxiety, 78–79, 115,
130–131, 132
concept of drives, 87
concept of psychic determinism, 73,
74, 75
concept of sexual orientation and
psychosexual development,
concept of unconscious, 70–72
“Mourning and Melancholia,” 79
“On a Scientific Psychology,” 386
“On Narcissism,” 63
psychoanalytic method of, 65
Frey, Uwe, 360
Functional magnetic resonance
imaging (fMRI), 3, 48, 298
Functional psychiatric syndromes, 21,
23, 47
Fuortes, Michael, 209
Furshpan, Edwin, 218
Fuster, Joaquin, 87
G protein–coupled receptors, 223,
GABA (γ-aminobutyric acid) receptors,
216, 217, 219
GAD (generalized anxiety disorder),
109, 110
Gap junctions, 175
Gender, genotypic vs. phenotypic, 88,
89, 90
Gender identification, 88
Generalized anxiety disorder (GAD),
109, 110
Genes, 377
for acetylcholine receptor, 168
alleles of, 46
for apoptosis, 248
behavioral modification of, 39,
brain expression of, 28, 187
chronic anxiety and altered
expression of, 139–146, 142,
cloning of, 164–165
in control of neuronal identity,
determination of neural
connections by, 27, 28, 39
effects on behavior, 39, 44–46
epigenetic regulation of, 42–43
of human genome, 28, 187
learning and altered expression of,
27, 28, 39, 42–43, 93, 112,
structural alterations in brain
circuitry and, 49–51, 51
mapping of, 28, 29
mental illness and, 44–45
molecular biology of disease,
398 Psychiatry, Psychoanalysis, and the New Biology of Mind
Genes (continued)
mutations of, 42, 46, 141, 142,
neural subtype–specific, 262
penetrance of, 46
alterations of, 51–52
polymorphisms of, 378–379, 381
psychotherapy-induced alterations
of, 27, 28, 39, 51–52
regulation of transcription of,
structure of, 42, 43
template function of, 42, 44
transcriptional function of, 42, 43
environmental factors and, 48–49
transposon tagging of, 174
Genetic medicine, 373, 376, 381–382
Genetic studies, 27–29, 30, 37, 38
of Alzheimer’s disease, 239–240
of amyotrophic lateral sclerosis, 238
of attention deficit disorder, 314
of Becker’s muscular dystrophy, 237
of behavior, 44
of channelopathies, 239
in Drosophila, 244
of Duchenne’s muscular dystrophy,
of fragile X syndrome, 237, 238
of Huntington’s disease, 46, 237,
of inherited and acquired
psychiatric disorders, 48–49,
of monogenic neurological diseases,
of Parkinson’s disease, 238
of schizophrenia, 38, 44–46, 141,
of sexual orientation, 91
of trinucleotide repeat disorders,
of vulnerability to psychiatric
disorders, 29, 48
Genetic testing, 379
biological basis of uniqueness,
individuality of mental life, 381
personalized medicine based on,
376, 377–378, 381–382
revolution in, 375–379
Genotypic gender, 88, 89, 90
Genotyping, 28
Gestalt psychology perspective
of learning, 347
of perception, 204–205
of visual function, 283–284
Gettner, Sonya, 296
Gilbert, Charles, 293–294
Glanzman, David, 350
elevation of
in Cushing’s syndrome, 84
in depression, 84
hippocampal function and, 83,
receptors for, 83, 84
stress-induced release of, 82, 83, 85
Glucose metabolic rate, cerebral, in
obsessive-compulsive disorder, 52
Glutamate, 350
Glutamate receptors, 216, 217, 219,
222, 232, 234, 236, 256
N-methyl-D-aspartate receptors,
219, 236, 256
in long-term potentiation, 234,
360, 361
Glycine receptors, 216, 217, 219
Goelet, Philip, 352
Goldberg, Michael, 299
Goldberger, Marianne, 72
Goldman-Rakic, Patricia, 87
Golgi apparatus, 175
Golgi, Camillo, 206, 207, 264
Gonadal development, 88–89
Goodman, Corey, 250
Gorski, R. A., 90
Gottesman, I. I., 45
Gouaux, Eric, 222
Index 399
Gradient morphogen signaling, 243
Gray, Charles, 288–289
Graziano, Michael, 296–297
Greengard, Paul, 157, 222, 226, 350
Grobstein, Clifford, 242
Gross, Charles, 277, 296–297
Gusella, James, 237
Habituation, 1, 16–19
animal studies of flexionwithdrawal
reflex, 16–18, 18
definition of, 16
long-term, 18–19, 20
sensitization after, 19, 22
memory for, 18
of orienting response to new
stimulus, 16
short-term, 18, 24
Hallucinations, 307
Hamburger, Viktor, 185, 245–246
Haplotype mapping, 28, 29
Harlow, Harold, 1, 10, 80–81, 82, 111
Harlow, Margaret, 10
Harnish, R., 305
Harrison, Ross, 207
Hartline, Keefer, 267–268
Hartmann, Ernest, 36
Hartmann, Heinz, 68
concept of “average expectable
environment,” 8, 80
Hawkins, Robert, 136, 345, 346, 347,
350, 361
Hebb, Donald, 288, 347
Hedgehogs, 243, 248
Heineman, Steven, 218
Hendricks, Ives, 2
Heston, L. L., 45
Hille, Bertil, 213, 221
atrophy of
in Alzheimer’s disease, 239
in Cushing’s syndrome, 83
in depression, 84
glucocorticoid exposure and, 83,
memory loss and, 83, 84
in posttraumatic stress disorder,
glucocorticoid receptors in, 83
neural circuitry of, 208
place cells in, 234, 235, 361
contextual learning and stability
of, 361–363, 362
pyramidal cells of, 342–343, 361
role in explicit memory, 56, 70, 83,
231, 232–234, 342–343,
359–363, 365
long-term potentiation, 232–
234, 233, 358, 360–363
murine model of, 359–363, 362
Schaffer collateral pathway of, 232–
234, 233, 358, 361
in trace conditioning, 76, 77
Hirschfeld, Magnus, 87
Hobson, Alan, 36
Hochner, Binyamin, 350, 352
Hodgkin, Alan, 167, 168, 212, 213,
215, 221, 234
Hokfelt, Thomas, 229
Holistic approach, 186, 201, 204–205,
Holmes, Gordon, 269, 279
Homeodomain proteins, 244–245
anatomical studies of, 90–91
Freud’s concept of, 87–88
twin studies of, 91
Horenstein, Sam, 2
Hormones, gonadal, 89
Horvitz, Robert, 248
“Hospitalism,” 9
HPA (hypothalamic-pituitary-adrenal)
in depression, 84
in panic attacks, 112
stress-induced activation of, 82, 83
Huang, Yan-You, 360
Hubel, David, 1, 11–12, 255, 265,
269–271, 270, 274, 277, 278, 281,
282, 283, 284, 316
400 Psychiatry, Psychoanalysis, and the New Biology of Mind
Human Genome Project, 28
Humanism, 383
Hunt, Thomas, 94
Huntingtin, 238
Huntington’s disease, 46, 237, 238
Huxley, Andrew, 167, 168, 212, 213,
215, 234
Hybridoma technology, 166, 192
Hyden, Holger, 347
5-Hydroxytryptamine. See Serotonin
Hyman, Steven E., 199–203
Hyperkalemic periodic paralysis, 239
Hypertension, 373
Hypothalamic-pituitary-adrenal (HPA)
in depression, 84
in panic attacks, 112
stress-induced activation of, 82, 83
anterior, interstitial nuclei of, 89–90
in fear response, 79
in sexual behavior, 89
sexual dimorphism of, 89–90
Id, 71
Imagery, 297, 304–308
effect of right posterior parietal
lesions on, 306–307
future studies of, 307–308
objective tests of, 305
pathological, 307
Imipramine, 109
Individuality of mental life, 381
Infantile amnesia, 81
Infantile sexuality, 64
Information-processing mechanisms,
Insel, Thomas R., 27–30
Integrins, 253–254
Interleukin-6, 248
International Haplotype Mapping
Project, 28
International Psychoanalytical
Association, 61
Interpretation, psychoanalytic, 65, 111
Ion channels and channel proteins,
166–175, 213–214, 234, 236
acetylcholine receptor, 168–172
calcium channels, 158–159,
cAMP effects on, 350, 351
in channelopathies, 239
chloride channels, 167, 172
conductance of, 214
leakage channels, 167
ligand-gated, 167, 215–219, 216
cloning of, 218
in prokaryotes, 222
superfamilies of, 219
nongated vs. gated, 166
potassium channels, 167, 171,
172–174, 212, 213, 219–222
second messengers in regulation of,
selectivity filter of, 172, 213,
sodium channels, 165, 167, 168,
171, 172–173, 180, 212,
structure of, 214, 236
voltage-gated, 167, 213, 216
cloning of, 218–219
patch-clamp technique for study
of, 215
voltage-clamp technique for
study of, 214–215
Ionic hypothesis of neuronal signaling,
167, 168, 212–213, 215
Ionotropic receptors, 222–223, 224,
Jacob, Francois, 69
James, William, 120, 125
Jamison, Kay, 96, 386
Jan, Lily Yeh, 218
Jan, Reinhard, 226
Jan, Yuh Numg, 218
Jeffrey, D. W., 96
Jessell, Thomas M., 203–317
Jung, Carl, 34
Index 401
Kaas, John, 271
Kahana, Ralph, 34
Kallmann, Franz, 37, 44–45
Kamin, Leon, 75, 126
Kandel, Eric
“Biology and the Future of
Psychoanalysis,” 63–99
commentary on, 59–62
“Genes, Brains, and Self-
Understanding,” 375–383
commentary on, 373–374
“From Metapsychology to
Molecular Biology,” 117–151
commentary on, 107–112,
“The Molecular Biology of Memory
Storage,” 341–366
commentary on, 337–339
“Neural Science,” 203–317
commentary on, 199–202
“Neurobiology and Molecular
Biology,” 161–197
commentary on, 157–159
“A New Intellectual Framework for
Psychiatry,” 33–56
commentary on, 27–30
“Psychotherapy and the Single
Synapse,” 5–24
afterword on, 385–388
commentary on, 1–3
three laws of, 382–383
Kantrowitz, J. L., 98
Kaplan-Solms, Karen, 95
Karlin, Arthur, 217
Katz, Bernhard, 167, 168, 212, 213,
215, 217, 224, 234, 356
Katz, Lawrence, 291
Kefauver-Harrison Drug Amendments
of 1962, 108
Kelly, Regius, 163
Kentros, Cliff, 361, 363
Kety, Seymour, 37, 38
Klein, Donald F., 107–112
Klein, Mark, 136, 350
Kleinian theory, 59
Koch, Christof, 284, 290
Koffka, Kurt, 283
Köhler, Wolfgang, 283, 347
Konorski, Jerzy, 274
Krafft-Ebing, Richard, 87
Kris, Anton, 36, 65
Kuffler, Stephen, 268, 269, 277
Kunkel, Louis, 237
Kupfermann, Irving, 345, 346, 347
Landmesser, Lynn, 249
Langley, John, 211, 212, 249
Language, 29, 128
brain region for, 55, 204
Lashley, Karl, 35, 347
Lateral geniculate nucleus (LGN), 269,
271, 278, 279
Lear, Jonathan, 68, 69
Learned helplessness, 110
Learning, 2, 8. See also Memory
in adaptation to evolutionary
pressure, 124–125
in adult, 12–21
alterations of gene expression by,
39, 42–43, 46–48, 93, 112, 140
animal studies of, 49
maintenance by structural
alterations in neural
circuitry, 49–51, 51
in Aplysia, 195–196, 201–202, 230,
231, 232, 337–339, 343–347,
appetitive, 78, 123
associative, 74, 75, 110, 120,
aversive, 110, 122, 123
classical conditioning, 74–78
cultural evolution and, 43
definition of, 13, 195
in Drosophila, 195–196, 232
effects on neural connections, 1–2,
17–21, 23, 47, 338, 347,
from experience, 15
fear, anxiety and, 114, 120, 122
402 Psychiatry, Psychoanalysis, and the New Biology of Mind
Learning (continued)
gene products involved in, 115
habituation in, 1, 16–19
neurobiology of, 195–196
operant conditioning, 35
perceptual, 293
psychotherapy-induced behavioral
change and, 39
reductionist approach to, 342–343,
sensitization in, 1, 19–21, 346
synaptic transmission and, 1–2,
17–21, 23, 47, 338
LeDoux, Joseph, 79, 114–116
Letourneau, P. C., 183
Lettvin, Jerome, 274
LeVay, Simon, 90, 256
Levi-Montalcini, Rita, 185, 245–246
Levine, S., 82
Levinthal, Cyrus, 163
Lewis, Edward, 244
LGN (lateral geniculate nucleus), 269,
271, 278, 279
Linkage analysis, 29
Lithium, 2, 96–97, 386
Llinas, Rodolfo, 224
Locke, John, 74
Loewi, Oto, 212
Lømo, Terje, 232, 360
Long-term potentiation (LTP), 83,
232–234, 233, 358, 360–363
early phase of, 358, 360
late phase of, 358, 360
explicit memory and, 360–363,
N-methyl-D-aspartate receptors in,
234, 360, 361
murine model of, 359–363, 362
protein kinase A in, 360–363
Lorente de Nó, Rafael, 230, 281,
Lou Gehrig’s disease, 238
Love, neurobiology of, 29
LTP. See Long-term potentiation
Luzzatti, C., 306
Mach, Ernst, 283
MacKinnon, R., 221–222
Mackintosh, N. J., 126
Magnetic resonance imaging,
functional (fMRI), 3, 48, 298
Malleret, Gael, 363
Managed care, 2, 54
Mangold, Hilde, 242
Manic-depressive disorder. See Bipolar
Mansuy, Isabelle, 363
MAOIs (monoamine oxidase
inhibitors), 107
for panic attacks, 120
MAPK (mitogen-activated protein
kinase), 231, 351, 352, 357, 358,
360, 364
Marr, David, 273
Marshall, Wade, 279
Martin, Kelsey, 353
Maternal deprivation, 8
Harlow’s monkey studies of, 10
Maturana, Humberto, 274
Mayeri, Earl, 179
Mayford, Mark, 292
Mayr, Ernst, 37
McEwen, Bruce, 83
Medial temporal lobe
in delay conditioning, 76
in explicit memory, 56, 70, 231
Memory, 2, 15, 29, 55, 114.
See also Learning
in Aplysia, 195–196, 201–202, 230,
231, 232, 337–339, 343–347,
brain regions involved in, 56, 70,
cognitive neural science of, 55–56
definition of, 195
in Drosophila, 195–196, 232
explicit (declarative), 55–56, 70, 71,
231, 356
complexity of, 364
inhibitory constraints on,
Index 403
late phase of long-term
potentiation and, 360–363,
medial temporal lobe and
hippocampal mediation of,
56, 70, 231, 232–234,
342–343, 359, 365
short- and long-term phases of,
slow development of, 81
storage of, 356–363, 358
for habituation, 18
implicit (procedural), 55, 56, 70,
71, 231, 356
fear conditioning and, 78
in infancy, 81
maternal separation–induced
stress and, 84
moral development and, 72
priming in, 56, 70, 71
procedural unconscious and, 72
psychoanalysis and, 72–73
storage of, 341–356
long-term, 144, 231–232, 364
CREB-1 in, 352–353
inhibitory constraints on, 353
psychic determinism and, 73–78
reductionist approach to, 342–343,
short-term, 144, 195–196, 231, 364
systems problems in study of, 240
taxonomy of, 71
working, 86–87
Memory loss
in Alzheimer’s disease, 239
in depression, 84
hippocampal atrophy and, 83, 84
infantile amnesia, 81
in posttraumatic stress disorder, 84
Memory storage, 341–366
alternative theories of, 347
in Aplysia, 343–347, 344, 348
conservation across species, 338
of explicit memory, 356–363
attention and, 363
calcineurin as negative regulator
of, 358, 363
long-term potentiation in
hippocampus, 83, 232–234,
233, 358, 360–363
mechanisms of, 229–232
overview of, 363–365
phases of, 346, 364
presynaptic facilitation in, 1, 19, 22,
110, 350–352
reductionist approach to, 342–343
of short- and long-term memory,
molecular biology of,
CREB-1 mediated transcription,
inhibitory constraints, 353
signaling pathways for, 364
synapse specificity of long-term
facilitation, 353–356, 354
neurotransmitter regulation of
local protein synthesis, 356,
Mendel, Gregor, 94, 377
Mental processes, 35, 38
genetic contribution to, 381
individuality of mental life, 381
localization of biological source of,
neural activity and, 28–29
as reflection of brain functions, 39,
40–43, 380
unconscious, 29, 34, 35, 56, 60, 64,
69, 70–73 (See also
Mentalization, 61
Merzenich, Michael, 293
Mesoderm-inducing signals, 242
Messenger RNA (mRNA), 42, 48, 141,
in brain, 187, 188
polyadenylated, 187
Metabotropic receptors, 222–223, 224,
Metapsychology, 64
404 Psychiatry, Psychoanalysis, and the New Biology of Mind
N-Methyl-D-aspartate (NMDA)
receptors, 219, 236, 256
in long-term potentiation, 234, 360,
Miledi, Ricardo, 168, 224
Miller, N. E., 123
Milner, Brenda, 70
biological basis of uniqueness,
psychoanalysts’ arguments for and
against a biology of, 67–69
psychoanalytic method and
psychoanalytic view of, 6,
Mind–brain relationship, 28–29, 39,
Mishkin, Mortimer, 279
Mitogen-activated protein kinase
(MAPK), 231, 351, 352, 357, 358,
360, 364
Molecular biology
background of, 162–165
of disease, 237–240
of memory storage, 341–366
from metapsychology to, 117–151
neurobiology and, 161–197
Molecular genetics, 94
Molecular neurobiology, 166–197
channel proteins, 166–175
generation of macromolecular
complexity in brain, 186–193
nervous system development,
perception, behavior, and learning,
synaptic transmitters, 175–180
Moments of meaning, in
psychoanalysis, 72, 78
Monoamine oxidase inhibitors
(MAOIs), 107
for panic attacks, 120
Monoclonal antibodies, 164, 166
to acetylcholine receptor subunits,
to block neurite outgrowth, 185
detection of molecular
heterogeneity with, 192–193
Montarolo, Pier Giorgio, 352
Moral development, 72
Morgan, Thomas Hunt, 162, 377
Morris, Richard, 360
Mother, surrogate, 10, 81
Mother–infant interaction
attachment and, 80, 81
separation response and, 81–83
Motivation, 55, 374
drives and, 87
psychoanalytic concept of, 64, 68
Motor neurons, 209–211
Mountcastle, Vernon, 265, 281, 316
Mowrer, O. H., 123
MRI (magnetic resonance imaging),
functional, 3, 48, 298
mRNA (messenger RNA), 42, 48, 141,
in brain, 187, 188
polyadenylated, 187
Muller, Robert, 361
Munk, Hermann, 267, 269, 279
Muscarinic receptor, 171–172
Muscular dystrophies, 237
Myasthenia gravis, 170
Myotonias, 239
Nachmansohn, David, 163
Nagel, T., 316
Natural selection, 44, 377, 381
NCAMs (nerve cell adhesion
molecules), 183–184, 251–253
Neely, James, 310
Neglect, visual, 302, 303, 306
Neher, Erwin, 168, 215, 218
Neilson, D. R., Jr., 17
Nemeroff, Charles, 83, 84
Nerve cell adhesion molecules
(NCAMs), 183–184, 251–253
Nerve growth factor (NGF), 185–186,
246, 247
Nervous system
Index 405
antigenic heterogeneity of cells of,
cell types in, 190–192
development of (See Neural circuit
diversity of cells of, 188–190,
integrative function of, 209–211
single unit analysis in, 211
Nestler, Eric J., 157–159
Neural circuit assembly, 180–186, 240–
263. See also Synaptic connections
axonal growth and selective
connections in, 182–183, 229,
240, 249–251, 250
chemoaffinity hypothesis of, 249
molecular era of, 251–255, 252
resonance hypothesis of, 249
cell lineage and, 181–182
control of neuronal survival and,
245–248, 246
extracellular matrix proteoglycan
in, 185, 253
function and, 262–263
future studies of, 259–263
in hippocampus, 208
inductive signaling, gene
expression, and control of
neuronal identity in, 241–245
learning and structural alterations
in, 49–51, 51
nerve cell adhesion molecules in,
183–184, 251–253
neurological disease and, 260–261
neurotrophins and, 185–186, 246,
selection and refinement of
neuronal connections in,
steps in, 241
Neural inducing factors, 242–243
Neural plasticity, 229–234
Neuregulins, 259
Neurexins, 258
Neuroanatomy, 264. See also Brain
Neurofibrillary tangles, in Alzheimer’s
disease, 239
Neuroimaging. See Brain imaging
Neuroleptics, 2, 107
Neuroligins, 258
Neurological disorders, 30
Neuromuscular junction, 258–259
Neuron doctrine, 206–207, 265, 286
Neuronal diversity, 188–190, 241–243
Neuronal recognition, 189–190
Neuronal regeneration, 261
Neuronal structure, 206–207
Neuropathology of psychiatric illness,
48, 385
Alzheimer’s disease, 239
Neuropeptide Y (NPY), 229
Neurophysiology, 265
Neuropsychology, 263–264
Neuroscience, 203–317
biological basis of uniqueness,
cognitive, 55, 200
of memory, 55–56
psychoanalysis and, 64–65
of consciousness, 297–315
future of, 315–317
genomics and, 376
goal of, 203
holistic approach to, 186, 201,
misuse of, 41, 381
reductionist approach to, 186, 201,
204, 205, 234–236, 283, 337,
systems, 28, 29, 263–297
computation in, 266
definition of, 263
neuroanatomy in, 264
neurophysiology in, 265
neuropsychology in, 263–264
psychophysics in, 265–266
vision as model system of,
266–267 (See also Visual
training in, 3, 53
406 Psychiatry, Psychoanalysis, and the New Biology of Mind
Neuroscience–psychiatry relationship,
6–7, 30, 34–35, 385–388
antidisciplines and, 7–8
biology and the future of
psychoanalysis, 54–56, 63–99
(See also Psychoanalysis)
complementarity of, 7
new framework for, 39–51
implications for psychiatric
practice, 52–54
psychiatric competence and, 54
rapprochement of, 38–39, 53, 114
synergism in, 8
Neuroses, 21, 34, 47, 109
Neurotransmission, 175
Neurotransmitters, 28, 175–180. See
also specific neurotransmitters
exocytic release from synaptic
vesicles, 224–226
habituation and release of, 17–19
inactivation of, 228–229
ionotropic receptors for, 222–223,
224, 356
membrane transporters for,
metabotropic receptors for,
222–223, 224, 356
modulation of ion channels by, 173
peptide, 176–180, 229
quantal release from presynaptic
terminals, 224–228, 225
in regulation of learning, 364
regulation of local protein synthesis
by, 356, 357
sensitization and release of, 19–21
small molecule, 176
synthesis of, 176, 189
Neurotrophins, 185–186, 246, 247,
New York Psychoanalytic Institute, 95,
NGF (nerve growth factor), 185–186,
246, 247
Nguyen, Peter, 360
Nicoll, Roger, 232
Nicotinic receptor, 163, 170, 171, 174,
Nieuwkoop, Peter, 242
Nirenberg, Marshall, 163
NMDA (N-methyl-D-aspartate)
receptors, 219, 236, 256
in long-term potentiation, 234, 360,
Norepinephrine, 176, 228, 229
Norman, D. A., 309
NPY (neuropeptide Y), 229
Nucleus basalis, in Alzheimer’s disease,
Numa, Shosaku, 169–170, 218
Nüsslein-Vollhard, Christine, 244
Obsessive-compulsive disorder (OCD),
cortical-striatal-thalamic brain
system in, 52
exposure therapy for, 387
neuroimaging in, 48, 386–387
pharmacotherapy for, 61, 386,
symptoms of, 52
treatment of, 52
Odorant receptors, 258
Oedipal complex, 81
O’Keefe, John, 361
Oldham, John M., 373–374
Olds, David, 64, 95
Olson, Carl, 296
Operant conditioning, 35
Opioid peptides, 178, 179, 188
Organic mental illness, 21–23, 47
Ostow, Mortimer, 93, 95
Palade, George, 207
Palay, Sanford, 207
Panic attacks, 109, 111, 112, 120
Parada, Luis, 247
Paramyotonia congenita, 239
Parietal lobe
localization of alerting and emotion
function in, 308
Index 407
neglect syndrome due to lesions of,
302, 303
imagery and, 306
Parkin, 238
Parkinson’s disease, 30, 238
Patch-clamp technique, 215
Pavlov, I., 74, 75, 78, 108, 110, 115,
120, 121, 122–123, 130
Peptide transmitters, 176–180, 229
Perception, 55, 380
Gestalt psychology view of,
neuronal substrates for, 284–285
visual (See Visual processing)
Perceptual learning, 293
Periodic paralysis, 239
Personality disorders, 200
Personalized medicine, 376, 377–378,
Perutz, Max, 238
PET (positron emission tomography),
29, 48, 298
Phenotype, 42
Phenotypic gender, 88, 89, 90
Phenylethanolamine Nmethyltransferase
(PNMT), 176
Phobias, 109, 111
Pinsker, Harold, 345
PKA (protein kinase A), 231, 350, 351,
352, 356, 358, 364
in long-term potentiation, 360–363,
calcineurin effects on, 358, 363
PKC (protein kinase C), 231
Placebo treatment, 107–108
Plotsky, Paul, 83
PNMT (phenylethanolamine N-
methyltransferase), 176
Polymorphisms, genetic, 378–379, 381
Polyproteins, 177–179
Positron emission tomography (PET),
29, 48, 298
Posner, Michael I., 203–317
Posttraumatic stress disorder (PTSD),
fear conditioning and, 78
hippocampal atrophy and memory
loss in, 84, 85
inherited vulnerability to, 48
pharmacotherapy for, 386
stressful experience and incidence
of, 79–80
Potassium channels, 167, 171,
172–174, 212, 213, 219–222
calcium-activated, 167
cAMP effects on, 350, 351
characterization of, 174
in chronic anxiety, 135–138, 137,
146, 147
cloning of, 218
delayed, 167, 350
early, 167, 173–174
inward-rectifying, 219–221, 220
S-type, 350
selectivity filter of, 213, 221–222
structure of, 219–222, 220
Potter, David, 218
Preconscious unconscious, 71–72, 78,
Prefrontal cortex
in fear response, 79
preconscious unconscious and,
Presenilin 1 and presenilin 2, in
Alzheimer’s disease, 239
Presynaptic facilitation, 1, 19, 22, 110,
in Aplysia, 134–136, 135, 137
cAMP and protein kinase A in,
serotonin in, 19, 134, 136, 150, 231,
350–352, 351
automatic, 310
in procedural memory, 56, 70, 71
reaction time and, 309–310
semantic, 310–311
Principle of dynamic polarization,
Pro-opiomelanocortin, 178
408 Psychiatry, Psychoanalysis, and the New Biology of Mind
Procedural unconscious, 71, 72–73
moral development and, 72
psychic determinism and, 73–78
psychoanalysis and, 72–73
Prodynorphin, 178
Proenkephalin, 178
Progesterone, 89
Prokasy, W. F., 126
Propranolol, for anticipatory anxiety,
Prosopagnosia, 274–277
Protein kinase A (PKA), 231, 350, 351,
352, 356, 358, 364
in long-term potentiation, 360–363,
calcineurin effects on, 358, 363
Protein kinase C (PKC), 231
Protein phosphorylation, 159, 173,
188, 196, 223, 363
Psychiatric disorders
classification of, 21, 47
diagnosis of, 108–109, 200
early experience and predisposition
to, 79–86
functional, 21, 23, 47
genetic studies of, 44–45
inherited vulnerability to, 29, 48,
79, 141
neuropathology of, 48, 385
organic, 21–23, 47
psychic determinism in, 73, 75
psychological causality and, 78–79
relationship between inherited and
acquired disorders, 48–49, 50
stigmatization of, 35
synaptic connections in, 21–23, 141
therapeutic approaches to, 385
Psychiatric genetics, 29, 30, 37–52.
See also Genes; Genetic studies
Psychiatric training, 3, 27, 29–30, 34,
biological orientation in, 53–54
neurobiology as antidiscipline of,
new intellectual framework for,
commentary on, 27–30
tension within, 6–7
Psychic determinism, 64, 69, 73–78
of classical conditioning, 75–78
definition of, 73
in psychopathology, 73, 75
Psychoanalysis, 3, 6, 7, 34–36, 38
Analysts’ arguments for and against
a biology of mind, 67–69
antidisciplines of, 7
biology and the future of, 54–56,
commentary on, 59–62
biology in the service of, 69–93
early experience and
predisposition to
psychopathology, 79–86
outcome of therapy and
structural brain changes,
91–93, 92
preconscious unconscious and
prefrontal cortex, 86–87
psychic determinism, 73–78
psychological causality and
psychopathology, 78–79
psychopharmacology and
psychoanalysis, 93
sexual orientation and biology of
drives, 87–91
unconscious mental processes,
70–73, 71
concept of anxiety, 111
declining influence of, 64, 68, 127
dialogue between biology and,
evaluating outcome of, 96–98
hermeneutic view of, 68
history of, 64
interpretation in, 65, 111
isolation from academic setting, 61
lack of scientific methodologies in,
64, 66–67, 127
limitations and decline of, 35–36
Index 409
in medical illness, 34
moments of meaning in, 72, 78
privacy of communication in, 66
psychoanalytic method and
psychoanalytic view of the
mind, 65–66
psychopharmacology and, 61, 93
research in, 61
separation from neuroscience, 35
susceptibility to observer bias in,
theoretical pluralism of, 59
therapeutic effect of, 69
training in, 67, 98–99
Psychoanalytic institutes, 61, 66, 67,
95, 96, 98–99
Psychodynamic therapy, 3, 29, 34
outcome studies of, 61
Psychological causality, 78–79
Psychology, 7–8
behaviorist, 55, 118, 126
cognitive, 38, 55, 117–118, 127,
204, 205
psychoanalysis and, 64–65
evolution of, 55
experimental, 265
Psychopharmacology, 2, 6, 29, 38,
107–108, 200, 386
alterations in gene expression,
psychoanalysis and, 61, 93
psychotherapy and, 52, 386, 388
Psychophysics, 265–266
altered gene expression in, 141
heritability of, 111, 141
toxic, 21, 47
Psychosexual development, 87–88
Psychosomatic illnesses, 34
Psychotherapy, 200, 385–387
alterations in gene expression by,
27, 28, 39, 51–52
brain imaging to monitor progress
of, 52
brain structural changes induced by,
3, 60, 91–93, 387
empirical studies of, 387
mechanisms of effect of, 386–387
by nonmedical specialists, 54
pharmacotherapy and, 52, 386, 388
psychodynamic, 3, 29, 34
and the single synapse, 5–24
commentary on, 1–3
PTSD. See Posttraumatic stress disorder
Purkinje cells, 190
Racism, 41
Ramachandran, V. S., 285
Ramón y Cajal, Santiago, 206–209,
208, 212, 229–230, 249, 281, 347
Rapoport, Judith L., 1–3, 36
Rayner, R., 123
Receptors. See specific receptors
Reductionist approach, 337, 339
to consciousness, 297
to learning and memory, 342–343,
to signaling in nerve cells, 186, 201,
204, 205, 234–236, 283
Reichardt, Louis, 163
Reiser, M., 68
Relational theory, 59
Religion and biology, 377
Repression, neurobiology of, 29
Rescorla, Robert A., 126
Resonance hypothesis of axonal
growth, 249
Retinal ganglion cells, 278, 279
Retinotectal projections, 249, 250,
253, 254
Riesen, Austin, 11
RNA, 28
messenger (mRNA), 42, 48, 141,
in brain, 187, 188
polyadenylated, 187
RNA polymerase, 42
Robbins, Lew, 107
Robins, Eli, 37, 38
410 Psychiatry, Psychoanalysis, and the New Biology of Mind
Romanes, G. J., 120
Rosenthal, D., 45
Rotenberg, Alex, 361
Rothman, James, 226
Rutishauser, Urs, 251
Sachar, E. J., 84
Sakmann, Bert, 168, 215, 218
Sander, Louis, 72, 93
Sapolsky, Robert, 83
Schacher, Samuel, 352
Schafer, Edward, 267, 269
Scheller, Richard, 226
Schildkraut, Joseph, 36
Schizophrenia, 30, 34, 36, 315
anterior cingulate cortex in, 314
as functional disorder, 21, 47
genetics of, 38, 44–46, 141, 314
incidence of, 45
inherited susceptibility to, 48, 141
neuroimaging in, 48
pathological imagery in, 307
pharmacotherapy for, 386
twin studies of, 45
School phobia, 111
Schwartz, James H., 95, 111, 136, 140,
146, 349, 350
Scientific literacy, 382
Searle, J. R., 316
Second-messenger pathways, 223, 349,
Seeburg, Peter, 218
Selective deafferentation, 293
Selective serotonin reuptake inhibitors
(SSRIs), 229
for obsessive-compulsive disorder,
Self-injurious behaviors, 374
Self-mutilation, 10
Self psychology, 59
Seligman, M. E. P., 123
Seltzer, B., 47
Selye, Hans, 82
Semrad, E. V., 11, 36
Sensitization, 1, 19–21
definition of, 19, 346
long-term, 144
in Aplysia, 129, 129–130,
133–134, 134, 140, 344,
in chronic anxiety, 110, 115,
122, 123, 138
morphological correlates of,
138–139, 139, 140
presynaptic facilitation in, 1, 19,
350–352, 351
short-term, 19, 24, 144, 146, 346,
Sensory attention, 297, 298–304
in Alzheimer’s disease, 302
brain networks concerned with,
301–304, 303
focal vs. ambient, 301, 314
in neglect syndrome, 302, 303
origins of modern study of,
visual areas biased by shift of,
299–301, 300
Sensory deprivation in infancy, 8–12
effects on synaptic connections, 1,
8–12, 13, 14, 16, 23, 47
social, 8–10
visual, 1, 11–12, 13, 14, 16, 282
Separation response, 81–84
Bowlby’s phases of, 81–82
hypothalamic-pituitary-adrenal axis
activation and, 82, 83
procedural memory and, 84
rodent model of, 82, 83, 84
Serotonin, 28, 176, 228, 229, 349
inhibition of reuptake of, 229
in long-term synapse-specific
facilitation, 353–356, 354
in maintenance of anxiety, 146
in presynaptic facilitation, 19, 134,
136, 150, 231, 350–352, 351
receptors for, 219
regulation of local protein synthesis
by, 356, 357
trauma-induced release of, 110
Index 411
Sexual behavior, 89
Sexual dimorphism, 89–90
Sexual orientation
anatomical studies of, 90–91
biological basis for, 90
definition of, 88
Freud’s concept of, 87–88
as inborn vs. acquired, 88, 91
Sexuality, infantile, 64
Shallice, T., 309
Shatz, Carla, 256
Sherrington, Charles, 16–17, 209, 212,
249, 287, 289
Shevrin, Howard, 96
Shooter, Eric, 247
Shuster, Michael, 350
Siegelbaum, Steven, 136, 350
Signal (anticipatory) anxiety, 78–79,
109–110, 115, 118, 120–124
animal models of, 109–110,
114–116, 121–124, 128
in Aplysia, 128, 129
behaviorists’ vs. Freud’s views of,
131, 132
in biological adaptation, 123
vs. chronic anxiety, 123–124
learned, 120, 122, 122–123
molecular model for, 148–150
pharmacological treatment of, 120
shared molecular components with
chronic anxiety, 147–148, 149
as stimulus response, 120–121
stimulus substitution and, 130
Signaling in nerve cells, 166–175, 189,
action potentials, 134–136, 137,
149, 167, 175, 189, 205, 210,
cloning of ion channels involved in,
218–222, 220
future for study of, 234–236
G protein–coupled receptors in,
223, 234–236
ionic hypothesis of, 167, 168,
212–213, 215
peptide transmitters, 176–180,
plastic properties of synapses,
229–234, 233
pre- to postsynaptic pathways of,
principle of dynamic polarization
and, 207
quantal transmitter release from
presynaptic terminals,
224–228, 225, 227
reductionist approach to, 186, 201,
204, 205, 234–236, 283
synaptic transmission, 167,
175–180, 212, 215–218
Singer, Wolf, 288–289
Skinner, B. F., 35, 110, 126
Smith, Jim, 242
Smithies, Oliver, 359
SNAP-25, 226
Social Darwinism, 41, 381
Social deprivation in infancy, 8–10
Harlow’s monkey studies of, 10
Spitz’s studies of, 9–10
Social phobia, 109, 111
Sociobiology, 40–41
Sodium channels, 165, 167, 168, 171,
172–173, 180, 212, 213
in chronic anxiety, 136
cloning of, 218, 219
closed, open, and inactivated, 172,
ion selectivity of, 172, 213
molecular biology of, 173
mutations of, in channelopathies,
subunits of, 173, 216, 219
Solms, Mark, 87, 95
Spemann, Hans, 242
Spencer, Alden, 342
Spencer, W. A., 17
Sperry, Roger, 249
Spitz, René, 1, 9–10, 80, 81
Spitzer, Robert, 38
Squire, Larry, 359
412 Psychiatry, Psychoanalysis, and the New Biology of Mind
SSRIs (selective serotonin reuptake
inhibitors), 229
for obsessive-compulsive disorder,
Starkman, M. N., 84
Stent, Gunther, 163
Stern, Daniel, 72, 93
Stevens, Charles, 168, 228
Steward, Oswald, 356
Stigmatization of mental illness, 35
anxiety response to, 120–121
aversive, 123
classical conditioning to, 74–75,
conditioned (cue), 122–123, 125,
126, 129
habituation to, 16–19
sensitization to, 19–21, 346
(See also Sensitization)
sensory, orienting of attention to,
297, 298–304 (See also Sensory
unconditioned (reinforcing),
122–123, 125, 126
types of learning produced by,
visual, orientation to, 301–304, 303
Stoner, Gene, 285–286
Stress response
hypothalamic-pituitary-adrenal axis
activation in, 82, 83
to maternal separation, 81–83
posttraumatic stress disorder and,
Stroke, 302
Stroop effect, 311, 313
Stryker, Michael, 256, 257
Substance-induced disorders, 47
Südhof, Thomas, 226, 228
Suomi, S. J., 10
Superego, 87, 111
Synaptic connections, 5–24, 27, 207
assembly of, 240–263 (See also
Neural circuit assembly)
in chronic anxiety, 138
connection specificity of, 229, 240
effects of learning on, 230
formation of, 181
information processing and patterns
of, 264
learning and, 1–2, 17–21, 23, 47,
338, 347, 352–353
habituation, 17–19, 18, 20, 24,
long-term potentiation and, 360
sensitization, 19–21, 22, 24
synapse specificity of long-term
facilitation, 353–356, 354
long-term efficacy of, 23
plastic properties of, 229–234
in psychiatric disorders, 21–23, 141
psychotherapy and, 23
sensory deprivation and, 1, 8–12,
13, 14, 16, 23, 47
Synaptic transmission, 167, 175–180,
205, 212, 215–218, 216.
See also Neurotransmitters
electrical vs. chemical, 217–218
G protein–coupled receptors in, 223
learning and, 338
Synaptic vesicles, 224
proteins involved in release cycle of,
226–228, 227, 236
Synaptobrevins, 226
Synaptotagmins, 228
Syntaxin, 226
Synuclein, 238
Systems biology, 28
Systems neuroscience, 28, 29,
computation in, 266
definition of, 263
neuroanatomy in, 264
neurophysiology in, 265
neuropsychology in, 263–264
psychophysics in, 265–266
vision as model system of, 266–267,
266–297 (See also Visual
Index 413
Takeichi, Masatoshi, 253
Takeuchi, A., 167, 168, 217
Takeuchi, N., 167, 168
Talbot, Samuel, 279
Taub, Edward, 91, 92
Testa, T. J., 124, 125
Testis determining factor, 89
Testosterone, 89, 90
TGF-β (transforming growth factor-β),
242, 243, 247
TH (tyrosine hydroxylase), 176
Therapeutic duality, 2
Thoenen, Hans, 247
Thompson, R., 17, 76
Tic disorders, 30, 61
Tourette’s syndrome, 30
Trace conditioning, 76–77, 77, 87
Training, professional
in neuroscience, 3, 53
psychiatric, 3, 27, 29–30, 34, 36–37
biological orientation in, 53–54
in psychoanalysis, 7, 98–99
Transcription factors, as determinants
of neuronal identity, 243–244, 260
Transference, 78
Transforming growth factor-β (TGF-β),
242, 243, 247
Trinucleotide repeat disorders,
type I, 237–238
type II, 238
Trust, secure attachment and, 80
Twin studies
of behavior, 44
of homosexuality, 91
of schizophrenia, 45, 46
Tyrosine hydroxylase (TH), 176
Unconscious, 34, 35, 56, 60, 64, 69,
70–73, 387
Freud’s concept of, 70–72
methods for study of, 73
motivation and, 374
neurobiology of, 29
preconscious, 71–72, 78, 86–87
procedural, 71, 72–73
moral development and, 72
psychic determinism and, 73–78
psychoanalysis and, 72–73
repressed or dynamic, 70–71
Ungerleider, Leslie, 279
Vaillant, George, 36
VAMP (vesicle-associated membrane
protein), 226, 227
Vasoactive intestinal peptide (VIP),
Vesicle-associated membrane protein
(VAMP), 226, 227
VIP (vasoactive intestinal peptide), 229
Visual deprivation in infancy, 1, 11–12,
cerebral cortical structural effects
of, 11–12, 13, 14, 16
children with congenital cataracts,
primate studies of, 11–12, 13, 14
Visual imagery, 297, 304–308
effect of right posterior parietal
lesions on, 306–307
future studies of, 307–308
objective tests of, 305
pathological, 307
Visual neglect, 302, 303, 306
Visual processing, 193–194, 205–206,
255–257, 266–297, 380
attentional control and, 299–301,
binding problem in, 287–289
changes in cellular representations
with visual experience,
consciousness and, 290
contextual interaction in, 283–287
detection of behaviorally significant
visual features, 274–277
early explorations of, 267
form-cue invariance in, 286
future study of, 282–297
Gestalt theory of, 283–284
414 Psychiatry, Psychoanalysis, and the New Biology of Mind
Visual processing (continued)
how sensory representations lead to
perception, 283–287
influence of local sensory context
on visual perception, 279, 280
links with other brain systems,
local cellular mechanisms of,
ocular dominance columns in
primary visual cortex in, 271,
in prosopagnosia, 274–277
receptive field for, 267–268, 269
single-neuron studies of, 267–268
specialized functions of higher
cortical visual areas in,
271–274, 275, 277
middle temporal area, 273–274,
striate cortex (area V1) in, 268–271,
visual guidance of behavior, 295–297
Visual system organization, 275,
hierarchical, 278
modification by early postnatal
experience, 281–282
in parallel processing streams,
topographical, 279–280
in vertical columns, 280–281
Vogt, Cecile, 264
Vogt, Oscar, 264
Volition, 297, 309–314, 316
Voltage-clamp technique, 213–214
von Bekesy, Georg, 266
von der Malsburg, Christoph,
von Economo, Constantin, 281
von Gerlach, Joseph, 206, 207
von Helmholtz, Hermann, 267
von Senden, M., 11
Wagner, A. R., 126
Wallerstein, R. S., 98
Walters, E. T., 128, 129, 131
Watson, J. B., 123, 126
Watson, James D., 162, 377
Weber, R. J., 305
Weiss, Paul, 249
Weisskopf, V. F., 124
Wender, Paul, 36
Wertheimer, Max, 283
Wessells, Norman, 242
Wexler, Nancy, 237
Wieschaus, Eric, 244
Wiesel, Torsten, 1, 11–12, 255, 265,
269–271, 270, 274, 277, 278, 281,
282, 283, 284, 316
Wilson, E. O., 6, 7
Winder, Danny, 363
Winnicott, D., 68
Wundt, Wilhelm, 267
Wurtz, Robert, 299
X chromosome, 89
Y chromosome, 89
Zeki, Semir, 271
Zorumski, Charles F., 337–339

Pesquisar este blog

arte (407) pintura (248) filosofia (102) fotografia (89) literatura (84) psicanálise (57) morte (37) HQ (28) freud (28) peanuts (27) skull (27) antropologia (26) comic (24) poesia (23) lacan (22) PSYCHOANALYSIS (20) Desenho (17) Picasso (15) o seminário (15) Bresson (13) cat (12) oriente (12) borges (11) kant (11) psicologia (11) foucault (10) levi-strauss (10) platão (10) SONHOS (9) religião (9) Kirchner (8) biografia (8) kafka (8) love (8) Ernest Max (7) Guimaraes Rosa (7) Ken Rosenthal (7) Mark Eshbaugh (7) NIETZSCHE (6) Salvador Dali (6) aristóteles (6) manet (6) snoopy (6) sociologia (6) Animais (5) Aristotle (5) Escher (5) Geertz (5) Hundertwasser (5) Lauren Simonutti (5) Sommer (5) medicina (5) munch (5) Arthur DOVE (4) CINEMA (4) Carl LARSSON (4) Cézanne (4) DICIONARIO (4) Descartes (4) Doré (4) Ernest Jones (4) Ernst HAAS (4) Guido Crepax (4) H. Bergson (4) Julio Cortázar (4) Kacere (4) Locke (4) Mucha (4) Richter (4) Van Gogh (4) Will Barnet (4) alexandre koyrè (4) caveira (4) drummond (4) gravura (4) hegel (4) história (4) linguística (4) monet (4) música (4) sartre (4) teatro (4) televisão (4) universo (4) verdade (4) Abbas (3) Arthur Rackham (3) Avigdor (3) Blake (3) CORINTH (3) Cambon (3) Connie Imboden (3) David Hockney (3) F. Hodler (3) Frida KAHLO (3) GEORGES BATAILLE (3) James Joyce (3) Kojeve (3) Konrad LORENZ (3) Lori Nix (3) M. J. A. Eguiño (3) Marcel MAUSS (3) Marqués de Sade (3) Marx (3) Memling (3) Pierre Bourdieu (3) Psychiatry (3) SENECA (3) Schopenhauer (3) Tom Chambers (3) Winnicott (3) arroyo (3) autobiografia (3) baby (3) caravaggio (3) cristianismo (3) dickens (3) einstein (3) erwitt (3) etologia (3) fisica (3) magia (3) planetas (3) B. F. Skinner (2) BACHELARD (2) Birman J. (2) CERVANTES (2) Castillo (2) Dix (2) Dulac (2) E. HOPPER (2) E. Nodel (2) ETNOLOGÍA (2) Ernest HAAS (2) Ferenczi (2) G. JOHN (2) GEORGE CANGUILHEM (2) Gustav Caillebotte (2) Hipocrates (2) J. Ensor (2) J. J. Tissot (2) JUNG (2) John Donne (2) KARL JASPERS (2) KIERKEGAARD (2) Kandinsky (2) Klimt (2) L. da VINCI (2) LOUIS ALTHUSSER (2) Lewis Carroll (2) M. Dzama (2) MAUGHAM (2) MERLEAU - PONTY (2) Mann (2) Melanie Klein (2) Neil Welliver (2) Norman Rockwell (2) Pascal (2) Piaget (2) Pollock (2) Quino (2) Roland Barthes (2) Sahlins (2) Serge Leclaire (2) St. Agostinho (2) Stratton (2) Vinicus de Moraes (2) Vuillard (2) WITTGENSTEIN (2) amor (2) beauvoir (2) biology (2) cogito (2) critica (2) cultura (2) diabo (2) erotic (2) estruturalismo (2) gide (2) guerra (2) loucura (2) lua (2) mind (2) mitologia (2) mitos (2) rilke (2) salomão (2) saturno (2) sono (2) sócrates (2) vida (2) ética (2) A. Comte (1) A. Warhol (1) Alix Malka (1) Andreas Gursky (1) Anticristo (1) Arcimboldo (1) Aristófanes (1) Augusto dos Anjos (1) B. Barbey (1) B. Jacklin (1) Bagheria (1) Barbara Morgan (1) Basquiat (1) Berkeley (1) Bhagavad-Gita (1) Bhopal (1) Bocklin (1) Bouguereau (1) Brauner (1) Bruegel (1) Brueghel (1) Brueguel (1) Burt GLINN (1) CALOUSTE GULBENKIAN (1) CAZOTTE (1) CRVANTES (1) Charles S. Peirce (1) Chavannes (1) China (1) Claesz (1) Confucius (1) Confúncio (1) D. (1) DAVIS Stuart (1) DEGAS (1) DELACROIX (1) Dalton Trevisan (1) Deleuze (1) Denis (1) Design (1) Diebenkorn (1) Diógenes (1) E. (1) ERNST CASSIRER (1) Emile Durkheim (1) Empédocles (1) Epimenides (1) F. Vallotton (1) FERDINAND DE SAUSSURE (1) Feuerbach (1) Feyerabend (1) Florbela Espanca (1) Franceco Clemente (1) Franz Marc (1) GROOT (1) GUSTON (1) Galileu (1) Gestalt (1) Graham (1) Grécia (1) Guercino (1) H. Arendt (1) H. MARCUSE (1) Hals (1) Helmut Newton (1) Holbien (1) Hume (1) J. Derrida (1) J.-F. Millet (1) Jan Van KESSEL (1) Jean Laplanche (1) KROYER (1) Kandel E. (1) Keane (1) Kim (1) Kitaoka (1) Klee (1) Knight (1) Korand Von SOEST (1) Kôhler (1) Liev Tolstói (1) M. Mead (1) Malinowski (1) Mantegna (1) Mark Petrick (1) Max Weber (1) Mário Quintana (1) Münter (1) N. Chomsky (1) NEIL GAIMAN (1) Nasio (1) Otto Rank (1) Ovídio (1) Palencia (1) Parmênides (1) Paul DELVAUX (1) Peter HILLE (1) Raduan Nassar (1) Ron Van Dongen (1) S. Franklin (1) Sandman (1) Satrapi (1) Schiele (1) Shakespeare (1) Silvers (1) Siqueiros (1) Spinoza (1) St. T. de Aquino (1) TELEPATIA (1) TODOROV (1) Tarsila do Amaral (1) Taschen (1) Thomas HOPKER (1) Truffaut (1) Tycho (1) Uccello (1) Velvet underground Nico (1) Verne (1) Victor Brochard (1) W. Metcalf (1) Web (1) Weinberg (1) William Bailey (1) Woody Allen (1) Xenofonte (1) Y. Utagawa (1) Yoshitoshi (1) alessandro gottardo (1) arcoiris (1) armour (1) arquitetura (1) asselyn (1) ate (1) bassano (1) biblia (1) breton (1) cartoon (1) ceticismo (1) cocaina (1) conto (1) criança (1) dança (1) direito (1) dor (1) eclesiastes (1) economia (1) edgar allan poe (1) edgar morin (1) ego (1) ensaio (1) epicurus (1) espelho (1) estações (1) eu (1) fala (1) feed (1) foto (1) frr (1) física (1) game (1) gato (1) giger (1) girafa (1) goya (1) hamlet (1) hoffmann (1) humor (1) identificação (1) impressionismo (1) intuição (1) jakobson (1) japan (1) krsna (1) kundera (1) lacn (1) leminski (1) lévi-strauss (1) mafalda (1) magritte (1) miró (1) moda (1) moral (1) mundo (1) mãe (1) narrativa (1) nausea (1) neruda (1) nokides (1) ocultismo (1) perguntas (1) poeso (1) poker (1) política (1) praia (1) sabedoria (1) sapatos (1) saramago (1) semina (1) semiótica (1) shopenhauer (1) soutine (1) suicidio (1) swan (1) sêneca (1) tatoo (1) tatuagem (1) tese (1) titã (1) touro (1) umberto eco (1) valentina (1) venus (1) virtude (1) war (1) weeks (1)

sobrefulanos by nokides