PATHWAYS of DISCOVERY "Science Wars"

Home , James Rothman, Stephen Kuffler, Thomas Jessell, Timothy Bliss

JANUARY PATHWAYS OF DISCOVERY "Science Wars"

Neuroscience: FEBRUARY Planetary Breaking Down Scientific Barriers Sciences

to the Study of Brain and Mind MARCH Eric R. Kandel and Larry R. Squire Cenomics

During tlic latter part of the 20th century,, the study of the brain moved from a peripheral position within both the biological and psychological sciences to become an interdisciplinary field called neu- APRIL roscience that now occupies a central position within each discipline. This realignment occurred because the biological study of the brain became incorporated inlo a common framework with ceil and Infectious molecular biology on the one side and with psychology on the other. Within this new framework, the Diseases scope of neuroscience ranges from genes to cognition, from molecules to mind. What led to the gradual incorporation of neuroscience into the central core of biology and to its alignment with psychology? From the MAY perspective of biology at the beginning of the 20th century, the Materials task of neuroscience—to understand how the brain develops Science and then functions to perceive, think, move, and remember— seemed impossibly difficult. In addition, an intellecmal barrier separated neuroscience from biology, because the lan- JUNE guage of neuroseience was based more on neuroanatomy Ctoning and and electrophysiology than on the universal biological language of biochemistry. During the last 2 decades this Stem Cells barrier has been largely removed. A molecular neuroseience became established by focusing on simple systems where anatomy and physiology were tractable. As a juty. result, neuroseience helped delineate a general plan for Communications I neural cell function in whieh the cells of the ner\'ous sys- and Science tem are understood to be governed by variations on universal biological themes. From the perspective of psychology, a neural approach to AUGUST mental processes seemed too reductionistic to do justice to Quantum the complexity of cognition. Substantial progress was re- Artful brain. Cross anatomy made Physics beautiful in a Christopher Wren illus- quired to demonstrate that some of these reductionist goals tration from a 1664 anatomy text. were achievable within a psychologically meaningful framework. The work olVemon Mountcastle, David Hubci. Torsten Wiesel, and Brenda Milner in the 1950s and 1960s, and the advent of brain Imaging in the 1980s, SEPTEMBER showed what could be achieved for sensory processing, perception, and memory. As a result of these The Cell Cycle advances, the view gradually developed that only by exploring the brain could psychologists fully satisfy their interest in the cognitive processes that intervene between stimulus and response. Here, we consider several developments that have been partieularly important for the maturation of neuroseience and for the restmcturing of its relationship to biology and psychology. OCTOBER Atmospheric The Emergence of a Cellular and Molecular Neuroseience Sciences The modern cellular science of the nervous system was founded on two important advances: the neuron doctrine and the ionic hypothesis. The neuron doctrine was established by the brilliant Spanish anatomist Santiago Ramón y Cajal (1). who showed that the brain is composed of discrete cells, called neurons, and that these likely serve as elementary signaling units. Cajal also ad- NOVEMBER vanced the principle of connection specificity, the central tenet of which is that neurons form Neuroseience highly specific connections with one another and that these connections are invariant and defining for each species. Finally, Cajal developed the principle of dynamic polarization, according to which information flows in only one direction within a neuron., usually from the dendrites (the neuron's input component) down the axon shaft to the axon tenninals {the output component). Al- DECEMBER though exceptions to this principle have emerged, it has proved extremely influential, beeause it Astrophysics and tied structure to function and provided guidelines for eonstrueting circuits from the images pro- Cosmology vided in histological sections of the brain.

www.sciencemag.org SCIENCE VOL 290 10 NOVEMBER 2000 1113 PATHWAYS OF DISCOVERY Cajal and his contemporary Charles Sherrington (2) fur- pores and the biophysical basis for their selectivity and gat- A Timeline of ther proposed that neurons contact one another only at spe- ing^how they open and close. For example, transmitter Neuroscience cialized points called synapses, the sites where one neuron's binding sites and their ion channels were found to be em- processes contact and communicate with another neuron. We bodied within different domains of multimeric proteins. Ion now know that at most synapses, there is a gap of 20 nm— channel selectivity was found to depend on physical- 2nd Century the synaptic cleft—between the pre- and postsynaptic cell. In chemical interaction between the channel and the ion. and A.D. the 1930s, Otto Loewi. Henry Dale, and Wilhelm Feldberg channel gating was found to result from conformational Galen of Perga- established (at peripheral neuromuscular and autonomie changes within the channel (5). mum identifies synapses) that the signal that bridges the synaptic cleñ is usu- The study of ion channels changed radically with the de- the brain as the organ of the ally a small chemical, or neurotransmitter. which is released velopment of the pateh-clamp method in 1976 by Erwin Ne- mind. from the presynaptic temiinal, diffuses across the gap. and her and Bert Sakmann (6), which enabled measurement of binds to receptors on the postsynaptic target cell. Depending the current flowing through a single ion channel. This pow- 17th Century on the specific receptor, the postsynaptic cell can either be erful advance set the stage for the analysis of ehannels at the The brain becomes accepted excited or iniiibited. It totik some time to establish that chem- molecular level and for the analysis of functional and con- as tKe substrate ical transmission also occurs in the central nervous system, formational change in a single of mental life but by the 1950s the idea had become widely accepted. membrane protein. When applied rather than Its ventricles, as Even early in the 20th century, it was already understood to non-neuronal cells, the method earty writers that nerve cells have an electrical potential, the resting mem- also revealed that all cells—even had proposed. brane potential, across their membrane, and that signaling bacteria—express remarkably sim- 1664 along the axon is conveyed by a prop- ilar ion channels. Thus, neuronal Thomas Willis agated electrical signal, the action po- signaling proved to be a special publishes Cerebri tential, which was thought to nullify case ofa signaling capability in- anatome, with it- herent in most cells. lustrations of the the resting potential. In 1937 Alan brain by Christo- Hodgkin discovered that the action The development of patch pher Wren. It is potential gives rise to local current clamping coincided with the ad- the most com- prehensive trea- How on its advancing edge and ihat vent of molecular cloning, and Ihesc tise on brain this current depolarizes the adjacent two methods brought neuroscientists anatomy and region of the axonal membrane suffi- new ideas based on the first reports of function ^published up to ciently to trigger a traveling wave of the amino acid sequences of ligand- and that time. depolarization. In 1939 Hodgkin and voltage-gated channels. One of the key .'Andrew Huxley made the surprising insights to emerge from molecular 1791 discovery that the action potential cloning was that amino acid sequences Luigi Calvani re- veals the electric more than nullifies the resting poten- contain clues about how receptor pro- nature of nervous tial—it reverses it. Then, in the late teins and voltage-gated ion channel pro- action by stimu- 1940s, Hodgkin, Haxley. and Bernard teins are arranged across the cell mem- lating nerves and muscles of frog Katz explained the resting potential brane. The sequenee data also often tegs. and the action potential in terms of pointed to unexpected structural rela- the movement of specific ions— tionships (homoiogies) among proteins. 1808 potassium (K"^), sodium (Na*). and These insights, in turn, revealed similar- Franz Joseph Gall Seeing neurons. Anatomist Ramón y Cajal ehloride (Cl)—through pores (ion proposes that used Colgi's stain to examine individual ities between molecules found in quite specific brain channels) in the axonal membrane. nerve cells and their processes, different neuronal and non-neuronal regions control This ionic hypothesis unified a targe contexts, suggesting that they may serve specific func- > tions. body of descriptive data and offered the first realistic promise similar biological flinctions. that the nervous system could be understood in terms of By the early 1980s, it became clear that synaptic ac- I 1852 physicoehemical principles common to all of cell biology (3). [ Hermann von tions were not always mediated directly by ion channels. f Helmholtimea- The next breakthrough came when Katz. Paul Fatt. and Besides ionotropic receptors, in which ligand binding di- • sures the speed John Eccles showed that ion channels are also ftindamental to rectly gates an ion channel, a second class of receptors, the lofa nerve im- ^ fHjbe in the frog. signal transmission across the synapse. However, rather than metabotropic receptors, was discovered, Here the binding being gated by voltage like the Na"" and K* channels critical of the ligand initiates intracelluiar metabolic events and Í 1879 for action potentials, excitatory synaptic ion channels aie gat- leads only indirectly, by way of "second messengers," to : Wilhelm Wundt ' establishes the ed chemically by ligands such as the transmittei- acetylcholine. the gating of ion channels (7). ' first laboratory During the 1960s and 1970s, neuroseientists identified many The cloning of fnetabotropic receptors revealed that man\ * of experimental iunino acids, peptides. and otlier small molecules as chemical of them have seven membrane-spiinning regions and are ho- f psychology in transmitters, including acetylcholine, glutamate. GABA, ' [, Germany, mologous to bacterial rhodopsin as well as to the photorecep- glycine, serotonin, dopamine. and norepinephrine. On the or- tor pigment of organisms ranging from fruit flies to humans. der of 100 chemical transmitters have been discovered to date. Further, the recent cloning of reecptors for the sense of smell In the 1970s, some synapses were found to release a peptide (S) revealed that at least 1000 metabotnipic receptors are ex- cotmnsmittcr that can modify the action of the classic, small- pressed in the mammalian olfactory epithelium and that sim- moleeule transmitters. Tlie discovery of chemical neurotrans- ilar receptors are present in flies and worms. Thus, it was in- mission was followed by the remarkable discovery that trans- stantly understood thai tlie class of receptors used for photo- mission between neurons is sometimes electrical (4). Electri- transduction, the initial step in visual perception, is also used cal synapses have smaller synaptic eiefts. which are bridged for smell and aspects of taste, and that these receptors share by gap junctions and allow cunent to flow between neurons. key features with many other brain receptors that work In the late 1960s information began to become available through second-messenger signaling. These discoveries ¡ibout the biophysical and biochemical structure of ionic demonstrated the evolutionary conservation of receptors and

1114 10 NOVEMBER 2000 VOL 290 SCIENCE vww.sciencemag.org PATHWAYS OF DISCOVERY emphasized the wisdom of studying a wide variety of experi- A Mechanistic View of Brain Development mental systems^—vertehrates. invertebrates, even single- The discoveries of molecular neuroscience have dramatical- 1891 Withelm von celled organisms-^o identify broad biologieal prineiples. ly improved the understanding of how the brain develops its Waldeyer-Hartz The seven transmembrane-spanning receptors activate complexity. The modern moleeular era of developmental introduces the ion channels indirectly through coupling proteins (G pro- neuroscience began when Rita Levi-Montai ein i and Stanley term neuron. teins). Some G proteins have been found lo activate ion Cohen isolated nerve growth factor (NGF). the first peptide 1897 channels directly. However, the majority of G proteins acti- growth factor to be identified in the ner\'ous system (12). Chartes vate membrane enzymes that alter the level of second mes- They showed that injection of antibodies to NGF into new- Sherrington introduces the sengers, such as cAMP, cGMP, or inositol triphosphate, bom mice caused the death of neurons in sympathetic gan- term synapse. wliich initiate complex intracellutar events leading to the ac- glia and also reduced the number of sensory ganglion cells. tivation of protein kinases and phosphatases and then to the Thus, the survival of both sympathetic and sensory neurons 1898-1903 modulation of channel permeability, receptor sensitivity; and depends on NGF. Indeed, many neurons depend for their Edward Thorndike and Ivan Pavlov transmitter release. Neuroscientists now appreciate that survival on NGF or related molecules, which typically pro- describe opérant many ol" these synaptic actions are mediated intraceilularly vide feedback signals to the neurons from their targets. Such and classical by protein phosphorylation or dephosphorylation (9), Nerve signals are important for programmed cell death—apopto- conditioning, two fundamental cells use such covalent modifications to control protein ac- sis—a developmental strategy which has now proved to be types of teaming. tivity reversibly and thereby to regulate function. Phospho- of general importance, whereby many more ceUs are gener- rylation is also eritical in other cells for the action of hor- ated than eventually survive to become functional units with 1906 Santiago Ramón mones and growth factors, and for many other processes. precise connectivity. In a major advance, genetic study of y Cajat sgmma- Directly controlled synaptic actions are fast, lasting mil- worms has revealed the ced genes and with them a universal riies compelting cascade critical for apoptosis in which proteases—the cas- evidence for the liseconds, but seeond-messenger actions last seconds to min- neuron doctrine, utes. An even slower synaptic action, lasting days or more, pases—are the final agents for cell death (13). that the nervous has been found to be important for long-tenii memory. In Cajai pointed out the extraordinary precision of neuronal system is com- this case, protein kinases activated by seeond messengers connections. Tlie first compelling insights into how neurons translocate to the nucleus, where they phosphorylate tran- develop their preeise connectivity came from Roger Sperry's scription factors that alter gene expression, initiate growtli of studies of the visual system of frogs and salamanders begin- Alois Aliheimef neuronal processes, and inerease synaptic strength. ning in the 1940s, which suggested that axon outgrowth is descritas the pathotogy of the guided by molecular cues. Sjierry s key find- neurodegenera- ing was that when the nerves from the eye tive disease that are cut, axons find their way back to their comes to bear his original targets. These seminal studies led Sperry in 1963 to formulate the chemoafFin- 1914 ity hypothesis (14).. the idea that neurons Henry Dale demonstrates form connections with their targets based on the physiologicat distinctive and matching molecular identities action of acetyl- that they acquire early in development. choline, which is later identified Stimulated by these early contributions, as a neuro- molecular biology has radically trans- transmitter. formed the study of nervous system de- 1929 Transmitter molecules bind velopment from a descriptive to a mecha- In a famous 10 excitatory receplors. receptor channels open, and Na* enters nistic field. Three genetic systems, the program of lesion tfie postsynaptic cetl worm Caenorhabditis elegam. the fruit fly experiments in Drosophila melanoga.^ter. and the mouse, rats. Karl Lashley The synapse. A presynaptic neuron propagates a signal by releasing neurotrans- attempts to mitter molecules that diffuse across the synaptic cleft to bind to receptors on the have been invaluable; some ofthe mol- localize memory postsynaptic cell. ecules for key developmental steps in the in the brain. mouse were first characterized by genetic Hans Berger usés Ionotropic and metabotropic receptors have helped to screens in worms and flies. In some cases, identical human scatp explain the postsynaptic side of synaptie transmission. In molecules were found to play an equivalent role throughout electrodes to the 1950s and 1960s. Katz and his colleagues turned to the phylogeny. The result of this work is that neuroscientists demonstrate etectroen- presynaptic terminals and discovered that chemical trans- have achieved in broad outline an understanding ofthe cephatography. mitters, such as acetylcholine. are released not as single molecular basis of nervous system development (15). A molecules but as packets of about 5000 molecules called range of key molecules has been identified, including spe- 1928-32 cific inducers. morphogens, and guidance molecules impor- Edgar Adrian quanta (10). Each quantum is packaged in a synaptic vesi- describes method cle and released by exocytosis at sites called active zones. tant for differentiation, process outgrowth, pathfinding. and for recording The key signal that triggers this sequenee is the influx of synapse formation. For example, in the spinal cord, neurons from single Ca"' with the action potential. aehieve their identities and characteristic positions largely sensory and motor axons; In recent years, many proteins involved in transmitter re- through two classes of inductive signaling moleeules ofthe H. Keffer Harttine lease have been identified (II). Their functions range (Vom Hedgehog and bone morphogenic protein families. These applies this two groups of molecules control neuronal differentiation in method to the targeting vesicles to active zones, tethering vesicles to the recording of cell membrane, and fusing vesicles with the cell membrane the ventral and dorsal halves ofthe spinal cord, respeetively. single-cetl activ- so that their contents can be released by exocytosis. These and maintain this division of labor through most ofthe ros- ity in the eye of moieeular studies reflect another example of evolutionary trocaudal length ofthe nen'ous system. the ho^se5hoe^ - crab. '! conservation: The molecules used for vesicle fusion and The process of neuronal pathfinding is mediated by exocytosis at nerve terminals are variants of those used for both short-range and long-range cues. An axons growth vesicle fusion and exocytosis in all cells. cone can encounter cell surface eues that either attract or

www.sciencemag.org SCIENCE VOL 290 10 NOVEMBER 2000 1115 PATHWAYS OF DISCOVERY repel it. For example, ephrins are membrane-bound, are and risk of a degenerative disease (¡8). One alíele (APO E4) 1940s is a significant risk factor for late-onset Alzheimers disease. Alan Hodgkin, distributed in graded fashion in many regions of the ner- Andrew Huxley, vous system, and can repel growing axons. Other cues, Conversely, the APO E2 alíele may actually be protective. A and Eternard Katz such as Uie netrins and the semaphorins, arc secreted in second risk factor is a^-macroglobulin. All the Alzheimer's- diffusible form and act as long-range chemoattractants or related genes so far identified participate in eitlier generating ehemorepellents. Growth cones ean also reaet to the same or scavenging a protein (the amyloid peplide), which is toxic cues diiferently at different developmental phases, for ex- at elevated levels. Smdies directed at this peptide may lead to ample, when crossing the midline or when switching from ways to prevent the disease or halt its progression. Similarl>. pathfinding to synapse formation. Finally, a large number the discovery of [i-secrctase and perhaps y-secretase, the of molecules are involved in synapse formation itself. etizymes involved in the processing of ß amyloid, represent Some, such as neuregulin, erbB kinases, agrin, and MuSK. dramatie advances that may also lead to new treamients. organize the assembly of the postsynaptic machinery, With psychiatric disorders, progress has been slower for whereas others, such as the laminins, help to organize the two reasons. First, diseases such as schizophrenia, depres- presynaptic differentiation of the active zone. sion, obsessive eompulsive disorders, anxiety states, and These molecular signals direct differentiation, migration, drug abuse tend to be complex, polygenic disorders that are process outgrowth, and synapse fonnation in the absence of significantly modulated by environmental factors. Second in neural activity. Neural activity is needed however, to refine contrast to neurological disorders, little is known about the the connections further so as to forge the adult pattern of anatomical substrates of most psychiatric diseases. Given the connectivity (¡6). The neural activity may be generated difficulty of penetrating the deep biology of mental illness, it spontaneously, especially early in development, but later de- is nevertheless remarkable how much progress has been pends importantly on sensory input. In this way. intrinsic ac- made during the past 3 deeades (19). Arvid Carlsson ;md tivity or sensory and motor experienee can lielp specify a Julius Axelrod carried out pioneering studies of biogenic precise set of functional connections. amines, which laid the foundation for psychopharmacology, and Seymour Kety pioneered the genetic study of mental ill- The Impact of Neuroscience on Neurology and Psychiatry ness (20). Currently, new approaches to many conditions, Molecular neiuoscience has also reaped substantial benefits such as sleep disorders, eating disorders, and drug abuse., are for clinical medicine. To begin with, recent advances in the emerging as the result of insights into tlie cellular and molec- «Kpiain eiKtrical ular machinery that regulates specific behaviors (2¡). More- activity of neu- study of neural development have identified stem cells, both rons by concen- embryonic and adult, which offer over, improvements in diagnosis, the bet- tration gradients promise in cell replacement ther- ter delineation of genetic contributions to of ions and n^ove- apy in Parkinson's disease, de- psychiatric illness (based on twin and ment of ions adoption studies as well as studies of af- "" " " ugh pores. myelinating diseases, and other conditions. Similarly, new in- fected families), and the diseovery of spe- sights into axon guidance cific medications for treating schizophre- th Cole de- nia, depression, and anxiety states have s th« volt- molecules offer hope for nerve ,*'-clamp tech- regeneration after spinal cord in- transformed psychiatry into a therapeuti- nique to measure jury. Finally, because most neuro- eally effeetive medical specialty that is current flow now closely aligned with neuroscience. across the cell logical diseases are associated " jibrane. witlî ceil death, the discovery in worms of a universal genetic pro- A New Alignment of Neuroscience and ' Donald Hebb in- gram for cell death opens up ap- Psychological Science troduces a synap- proaches fbr cell rescue based on, The brain's computational power is con- tlc [earning rute, for example, inhibition of the ferred by interactions among billions of which becomes caspase proteases. known as the nerve cells, which are assembled into net- Hebb rute. Next, consider the impact of works or circuits that cany out specific molecular genetics. Hunting- operations in support of behavior and 19305 to 19505 The chemical na- ton's disease is an autosomal Little man inside. This 1950 homunculus sum- cognition. Whereas the molecular mature of synaptic dominant disease marked by marized studies of the cerebral localization of chinery and electrical signaling properties iransmission is es- progressive motor and cognitive motor function. of neurons are widely conserved aeross tablished by Otto Loewi. Henry Dale, impairment that ordinarily animal species, what distinguishes one Wilhelm Feldberg, manifests itself in middle age. The major pathology is eell species from another, with respect to their cognitive abilities, Stephen Kuffler, death in the basal ganglia. In 1993, the Huntington's Dis- is the number of neurons and the details of their connectivity. and Bernard Katz at periphieral ease Collaborative Research Group isolated the gene re- Beginning in the 19th century there was great interest in synapses and is sponsible for the disease (17). It is marked by an extended how these cognitive abilities might be localized in the brain. extended to the series of trinucleotide CAG (eytosine. adenine. guanine) One view, first championed by Franz Joseph Gall, was that spinal cord by John Eccles and others repeats, thereby placing Huntington's disease in a new the brain is eomposed of specialized parts and thai aspects class of neurological disorders—the trinucleotide repeat of perception, emotion, and language can be locaiized to Wilder Penfield diseases—that now constitute the largest group of domi- anatomically distinct neural systems. Another view, champi- and Theodore nantly transmitted neurological diseases. oned by Jean-Pierre-Marie Flourens. was that cognitive Rasmussen map the motor and The molecular genetic analysis of more complex degener- functions are global properties ari.sing from the integrated sensory ho- ative disorders has proceeded more slowly. Still, three genes aetivity of the entire brain. In a sense, the history of neuro- munculus and il- science can be seen as a gradual ascendancy of the localiza- lustrate localiza- associated with familial Alzheimer's disease—those that tion of function in code for the amyloid precursor protein, presenilin I. and pre- tionist view. the human brain. senilin 2^havc been identified. Molecular genetic studies To a large extent, the emergence of the localizationist have also identified the first genes that modulate the severity view was built on a eentury-old legacy of psychological sci-

1116 10 NOVEMBER 2000 VOL 290 SCIENCE www.sciencemag.org PATHWAYS OF DISCOVERY cnce. When psychology emerged as an experimental science the two eyes. These nsults provided an entirely new view of in the late mtli century, its founders, Gustav Fechner and the anatomical organization of the cerebral cortex. 1950s Wilhelm Wundt. focused on psychophysics—the quantita- Karl von Frisch, Wiesel and Hubel also investigated the effects of early sen- Konrdd Loreni, tive relationship between physical stimuli and subjective sory deprivation on newborn animals. They found that visual and Nikolaas sensation. The success of this endeavor encouraged psychol- deprivation in one eye profoundly alters the organization of Tinbergen ogists to study more complex behavior, which led to a rigor- establish the oeular dominance columns (23). Columns receiving input science of ous, laboratory-based tradition termed behaviorism. from the closed eye shrink, and those receiving input from the ethology (animal Led by John Watson and later by B. F. Skiimer, behavior- open eye expand. These studies led to the dlscover> that eye behavior in natu- closure alters the pattem of synchronous activity in the two rat contexts) and ists argued that psychology should be concerned only with lay the founda- observable stitïiuli and responses, not with unobservable eyes and that tliis neural activity is essential for fine-tuning tion for neiiro- processes that intervene between stimulus and response. synaptic connections during visual system development (16). et hoi ogy. This tradition yielded lawful principles of behavior and In the extrastriate cortex beyond area VI, continuing learning, but it proved limiting. In the 1960s, behaviorism 1955-60 electrophysiological and anatomical studies have identified Vernon Mount- gave way to a broader approach concerned with cognitive more than 30 distinct areas important for vision (24). Fur- castle, David processes and internal representations. This new emphasis ther, visual information was found to be analyzed by two Hübet, and focused on precisely those aspeets ot" mental life—from per- Torsten Wiesel parallel processing streams (25). The dorsal stream, con- pioneer single- ception to action^that had long been of interest to neurolo- cerned with where objects are located in space and how to cell recording gists and other students of the nervous system. reach objects, extends írom area VI to the parietal cortex. from mammalian sensory cortex; The first cellular studies of brain systems in the 1950s il- The ventral stream extends from area VI to the inferior tem- NilS'Ake Hillarp lustrated dramatically how much neuroscience derived from poral cortex and is concerned with analyzing the visual form introduces psychology and conversely how much psychology could, in and quality of objects. Thus, even the apparently simple task fluorescent microscopic turn, inform neuroscience. In using a cellular approach, neu- of perceiving an object in space engages a disparate collec- methods to study roscientists relied on the rigorous experimental methods of tion of specialized neural areas that represent different as- celtiitar distribu- psychophysics and behaviorism to explore how a sensory pects of the visual information—what the object is. where it tion of biogentc stimulus resulted in a neuronal response. In so doing, they is located and how to reach for it. found cellular support for localization of ftinclion: Different 1956 brain regions had difierent cellular response properties. Thus. A Neuroscience of Cognition Ritalevi- it became possible in the study of behavior and cognition to The initial studies of the visual system were perfomied in Montalclnl a move beyond description to an exploration of the meehanisms anaesthetized cats, an experimental preparation far removed underlying the internal representation of the extemal world. from the behaving and thinking human beings that are the In the late 1950s and 196ns Mountcastle. Hubel. and focus of interest for cognitive psychologists. A pivotal ad- Wiesel began using cellular approaches to analyze sensory vance occurred in the late 1960s when single-neuron record- processing in the cerebral coriex of cats and monkeys (22). ings were obtained from awake, behaving monkeys that had Their work provided the most fundamental advance in under- been trained to perform sensory or motor tasks (26). Witii standing the organization of the brain since the work of Cajal these methods, the response of neurons in the posterior pari- at the turn of the century. The cellular physiological tech- etal cortex to a visual stimulus was found to be enhanced niques revealed that the bniin both i'ilters and transfomis sen- when the animal moved its eyes to attend to the stimulus. sory information on its way to and within the cortex, and that This moved the neurophysiological study of single neurons these transfomiations are critical for perception. Sensory sys- beyond sensory processing and showed that reductionist ap- lems analyze, decompose, and then restructure raw sensory proaches could be applied to higher order psychological pro- information according to built-in connections and rules. cesses such as selective attention. Mountcastle found Iliat single nerve cells in the primary so- It is possible to correlate neuronal firing with perception matic sensory cortex respond to st-ieeific kinds of touch; Some rather direetiy. Thus, building on earlier work by Mountcas- respond to superficial touch iind others to deep pressure, but tle. a monkey's ability to discriminate motion was found to cells almost never respond to btith. Tlie dißerent ceU types are closely match the performance of individual neurons in area segregated in vertical columns, which comprise thousands of MT, a cortical area concerned with visual motion processing. isolate and purify neurons and extend about 2 mm ftom the cortical surface to Further, electrical microstimulation oi' small clusters of neu- nerve growth factor. the wliite tnatter below it. Mountcastle proposed that each col- rons in MT shins the monkey's motion judgments toward fhe umn serves as an integrating unit, or logical mLxiule. and tliat direction of motion that the stimulated neurons prefer (27). these columns are the basic mode of cortical organization. Thus, activity in area MT appears sufficient for the percep- Single-cell recording was pioneered by Edgar Adrian and tion of motion and for initiating perceptual decisions. applied to the visual system of invertebmtes by M. Keffer Hart- line and to the visual system of mammals by Stephen KufFler, the mentor of Mubel and Wiesel. In recordings from the retina, Kuffler discovered that, rather tlian signaling absolute levels of light, neurons signal contrast between spots of light and dark. hi the visual cortex, Hubel and Wiesel found that most cells no longer respond to spots of light. For example, in ana VI at the occipital pole of the cortex, neurons respond to specific visual features such as lines or bars in a particular orientation. More- over, eells witli similar orientation preferences were found to group together in vertical colmnns similar to those that Mount- castle had found in sotnatosensory cortex. Indeed an indepen- dent system of vertical columns—the ocular dominance Single-cell recording. Vernon Mountcastle. David Hubel, and columns—was found to segregate infonnation arriving from Torsten Wiesel pioneered cellular studies of sensory cortex.

Www.5Ciencemag.org SCIENCE VOL 290 10 NOVEMBER ZOOO ni7 PATHWAYS OF DISCOVERY These findings, based on recordings from small neuronal range. Recent sueeess in obtaining fMRI images from populations, have illuminated important issues in perception awake monkeys, combined with single-cell recording, and action. They illustrate how retinal signals are remapped should extend the utility of functional neuroimaging by per- from retinotopic space into otli- mitting parallel studies in humans and nonhuman primates. er coordinate frames that can One example of how parallel studies of humans ;md non- guide behavior; how attention human primates have advanced the understanding of brain can modulate neuronal activity: systems and cognition is in the study of memory. The neuro- and how meaning and context seience of memory came into focus in the 1950s when the influence neuronal aetivity. so noted amnesic patient H.M. was first described (32). H.M. that the same retinal stimulus developed profound tbrgetfulness after sustaining a bilateral can lead to different neuronal medial temporal lobe resection to relieve severe epilepsy. Yet responses depending on how he retained his intelligence, perceptual abilities, and person- the stimulus is perceived (28). alit>. Brenda Milner's elegant studies of H.M. led to several This same kind of work (relat- important principles. First, acquiring new memories is a dis- ing cellular activity directly to tinct eerebral funetion, separable from other perceptual and perception and action) is cur- cognitive abilities. Second, because H.M. could retain a rently being applied to the so- number or a visual image for a short time, the medial tem- called binding problem—how poral lobes are not needed for immediate memory. Third the multiple features of a stimu- these structures are not the ultimate repository of memor>. lus object, which are represent- because H.M. retained his remote, childhood memories. ed by specialized and distribut- It subsequently became clear that only one kind of mem- ed neuronal groups, are synthe- ory, declarative memory, is impaired in H.M. and other am- sized into a signal that repre- nesic patients. Thus, memory is not a unitary faculty of the sents a single percept or action mind but is composed of multiple systems that have differ- and to the fimdamental question ent logie and neuroanatomy (33). The major distinction is Big picture. Brain maps, of what aspects of neuronal ac- between our capacity for conscious, declarative memor\' like this one of the ma- tivity (e.g., firing rate or spike about facts and events and a collection of unconscious, non- caque visual system, reveal timing) constitute the neural dechiTdtive memory abilities, sueh as skill and habit learning multiple brain regions codes of infonnation process- and simple forms of conditioning and sensitizaron. In these working together as a func- ing (29). tional system. eases, experience modifies performance without requiring Striking parallels to the orga- any conscious memory content or even the experience that nization £ind lunction of sensory cortices have been lbund in memory is being used. the cortical motor areas supporting voluntary movement. An animal model of human amnesia in the nonhiiman pri- Tlius. there are several cortical areas directed to the planning mate was achieved in the early 1980s, leading ultimately to and execution of voluntary movement. Primary motor cortex the identification of the medial temporal lobe structures that has columnar organization, with neurons in each column gov- support declarative memory—the hippocampus and the adja- erning movements of one or a few joints. Motor areas receive eent entorhinal. perirhinal, and parahippocampal cortices input from other cortical regions, and information moves (34). The hippocampus has been an especially active target of through stages to the spinal eord, where the detailed eircuitry study, in part because this was one of the structures damaged that generates motor patterns is located (30). in patient H.M. and also beeause of the early discovery of Although studies of single eells have been enormously in- hippocampal place cells, which signal the location of an ani- formative, the functioning brain consists of multiple brain mal in space (35), This work led to the idea that, onee learn- systems and many neurons operating in concert. To monitor ing occurs, the hippocampus and other medial temporal lobe aetivity in large populations of neurons, multielectrode arrays structures permit the transition to long-term memory, per- as well as cellular and whole-brain imaging techniques are haps by binding the separate cortical regions that together now being used. These approaches are being supplemented store memory for a whole event. Thus, long-tenn memory is by studying the effect of selective brain lesions on behavior thought to be stored in the same distributed set of cortical and by molecular methods, such as the delivery of markers or struetiires that perceive, process, and analyze what is to be re- other molecules to specific neurons by viral transfection. membered, ;md aggregate changes in large assemblies of eor- which promise fine-resolution tracing of anatomical connec- tical neurons are the substrate of long-term memory. The tions, activity-dependent labeling of neurons, and ways to frontal lobes are also thought to influence what is selected for transiently inactivate speeific components of neural circuits. storage, the ability to hold information in mind for the short Invasive molecular manipulations of this kind cannot be term, and the ability later on to retrieve it (36). applied to humans. However, functional neuroimaging by Whereas declarative memory is tied to a particular positron emission tomography (PET) or functional magnetic brain system, nondeclarative memory refers to a collection Drosophlia, resonanee imaging {¡"MRI) provides a way to monitor large of learned abilities with ditTerent brain substrates. For ex- and C. eiegans, are introduced neuronal populations in awake humans while they engage in ample, many kinds of motor learning depend on the cere- to analyze cognitive tasks (31). PET involves measuring regional blood bellum, emotional learning and the modulation of memory elementary tlow using H; '•"'0 and allows for repeated measurements on strength by emotion depend on the amygdala, and habit aspects of behavior and learn- the same individual. fMRI is based on the fact that neural learning depends on the basal ganglia (37). These forms of ing at the activity changes local oxygen levels in tissue and that oxy- nondeclarative memory, which provide for myriad uncon- cellular and genated and deoxygenated hemoglobin have different mag- scious ways of responding to the world, are evolutionarily lecular netic properties. It is now possible to image the second-by- aneient and observable in simple invertebrates such as second time course of the brain's response to single stimuli Aply.sia and Dro.sophikt. By virtue of the unconscious sta- or single events with a spatial resolution in the millimeter tus of these forms of memory, they create some of the

1118 10 NOVEMBER 2000 VOL 290 SCIENCE www.sciencemag.org PATHWAYS OF DISCOVERY mystery of human experience. For here arise the disposi- brain regions and also to rum genes on and off. Such genetie tions, habits, attitudes, and preferences that are inaccessi- and pharmacological experiments in intact animals suggest 1962-63 Brain anatomy in ble to conscious recollection, yet are shaped by past events. that interference with LTP at a specific synapse—the Schaf- rodents is found influence our behavior and our mental life, and are a fun- fer collateral-CAl synapse^—commonly impairs memory to be altered by damental part of who we are. for space and objects. Conversely, enhancing LTP at the experience; first evidence for rote same synapse can enhance memory in these same declara- of protein syn- Bridging Cognitive Neuroscience and Molecular Biology in tive memory tasks. The findings emerging from these new thesis in meai the Study of Memory Storage methods (42) complement those in Aplysia and Drosophiia formation. The removal of scientifie barriers at the two poles of the and reinforce one of Cajal's most prescient ideas: Even 1963 biologicLil sciences—in the cell and molecular biology of though the anatomical connections between neurons develop Roger Sperry pro- nerve cells on the one hand, and in the biology of cognitive according to a definite plan, their strength £tnd effectiveness poses a precise are not predetermined and can be altered by experience. system of chemi- processes on the other—has raised the question: Can one cat matching be- antieipate an even broader unification, one that ranges Combined behavioral and molecular genetic studies in tween pre- and from molecules to mind? A beginning of just such a syn- Drosophila, Aplysia. and mouse suggest that, despite their postsynaptic neuronal partners thesis may be apparent in the study of synaptic plasticity different logic and neuroanatomy. declarative and non- (the chemoaffini- and memory storage. declarative forms of memory share some common cellular ty tiypothests). For all of its diversity, one can view neuroseience as being and molecular features. In both systems, memory storage concerned with two great themes—the brain's "hard wiring" 1966-69 depends on a short-term process lasting minutes and a Ed Evarts and and its capacity for plastieity. The fonner refers to how con- long-term process lasting days or longer. Short-term memo- Robert Wurti de- nections develop between cells, how cells ftinction and com- ry involves covalent modifications of preexisting proteins, velop methods municate, and how an organism's inbom ftinctions are orga- leading to the strengthening of preexisting synaptic connec- tor studying movement ana nized—^its sleep-wake cycles, hunger and thirst, and its abil- tions. Long-term memory involves altered gene expression, perception with ity to perceive the world. Thus, through protein synthesis, and the growth of single-cetl evolution the nervous system lias inherited new synaptic connections. In addition. recordings from äwake, behaving many adaptations that are too important to a number of key signaling molecules monkeys. be Icíí to the vagaries of individual experi- involved in converting transient short- ence. In contrast, the capacity tor plasticity term plasticity to persistent long-term 1970 refers to the fact that nervotLs systems can memory appear to be shared by both Synaptic chai are related to adapi or change as the result of the experi- declarative and iiondeclarative memo- Learning and ences that occur during an individual life- ry. A striking feature of neural plastici- memory storage time. Experience can modify the nervous ty is that long-term memory involves in Apiyiia. system, and as a result organisms can learn structural and functional ehange (38. Mid-1970s and remember. 43). This has been shown most directly Paul Creengard in invertebrates and is likely to apply to shows that many The precision of neural connections pos- neuro transmit- es deep problems for the plasticity of behav- vertebrates as well, including primates. ters work by ior. How does one reconcile the precision It had been widely believed that the means of protein and specificity of the brain's wiring with the sensory and motor cortices mature ear- phosphorylation. known capability of humans and animals to Real-time brain. Imaging methods can ly in life and thereafter have a fixed or- 1973 acquire nev^' knowledge? And how is knowl- reveal those brain areas that are active ganization and connectivity. However, Timothy Bliss and edge, once acquired, retained as long-term it is now clear that these cortices can be Terje Lomo dis- during specific cognitive tasks. cover long term memory'.' A key insight about synaptic trans- reshaped by experience (44). In one ex- potentiation, a mission is that the precise eonneetions between neurons are not periment, monkeys learned to discriminate between two vi- Candidate synap- fixed but are modifiable by experience. Beginning in 1970, brating stimuli applied to one finger. After several thousand tic mechanism for trials, the cortical representation of the (rained finger be- tong-term mam- studies in invertebrates such as Aplysia showed thai simple maiian memory, forms of leiiming—liabituation. sensitization, and ckissicaJ con- came more than twice as large as the corresponding areas ditioning—result in functional and structural changes at for other fingers. Similarly, in a neuroimaging study of 1976 ^ right-handed string musicians the cortical representations of Erwin Neher a .synapses between the neurons that mediate the behavior being Sert Sakmann de- modified. These changes can persist tor days or weeks and par- the fingers of the left hand (whose fingers are manipulated velop the patch- allel the time courîie of tlic memory process (3H). Tliese cell bi- individually and are engaged in skillful playing) were larger clamp technique than in nonmusicians. Thus, improved finger skills even in- for recording the ological studies have been complemented by genetic sUidies in activity of 5ingle Dfusophila. As a result, studies in .Aplvsia and Dmsophila have volve changes in how sensory cortex represents the fingers. ion channels. identified a niunber of proteins importiint for memory (39). Because all organisms experience a different sensory envi- Late 19701 In his now-famous book. The Organization of Behavior, ronment, each brain is modified differently. This gradual ereation of unique brain architecture provides a biological Neuroimaging by Donald Hebb proposed in 1949 that the synaptic strength positron emission between two neurons should increase when the neurons ex- basis for individuality. tomography Is hibit coincident activity (40). In 1973, a long-lasting synap- developed, tic plasticity of this kind was discovered in the hippocampus Coda 1980s I (a key structure for declarative memory) (41). In response to Physicists and chemists have often distinguished their disci- Expérimentât evidence becomes ü a burst of high-frequency stimuli, the major synaptic path- plines from the field of biology, emphasizing that biology available for the ^ ways in the hippocampus undergo a long-term change, was overly descriptive, atheoreticai, and lacked the coher- divisibility of I known as long-term potentiation or LTP. The advent in the ence of the physical sciences. This is no longer quite true. memory into î 1990s of the ability to genetically modify mice made ii pos- multiple syst«— In the 20th eentury. biology matured and became a coherent an animal mo' « sible to relate specific genes both to synaptic plasticity and discipline as a result of the substantial achievements of of human ami t to intact animal behavior, including memory. These tech- molecular biology. In the second half of the eentury. neuro- sia is developed. S niques now allow one to delete specific genes in specific seience emerged as a discipline that concerns itself with

vww.sciencemag.org SCIENCE VOL 290 10 NOVEMBER 2000 1119 PATHWAYS OF DISCOVERY

both biology and psychology and that is Each month, 8ritannica.com enhances the 423 (1978); J. Kaas and P. Garraghty, Curr. Opin. 1986 beginning to achieve a similar coher- Pathways of Discovery essay with links to Neurobioi 4. SZZ (1992). H.Robert Horvitz relevant items within and without Encylope- 25. L Ungerleider and M. Mishkin, in Tïie Analysis of discovers the ced ence. As a result, fascinating insights dia Britannlca's vast stores of information. To Visual Behavior. D. ). Ingle etat.. Eds. (MIT Press. genes, which ar« into the biology of cells, and remark- access this month's Pathways essay and all Cambridge, MA, 1982), pp. 549-586: A. Milner critical for pro- able principles of evolutionary conser- previous ones, go to www.britannica.com. and M. Coodale, The Visual Brain in Action grammed cetl ctick on the "Science" channel, and select (Oxford, New York, 1995). vation, are emerging from the study of "Features Archive." 26. R. H.Wurtz,/ Neurophysioi. 32, 727 (1969): E.V. Ï nerve cells. Similarly, entirely new in- — Evarts, in Methods in Medical Research, R. F. Rush- Patient R.B.«- sights into the nature of mental process- man, Ed. (Year Book. Chicago, 1966), pp, 241-250 tabtishes the im- es (perception, memory, and cognition) are emerging from 27. W.T.Newsome eta/., Nature 341, 52 (1939): C. D. Salzman et ai.. Nature 346,174(1990). portance of the the study of neurons, circuits, and brain systems, and com- hippocampus for 28. R.A.Andersen eta/., Anou. Rev. Neurose/. 20,303 (1997); R. Desimone and ' human memory. putational studies are providing models that can guide ex- J. Duncan, Annu, Rev. Neurosci. 18, 193 (1995); M. I. Posner and perimental work. Despite this remarkable progress, the neu- C. D. Gilbert, Proc. Nati Acad. Sd. U.S.A. 96. 2585 (1999); T. D. Albright 1990 and C. R. Stoner, Proc. Nati Acad Sd. U.S.A. 92, 2433 (1995); C. D. Gilbert, it$6gj Ogawa ano roscience of higher cognitive processes is only beginning. Physioi Rev. 73, 467 (1998); N. K. Logothetis. Philos. Trans. R. Soc. iondon. ; colleagues deveU For neuroscience to address the most challenging problems Ser.ß 353,1801 (1998). I op functional conlronting the behavioral and biological sciences, we will 29. For the binding problem, see Neuron 24 (i) (1999); for neural codes, see t magnetic reso- need to continue to search for new molecular and cellular M. N. Shadlen and W. T. Newsome, Curr. Opin. Neurobioi. 4. 569 (1994): J, nance Imaging. W. R. Softky, Curr. Opin. Neurobioi S. 239 (1995). approaches and use them in conjunction with systems neu- 30. S. Griilner et ai. Eds., Neurobiology of Vertebrate Locomotion. Wenner- , Mario Capetchi roscience and psychological science. In this way, we will Cren Center tntemational Symposium Series, vol. 45 (Macmillan, London, and Oliver best be able to relate molecular events and spécifie changes 1986); A. P. Georgopoulos, Curr. Opin. Neurobioi 10, 238 (2000). • Smythtes develop 31. L Sokoloff etai.J. Neumchem. 28,897 (1977): M. Reivich et ai. On:. Res. ^ gene knockout within neuronal circuits to mental processes such as percep- 44 127 (1979); M. I. Posner ant) M. E. Raichle, Images of Mind (Scientific \\ technotogy, which tion, memory, thought, and possibly consciousness itself. American Library, New York, 1994); S. Ogawa ef ai. Proc. Nati Acad. Sd. I Is soon applied to U.S.A. 87,9868 (1990); B.R. Rosen e( a/., Proc. Nat/./icad. Sd. U.S.A. 95, I ncuroscwnce. 773 (1998). References and Notes 32. W. 8. Scoville and B. Milner,/ Neuroi, Neurosurg.. Psychiatr. 20, 11 1991 1. S. Ramón y Cajal. Nobel Lectures: Physiology or Medicine (19O1-19¿1) (El- (1957); B. Milnet etat..Neuron 20, 445 (1998). lind« Buck and sevier, Amsterdam, 1967), pp. 22Û-2S3. 33. L R. Squire, Psychol. Rev. 99, 195 ( 1992); D. L Schacter and E. Tulving, Eds., ' Richard Axel dis- 2. C. S. Sherrington, The Central Nervous System, vol 3 of A Textbook of Phys- MemorySystems (MIT Press. Cambridge, MA, 1994). ' cover that the ol- iology. M. Foster, Ed. (MacMillan, London, ed. 7, 1897). 34. M. Mishkin, Phitos. Trans. R. Soc. London. Ser B 298, 85 (1982); L R. Squire factory receptor 3. A. L Hodgkin and A. F. Huxley, Nature 144, 710 (1939); A. L Hodgkin et ai. and S. Zola-Morgan, Science 253, 1380 (1991). family consists of J. Physioi (Lond.) 116. 4Z4 (1952); A. L Hodgkin and A, F. Huxley, J, Physi- 35. J. O'Keefe and J. Dostrovsky, Brain Res. 34, 171 (1971); H. Elchenbaum et over TOOO differ- oi (iond.) 117,500(1952). a/.,NeuTOn23,2O9(1999). ent gerws. The 4. E. J. Furshpan and D. D. Potter, Nature 180, 342 (1957); M, V. L Bennett, in 36. L. R. Squire and E. R. Kandel, Memory: From Mind to Molecules (Scientific anatomical com- Structure and Function of Synapses. C. D. Pappas and D. P. Purpura, Eds. American Library, New York, 1999); H. Eichenbaum, Nature Rev. Neurosa ponents of the (Raven Press, New York, 1972), pp. 221-256. 1. 1 (2000): P. Goldman-Rakic. P>J//OÏ. Trans. R. Soc. iondon. Ser. B 351, medial temporal 5. C. M. Armstrong and B. Hille, Neuron 20, 371 (1998); W. A. Catterall, Neu- 1445 (1996): R. Desimone, Proc. NatiAcad, Sd. U.S.A. 93, 13494 (1996); lobe memory sys- ron. \n press. B.Hitie et ai. Nature Medicine S, 1105 (1999); D.A. Doyle el S. Higuchi and Y. Miyashita, Proc. Nat/. Acad. Sei U.S.A. 93,739 (1996). tem «re identified. ai. Science 280, 69 (1998); J. P. Changeuy and S. ). Edelstein, Neuron 21, 37. R. F, Thompson and D. \. Krupa, Annu. Rev. Neurosci. 17. 519 (1994); 959 (1998); A. Karlin, Harvey Lecture Series 85, 71 (1991). J. LeDoux, The ¿motional Brain (Simon & Schuster, New York, 1996); 1993 6. E. Neher and B. Sakmann, Nature 260, 799 (1976). ). L McGaugh, Science 287, 248 (2000): M. Mishkin et ai. in Neurobiotogy The Huntington's 7. R. J. Lefkowitz, Nat CeU Bioi 2. E133-6 (2000), of Learning and Memory. G. Lynch et ai.. Eds. (Guilford, New York, 1984), Disease Collabo- 8. L 8uck and R. Axel, CeH 65,175 (1991). pp. 65-77; D. L. Schartet and R. L Buckner, Neuron 20,185 (1998). !• rative Research 9. E. J. Nestler and P. Creengard, Protein PhosphorylatJon in the Nermus fys- 38. V. Castelluccieta/., Sdence 167, 1745 (1970); M, Brunelli et ai. Science , Croup identifies tem (Wiley, New York, 1984). 194, n78(1976);C. Bailey eta/.. Proc. Nat. Acad. Sc/. U.S.A. 93, 13445 the gene respon- 10. J, Del Castillo and B. Kati, / Physioi 124, 560 (1954); B. Kati, in 7ïie Xth (1996). . siblefor Hunting- Sherrington Lecture (Thomas, Springfield, IL, 1969). 39. S. 8enzer, Sd. Am. 229, 24 (1973): W. G. Quinn and R. J. Greenspan, Annu. 11. T. Sudhof, Nature 375, 645 (1995); R. Scheller, Neuron 14. 893 (1995): Rev. Neurosci. 7. 67 (1984); R. L. Davis, Physioi Rev. 76, 299 (1996): J. Vin J. A. McNew et ai. Nature 407, 153 (2000). and T.Tully Curr. Opin. Neurobioi. 6,264 (1996). 1990s 12. S. Cohen and R. Levi-Montalcini, Proc. Nati Acad. Sei. U.S.A, 42, 571 40. D. O. Hebb. The Organization of Behavior A Neuropsychologkat Theory Neural develop- (1956); W. M, Cowan, Neuron 20, 413 (1998). (Wiley, New York, 1949). - ment is trans- 13. M. M. Metzstein e( ai.. Trends Genet. 14, 410 (1998). 41. T. V P. Bliss and T. Lomo,/ Physio/, (iond.) 232,331 (1973). formed from a de- 14. R. W. Sperry, Proc. Nati Acad. Sei. U.S.A. 50, 703 (1963); R, W. Hunt and 42. M. Mayfofd etal.. Science 274, 1678 (1996); J.Tsien eta/., CeH 87, 1327 scriptive to a W. M. Cowan, in Brain. Circuits and Functions of Mind. C, B. Trevarthen, Ed. (1996);A. Silva eta/,. A/inu.ftev.Neuroin. 21, 127 (1998); S. Martin et a/.. molecular disci- (Cambridge Univ. Press, Cambridge. 1990), pp, 19-74. Annu. fíev. Neurosci 23, 613 (2000); E. P. Huang and C. F. Stevens, fssays pline by Gerald 15. M.Tessier-Lavigne and C. S. Goodman, Sdence 274, 1123 (1996);T. M. Jes- Biochem. 33,165 (1998); R. C. Matenka and R.A. Nicoll, Sc/ence 28S, 1870 Fischbach, jack seil, Nature Rev. Genet. 1, 20 (2000); T. M. Jessell and ]. R. Sanes. Curr. (1999); H. Korn and D. Faber, CRAcad. Sd. Ill 321,125 (1998). McMahan.Tom Opin. Neurobioi. in press. 43. W.T.Creenough and C.H.Bailey, TrendsNeuTOid, 11,142(1998), Jessell. and Corey 16. L C. Katz and C. J. Shatz, 5c/ence 274, 1133 (1996). 44. D. V. Buonomano and M. M. Merienich, Annu. Rev, Neurosci. 21, 149 Goodman; neu- 17. Huntington's Disease Collaborative Research Croup, Ceíí 7Z, 971 (1993): (1998); C. Gilbert, Proc. Nat. Acad. Sei. U.S.A. 93 10546 (1996);T. Elbert et roimaging is ap- H. L. Paulson and fC. H. Fischbeck, Annu. Rev. Neurosci 19, 79 (1996). al.,Science 270, 305(1995). plied to problems 18. D. M. Walsh et ai, Biochemistry 39. 10831 (2000); W, ), Strittmatter and 45. The work of E.R,K. is supported by NIMH, the C. Harold and Leila Y. Math- of human cogni- A. D. Roses, Annu. Rev. Neurosci 19. 53 (1996); D. L Price. Nature 399 ers Foundation, the Lieber Center for Research on Schizophrenia, and the tion, inciudirtg (6738) (Suppl). A3-5 (1999). Howard Hughes Medical Institute. The work of L.R.S. is supported by the perception, atten- Medical Research Service of the Department of Veterans Affairs, NIMH, tion, and memory 19. S. E. Hyman, Arch. Gen. Psychiatr. 157, 88 (2000): D. Charney et ai,. Eds,, Neurobiotogy of Mental Illness (Oxford, New York. 1999); S. H. Barnodes, and the Metropolitan Life Foundation. We thank Thomas Jessell and Molecules and Mental Illness (Scientific American Library, New York, 1993); Thomas Albright for their helpful comments on the manuscript. Reinhard Jahn, S. Snyder, Drugs and the Brain (Scientific American Library, New York. James Rothman, 1986). Richard Schelter, 20. A. Carisson, Annu. Rev. Neurosci. 10, 19 (1987); J. Axelrod, Science 173, and Thomas Sud- Eric R. Kandel is University Professor, Columbia University, New York, and 598 (1971); S. S. Kety.4m./ Psychiatr. 140, 720 (1983); N. A. Hillarp et ai, hof delineate Senior Investigator at the Howard Hughes Medical Institute. He is a mem- Pharmacoi Rev. 18, 727 (1966). molecules critical ber of the National Academy of Sciences and the Institute of Medicine and 21. t S. Kilduff and C. Peyron, Trends Neufosd. 23, 359 (2000); K. L House- ' exocytosis a past president of the Society for Neuroscience. This past October, he was knecht et ai.,/ Anim. Sd. 76, 1405 (1998); G. F. Koob, Ann. N.Y. Acad. Sd. named a co-recipient of the Nobel Prize in physiology or medicine, 909,17 (2000); E. J. Nestler, Curr. Opin. Neurobioi 7, 713 (1997). 22. V. B. Mountcastle, y. Neurophysioi. 20, 408 (1957); D. H. Hubel and Fir« 3D structure T. N. Wiesel,/ Physioi 148, 574 (1959): D. H. Hubel and T. N. Wiesel, Weu- Larry R. Squire is Research Career Scientist at the Veterans Affairs San of an Ion channel ron 20. 401 (1998). Diego Healthcare System and Professor of Psychiatry, Neurosciences, and ¡< is revealed by Z3. T. Wiesei and D. Hubel,/ Neurophysioi 26,1003 (1963). Psychology, University of California, San Diego. He is a member of the Rod MacKinnon. 24. D. Van Essen, in Cerebral Cortex. A. Peters and E. G. Jones. Eds. (Plenum National Academy of Sciences and the Institute of Medicine and a past Publishing Corp.. New York. 1985), vol. 3, pp. 259-327; S. Zeki, Nature 274. president of the Society for Neuroscience,

1120 10 NOVEMBER 2000 VOL 290 SCIENCE www.sdencemag.org Copyright of Science is the property of American Association for the Advancement of Science and its content may not be copied without the publisher's express written permission except for the print or download capabilities of the retrieval software used for access. This content is intended solely for the use of the individual user. Copyright of Science is the property of American Association for the Advancement of Science and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.