National Academies Press: OpenBook

Opportunities in Biology (1989)

Chapter: 6. The Nervous System and Behavior

« Previous: 5. Development
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 175
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 176
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 177
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 178
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 179
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 180
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 181
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 182
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 183
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 184
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 185
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 186
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 187
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 188
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 189
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 190
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 191
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 192
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 193
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 194
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 195
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 196
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 197
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 198
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 199
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 200
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 201
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 202
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 203
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 204
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 205
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 206
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 207
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 208
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 209
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 210
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 211
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 212
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 213
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 214
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 215
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 216
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 217
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 218
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 219
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 220
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 221
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 222
Suggested Citation:"6. The Nervous System and Behavior." National Research Council. 1989. Opportunities in Biology. Washington, DC: The National Academies Press. doi: 10.17226/742.
×
Page 223

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

6 The Nervous System and Behavior The Objective of Modern Neuroscience Is to Understand How the Nerve Cells of the Brain Direct Behavior Many central issues with which neurosciences is concemed, such as how we perceive the world around us, how we learn from experience, how we remember, how we direct our movements, and how we communicate with each other, have commanded the attention of thoughtful men and women for centuries. But it was not until after World War II that neuroscience began to emerge as a separate and increasingly vigorous scientific discipline that has as its ultimate objective pro- viding a satisfactory account of animal (including human) behavior in biological terms. This ambitious goal has as its basis the central realization that all behavior is, in the last analysis, a reflection of the function of the nervous system. It is the organized and coordinated activity of Me nervous system that ultimately mani- fests itself in the behavior of the organism. The challenge to neuroscience then, is to explain, in physical and chemical terms, how the nervous system marshalls its signaling units to direct behavior. The real magnitude of this challenge can perhaps be best judged by consider- ing the structural and functional complexity of the human brain and the bewilder- ing complexity of human behavior. The human brain is thought to be composed of about a hundred billion (10~) nerve cells and about 10 to 50 times that number of supporting elements or glial cells. Some nerve cells have relatively few connections with other neurons or with such effecter organs as muscles or glands, but the great majority receive connections from thousands of other cells and may themselves connect with several hundred other neurons. This means that at a fairly conservative estimate the total number of functional connections (known as synapses) within the human brain is on the order of a hundred trillion (10~4~. But what is most important is that these connections are not random or indiscriminate: 175

176 OPPORTUNITIES IN BIOLOGY They constitute the essential "wiring" of the nervous system on which the extraor- dinarily precise functioning of the brain depends. We owe to the great neu- roanatomists of the last century, and especially to Ramon y Cajal, the brilliant insight that cells with basically similar properties are able to produce very differ- ent actions because they are connected to each other and to the sensory receptors and effecter organs of the body in different ways. One major objective of modem neuroscience is therefore to unravel the patterns of connections within the nerv- ous system in a word, to map the brain. A second, related objective is to identify the differences that exist between nerve cells. For although nerve cells have a number of properties in common especially their abilities to respond to signals from other cells and to conduct signals along their processes-on morphological grounds alone, thousands of different classes of nerve cells are evident. The morphological differences were the first to be recognized once techniques had been developed that reveal the form of individual neurons. Some cells were found to have only a single process, others just two processes, and still others-including the overwhelming majority of neurons in the brains of vertebrates-have several, often scores, of processes. In most cases we can recognize a single process, the axon, that serves to conduct information-usually in the form of all-or-none signals known as action poten- tials or nerve impulses to other cells. Variable numbers of receptor processes or dendrites receive information from other cells, integrate it, and relay it to the nerve cell body and beyond it to the axon. But it is not only in the morphology of the processes that nerve cells differ. We now know that dozens of different classes of neurons can be recognized on the basis of the chemical messengers or neurotransmitters that they use to communi- cate with other cells. The discovery in the early 1950s that almost all nerve cells communicate with each other through the release of chemical neurotransmitters at specialized sites along the course and at the ends of their axons was one of the major events that marked the beginning of modern neuroscience. Only in the past decade, however, have we come to realize that there may be not just a handful of chemical transmitters as was once thought, but perhaps a hundred or more, and that it is the subtle and distinctive actions of these transmitters that account for much of the functional complexity of the nervous system. We have also come to realize that neurotransmitters can act upon other cells only if the cells have the necessary receptors to selectively bind the neurotransmitter. The interaction of the neurotransmitter with its appropriate receptor is what initiates the response of the target cell. Again, it is only in the past few years that we have come to appreciate that the target cells can respond in several different ways depending on the nature of the transmitter, the types of receptors involved, and the mechanisms that the transmitter-receptor interaction activates. In some cases the response of the target cells is rapid and transitory, with a time course of just a few thousandths of a second; in other cases the cell responds over a fairly long period perhaps many seconds; and in certain situations the behavior of the target cell may be

THE NERVOUS SYSTEM AND BEIlAVIOR 177 modified for many hours, or even days and weeks. A third major task of neuroscience, therefore, is to understand how nerve cells generate signals, often over long distances, and how these signals change the various target cells with which the neurons are in functional contact. The cellular and molecular mechanisms involved in nerve signaling and synaptic transmission are currently among the most intensively studied and best understood aspects of neuroscience. Less well understood, but no less important, are the longer term changes in nerve cells that must underlie the acquisition and storage of information that we commonly refer to as learning and memory. Although there is a vast body of literature on human learning and memory and on the effects of damage to various parts of the brain on its ability to acquire, store, and retrieve information, it is only relatively recently that the longer term effects of synaptic activity that must be involved in these processes have begun to be studied at the cellular and molecular levels. The first insights that we have gained into these processes suggest that a wide variety of behavioral phenomena may well prove to be explicable on the basis of just a few general mechanisms such as the covalent modification of particular molecules involved in nerve signaling or the activation of specific genes and the synthesis of new proteins. A fourth major objective of neuroscience is to account for the unusual cell biology of neurons. Although nerve cells share many properties with other cells, their special roles in the transduction of sensory information, in the transmission of signals over considerable distances, in being able to respond to signals from other cells, and, in turn, in being able to modify the activity of their target cells imposes on neurons a number of highly specialized functions. These considera- tions raise a number of intriguing questions including (1) how the enormous phenotypic diversity seen among nerve cells is generated, (2) how different parts of each neuron become specialized to either receive or transmit signals, (3) how nerve cells are able to maintain such lengthy processes given that the genetic information is confined to the cell nucleus and most of the synthetic machinery is confined to the relatively small cell body, (4) how communication is maintained between the nerve cell body and its various processes, and (5) what changes occur in the cells in response to "experience" and aging. The fact that most neurons have to survive and continue to function effectively throughout the life of the organism for 70 or more years in the case of neurons in the human brain is one of their most impressive characteristics. Recent developments in molecular and cell biology are beginning to influence the study of these phenomena, and there is every reason to be confident that they will soon be as well understood as the mechanisms involved in impulse conduction and synaptic transmission. Undoubtedly the greatest challenge to contemporary and future neuroscience is to understand what might be referred to as the "information-processing" capac- ity of the brain to determine how the various systems within the brain are organized and function to direct and mediate such behavioral phenomena as sensory perception, language function, motor actions, emotion, cognition, and

178 OPPORTUNITIES IN BIOLOGY thought. Again, although we know from clinical neurology and pathology that the destruction of certain areas of the brain seriously impairs or effectively abolishes these capacities, how these higher functions are normally carried out remains largely unknown. That it has been possible to produce machines that can dupli- cate some aspects of these higher brain functions has suggested that developments in computer science and especially artificial intelligence may inform our under- standing of how the brain functions in much the same way as molecular and cellular neuroscience have been informed by concurrent developments in molecu- lar and cell biology. But this remains to be seen, and for the present the single greatest challenge to neuroscience is to elucidate how the brain works. It Is Important to Use Both Reductionist and Synthetic Strategies for Studying the Nervous System The strategies that neuroscientists have adopted for studying the nervous system have varied over the years as new techniques and methods have been developed. But from the beginning they have all been based on a few general premises. Among the more important of these have been the following: (1) Animal behavior reflects the activity of the nervous system; (2) no matter how simple or complex the nervous system or the behavior, the essential units-the neurons are alike in most significant respects; and (3) because of the structural and functional similarity in neurons, it is important to select the neuronal system or the neurons that are most advantageous for study, regardless of where they are found. Before the introduction of micropipettes as recording electrodes, the giant axons that control the mantle musculature of the squid were the objects of choice in the study of the physico-chemical mechanism of nerve signaling: They were several centimeters long and up to 0.5 millimeters in diameter, they could be easily dissected out, they remained viable in a recording chamber for many hours, and long, insulated wire electrodes could be inserted for some distance down their length so the potential difference across the axonal membrane (axolemma) at rest and during the conduction of an impulse could be measured. Later when it became important to study the electrical events that occur in both the pre- and postsynaptic elements during synaptic transmission, an unusual "giant" synapse in the squid nervous system proved to be invaluable. Over the years the most useful preparation for studies of synaptic transmission in verte- brates has been the neuromuscular junction, because it is readily identifiable and easy to handle, and because it is relatively easy to record from the postsynaptic cells. Similarly, the modified muscle cells that compose the electric organs of certain fish have provided the richest source of the receptor for the neurotransmit- ter released at nerve-muscle junctions. The availability of the acetylcholine (ACh) receptor in such large amounts made possible, first the isolation and biochemical purification and characterization of the four subunits of which the receptor is composed, and later the cloning of the genes for each subunit.

THE NERVOUS SYSTEM AND BEHAVIOR 179 The earliest studies of behavior were largely descriptive, but in the hands of the neurophysiologists and ethologists they provided a wealth of information about reflex (including conditioned reflex) and "instinctive" behavior, some of which has since been analyzed electrophysiologically. Because of the difficulty of analyzing behavior in most complex organisms, considerable attention has been paid to the analysis of more simple, rigorously definable behaviors in simple organisms. For example, much of what we know about the way in which motor activity is programmed has come from the analysis of locomotor behavior in crayfish and leeches. More recently, scientists have taken advantage of certain large, readily identifiable neurons in the sea snail Aplysia to analyze the short- and long-term changes that occur in neuronal structural and function in certain well- defined behaviors including an especially useful model for nonassociative and associative learning. The Aplysia nervous system has also proved useful for studying gene expression in specific neurons and for determining the neuropep- tides that mediate even complex behavior such as egg-laying. In an attempt to work out the complete organization of a nervous system in an animal which is readily amenable to genetic manipulation, the nematode Caeno- rhabditis elegans, which has just over 300 neurons, has been studied. The lineage of each of these neurons has been determined (by direct inspection), specific lineage mutants have been identified, and the complete wiring pattern of the nervous system has been reconstructed from serial electron micrographs. Of even greater use for genetic studies is the nervous system of Drosophila. Mutations that affect different parts of the nervous system have been known for many years, but it is only in the past 15 years that a major effort has been made to identify behavioral mutants, to analyze the development of the nervous system (including the eye), and to clone the genes for a number of interesting neuronal proteins such as rhodopsin and one of the potassium (K+) channels. The power of this approach, which has already yielded so much in other systems, holds great promise, especially for understanding the molecular mechanisms involved in neural development. While we cannot emphasize too strongly the importance of this search for simple systems in which to analyze specific aspects of neuronal functions, in the last analysis the greatest interest lies in understanding the functions of the human brain. Until recently the opportunities for doing this were limited. Clinical neurologists led the way by analyzing brain function resulting from localized or more general brain pathologies. Much of what we know about the organization of the human brain has come from studies of this kind, and when patients with specific brain lesions (such as an interruption of the corpus callosum that unites the two cerebral hemispheres, or localized damage to the speech areas or to those concerned with memory processing) have been carefully studied by neuropsy- chologists, they have revealed aspects of brain function that could not have been determined in any other way.

180 oppo~ruNlTlEs INBlOLOGY In the past decade a number of noninvasive methods have been developed for studying the human brain. These include computerized axial tomography (CAT scanning), positron emission tomography (PET scanning), magnetic resonance imaging (MRI), event-related potential recordings from the scalp with computer- ized averaging techniques, and magnetoencephalography. The full impact of these technologies has yet to be felt, but it is already clear that they will enable us to study many aspects of brain function in the intact human brain that hitherto could be analyzed only in the brains of experimental animals. For the study of those distinctive aspects of human behavior, such as speech, they should prove invaluable. Useful as these noninvasive methods are proving to be, at present they suffer from severe limitations in spatial or temporal resolution (or both). They usually provide information only about the summed activity of larger numbers of neurons rather than about the functions of individual neurons. For studies of this kind, we must still turn to experiments on animals. The need for such experiments will continue, for, despite efforts to find alternatives, the only hope we have in the forseeable future of understanding the organized activity of the brain is by directly studying the brain itself. Simulations and computer modeling of brain functions are no substitute for direct observation. One of the most promising developments in this regard has been the perfect- ing of techniques for recording the activity of individual neurons in conscious, behaving primates. This approach, first introduced about 15 years ago, has become increasingly popular for studies of sensory perception and motor control. It has been possible in several instances to train the experimental animal to carry out a psychophysical task for which comparable human performance can be measured; investigators are now beginning to collect substantial data on the activities of neurons in parts of the brain (including the so-called association areas, which had hitherto defied analysis) that are likely to be directly applicable to human brain function. An extension of this approach that permits simultaneous multiple recordings from many neurons is one of the promising recent develop- ments in neuroscience. Nerve Cells Are the Signaling Units of the Brain As we have seen, almost all nerve cells have at least three or four main parts: (1) a cell body that contains the nucleus and most of the cell's biosynthetic machinery; (2) a number of relatively short processes, called dendrites, which extend from the cell body and provide the largest receptive surface for inputs to the cells; (3) an axon, which usually extends for some distance from the cell body and is used for long-range signaling; and (4) specialized regions, commonly at the end of axons, called synaptic boutons or synaptic endings, where communication with other nerve cells or special effecter tissues (such as gland or muscle cells) is carried out.

THE NERVOUS SYSIEM AND BEHAVIOR 181 The best way to understand how these various components of a neuron work is to consider them in the context of a simple behavior-for example, the reflex withdrawal of a hand that touches a very hot object. Contact with the hot object activates a group of sensory receptors in the skin that respond to heat and causes them to fire a burst of all-or-none signals called action potentials. These action potentials propagate along the length of the sensory neurons, past the cell bodies, and to the axons, which extend into the spinal cord. At the ends of the axons the action potentials cause a chemical transmitter to be released. The chemical transmitter released at the ends of several axons interacts with receptors on the surfaces of the dendrites of certain spinal cord neurons giving rise to an activating signal called an excitatory postsynaptic potential. If the excitatory potentials elicited by impulses in the sensory axons are of sufficient amplitude, they trigger a nerve impulse, or a group of impulses, in the spinal cord cells. These, in turn, through their axons, activate a group of motor or effecter nerve cells. The axons of the motor cells extend out from the spinal cord to the muscles in the forearm and hand, where again a chemical transmitter is released at the nerve-muscle junction. The binding of the transmitter to the appropriate receptor in the muscle causes a brief change in the surface membrane of the muscle cells that leads the muscles to contract and the hand to withdraw. Concurrent with the excitation of neurons in the spinal cord that activates motor neurons, some of the branches of the sensory axons contact yet other spinal neurons that, when activated, inhibit the activity of the motoneurons that normally cause the forearm and hand to extend. The activation of yet other neurons in the spinal cord leads to the propagation of information about the sensory stimulus (its location, nature, and intensity) to higher levels within the nervous system. These lead, among other things, to the conscious perception that the hand has been in touch with a hot object, and if the stimulus is severe enough, a generalized arousal of the individual that focuses attention on the stimulus and its behavioral significance. If the hand is jerked back with sufficient vigor, there may also be a number of reflex adjustments within the spinal cord to maintain the subject's balance and posture. This simple example serves to make several general points about the nervous system and its role in behavior. (1) Most behavior occurs in response to an external sensory stimulus of some kind; (2) sensory signals must be transduced into nerve signals; (3) nerve impulses travel along specific pathways to defined areas of the central nervous system; (4) nerve cells communicate with each other through specialized junctional zones known as synapses; (5) synaptic transmis- sion can be either excitatory or inhibitory depending on the chemical neurotrans- mitter involved; (63 most behavior manifests itself in the form of overt motor actions; and (7) many sensory stimuli are also consciously perceived as a result of the transmission of information to higher brain centers including the central cortex, and this perception may result in conscious arousal and the focusing of attention on the stimulus and its behavioral consequences.

182 OPPORTUNITIES IN BIOLOGY Considerable progress has been made in recent years in our understanding of all of these steps: sensory transduction, the nature of the nerve impulse, synaptic transmission, the anatomical pathways involved in a variety of sensory and motor mechanisms, and how these pathways are assembled during development. In the sections that follow we shall try to summarize what is known about these issues and to point out the directions in which future work seems to be headed. NERVE CELL COMMUNICATION Nerve Cells Communicate by Electrical awl Chemical Signals To understand how neurons and synapses work, we need to understand how a nerve impulse or action potential in a presynaptic neuron causes the release of a chemical neurotransmitter at the synapses formed by its axon. But fast we must focus on the ionic currents that produce the action potential in the presynaptic neuron and on the way these currents interact with the structures in the terminal parts of the axon to bring about the release of the transmitter. The use of certain naturally occurring neurotoxins that bind specifically to the sodium channel has made possible a preliminary molecular characterization of the sodium channels in muscle and brain. The channel is a large glycoprotein with a molecular weight of 270,000 whose amino acid sequence has been deter- mined from the corresponding complementary DNAs. The amino acid sequence has, in tum, suggested several ideas about the function and evolution of the different segments of the channel protein. For example, the molecule contains four similar sequences (homologous internal repeats), each about 150 amino acid residues long: These repeat sequences suggest that the channel may have evolved from a single ancestral DNA segment that was duplicated within the gene three times. By examining the distribution of specific amino acids within the entire peptide, it is possible to identify candidate domains concerned with various functional properties. In particular, each internal repeat has five long hydropho- bic areas that probably represent the transmembrane domain and a charged segment that is thought to serve in the Bating process (that is, the opening and closing of the channel) and in the selectivity of the channel for sodium. The channel probably responds to changes in membrane potential by undergoing a conformational change and a masking of the positive charges that bound its pore. The techniques of site~irected mutagenesis to modify specific sites in the chan- nel protein should make it possible to test these ideas directly. The analysis of the voltage-gated sodium channel has brought to light two features we shall encounter again when we consider the acetylcholine receptor. First, several stretches of hydrophobic amino acids seem to correspond to trans- membrane alpha helices. Second, the channel is symmetrical, consisting of similar subunits arranged in the plane of the membrane around a central aqueous pore. These early findings encourage us to think that all membrane channels are

THE NERVOUS SYSTEM AND BEHAVIOR 183 :: ~ : : ::: ~ ::~:::~RE:STINGAND~AGTION~POTE~NTIALS -I ~ ~ : :: ~ : ~ it: ~ ~ ~::~At~::rest, all nerve: cells: ~:have~an ~ el~r~al potential across :their plasma:: :~ :membranes~: of: abobt~50 ~mi:lluro~ks~ (mV)~, th~e~::inside~ Of :~the::~ - ll: being~n~g~tive~: :~1:~ ~: :::~:~:with respect tote outside Refigure 6-1~). An :~adtio~n~potential~:~'s :an: ~all~r-~none~ ::: signal that: not only ~:reduces~b~: actually reverses~th~e mem~b~rane:~potential,~ ::: i: ~1 Moving Ma total:of About 100 m:V~from Isis resting valu:e~of~-5Q: molto +50~ my/. :: The action ~p~ential~ is generated:: by the Movement ~ Fins through Go types ~ of ~ ~'ntr~ns~c membrane prawns ~ G alled~::' voltag~gated~': channels. ~ ~ ~those~for : I:: sodi u m and: those Far ~pot:assiu m. ~ ~ Each Of ~ these ~ channels ~ is Hosed At rent :: and o~pens~'n:::an ::~all~r-nan~e:: fashion as Alar result:: of: a reduction in :fhe mem horn :~tanti;al b~Iow:a~critica:l vie. When open, loach channel permits theta :: : :: it: rapid:~:~:~movement:~ ::~of~::~: ~:io;ns throug~h~::th~:~membran:e,:: givin 9 nse:~::to:: ,ouises~: Ot~:~ :cu~rrent:~ofYariable~d~u~ration~butOo~nsta:ntamplitude.:~::S6dium~,~beTng~:~a:~p~osi-~ ~ tivel~y ~:charged~ mn~ Which, i: ~ rest, finis i: at higher concentratio~r,~outs~ide~:~:;the i: ~ ~neuron: moves: :rapid:ly::into: the ::oell from ~th:e:::::outside: b (ingoing in positive t ~ ~ :c;harge,: this movement causes: ~:the rising :or :depolarizing ph~ase~:~6f::the ~ acting ::: :: : : ~ : :: pat~ential.:::~ln:turn, ~potassi~u:m: ions: carry~posKivQ charge outfox the cell; thirsts ::~ movement~is responsible for the declining or re~lar~zing phase The Ton:: potential. : ~ i:: : ~:~ likely to share certain common structural features. Once these are understood, it may turn out that all membrane channels work in much the same way. But at present, the conceptual gap between the primary structure of the channel proteins and their function is too large to allow us to make this prediction with any degree of confidence. Synaptic Transmission: The Nerve-Muscle Junction as a Prototypical Example The nerve impulse is essentially a form of electrical signaling, with the wave of ionic currents sweeping down the surface of the axon at speeds in the range of 1 to 100 meters per second. Communication between different neurons and be- tween neurons and other cells is chemical in most instances the release from the nerve endings of a small amount of a specialized neurotransmitter that diffuses across the space separating the two cells. The binding of the transmitter to receptor molecules in the membrane of the postsynaptic cell gives rise, in turn, to a new class of signals called synaptic potentials. Thus, whereas the action potential is a purely electrical signal, the synaptic potential is an electrical signal initiated by a chemical one. In the past two decades a large number of such

184 OPPORTUNITIES IN BIOLOGY On > +50 ._ - .m a) o Q Q -50 a) o FIGURE 6-1 Resting and action potentials. T. Ime chemical transmitters haves been identified: They range from small molecules such as ACh, glutamate, noradrenaline, and serotonin to much larger molecules including a rapidly growing number of peptides. Over the past 30 years, each of the steps involved in synaptic transmission has been characterized in considerable detail, primarily through the use of intra cellular microelectrodes and thin-section electron microscopy. The pace of research on synaptic biology has increased rapidly in recent years since the introduction of rapid freeze-fracture electron microscopy, patch-clamp micro electrode techniques, and the application of modern methods of protein chemis try, recombinant DNA probes, and monoclonal antibodies to the isolation and characterization of the molecular mechanisms involved. Because of its ease of access and because it was the first site at which chemical transmission was identified, the neuromuscular junction between the motor axon terminals and muscle cells-has been the most intensively studied synapse and illustrates the major features of synaptic transmission. When an action potential invades a motor nerve terminal, it releases the transmitter ACh after an irreducible delay of about 0.5 to 1.0 milliseconds. The transmitter then diffuses across the 50-nanometer synaptic cleft between the nerve terminal and the muscle cell in a matter of about 200 msec before binding to ACh receptors in the junctional region of the muscle membrane. The interaction of the transmitter with the receptor leads to a conformational change in the receptor and the opening of its channel. This is followed by an influx of sodium ions that depolarizes the postsynaptic membrane. In contrast to the all-or-none nature of action potentials, the depolarization of the muscle membrane the end-plate potential is a local response proportional to the amount of ACh released. Under

THE NERVOUS SYSTEM AND BEHAVIOR 185 normal ionic conditions the end-plate potential is usually large enough to trigger an action potential that is then conducted away from the end-plate region along the surface of the muscle cell. The action of the synaptically released ACh is brief: The end-plate current decays within 1.0 to 2.0 msec. The duration of action of the released ACh is limited by its breakdown by an extremely active enzyme, acetylcholinesterase, that is concentrated in the synaptic cleft. The discovery in the early 1960s of a-bungarotoxin, an 8,000-dalton peptide that binds specifically and with high affinity to ACh receptors in skeletal muscle, provided a crucial tool for biochemical studies of ACh receptors. The receptor, first purified from electric organs of the rays Torpedo California and Torpedo marmorata (uniquely rich sources that are embryonically related to muscle cells) is a 275,000-dalton glycoprotein made up of four subunits. Complementary DNA cloning techniques have revealed that each subunit contains four hydrophobic regions that presumably span the lipid bilayer. A fifth, amphipathic domain, located between the third and fourth hydrophobic regions, may also span the membrane. The ACh receptor has led the way because of the relative ease of its purifica- tion from receptor-rich electric organs, but complementary DNAs that code for other receptors or ion channels (ligand-gated and voltage-gated) have now been cloned and sequenced. It has been known since 1914 that ACh activates two broad classes of receptors. Nicotinic ACh receptors of the sort we have so far considered are present at motor end plates, at synapses within autonomic ganglia, and at a few synapses within the central nervous system (CNS). Muscarinic ACh receptors are found in various autonomic effecters including smooth muscle, cardiac muscle, and exocrine glands. Most ACh receptors within the brain are muscarinic. This receptor has now been cloned, as has the p-adrenergic receptor. Relatively Little Is Known About the Molecular Details of Transmitter Release at the Neuromuscular Junction In addition to the two major currents involved in the action potential (the sodium and potassium currents), a third, minor current is particularly important at the presynaptic terminals of the synapse: the calcium current. The calcium current is small, only about 1/100 of the sodium or the potassium current, and therefore it does not usually contribute importantly to the action potential per se. Rather, it serves as a messenger carrying into the cell information that is necessary for release of the chemical transmitter. The function of calcium in this context can best be understood if we shift our attention from the receptors on the postsynaptic cell the muscle to the presyn- aptic terminals of the motor neuron. The chemical transmitter is released from the axon terminals, not as isolated molecules, but in packets containing about 5,000 molecules of ACh. Enclosed in small subcellular organelles called synaptic vesicles, these packets of ACh are released from the expanded terminals of the

186 OPPORTUNITIES IN BIOLOGY axon by an exocytotic process, in which vesicles fuse with the external membrane of the synaptic terminal and extrude their contents into the synaptic cleft. Exocy- tosis occurs only at certain points, called active zones, within the axon terminal. At these active zones, a gridlike array of"dense projections" provides a cluster of docking sites for the vesicles. Calcium is thought to be essential for fusion of the synaptic vesicles with the external membrane of the axon, and hence its entry into the axon terminal is a necessary prerequisite for the exocytotic release of the transmitter. One of the best-characterized proteins associated with synaptic vesicles is synapsin I, a phosphoprotein with a globular head and an elongated, collagenlike tail. Synapsin I seems to be involved in transmitter release, although it remains to be determined whether it regulates vesicle-membrane fusion (directly or indi- rectly) or if it acts on another process such as vesicle mobilization. There Are a Variety of Neurotransmitters The number of putative neurotransmitters and neuromodulators of synaptic function has increased tremendously in the past decade, and many more probably remain to be identified. In addition to ACh, catecholamines (norepinephrine, dopamine), indolamines (serotonin), and amino acids (glutamate, y-aminobutyric acid or GABA, and glycine), attention has been directed recently to purines and to a wide variety of peptides, several of which were first identified and characterized in other tissues as possible local or more distally acting hormones. Although many of these peptides have been shown to excite or inhibit the firing of nearby neurons at relatively low concentrations, and while their selective distribution within the nervous system argues strongly that they may function as neurotrans- mitters, at present the great majority of neuroactive peptides must still be regarded as transmitter candidates. Neuropeptides tend to alter the membrane properties of target neurons on a slower time scale, causing them to respond differently to other ongoing synaptic drives; for this reason the term neuromodulator may be more appropriate than neurotransmitter. Yet another distinction between peptides and more conventional transmitters is the manner in which they are synthesized. Most conventional transmitters are synthesized by enzymes located in nerve terminals; peptides, on the other hand, are synthesized as parts of larger polyproteins on polynbosomes in the cell body. The polyprotein may contain several active peptides or multiple copies of the same peptide: To cite just one example, pro-opiomelanocortin gives rise to adrenocorticotropic hormone (ACI H), p-endo~phin, melanocyte-stimulating hormone, and another small peptide, all of which are cleaved from the parent protein by appropriate enzymes. Undoubtedly one of the greatest challenges to neurobiology is to determine the role of the more than 50 neuropeptides that have been identified and of the 100 or more that we believe are yet to be discovered.

THE NERVOUS SYSTEM AND BEHAVIOR Some Synaptic Receptors Activate Ion Channels by Means of Second Messengers 187 Ionophoric receptors and those for ACh function as ion channels and mediate fast synaptic transmission. Depending on which ions pass through the membrane, the effect on the postsynaptic cell will be excitatory or inhibitory. Whatever the sign, the effect is brief (usually just a few milliseconds) because it depends on a change in the conformation of the receptor protein molecules. In contrast to ionophoric receptors, other receptors are coupled to ion channels by guanosine triphosphate proteins that bind second messengers, such as cyclic adenosine monophosphate (cyclic AMP), cyclic guanosine monophosphate (cylic GMP), calcium, diacylglycerol, or arachidonic acid. Second messengers typically dif- fuse within the cell to deliver the information from the receptor to various target proteins, modifying these proteins' activities and thereby modulating the physio- logical responses of that cell. The information that a second messenger delivers can be amplified to alter both the magnitude and the duration of the signal. At their briefest, these changes persist for as long as the store of receptor-stimulated second messengers lasts- usually minutes to an hour. However, the activation of one second-messenger pathway may lead to changes in others, and cumulatively these changes can be prolonged for hours. If the cascade of receptor-stimulated reactions alters gene expression (for example, inducing the synthesis of new protein molecules within the stimulated cell), the effects may endure for days, weeks, or even months. All second messengers studied so far operate by modifying the activity of proteins within the stimulated cells. Although most of the modifications studied are produced ultimately through the activation of one or another enzyme, some second messengers (like cyclic GMP in photoreceptor cells of the retina) may act directly on their target molecules without the intervention of other enzymes. To date, the best-studied neuronal second messengers are calcium, cyclic AMP, and the products of phosphoinositol turnover that are released from the plasma mem- brane by receptor-mediated activation of the enzyme phospholipase C. THE CYTOSKELETON AND THE TRANSPORT OF MATERIALS WITHIN NERVE CELLS Among animal cells, neurons are distinguished by the number, length, and variety of their processes. Since some of the processes may be extremely long (the axons of motor cells in the spinal cord that innervate muscles in the human foot extend a meter or more), and since all the genetic and most of the biosyn- thetic machinery of the cell is confined to the cell body, an elaborate transport system has been evolved to support and maintain neuronal processes. Although

188 OPPORTUNITIES IN BIOLOGY the same transport system delivers material to the dendrites, it is usually referred to as axonal transport. The materials transported along neuronal processes move with different ve- locities over the range of about 0.5 to 500 millimeters per day. The more slowly moving materials are associated with the transport of cytoskeletal components- microtubules, intermediate filaments, and so forth while the more rapidly mov- ing are membrane and synaptic components. Among the key issues involved in axonal transport are: (1) What materials are transported at each rate? (2) What is the nature of the transport mechanism? (3) What role does axonal transport play during the growth of neural processes and during normal neuronal function? Slow Axonal Transport Several Components of Slow Axonal Transport Have Been Defined, and It Is Likely That They Will Be Understood in Terms of Molecular and Physical Mechanisms Witiun the Next 5 Years Slow axonal transport moves materials at a rate of between 0.5 and 5 mm/day and consists primarily of cytoskeletal proteins transported from the cell body to the end of neural processes. In mature neurons, slow axonal transport maintains the nerve cell processes, whereas in developing axons the rate of process exten- sion corresponds fairly closely to the rate of slow axonal transport. The finding in axons treated with imino-dipropionitrile that the intermediate filaments separate from the microtubules to form a peripheral cylinder suggests the presence of at least two motile complexes, which is consistent with an earlier characterization of axoplasmic transport. In order to directly measure the microtubule movement in slow axonal transport, fluorescently labeled tubulin has been followed as it migrates down a process after being microinjected into the perikaryon. Similarly, specific inhibitors of axonal transport can be injected into the cell, and their distribution, as well as their possible modes of inhibition, can be determined directly. The newer technologies of microinjection, quantitative fluorescence video microscopy, and light and electron microscopic labeling with antibody probes have to a large extent replaced earlier experimental approaches that depended on monitoring bulk movement of radioactivity down axons or the damming back of materials after nerve ligation. The recent cloning of many genes for the cytoskel- etal proteins opens the possibility of manipulating the molecular architecture of specific proteins and determining the subsequent alteration in function in vivo. Few examples exist at present, but improvements in expression systems should stimulate many such studies. Most of the clones have been used to follow the amounts of messenger RNA (mRNA) present during development or differentia- iion. For example, nerve growth factor (NGF) stimulation of certain responsive cells sets in motion the synthesis of new mRNA for the tau proteins that bind to

THE NERVOUS SYSTEM AND BEHAVIOR 189 microtubules and stabilize them. In this case switching cells from an undifferenti- ated state to a characteristic neuronal phenotype involves the coordinate modula- tion of the levels of a number of different proteins within those cells. Rapid Axonal Transport Recent Studies of Rapid Axonal Transport Exemplify How Current Technologies Can Be Used to Understand the Cell Biology of Neurons A broad spectrum of materials are transported along neuronal processes with velocities in the range of 20 to 500 mm/day. Included in these materials are several synaptic components including some of the synaptic vesicle membrane proteins and a number of transmitter-synthesizing enzymes. The general problem is how materials move from one point to another within cells and, in the case of neurons, how material can be rapidly moved from the perikaryon to the distal ends of processes. The application of modern video technologies to light microscopy has allowed investigators to follow the move- ments of transport vesicles within axons, and by modifying the buffer conditions, in dissociated axoplasm. Single filaments were seen to support the bidirectional transport of vesicles. Sequential light- and electron-microscopic examination of the transport filaments revealed that they were single microtubules; single micro- tubules could also move on glass in the presence of a soluble protein fraction from axoplasm. Through the use of microtubule movement as an assay of the "motor protein," a new microtubule-based molecule, kinesin, was purified. Considerable evidence now exists that the motor does indeed drive vesicle movements from the cell body to the synapse in neurons. A cytoplasmic dynein (an enzyme that converts chemical energy into movement in association with microtubules) pow- ers transport in the opposite direction, from the synapse to the cell body. The major questions of how vesicles are programmed to move in the retrograde or anterograde direction will also be addressed. Although several rapidly transported proteins have been identified and char- acterized, much remains to be done before we can completely account for the to- and-fro movements of the materials within nerve cells. However, it is already clear from studies of NGF that the retrograde movement of trophic molecules provides an essential link between events occurring in the vicinity of axon terminals and the regulation of genomic function in the cell nucleus. DEVELOPMENT OF THE NERVOUS SYSTEM Despite a Century of Research on Development, Many Important Questions Remain Unanswered As developmental biology itself matured, the development of the nervous system became a subject of interest to a generation of embryologists who were

190 OPPORTUNITIES IN BIOLOGY fascinated by the complex sequence of morphogenetic events that give rise to the brain and spinal cord and lead progressively to its elaboration and refinement. More recently, interest has been directed toward elucidating some of the develop- mental anomalies that affect the nervous system, including a wide spectrum of genetic disorders whose consequences are frequently devastating to the affected individual. The sorts of questions that have been addressed and that continue to attract the most attention are, What embryonic tissue gives rise to the nervous system? Where are the billions of nerve cells and tens of millions of glial cells generated? How do the cells reach their definitive locations, and how do they selectively aggregate with other cells of like kind? What molecular events underlie the differentiation of neurons and confer upon them their distinctive morphological, biochemical, and physiological features? How do axons find their way through the developing nervous system and finally identify the appropriate target struc- tures with which to establish connections? To what extent is the development of the nervous system genetically determined, to what extent are epigenetic factors involved, and to what degree is the immature nervous system capable of respond- ing to environmental factors? Considerable progress towards answering these questions has come from the identification of suitable model systems and from the recognition that, despite some differences in their patterns of development, vertebrates and invertebrates are remarkably similar. The main advantage of the invertebrate nervous systems that have been studied is that they tend to be rather stereotyped and often contain relatively small numbers of neurons. And, it is often possible to follow the fate of individual cells throughout their life history, and in some cases (Drosophila is a good example) to perform a variety of genetic manipulations that affect the nervous system. Against these advantages is the fact that development in most invertebrates seems to be rigidly programmed and shows little of the plasticity so characteristic of vertebrate development. Needless to say, among vertebrates, the greatest interest centers on understanding the development of the human brain, but as yet we have hardly progressed beyond the descriptive level. The development of the vertebrate nervous system consists of a number of interrelated steps beginning with the phenomenon of neural induction. Factors Affecting Neural Induction Need to Be Determined The emerging nervous system of vertebrates first appears as a thickening of the ectoderm (the outermost layer of the embryos in the dorsal midline. This thickened region, the neural plate, arises in response to the inductive influence of the underlying notochord and mesoderm, which during the process of gastrulation have invaginated from a region called Hensen's node or the dorsal lip of the blastopore and extend forward toward the future head-end of the embryo. Despite considerable effort, mainly in the period between the two World Wars, the nature

THE NERVOUS SYSTEM AND BEHAVIOR 191 of this inductive influence is still poorly understood. In large part this is because the amounts of tissue available for study are severely limited and because, until relatively recently, our knowledge of gene activation (which must underlie this process) was rudimentary. The availability of modern molecular genetic tech- niques should make it possible in the near future to identify the factors involved in neural induction and to isolate and characterize them. Coincident with the induction of the neural plate, the tissue along its margins is induced to form the presumptive neural crest. The neural crest itself is a transitory structure that is first recognizable as a longitudinal band on the dorsal surface of the neural tube. Almost immediately, the cells of the crest become widely dispersed, migrating along predetermined pathways to the skin, gut, head, and so forth, where they give rise to a remarkable number of different tissues. The enormous phenotypic diversity of the derivatives of the neural crest has made it a subject of special interest in recent years. Among other tissues it is known to give rise to nearly all pigmented cells, much of the mesenchyme and the skeletal components of the head and face, and certain of the endocrine glands, as well as to most peripheral sensory neurons, the neurons and supporting cells of the auto- nomic ganglia, and the Schwann cells of peripheral nerves. From a variety of ingenious experiments in amphibian and chick embryos it has become clear that the precursor cells in the neural crest are pluripotent, and the fate of their progeny is largely determined by the environment through which the the cells migrate and the regions in which they finally come to reside. At least some of the cells can change their phenotype relatively late in life given the right conditions. For example, cultured sympathetic neurons that normally synthesize only the neuro- transmitter noradrenalin can, in time, begin to synthesize and release acetylcholine. Proliferation of Both Neuronal and Glial-Cell Precursors Occurs in a Highly Programmed Manner Neuronal proliferation in the CNS occurs for the most part within the ven- tricular lining of the original neural tube or its later derivatives in distinct spatio- temporal patterns. In a few regions, secondary proliferative foci are set up in specialized areas referred to collectively as the subventricular zone. Unfortu- nately, despite a good deal of effort, we know comparatively little about the factors that regulate the patterned cell divisions in the nervous system, but it is these complex patterns that ultimately determine the numbers of neurons and glia found in different regions of the brain and establishes the initial size of each neuronal population. The technique of autoradiography with tritium-labeled thymidine has made it possible to establish, for a large number of structures, the times at which their constituent neurons lose their capacity for DNA synthesis and the orderly se- quence in which different neuronal types appear.

192 OPPORTUNITIES IN BIOLOGY One of the major limitations for our understanding of the factors regulating cell proliferation in the CNS is the absence of suitable markers for the stem cells of different neuronal lineages. Attempts to generate such markers were until recently largely unsuccessful, but in the past four or five years a number of monoclonal and polyclonal antibodies against the major classes of central and peripheral glial cells have become available. The introduction of identifiable genomic sequences into transgenic or chimeric animals should rapidly transform this area and lead to the sorts of insights that have been so valuable in the study of neuronal lineages in the invertebrate nervous system (especially in the nematode C. elegans). The Overwhelming Majority of Nerve Cells Must Undergo at Least One Major Phase of Migration in Order to Reach Their Definitive Locations The withdrawal of neurons from the cell cycle appears to be the trigger for their outward migration from the ventricular or subventricular zones. In most parts of the CNS the initial migration of neurons is more or less radial with respect to the ventricular lining of the neural tube, and it seems to occur mainly along the surfaces of radially oriented processes of glial cells whose bodies lie within the ventricular zone. Perhaps, not unexpectedly, in the course of migration some neurons become misdirected and end up in ectopic loci. If the ectopic cells are able to make the appropriate connections they are able to survive; if they do not they are usually eliminated by cell death. Selective Cell Aggregation Neuronal Aggregation Involves Specific Cell-Adhesion Molecules It has been known for almost 50 years that dissociated embryonic cells, if artificially mixed together, can sort themselves out in a tissue-specific manner, ectodermal cells associating with ectodermal cells, mesodermal with mesoder- mal, and so on. Recently, developmental neurobiologists have taken advantage of this phenomenon to explore the molecular basis for the selective associations of neurons during development. A number of cell-surface molecules that seem to mediate such cell-cell interactions have also been identified and the genes that encode them have been cloned. One such cell-adhesion molecule (or CAM) referred to as N-CAM because it is predominantly expressed in neural tissue, shows homophilic binding (two molecules on different cells stick to each other) has been referred to in Chapter 5. It is widely distributed on the surfaces of all neurons and occurs on certain nonneuronal cells as well. During development it undergoes a characteristic embryonic-to-adult modification with an increase in binding affinity. A second,

THE NERVOUS SYSTEM AND BElIAVIOR 193 termed NO-CAM because of a presumed role in neuron-glial interactions exhibits heterophilic binding. Recent immunocytochemical studies have established that NO-CAM is preferentially expressed on growing axons. Antibodies to N-CAM selectively prevent nerve cell aggregation in vitro and also perturb normal neurite fasciculation. The widespread distribution of both molecules at key stages in the development of the brain and spinal cord suggests that they may each play a critical role in several of the morphogenetic events being considered here. How- ever, there is at present no conclusive evidence that they are specifically involved in the selective aggregation of neurons to form the various nuclear groups and cortical layers that characterize the CNS or the various ganglia of the peripheral nervous system. However, rapid progress has been made in the past decade on these and several other related cell-cell and cell-substrate adhesion molecules, and this continues to be an area of promise for future investigation. Most Neural Cells Seem to Undergo Their Major Differentiation Only After Reaching Their Final Locations When cells migrate away from the ventricular zone, they display many of the differentiated features characteristic of neurons or glia, but their major phase of differentiation usually occurs only after they have reached their final destinations. Formally, one can recognize three aspects to this phase of neuronal differentia- tion. First, the cells acquire a distinctive morphology, usually characterized by the development of several dendrites and a single axon. Second, the cells acquire a number of distinctive membrane properties. These properties generally do not all emerge simultaneously, but appear over a period of time. The third aspect of neuronal differentiation is associated with adoption of a particular mode of synaptic transmission. In most neurons, the cells synthesize one or more neurotransmitters or neuromodulators and generate all the necessary cellular machinery for their transport to the axon terminals and for their exocy- totic release. Simultaneously, the cells express a variety of receptor molecules that become inserted into the appropriate postsynaptic sites on their own surfaces. Recently, researchers have focused on the cloning of the genes for (1) peptide transmitters, (2) enzymes involved in the synthesis of the more conventional neurotransmitters, and (3) receptor molecules involved in synaptic transmission. This work holds great promise that the regulation of these molecular aspects of neuronal differentiation will soon be well understood. Although Much Work Has Been Done on Axonal Outgrowth, the Matter Is Still Far from Being Resolved The central issue in developmental neurobiology is, "How are the highly specific patterns of connections that characterize the mature nervous system

194 OPPORTUNITIES IN BIOLOGY generated?" The general question encompasses three separate issues. The first concerns how the cells acquire individual "addresses" that define their position in the three-dimensional neuronal complex within which they lie. The second concerns the expression of this acquired "positional information" in the out- growth of the cells' axons and the identification of their appropriate targets. The final question is how the cells identify both the region in which they should terminate and the appropriate subset of neurons on which to form synapses. Several lines of evidence suggest that most cells acquire their address either at the time they are first generated or when they first assemble with their fellows to constitute the primordium of the definitive neuronal population. The nature of the addressing mechanism, however, is unknown. By contrast, the mechanism of axonal outgrowth is reasonably well understood. Axons are extensions of the cell that grow by the addition of new materials at their expanded ends, referred to as growth cones. These growth cones are highly motile structures that bear large numbers of delicate fingerlike processes lmown as filopodia; the filopodia are thought to play a key role both in the recognition of the axon's appropriate course and in the identification of its appropriate target. Most axons do not grow in isolation but in association with other axons from the same neuronal population, and it is becoming clear that they can use a variety of strategies to find their way. These include selective axon-axon interactions, selective axon-substrate adhesivi- ties, the identification of key landmark structures, chemical tropisms, and even simple mechanical factors. What is most impressive is that even radical experi- mental perturbations of a group of axons (such as deliberately forcing them to grow into an abnormal pathway) rarely succeed in preventing them from reaching their predestined targets. A different set of factors seems to be involved in the identification of the desired target. This issue has proved to be the most difficult to study. The most widely accepted view (the chemoaffinity hypothesis) was first put forward to account for the uncanny ability of regenerating axons in the mammalian periph- eral nervous system and in the CNS of fish and amphibians to '`home in" on their targets and to reestablish orderly connections with their original targets. Accord- ing to this hypothesis, each small group of neighboring neurons requires a distinc- tive cytochemical label that is also expressed on the surfaces of the growing axons; the presence of matching or complementary labels on the target cells enables the axons to recognize and form synapses with their appropriate partners. A good deal of experimental evidence is consonant with this hypothesis (and none yet contradicts it), but the nature of the proposed molecular labels has so far escaped identification. The final event in the establishment of the initial pattern of connections is the formation of synaptic contacts between the related populations of cells. To date, this problem has been most carefully studied at the neuromuscular junction. Here functional contacts can be established extremely quickly within minutes, in fact,

THE NERVOUS SYSTEM AND BEHAVIOR 195 of the axon's contacting the muscle cell. The assembly of the entire complement of presynaptic and postsynaptic components does not occur for some time, how- ever, and must involve a complex set of inductive interactions between the axon terminal and its target cell. Because growing axons have to make their way through a veritable jungle of other neuronal and glial processes, it is perhaps not surprising that some of them enter an incorrect pathway or grow to an inappropriate target. Most of these aberrant or erroneous connections are eliminated during the next two phases of neurogenesis, which are concerned with the progressive refinement of the initial pattern of connections. Nerve Cell Death During Neural Development Allows for the Fine-Tuning of the Nervous System It has been known for more than 80 years that some nerve cells die during normal neural development, but it was not until the late 1940s that the full significance of such cell deaths came to be appreciated. We now know that, in almost every part of the nervous system, neurons are initially overproduced; at some later period, between 15 and 85 percent of the initial population degenerate. In a few situations it has been possible to establish that the phase of cell death is temporally related to the period during which connections are being established within the target fields. And the finding that the number of neurons that finally survives is closely related to the size of the target field has led to the suggestion that the axons of the cells compete with each other for some entity (probably a trophic agent) that is normally available in the target area in only limited amounts. The cells that are successful in this competition survive; those that are unsuccess- ful die. To date the only well-characterized trophic agent is NGF, which is essential for the survival of many sensory neurons, sympathetic ganglion cells, and central cholinergic neurons. If an excess of NGF is made available to the axons of embryonic sensory ganglion neurons, the normally occurring death of 40 to 50 percent of the cells can be completely prevented. A vigorous search for other neuronal growth factors is under way; a number have been identified and partially characterized, and at least two (fibroblast growth factor and epidermal growth factor) are known to be present in the CNS and capable of maintaining dissociated neurons in culture. Naturally occurring cell death seems to serve at least three functions. (1) It matches the sizes of individual neuronal populations to each other and to the functional requirements of their targets. (2) It allows for the elimination of developmental errors, especially errors in connectivity. (3) It defines the limits of particular cellular lineages.

196 OPPORTUNITIES IN BIOLOGY A Second Regressive Phenomenon During Neural Development Selectively Eliminates Excess Connections Just as more neurons are generated than seem to be needed, there is generally an initial excess in the number of connections formed by each neuron; over a period of some days or weeks these excessive connections are progressively eliminated until the mature number is reached. For example, muscle cells and neurons that normally receive only one or a few more inputs frequently receive many synapses early in development and progressively lose the supernumerary contacts over a period of three or four weeks. In other cases, longer collateral pathways are selectively eliminated, and whole fiber systems come to be reorgan- ized. For example, all areas of the cerebral cortex initially send axon collaterals to the opposite hemisphere through the corpus collosum. Later, the callosal projec- tion becomes restricted to certain functionally defined zones, not through the death of the cells that initially project to the contralateral side, but rather through the selective loss of callosal collaterals. The mechanism underlying the selective elimination of particular axonal branches is far from clear, but may well also involve trophic factors. Certainly in the case of NGF, both the responsive cells and each of their processes require a continuous supply of the factor for their maintenance. Electrical activity also seems critical for the maintenance of individual axonal branches. Blocking activity with a drug such as tetrodotoxin (which blocks sodium channels) com- pletely prevents the refinement of connections in several developing (and regen- erating) neuronal systems. This is of considerable importance,. for the early excess of neuronal connectivity and the critical dependence of connections on the maintenance of appropriate patterns of activity must provide the substratum for much of the plasticity of the immature nervous system. A Rapidly Expanding Body of Literature Indicates how Widespread Plasticity Is in the Nervous System To the age-old question of whether the development of the brain is shaped more by nature or by nurture, the answer is now clear: Both are important at different times and in different ways. The early development of the nervous system including all the events that lead to the establishment of the initial patterns of connections seems to be determined largely by genetic and locally acting epigenetic factors. But once the initial neural framework has been laid down, environmental factors become increasingly important. In several instances we now recognize what are referred to as critical periods during which the relevant neural systems seem to be particularly susceptible to external environ- mental influences. The final form of the brain and its functional capacities are shaped by the interplay of intrinsic (genetic and epigenetic) and extrinsic (envi- ronmental) influences. Consider, for example, the importance of hormonal influences on the devel- opment of the brain and, in particular, the sexual determination of those parts of

THE NERVOUS SYSTEM AND BEHAVIOR 197 the hypothalamus that control the pattern of release of the gonadotrophin hormones from the anterior pituitary gland. Although an animal's sex is genetically deter- mined, the regulation of hypothalamic function is determined largely by the levels of circulating sex hormones during a critical period in development. If these hormonal levels are significantly perturbed, the pattern of gonadotrophin hormone release can be completely reversed from the male to the female pattem, or vice versa. Animals experimentally subjected to this type of sex reversal display all the behavioral characteristics normally associated with the opposite sex. Even more attention has been focused on the capacity of the major sensory systems to respond to normal or altered sensory experience. The most striking findings have come from work on the visual system, in which, during a critical period, any marked abnormality in the animals's visual experience can result in an Revocable alteration in its later visual behavior. For example, if the upper and lower eyelids of one eye are sutured closed throughout this critical period and then the eye is reopened, the animal subse- quently behaves as if it were blind in the deprived eye. When the visual cortex is explored neurophysiologically, the cells that would normally be activated by stimulating that eye are found to be either silent or dominated largely by inputs from the nondeprived eye. And when the cortex is examined anatomically, the regions in which the inputs from the deprived eye terminate are found to be substantially reduced in size, whereas those related to the nondeprived eye are correspondingly expanded. Equally striking are the findings in experiments in which animals have been reared in a structured visual environment. Kittens exposed throughout the first few weeks of life to only vertical stripes subsequently have great difficulty when they encounter horizontally oriented objects. Physiological recordings made from their brains show that the majority of the cells in the visual cortex respond only to vertically oriented stimuli. Experiments of this kind have already had a profound influence on the way in which human clinical problems such as strabis- mus are treated and have emphasized the importance of a rich environmental experience during early childhood. Of special interest for developmental psychologists is the enormous capacity of the brain to adapt to early injury, which to a large extent must depend on its morphological plasticity. Nowhere is this better demonstrated than in the brains of young children who have suffered substantial damage to one cerebral hemi- sphere. Even after complete removal of the language areas of the cerebral cortex in the normally dominant left cerebral hemisphere, children can learn to speak perfectly normally, language function having been taken over by the correspond- ing areas in the opposite hemisphere. The Ontogeny of Behavior Has Been a Neglected Area of Neuroscience Until recently, developmental neurobiology has been concerned largely with the initial assembly of the nervous system and to a distressing extent has ne- glected the equally important issue of the emergence of behavior. Inasmuch as

198 OPPORTUNITIES. IN BIOLOGY the goal of neuroscience is to provide a sound scientific basis for an understanding of all aspects of behavior, including such higher brain functions as thought, memory, perception, and feeling, the study of the ontogeny of behavior is likely to become increasingly important in the near future. One serious obstacle to progress so far has been the difficulty of identifying good animal models for experimental study. As a result, the problems that have been analyzed such as the emergence of motor behavior in salamanders and chicks or the development of visual behavior in kittens and young monkeys- have yielded results that are suggestive but hardly definitive. In contrast to most other fields of neuroscience, in this area human studies have led and continue to lead the way, even though by their nature they are usually constrained to observa- tion and description rather than experimental manipulation. NEURAL PLASTICITY AND ELEMENTARY FORMS OF LEARNING Learning Is a Major Vehicle for Behavioral Adaptation and a Major Force for Social Progress Learning is the process by which we acquire new knowledge about the world, and memory is the process by which we retain that knowledge. Most of what we know about the world we have learned, and, in a larger sense, learning goes beyond the individual to the transmission of culture from generation to genera- tion. Learning is thus the major vehicle for behavioral adaptation and also the major force for social change. The ability to acquire and retain new information (that is, to learn and to remember) characterizes all moderately complex animals, and their capacities for learning and memory correlate well with the complexity of their nervous systems. In human beings, these capacities have resulted in a completely new kind of evolution-the evolution of culture. How does learning occur? What changes occur in the brain when behavior is modified as a result of experience? In a prescient lecture in 1894, Ramon y Cajal proposed that learning might produce prolonged changes in the effectiveness of the synaptic connections between nerve cells and that these changes could serve as the mechanism for memory. More recent research suggested that sensory stimuli could produce two types of changes in nerve cells and their connections: (1) invariant and transient excitability change and (2) a facultative but enduring plastic change. To determine if synapses undergo plastic changes and to relate these changes to learning, it has been necessary to develop suitable cellular techniques and appropriate experimental systems. Throughout the past decade, significant prog- ress has been made along both lines. During the next decade, it should be possible to extend the analysis of mechanisms more fully to the molecular level and to apply similar strategies to more complex forms of learning.

THE NERVOUS SYSTEM AND BEHAVIOR The Most Instructive Forms of Learning Hoe Been the Simple Forms Commonly Referred to as Habituation, Sensitization, and Classical Conditioning 199 Habituation, the simplest behavioral modification, occurs in all animals. It is a process in which an animal learns, though repeated exposure, that the conse- quences of a weak stimulus are neither noxious nor rewarding, and so can be safely ignored. By not responding to stimuli that have lost their novelty or meaning, animals free their attention for stimuli that are rewarding or significant for survival. When Pavlov first described behavioral habituation, he assumed (on the basis of indirect evidence) that it would involve active inhibition. However, in those systems where it has been possible to analyze the synaptic mechanisms of habituation directly, neither presynaptic nor postsynaptic inhibition seems to be involved. Rather, habituation is produced by a simple depression of chemical transmitter released from excitatory synaptic connections. In certain cases, this depression has been related to a reduction in calcium ion influx, but second- messenger systems (Chapter 4) may also be implicated. Behaviorally, sensitization is the opposite of habituation: It results in a heightened responsiveness after exposure to novel or noxious stimuli. Often a single strong noxious stimulus applied to an animal will enhance its defensive reflexes for periods of up to several hours, and repeated stimulation may lead to reflex enhancement that persists for days or even weeks. Thus, whereas habitu- ation causes an animal to ignore innocuous or trivial stimuli, training that leads to sensitization causes it to attend to the relevant stimuli because they may prove to be either painful or dangerous. Short-term sensitization has been analyzed in detail in two invertebrate reflex systems, in which presynaptic facilitation is initiated by modulatory transmitters acting through two second-messenger systems: cyclic AMP-dependent protein phosphorylation and calcium ion concentration. The presynaptic facilitation involves a change in the potassium channel that broadens the action potential and increases calcium influx. In addition, an altered intracellular handling of calcium leads to more effective mobilization of transmitter vesicles. Habituation and sensitization are nonassociative forms of learning-the ani- mal learns about the property of a single event or stimulus. By contrast, classical and instrumental conditioning are associative forms of learning in which the animal learns about the relationship between two events. In classical condition- ing, an initially weak or ineffective conditioned stimulus becomes highly effec- tive in producing a behavioral response after it has been paired with a strong unconditioned stimulus. When an animal has been conditioned, it has learned that the CS predicts the US and, by inference, it learns about cause-and-effect relation- ships in the environment.

200 OPPORTUNITIES IN BIOLOGY In each invertebrate system in which the cellular basis of conditioning has been carefully analyzed, biogenic amines or peptides have been found to serve as the chemical signal for the unconditioned stimulus. Through second messengers, potassium channel function is modulated and the effect of the conditioned stimu- lus is enhanced. In these cases the effects of the second message are pleiotropic: Many proteins are covalently modified in parallel, and the effects last from minutes to hours. Similar conclusions have emerged from experiments on learning using a completely different, mutational, approach in Drosophila. In Drosophila, single- gene reaming mutants have now been isolated that are defective in biogenic amine synthesis or in one or another step in the cyclic AMP signaling systems. Thus, the dopadecarboxylase mutant blocks synthesis of the monoamine transmitters dopa- mine and serotonin, the turnip mutation affects calcium stimulation of the adenylcyclase activity, and dunce alters or eliminates a cyclic AM phospho- diesterase isoenzyme. These Drosophila mutants, which cannot be classically conditioned, also fail to exhibit sensitization. Long-Term Memory Involves Genes and Proteins Not Utilizedfor Short-Term Memory A variety of studies in animals and humans indicate that newly formed memories are easily disrupted for a short period of time after being formed. Consistent with this view is the well-established clinical observation that head trauma or epileptic seizures produce amnesia for events just preceding the trauma or seizure; memory for earlier events is not affected. Recently, animal experi- ments with inhibitors of protein synthesis have established that although short- term memory does not require the synthesis of new proteins, memories lasting days or weeks are disrupted by the inhibition of protein synthesis. Particularly important is the finding that long-term memory is most sensitive to disruption during and immediately after training: No deficit in long-term memory is ob- served if exposure to the protein synthesis inhibitor is delayed by as little as 1 hour after training. Studies of long-term sensitization in invertebrates have demonstrated that a narrow time window during which new proteins must be synthesized reflects a fundamental property in the storage of information by specific neurons and their connections. With the tools of modern molecular biology, it should now be possible to identify the genes and proteins necessary for long-term memory in simple animals and to use these genes and proteins as probes to explore tong-term memory in higher forms. We Are Beginning to Understand the Mechanisms Underlying Various Simple Forms of Learning and the Short-Term Memory to Which They Give Rise One of He major findings to emerge from this research is that short-term memory involves second-messenger systems similar to those used for other

TlIE NERVOUS SYSTEM AND BEHAVIOR 201 cellular processes. What is distinctive to nerve cells is the range of receptors that initiate the second-messenger signaling and, even more, the substrate proteins that act as effecters. It should be possible, therefore, to build a bridge between the study of simplified learning and behavior and more general studies of cell biol- ogy. The finding that long-term memory is likely to involve alterations in gene expression opens the parallel opportunity to bridge learning and molecular biol- ogy. This should make it possible to identify the genes and gene products important for the acquisition and retention of long-term memory in any system and to use those identified in simple systems for probing more complex systems. NEUROBIOLOGY OF PERCEPTION: VISION Vision Research Is in a Period of Rapid Progress In vision research, exciting developments have occurred at many levels of organization ranging from the molecular to the behavioral and a wide range of fundamental questions are being attacked, with an extensive arsenal of tech- niques. We will therefore use vision as an example to illustrate the principles emerging in the analysis of the mechanisms of perception of all types. This section tries to capture some of the highlights of recent progress in studies of the organization and function of the retina and visual cortex. The Photoreceptors of the Retina Transduce Light into Electrical Signals At the photoreceptor level, the most striking advances have been the cloning and sequencing of the opsin genes in several species (including humans) and the elucidation of the transduction process. The successful cloning of the opsins reflects the high degree of conservation in these genes and the relative ease of progressing from one set of genes to another on the basis of sequence homology. The human opsin story is particularly intriguing because it has definitely estab- lished the molecular basis of the various forms of color blindness. Our under- standing of transduction has been greatly facilitated by applications of patch clamping and related biophysical techniques to the photoreceptors. These meth- ods have provided strong evidence that the elusive second messenger in the transduction sequence is the cyclic nucleotide called cyclic guanosine mono- phosphate (cyclic GMP). Many critical steps in the overall cascade remain to be worked out in detail and analyzed quantitatively, but the next few years will see rapid progress toward a thorough understanding of the exquisite sensitivity, dynamic range, and other remarkable properties of both vertebrate and inverte- brate photoreceptors. Between the photoreceptors and the output stage of the retina are several classes of interneurons whose functional properties and chemical transmitters are well known because of intracellular labeling methods and immunocytochemistry. The actual output of the retina is carried by distinct classes of retinal ganglion cells that transmit qualitatively different Apes of information (for example, "on"

202 OPPORTUNITIES IN BIOLOGY and "off" channels, color channels, and black-and-white channels) to the brain. Substantial progress has been made in identifying these channels in different species and in delineating the underlying circuitry responsible for their character- istic features. Neural modeling techniques, combined with precise anatomical and physiological data, are beginning to contribute to the analysis of how specific receptive field properties are generated and how the first step in visual perception · · - IS organlzea. The axons of the visual ganglion cells form the optic nerve and conduct visual information from the eye to a number of subcortical visual centers in the brain, some of which are concerned with various visual reflexes, the control of eye movements, and the regulation of longer-term light-induced hormonal changes, including those responsible for circadian and other behavioral rhythms. Of greatest interest for visual perception is the relay of information to the primary visual cortex through the lateral geniculate nucleus, in which the different types of ganglion cell axons are spatially segregated. Visual Areas Occupy More Than Half of the Cerebral Cortex in Some Species and Consist of Two Major Subdivisions, the Striate and the Extrastriate Cortex The primary visual (or striate) cortex is the largest and best-understood of all cortical areas. It is surrounded by a mosaic of about 20 other visual areas, collectively termed extrastriate cortex, whose organization and function have only recently begun to be understood. Striate Cortex. The richness and diversity of neural response properties in striate cortex is remarkable. A single neuron can signal detailed information about more than 10 different stimulus features, including orientation, color, depth, motion, and texture. Current evidence suggests that several "functional streams" emanate from the striate cortex, each of which conveys information related specifically to form, color, or motion. The recent success in using optical techniques to monitor activity patterns across the surface of the cortex will provide a powerful tool in the further dissection of these streams. Despite much new information about cortical circuitry obtained with physio- logical, anatomical, and pharmacological approaches, we do not know the spe- cif~c wiring diagram responsible for even a single property of cortical neurons (for example, selectivity for orientation or binocular disparity). The difficultly of resolving these issues may largely reflect the fundamental complexity of cortical circuitry. Several lines of evidence suggest that cerebral cortex in general and visual cortex in particular operate much more as a richly interconnected, highly distributed neural network than has generally been appreciated. If so, neural modeling and other computational approaches will ultimately play an important role in the elucidation of cortical function. And we can look forward to even stronger interactions between experimentalists and theoreticians interested in "computational vision" in the near future.

THE NERVOUS SYSTEM AND BEHAVIOR 203 Extrastriate Cortex. In monkeys and cats, which have been intensively studied, evidence exists for as many as 20 extrastriate visual areas. Obtaining a precise count of the total number of areas along with reliable criteria for their identification remains a formidable challenge. Monoclonal antibodies and other brain-specific markers may provide powerful tools for achieving this goal. The availability of a variety of sensitive anatomical tracing techniques has led to the identification of a plethora of distinct visual pathways. Nearly 100 cortico-cortical connections alone have been identified in monkeys; with the complexity of the other cortical afferent and efferent connections and the intrinsic connections within each area, it is evident that computer data bases will soon be needed to handle these vast amounts of information. Equally important is the need to elucidate organizational principles that reflect the strategies used for distributing information among various cortical areas. Current evidence supports the notion that visual areas are arranged in such a way that each principally subserves one (or at most a few) specific functions such as motion or color perception. Thus, in the monkey, area MT is concerned mainly with motion- related information, while area V4 is concerned with information about color and form. It is proving difficult to determine exactly how these extrastriate areas process and transform visual messages, as opposed to simply relaying informa- tion already extracted in the striate cortex. However, several recent studies have provided intriguing hints about higher-order analysis of motion, form, texture, and color information. More rapid progress on this front can be anticipated as a result of stronger interactions between neurophysiology, psychophysics, and computational science. Other technical advances will also have major impacts. Positron emission tomography can be used to map functionally distinct regions of the human visual cortex, which will play an important role in linking human psychophysical studies to animal models. Neural recordings with multiple electrode arrays offer the prospect of understanding how neural ensembles process information whose representation at the level of single cells is not explicit. However, fundamentally new concepts in data analysis will be necessary if we are to learn how to interpret such experiments. The Development and Plasticity of the Visual System Raise Many Intriguing Questions Until now the major focus has been on the striate cortex, where much has been learned about the interplay of innate instructions and environmental influ- ences in generating the intricate architecture of the normal adult cortex. New principles may emerge from studies of development and plasticity in the extrastri- ate visual areas, which so far have received comparatively little attention in terms of their development. Again, new techniques, ranging from optical monitoring of activity to area- or cell-class-specific monoclonal antibodies, will greatly enhance the scope of questions accessible to experimental analysis.

204 OPPORTUNITIES IN BIOLOGY Behavioral Studies of Vision in Animals The effort involved in training animals to make visual discrimination is great. The most impressive contributions of behavioral analyses of visual performance have come in areas in which the neural substrate can be more or less directly linked to the behavior, such as the identification of neurons in the striate and immediate prestriate areas of the cortex that respond only when the images on the two retinas are displaced to an appropriate extent. The Studies of the Control of Eye Movements and Visual Attention Have Borrowed the Attentional Paradigms of Experimental Psychology and Applied Them to Experiments with Animals These studies have revealed the existence of complex mechanisms involved in the spatial distribution of attention; companion neurophysiological studies have contributed to our understanding of the neural mechanisms involved. As indicated earlier, the development of the awake, behaving monkey for central nervous system recording has been, and will continue to be, of the utmost value in studies of the oculomotor system and of the higher functions of the visual nervous system more generally. As we become increasingly able to pose sensible neurophysiological questions of higher cortical areas, these behavioral techniques will become increasingly important. Animal Models Are Also Used in Studying Visual Development Rearing animals under conditions of abnormal visual stimulation or visual deprivation leads to abnormal development of the functions of the central visual pathways (amblyopia). Parallel studies that relate the behavioral anomalies in these animals to underlying functional and morphological changes in the visual pathways can be useful in revealing relations between visual functions and the underlying neural substrate, as well as being of general developmental interest in themselves. One of the Oldest Questions in Neurobiology Concerns the Relation of a Particular Behavioral Performance to the Neural Machinery that Subserves that Performance The study of vision has often been concerned with this kind of question, particularly with respect to retinal mechanisms. There is now more interest in tackling questions of this kind in the central nervous system. By use of judi- ciously chosen parallel psychophysical and behavioral experiments, it has been possible to draw reasonably strong conclusions concerning the sites of brain mechanisms that subserve a variety of visual performance.

THE NERVOUS SYSIEM AND BEHAVIOR 205 A reasonably clear example concerns the locus of the spatial frequency chan- nels whose existence is supposed on purely psychophysical grounds. These channels are selective for both stimulus orientation and size; they are sensitive to the direction of stimulus motion; they are binocularly activated and may be sensitive to binocular disparity. Computational Approaches to Visual Function Although computer vision has developed without reference to biological vis- ual systems, it has become increasingly apparent that the types of algorithms used in machine vision are likely to throw light on how the brain processes visual information. Although most of the effort in computer vision has so far been devoted to the applied problem of programming computers to deal sensibly with image data, a growing number of workers are turning their attention to biological visual systems and are trying to model different aspects of visual behavior (and the underlying neural mechanisms). Still, relatively few process models take into account the structure and function of the nervous system; much more common are competence models, which attempt to mimic the biological function, but do so in a way that may be completely unrelated to the approach used in the actual biological system. No doubt this situation will eventually be reversed, given the richness of biological vision and the rather barren repertoire of most vision machines. The Use of the Awake, Behaving Primate and Computational Models Will Show Dramatic Results in the Next Decade The use of the awake, behaving primate, combined with more sophisticated behavioral and physiological approaches, will be important in increasing our understanding of visual function, especially the broad range of functions subserved by the extrastriate visual cortex. In addition, the application of computational models to the myriad problems of visual processing will increase the sophistication of our approach to studies of the function of the visual system at all levels, but especially at behavioral and psychophysical levels. NEUROBIOLOGY OF MOTOR CONTROL Our Understanding of How the Brain Controls the Movements of the Body Is Undergoing Dramatic Changes For more than a century, the primary goal of those interested in the control of movement was to map the areas of the brain concerned with movement. Now, at last, we are beginning to address such questions as, How does the brain decide when to move? How does the brain select its targets? What are the speed, accuracy, and force of particular movements? The gradual emergence of new

206 OPPORTUNITIES IN BIOLOGY techniques that allow effective study of the mammalian brain make it possible to develop increasingly sophisticated and realistic models of motor function. These new techniques include the study of the activity of single nerve cells and popula- tions of cells during real behavior, new anatomical methods for tracing the connections between specific nerve cells of the brain; imaging that allows one to visualize neural activity as it occurs in the living brain; and quantitative methods derived from biomechanics, control theory, and computer science, applicable equally to human subjects and experimental animals. Within the foreseeable future, it should be possible to determine which brain cells play what roles in planning, initiating, and carrying out movements. Until the recent development of these new methods, the study of motor control was restricted to an examination of primitive reflexes and postural control mechanisms. Experiments were performed on animals that either had large regions of their central nervous systems destroyed or that had been immobilized by anesthesia. Although these studies continue to be necessary to establish some of the fundamental ways the nervous system operates, they have provided little insight into how the normal nervous system operates. Research in motor control now focuses on intact, awake animals and, where possible, on human beings. Current approaches to the study of motor control derive from techniques for recording from single nerve cells in awake animals performing skilled motor tasks. This method allows us to search for control signals used by different motor regions of the brain. As a result, we now have a general idea of the roles of venous regions in producing voluntary movement. Even simple movements require the cooperation of many motor centers in the brain. In controlling movement, the brain acts as a complex parallel computer, which decomposes motor plans into several components, which are then reassembled into a final, effective movement. As a result of these studies on single neurons, neuroscientists are developing a new model of how the brain controls behavior. This model emphasizes the brain's ability to adapt to new situations and to form new strategies to solve them effectively. It marks a significant advance from earlier views that held the brain to be merely the center for coordinating automatic reflexes. A second important insight derived from these studies has been the realiza- tion that the activity of neurons within a region varies with the task to be performed. The coding of information by single neurons often depends on the behavioral context. For example, one neuron may respond to a specific signal, such as a light, but only if this light tells the animal to expect a food reward. Such selective responding has been observed in nerve cells not only in the brain, but also in the spinal cord. These findings tell us that the awake brain is continuously modulating its activity in meaningful ways to respond to different situations. While it has become standard to record from single neurons in behaving animals, it remains difficult to simultaneously observe the behavior of different regions of the several neurons in different regions of the brain as they cooperate to

TlIE NERVOUS SYSTEM AND BEHAVIOR 207 shape behavior. The first important techniques used to allow this type of analysis were anatomical ones that can map specific connections between one brain region and another. Using suitable radioactive labels or other markers, it is possible to trace the connections made by individual nerve cells from one part of the brain to another. The selectivity of these tracers makes it possible to examine only those specific classes of brain cells of interest. The introduction of imaging methods has been particularly important be- cause they can be used in alert human subjects. When the activity of a region of the brain increases, the blood flow to that region also increases and the region uses more glucose, the cellular fuel of nerve cells. By injecting minute amounts of radioactive compounds and then measuring the radioactivity over the scalp, it is possible to measure the blood flow and glucose consumption in different areas of the brain and thereby to reveal, with remarkable clarity, changes in activity of different brain regions when subjects perform different movements. It is even possible to "see" the brain planning a movement. Especially striking has been the finding that the premolar areas of the cerebral cortex become active when subjects mentally rehearse or plan complex motor actions; other areas become active only when the movement is performed. These results are now guiding studies of the activity of single neurons in experimental animals performing similar tasks. New Techniques Make Possible Collaborative Ventures That Increase the Yield of Scientific Studies Some of the new imaging techniques have been used to map the location of specific neurotransmitters. Not only can normal brain connections be mapped in this way, but these techniques allow us to identify the abnormalities of neuro- transmitters in diseases such as Parkinsonism, Huntington's chorea, and dystonia. These studies reveal the chemical changes of these diseases in the living patient with a minimum of pain or risk and in a manner that goes far beyond the reach of the finest autopsy as carried out even a decade ago. At the same time as some neuroscientists have been shedding new light on how the brain controls movement, others have been studying the primary effecter systems of movement the muscles and joints. Because muscles have certain intrinsic properties, the brain must adapt its neural language to a dialect the muscles can understand and obey. One major problem for the nervous system, including the brain and spinal cord, is that the muscles distort the commands of the brain. The nervous system has evolved several mechanisms for overcoming these distortions. For example, to produce brief, accurate movements in one direction, the nervous system has developed a strategy in which opposing muscles are precisely activated in rapid succession. This requires coordination of nerve cell signals in hundredths of seconds. Because movements can be impeded or interrupted, the nervous system has also developed means for rapidly correcting errors. Certain automatic reflexes, such as the familiar knee jerk that is tested

208 OPPORTUNITY IN BIOLOGY during routine physical examinations, prevent muscles from distorting neural commands and allow for smooth mechanical action which otherwise would be jerky and intermittent. Because the tasks performed by the nenous system to control movement are complex, it has been useful to apply techniques developed in engineering to understand what the nenous system is doing. One such technique is control theory, derived from the study of feedback in electrical and mechanical systems. After some initial disappointments, approaches based on control theory have now provided crucial insights into how eye movements are controlled. This theory has predicted the existence of nene cells with specific properties that were subse- quently discovered by experiments in animals. Control theory has also provided a better understanding of disorders of eye movement in human patients. In the future, control theory should provide similar insights into how movements of the arms and legs are achieved. Artificial Intelligence, Particularly Robotics, Has Also Advanced Our Understanding of Movement Control Ideas derived from artificial intelligence are currently stimulating research on the role of the cerebellum in the control of movement. This par' of the brain is crucial for normal motor coordination, and its regular, modular structure has fascinated brain scientists for more than a century. Recent work in experimental animals has confirmed that the cerebellum helps correct errors in ongoing move- ment. Artificial intelligence has also influenced studies of the role of the vestibu- lar system, that part of the motor system which orients the head, eyes, and the body and helps to maintain balance. While current attempts to apply models derived from control theory, robotics, or artificial intelligence are unlikely to be the final word in motor control, they are now making an important contribution by sensitizing neuroscientists to the importance of generating testable models of motor control that make explicit performance predictions. These are a necessary step in reaching a more complete understanding of the whole motor system. An important collaboration has also developed between physiologists and engineers in robotics. Physiologists have learned how to better define the pro- cesses necessary to control the human body, which is a complex system with multiple interdependent components. In turn, attempts are being made to apply what has been learned about muscular control to the design of robotic limbs for manufacturing purposes and to the construction of artificial walking machines that can maneuver through rough terrain. Prosthetic limbs for impaired patients may eventually allow some people who currently use wheelchairs to walk again. The development of complex armlike devices has stimulated work aimed at how the brain moves the arm through a specific path in three-dimensional space. This work shows that merely to fix the direction of a movement, many millions of brain cells in several different areas must work together.

THE NERVOUS SYSTEM AND BElIAVIOR ::: : :: ~ :: ~ ~NFW:TECHNIQUES AND PAR:KINSON'S Dl:SEASE : :: :: ~ :~ :~ :: :: ~::: 1 ~ ~ ~One~:example may clarify the waif: ln~w~h~:h new techniques have cam- I: bined in recent :years to help both our base understanding of Chows the brain co~rals~movements~: and our ~abii~y to~h~elp~p~ients with ~motori~:di:sorders. Padtinson'~s~d~isease,~ a relatively common negator disorder of ~the~elderly, Was::: i first~:d:e:scribed~ in the nineteenth: century, but~:on~iy~ in the 196:0s:~ was:::it: : discovery cheat a: specific chemical: neurdt~ra~nsmitter, ~dopam~ine, was :d~i-: cient~i~n these patients.: This disco~ry~:q~uick~ led to: a ~latively~successfu~l :~ ~ therapy Of replacing: the nelJrotransmitter,~ thereb~r~gire&:tly ~mproving~ they ~ it: quaity Of life~:and: Seven the::~su~ivai: of these pat~;ents~.: ~l~e:~:most exciting recent oontin~uat~n of this wow is the~ssibilKy of brain tissue transplants that ma ~restbre~natural ~pamihe function. ~ Much transplar~ts~ have ~en~ successful i n Animals depleted of ~ dopamine Andre have ~ already been term abroad in far small group of human ~patients. ~ In the meantime, new neu-~ ro~anatomical techniques have greatly increase our knowledge about how the parts of brain di~so~er~ in Parkinson is disease acre ~interoonnedted. Changes in brain actwity~ behave been examined radiologically And the: ~dbfi ciOnGies: of dopamine shown quantitatively in :~living patients. ~ Studies of the ;~a~ct~n~of single nerve Reilly A animals havens begun to elucidate how disor dered nerve ~cel I movement produces the~1rozen. limbs Of these ~ patients. 209 Increasingly sophisticated ~studies, based on iasights~from engineering, have produced a far more accurate~character~ation of the~nature Of the motor deficit. Several years ago, the chemical known colloquial as MPTE, which Lisa by-p~ubt~of the:~synthe~sis of certain ill it drugs, was~fo~und~to cause a toxic form of Parkinson s disease in ma numb of drug users. This unfort~u-~ nate circumstance has Men quickly exploited to gain further insight into~the mechanism of Parkinson s diseases and its possible cause. Already, the study of these patients has~im:~e clearer the brain regions damaged in Parkinson s disease and~sugg~ested that~the disease self, whose cause remains unknown,: might be produced by a subtle toxin and thrust might be preventable. it has now bben~convincingly~shown~that~some animals given the toxin~devetop a condition virtually identical to Paricinsons disease. Neuroscientists are now Applying the new biochemical, anatomical, and physiological techniques to study these animals, and there is reasonable hope~that They may sold Some of The remaining mysteries of Parkinson s disease in the near future. :: :: :: j

210 OPPORTUNITIES IN BIOLOGY NEUROBIOLOGY OF COGNITION The relation between cognition and the brain has been a topic of philosophi- cal speculation for millenia and a focus of scientific study for more than a century. Recent developments in our ability to monitor brain function and the development of computational models of cognition have abruptly altered the pace of scientific progress in this area. Methodological Developments Provide Unprecedented Opportunities for Exploring the Neurobiology of the Human Brain Methods for imaging metabolic correlates of brain function provide a new basis for localization of function. New techniques for analyzing electrical and magnetic signals from noninvasive probes allow analysis of the dynamics of neural activity. These two types of methods have made it possible to relate changes in cognitive function to discrete structural areas in the living expenmen- tal subject. They have also made it possible to analyze disease. Perhaps even more important, they are allowing exploration of regional changes in brain activ- ity during perception, thought, and action. There is every reason to suppose that some of those methods will become more precise in the coming years and that additional new methods (for example, magnetic and electron spin spectroscopy) can be applied. As Our Ability to Explore the Human Brain Has Changed, So Has Our Definition of What Cognition Is and Our Understanding of How It Can Be Studied Experimentally Computational models of cognition arising from efforts to develop artificial intelligence can provide an analytical basis for neural studies of mental processes. These models picture the nervous system in terms of the sequence of operations necessary to carry out cognitive functions and allow us to view cognition in terms of the elementary mental operations of which it is constituted. Experimental studies of human beings carrying out these operations reveal the dynamics of simple computations with time resolution in the millisecond range. For example, studies of the time needed to determine whether a target item is a member of a stored set indicate a serial search process with a comparison time of 30 millisec . · - onus per item In memory. We can now relate an individual's complex overt behavior to two different models of how the nervous system functions during thought. One model stresses the time dynamics of serial and parallel computations that occur when human beings execute elementary tasks. The other stresses the anatomical systems in the human brain that become active during thought. A fundamental goal is to understand how computational models of tasks such as reading, listening, imaging, and problem-solving relate to the anatomical

THE NERVOUS SYSTEM AND BEHAVIOR 211 structures and wiring diagrams of regions of the brain known to be involved in these processes. Such an understanding of the complex relations between mental computations and their underlying neural bases seems critical for illuminating the physical basis of changes in mental and emotional life that occur with either normal development or neurological and psychological disorders. Our current ability to relate images of brain activity to the mental computa- tions found in cognitive models contains large gaps. For example, we do not yet understand how the brain's electrical activity relates to changes in cerebral blood flow. This prevents us from taking full advantage of the opportunity to combine findings based on the more spatially specific neural imaging techniques with those derived from the more accurate, temporally precise recordings of event- related activity. Nor do we know how the activity of individual cells or of such cellular configurations as cortical columns relates to the computations described by computational models. These basic questions of method are common to all areas in which one hopes to understand the relations between brain activity and function. Several areas of investigation in the study of cognition have already begun to use the new methods to produce important findings relating cognition to neurobi- ology. These areas include learning and memory, attention, language, and the psychobiology of development. At Least Two Types of Memory Differ on the Basis of What Is Learned and Where the Memory Is Stored Cognitive psychology has emphasized that memory for different tasks varies according to the type of knowledge acquired by the subject and on how the subject encodes and recalls the information learned. The two categories are often distin- guished as procedural (or reflexive) and declarative. Recent studies suggest the intriguing hypothesis that each of these two forms of memory is processed by a different neural circuit. Procedural memory is acquired in an automatic or reflexive way without awareness or cognitive processes such as comparison and evaluation. It includes perceptual and motor skills and the reaming of procedures and rules. This form of memory is thought to use elementary forms of plasticity and to be stored within the sensory and motor systems employed for the expression of that particular task. For example, the classically conditioned eyeblink response in rabbits can be abolished by specific lesions of certain park of the cerebellum. When these areas are destroyed, the effective conditioned auditory stimulus no longer produces an eyeblink, although the unconditioned eyeblink response that follows the US (air puff) remains intact. Cells in regions of the cerebellum also show learning- dependent increases in neuronal activity that closely parallel the development of the conditioned behavioral response. The results of these experiments indicate that the cerebellum plays an important role in mediating the conditioned eyeblink and perhaps other simple forms of classical conditioning.

212 OPPORTUNITIES IN BIOLOGY Declarative memory depends on conscious reflection and on such cognitive processes as evaluation, comparison, and inference for its acquisition and recall. Declarative memory encodes information about specific autobiographical events as well as the temporal and personal associations for those events. It is often established in a single trial or experience, and it can be concisely expressed in verbal declarative statements, such as "I read a fascinating book last week." Declarative memory involves the processing of bits and pieces of information that the brain can then use to reconstruct past events or episodes. By repetition, declarative memory may at times be tranformed into the reflexive type. Learning to drive a car at first involves conscious cognitive processes, but eventually driving becomes more automatic and reflexive. Thus, even certain verbal learn- ing tasks, if repeated often enough, are thought to assume the characteristics of reflexive learning because they can be performed without the participation of other cognitive strategies. Studies in humans suggest that the temporal lobe and closely associated structures of the limbic system including, in particular, the hippocampus, may be critically involved in the acquisition of declarative memory. The structures are not themselves thought to be sites for memory storage, but are somehow involved in the process by which memories are placed into storage or are retrieved and read out from storage. Some of the first evidence for a role for the temporal lobe in memory came from the study of a few epileptic patients who underwent bilateral removal of the hippocampus and associated structures in the temporal lobes to relieve their epileptic symptoms. These patients are amnesic in the sense that they have lost the ability to store new memories although, for the most park their early-formed memories are intact. Such patients cannot master tasks requiring declarative memory, but they perform well on procedural tasks. A given learning task often involves aspects of both types of learning, and in these instances patients remem- ber some aspects of the problem, but not others. Thus, if the patient is given a highly complex mechanical puzzle to solve, the patient may learn it as quickly as a normal person but on questioning will not remember seeing the puzzle or having worked on it previously. In other words, amnesic patients can learn a complex skill and yet cannot recall the specific events that allowed them to learn the rules and procedures that make up the skill. This idea helps explain why amnesic patients, when they perform a particular task, are often not aware that they had actually learned it just a few days earlier. Attention Visual Spatial Attention Will Probably Be the First Cognitive System to Be Understood in Terms of the Circuitry That Supports It Highly parallel computations of visual and auditory information have now been described. We are also beginning to understand the neural and cellular bases

TlIE NERVOUS SYSTEM AND BEHAVIOR 213 for selection of sensory information. Clear electrical signs of selection observable even with scalp electrodes can separate the messages that are being attended to from those that are being ignored, within the first 100 msec after presentation. Recording from single cells in alert monkeys has provided a great deal of infor- mation about the anatomy of the system that selects information from visual space. We know that an area of the midbrain (the superior colliculus) is important for selection when the animal attends by making eye movements, whereas tha- lamic (pulvinar) and cortical (parietal lobe) areas are involved when the animal attends covertly to an area of the visual field not currently being fixated. Lesions of these areas produced by strokes or tumors in humans produce deficits similar to those described in the monkey. We are beginning to relate the detailed computa- tions performed when attention is moved from one visual location to another to these anatomical areas. These signs of selective attention depend on the integrity of the prefrontal cortex. In patients with lesions in this area, electrical activity related to early selection is reduced and performance is impaired. Similar deficits in event- related potentials have been found in schizophrenic subjects who are often de- scnbed as lacking higher levels of attentional control. Furthermore, blood-flow studies of schizophrenic subjects who are performing tasks that require shifting of attention among different stimulus dimensions of color, form, and number show a deficit in flow in the prefrontal cortex. The deficits in adult patients with frontal lesions are often characterized by difficulty in maintaining coherent programs designed to reach a goal. These same patients are frequently distracted from their goals by sensory events, as if they were less able than normal subjects to control sensory activation. Animals with lesions in frontal areas have difficulty in responding correctly when a delay is imposed between the stimulus presentation and their response. These animals have trouble whenever they are required to select a novel or less typical response. They seem to have difficulty in resisting the momentarily strongest response in order to pursue a goal. Psychobiology of Development Studies on the Psychobiology of Development Have Transformed Our Understanding of the Capabilities of the Newborn The genetic endowment of human newborns provides a considerable capabil- ity for perception, learning, and even such higher-level concepts as number. Applications of simple conditioning and habituation methods to infants is provid- ing a basis for exploring differences in cognitive ability. There seems to be some stability from measures in early infancy to later achievement as measured by standardized tests. In addition, temperamental differences among infants in reactivity to external events, emotionality, and inhibitory control are providing a new impetus toward understanding the biological development of personality. Important shifts in temperament seem to occur during definable time periods, in which the maturation of neural systems change the capability of the developing

214 OPPORTUNITIES IN BIOLOGY infants to regulate their own behavior. These critical periods provide important clues to the changes in behavior resulting from maturation of brain regions during development. Studies of the development of monkeys have already enlarged our understanding of the slow maturation of some areas of the brain. The frontal lobes, for example, continue to develop for some years afterbirth. These develop- mental processes can now be studied in human infants by observing changes in metabolic activity within regional brain areas. The results so far reported with these techniques fit the time course of shifts found by behavioral studies. Behavioral changes seem to occur in parallel for monozygotic twins, who become increasingly concordant with age. Thus, at least some of the shifts observed in development seem to have a genetic basis. These new findings set the occasion for reexamining the fundamental issue of how genetic and environ- mental influences work in concert to shape the social and cognitive development of infants and children. The genetic analysis of the development of behavior promises to provide insight into some disorders. For example, genetic analysis of developmental dyslexia has suggested not only the inheritance of one form of the disorder, but also through linkage analysis provides suggestive evidence for an autosomal dominant locus on chromosome 15. Fundamental Understanding of the Neurobiology of Cognition Will Have Important Practical Applications The analysis of reading in terms of elementary cognitive operations has already begun to guide efforts to produce specific remediation techniques in developmental or acquired dyslexia. Of more importance in terms of public health are the efforts to apply these concepts to closed head injures. Although it is still unclear how successful this kind of cognitive remediation is, the potential benefits are great. Neural imaging techniques have revolutionized the practice of clinical neu- rology. In the near future, our understanding of the neural mechanisms underly- ing selective attention and language should assist the neurosurgeon in the delicate task of avoiding the most critical areas when performing needed surgery. Im- proved assays of cognitive function should also allow better tuning of drug therapies and replacement or transplant methods. The combination of great intellectual interest and obvious practical impor- tance makes this area a central one for the future of neurobiology. BEHAVIORAL ECOLOGY Behavioral Ecology and Sociobiology Encompass the Study of the Evolutionary Adaptiveness of Behavior Evolutionary adaptation refers to differences in structure, physiological pro- cesses, behavioral patterns, or complexes of traits that increase the inclusive

THE NERVOUS SYSTEM AND BElIAVIOR 215 fitness of one organism over that of another organism of the same species. But because the effect of specific traits on the inclusive fitness of organisms is difficult to measure directly, evolutionary adaptations are usually inferred from the organisms' '~goodness of fit" to their environments. One of the tasks of behavioral ecology is to investigate this particular correlation: to understand how selection pressure, exerted by the ecological and social environment, favors one behavior over another. As a consequence of this evolutionary approach, the closely interlocked fields of behavioral ecology and sociobiology have been revitalized during the past 10 to 15 years. The fields have been energized by the merging of ethology, population genetics, and modern evolutionary theory in a manner that has proven effective in generating new hypotheses about the adaptiveness and evolution of behavior. In the past decade or so, numerous mathematical models have structured the theoretical framework of behavioral ecology. Among these models, which are derived from theoretical population genetics, the most prominent are inclusive fitness theory (now better known as kin selection theory), optimization models (derived from microeconomics), and evolutionary stable-strategy models (de- rived from game theory). These theoretical concepts must now be tested by much more extensive and rigorous experimentation. To that end, we need the techniques and methodologi- cal approaches developed in experimental ethology, psychobiology, and neurobi- ology. We need to understand the neurobiological mechanisms underlying be- havioral expressions to appreciate the framework that defines the animal's re- sponse to environmental conditions. A knowledge of the morphological features and physiological mechanisms underlying behavioral patterns is crucial to our understanding of the evolutionary constraints on behavioral~cological adapta- tions. Without an appreciation of the behavioral mechanisms involved in such key phenomena as competition, parent-offspring relationships, communication, and interspecific interaction within ecological communities, an adequate and precise description of ecological organization is not possible. Ecologists need to appreciate more fully the function of behavior as one of the major keys for understanding ecological systems. Conversely, neurobiologists (despite their recent advances at the molecular level) must not forget that these mechanisms are the products of evolution and, in particular, of natural selection acting in specific environmental and social settings. One of the Most Rewarding Trends in the Study of Behavior Is the Convergence of Field and Laboratory Approaches As ecologists increasingly realize the importance of behavior, they have begun to turn to laboratory techniques developed by experimental behavioral biologists. At the same time, psychobiologists have increasingly applied their experimental methods to investigations of naturally occurring behavior. They

216 OPPORTUNITIES IN BIOLOGY now more fully appreciate the biological and specifically ecological boundaries that affect learning patterns and the development of behavior in general. The "Umwelt" concept, first presented in 1921, has gained new meaning. We now ask with new methods and vision, What is the perceptual environment of an animal? How is it affected by the animal's developmental stage, social status, motivation, and other behavioral contexts? How does an animal filter out extra- neous stimuli or select particular cues from an indescribably rich palette of environmental stimuli? And which cues trigger an animal's predisposition to learn and to store the things learned in its memory? These are important questions not only for understanding adaptive learning mechanisms and sensory-neurobiol- ogical processes, but for understanding the significance of the hierarchical organi- zation of cues in such behaviors as orientation, habitat choice, mate selection, kin recognition, and the identification of competition and enemies. A particularly interesting illustration of the principle is kin recognition. Much of sociobiological theory predicts that animals will behave differently toward close genetic relatives and nonrelatives. As predicted, most instances of apparently altruistic cooperation that have been analyzed reveal nepotism at work. Clearly, animals must be able to recognize their close kin. Animals from many different taxa have this capacity. How they accomplish such often fine-tuned discrimination is now an active area of investigation, and most of the major hypotheses suggest roles for learning, memory, and specific sensory cues. Be- cause adequate tests of these hypotheses will require careful experiments, an array of laboratory techniques in experimental behavioral physiology and psychobiol- ogy are being developed and applied. Studies of Kin Assemblages and Kin Recognition Lead to a New Understanding of Population Structures and Mating Strategies Kin recognition not only makes nepotistic behavior possible, but bears the responsibility for the avoidance (or optimization) of inbreeding. In highly evolved social systems, such as the eusocial insect societies, kin recognition functions as a social immune system, which accepts individuals that carry the right family label and rejects those labeled with foreign markers or lacking the familiar markers. The strategy of recognition of "self" and rejection of "alien" in such societies, which have been called superorganisms, resembles the strategy metazoan organ- isms use to protect bodily integrity. Interesting evolutionary parallels can be drawn between kin recognition systems and the immune system. In most organisms studied to date, kin recognition labels seem to be chemi- cal probably complex blends of specific chemical compounds that are ulti- mately genetically determined. These labels apparently have to be learned by kin- mates, but the learning process also seems to contain specific temporal patterns. Also, learning seems to be programmed and constrained by templatelike neural mechanisms. Substantial evidence has been adduced of similar mechanisms in

THE NERVOUS SYSTEM AND BEHAVIOR 217 invertebrates, lower and high vertebrates, including primates, and even in human infants and children. A second advance in our understanding of social groups centers on conflicts of interest. A counterpoint to the documentation of cooperation between relatives is the discovery of many instances of subtle and not-so-subtle disharmony in apparently cooperative groups. It follows from the neo-Darwinian "selfish gene" view of evolution that cooperation between individuals should reflect a delicate balance between costs and benefits that could easily tip toward conflict. Here, too, recognition systems play a crucial role. Conflict and competition appear to constitute a major force in structuring ecologial communities, but little is known about the behavioral mechanisms underlying competition. Central questions that should be addressed concern the role of learning in competitor recognition and the comparative assessment of "self" versus "opponent" in competitive interactions. As in the study of coopera- tive and competitive interactions within a single species, the role of learning and memory is a central topic for understanding these naturally occurring behaviors. Direct Links Are Being Made Between Behavioral Ecology and Development Psychobiology The only significant result of evolution by natural selection is the determina- tion of what genes are preserved, or what new genetic variants persist, and which disappear after numerous generations. But the phenotype on which selection acts is not merely an adult, but a life cycle, and therefore behavioral ecologists are not concerned solely with the genes that affect adults but also with those that affect the whole of development Such behaviors as foraging, mating, nursing, helping, and fighting are based on short-term decisions. The evolutionary significance of these behaviors will be fully understood only if they can be related to long-term life-history patterns. Life-history theory deals with questions such as how an individual should allocate resources to growth versus reproduction to achieve the greatest fitness. Attempts to integrate typical behavioral-ecological analyses of short-term decisions with long-term approaches of life history theory should increase. All Social InteractionsInvolve Communication The study of communication will continue to be a major topic in behavioral biology and will entail, on the one hand, the investigation of the signal-receptor systems and the neural mechanisms of information processing and, on the other, the comparative study of the evolution and ecological adaptation of communica- tion strategies. The two main themes of evolutionary biology are adaptation and phylogeny. Both are best examined by comparative methods. Methods developed by taxono

218 OPPORTUNITIES IN BIOLOGY mists are now being applied to the comparative study of behavior. Adaptive strategies can often be deduced from analogous mechanisms found in phylogen- etically diverse groups of species. As behavioral fossils do not exist, the recon- struction of the most likely history or phylogeny of animal communication is based on comparative studies of organizational levels of communication mecha- nisms in closely related species. A current topic of debate is the question of whether signals, or communica- tion displays, were selected during evolution for their efficacy in transferring information or for their effectiveness in persuading or manipulating others. Only detailed behavioral-ecological analyses of communication strategies in animals can provide answers to these questions. In recent years it has become increasingly clear that communication is rarely characterized by a direct all-or-none response. Communication is not always a deterministic releasing process, but sometimes plays a different and more subtle role, modifying the behavioral properties of the receiver and alerting and focusing the receiver's attention on the situation context. This kind of system has been called modulatory communication. In it, signals do not release specific behavior patterns, but rather modulate the probability of reactions to other stimuli by influencing the motivational state of the receiver. We should expect such modu- latory communication to be most frequent in complex animal societies, where many members perform many different tasks at the same time, and where an economically efficent organization of behavior requires that the work force dis- tribute its energy investment among different tasks through an optimum division of labor. It has recently been argued that the social system itself, by communicators processes, can develop the properties of problem solving; it can develop what amounts to a cognitive system that encompasses but also exceeds the cognitive capabilities of the individual components. It has even been suggested that we compare the coordinating mechanism active in such superorganisms with the interactive neuronal processes that endow central nervous systems with their acknowledged cognitive capacities. It is remarkable that in the brain, as in highly social systems, we find mechanisms that set the overall level of arousal. Recent examples in social insect communication illustrate this point impressively. For example, tonic sensory input from a variety of sense organs and spontaneous activity of neural arousal systems perform in the nervous system the functions that unspecific modulating signals serve in social organizations. In both forms of organization we further find more specific regionalized or addressed mechanisms of focusing the atten- tion to a specific subset of stimuli in a given context We find that, within the larger systems, mechanisms exist that modulate in graded fashion the activity probability of small dedicated subpopulations of neurons or individuals that are thus recruited to perform specific tasks. It is probably more than chance that the neurophysiologist arrives at describing these mechanisms as local modulating

THE NERVOUS SYSTEM AND BElIAVIOR 219 interactions between neurons just as the student of social communication in animals independently finds it appropriate to qualify basic processes of social organization in this way. ABNORMALITIES OF BEHAVIOR Central to the Study of the Nervous System Is the Desire to Understand the Abnormalities of Behavior Produced by Various Neurological and Psychological Disorders The goal of the modern study of the nervous system is to understand human behavior: how we sense objects in the world around us, execute skilled move- ments, feel, think, learn, and communicate with one another. Study of the nervous system has traditionally provided the scientific and therapeutic underpinnings for neurology and psychiatry. We illustrate this point with two examples: (1) the application of modern molecular genetic approaches to diagnose neurological diseases and (2) the application of modern biochemical and imaging techniques to diagnose and treat psychiatric disorders. Molecular Genetics and the Diagnosis of Neurological Disorders A surprising number of serious neurological diseases have a genetic origin. These include neurof~bromatosis, Huntington's disease, a subform of Alzheimer's disease, retinoblastoma, and various congenital diseases of muscle. The devasta- tion these diseases produce is great. For example, John Merrick, the "Elephant Man" who lived in the late 1880s, was relegated to a life as a side-show circus freak. He suffered from severe neurofibromatosis. The more than 100,000 sufferers of neurofibromatosis today receive no better treatment than Merrick, other than frequent surgeries (as many as 100 a year in some cases) to remove the offending neurof~bromas that can disfigure the entire body, occlude the auditory canal, and extend into the brain and spinal cord to pose an immediate life threat. Neurofibromatosis is an autosomal dominant disorder, which means that each child of a parent with neurof~bromatosis has a 50 percent chance of inheriting the disease. Today, more than half of all nursing homes beds in America are occupied by patients with Alzheimer's disease, an illness considered the fourth leading cause of death in this country. At least 10 percent (and by some estimates a much higher percentage) of cases are autosomal dominant a 50 percent risk to offspring. There is no treatment for Alzheimer's disease, only a growing number of elderly people in the United States who are at risk. Huntington's disease is the genetically programmed loss of nerve cells im- portant for mental and motor function, which usually has its onset in midlife. It, too, is an autosomal dominant disorder for which no effective therapy exists. Folk

220 OPPORTUNITIES IN BIOLOGY singer Woody Guthrie, who wrote 'This Land is Your Land," died of Hunting- ton's disease. He also wrote from Brooklyn State Hospital a poem ending "there is no hope lulown.'' But with the advent of recombinant DNA technology, there is, for the fast time, real hope. Within the next decade, it should be possible to know the chromosomal assignment and exact DNA sequence of the genes that cause these and some other pernicious diseases that affect the nervous system. It should be possible to trace from a genetic lesion to the biochemical or regulatory disturbances it produces, from DNA through anatomical and physiological tracts to the expression of an aberrant gene in disordered thought or action. We should be able to chart the pathway from gene to brain to behavior, learning how the tiniest defects can cause the wildest movements, severe memory loss, suicidal depression, or the capacity to hear or see what does not exist. We should also be able to identify the genes responsible for normal brain functioning, which will provide powerful new in- sights into mental functioning. Molecular Pharmacology, Modern Imaging, and the Diagnosis and Treatment of Psychiatric Disorders Diagnosis in psychiatry is less precise than in neurology or in the rest of medicine because most psychiatric diseases cannot as yet be localized to specific regions of the brain, much less to particular proteins in specific nerve cells. Thus, the diagnosis of psychiatric disorders must rely primarily upon clinical symp- toms. One major way of grouping psychiatric disturbances is into those which are psychotic and those which are not psychotic. The term psychotic can be loosely defined as reflecting a major loss of contact with reality. Nonpsychotic distur- bances include anxiety, neurosis, and character disorder. The major psychotic disturbances are schizophrenic and affective illness, comprising mania and de- pression. In terms of human suffering and public expense, psychotic disorders present a more serious problem for society than nonpsychotic illness because of the much greater disability caused. Since the major psychoses often commence in early adulthood and persist throughout life, their total cost to society greatly exceeds that of cancer and heart disease. For example, at least 1 percent of the population is schizophrenic, an incidence comparable to that of diabetes. It is likely that the incidence is substantially higher, since many individuals who seem to be schizophrenic are not subjected to rigorous diagnosis. Recent pharmacological studies have provided important insights into schizo- phrenia. The effects of drugs have permitted the development of hypotheses about specific neurochemical abnormalities. Many of the drugs that influence schizophrenic symptoms affect the neurotransmitter dopamine. The neuroleptic antipsychotic drugs act by blocking dopamine receptors. Reserpine, which has antischizophrenic effects, depletes the brain of dopamine. Amphetamines, which often exacerbate schizophrenic symptoms, release dopamine. These findings

TlIE NERVOUS SYSTEM AND BEHAVIOR 221 have suggested that an excess of dopamine might be relevant to schizophrenic pathophysiology. Postmortem studies consistently show increased numbers of dopamine receptors in the schizophrenic brain. This result has been reinforced by recent PET studies in patients, which show that schizophrenics have almost twice as many dopamine receptors as control subjects. It has often been suggested that schizophrenia is a family of diseases that may have different etiologies. One recent classification focuses on the presence of positive or negative symptoms. Positive symptoms refer to florid delusions and hallucinations, while negative symptoms reflect autism and general withdrawal, a "wall-flower" type of behavior. Most neurological drugs are more effective in relieving the positive than the negative symptoms of schizophrenics. Chronic "burnt-out" schizophrenics often display primarily negative symptoms. That patients with chronic schizophrenia have enlarged cerebral ventricles relative to those with the acute forms of the disorder, along with differential drug responses, has prompted the definition of two subtypes of the illness, type I and type II schizophrenia. Type I manifests positive symptoms, a good response to neurolep- tic drugs, and no enlargement of the cerebral ventricles, whereas type II schizo- phrenia is characterized by negative symptoms, a poor drug response, and en- larged ventricles. It is likely that the next five years will witness more characteri- zation of these symptomatic subtypes of the disease and linkages to laboratory abnormalities. The advent of CAT scanning has led to a greater clarification of the enlarged cerebral ventricles in schizophrenics first noted by pneumoencephalography. The recent development of MRI should permit far more extensive delineation of the ventricular enlargement. Novel therapeutic approaches to schizophrenia are likely to focus differen- tially on the positive and negative symptoms. Of the neuroleptics in common use, only the diphenylbutylpiperidines show selective efficacy in relieving negative symptoms. The diphenylbutylpiperidines are just as potent calcium antagonists as dopamine antagonists. Conceivably, centrally active, selective calcium an- tagonists may have utility in the specific therapy of negative symptoms. So far, the most direct insight into the genetic contribution to schizophrenia has come from twin and adoption studies. These studies have ruled out the possibility that environmental factors artifactually account for hereditary patterns. For instance, among schizophrenics who had been adopted at birth, the biological parents display a high incidence of schizophrenia, whereas the incidence in the adoptive parents matches that of the general population. Twin studies, however, show that environmental factors must play some role in the expression of the genetic tendency. The concordance rate for schizophrenia in identical twins is about 50 percent and not 100 percent, as would be expected if genetic factors alone accounted for the disease. Several studies examining identical twins discor- dant for schizophrenia reveal environmental trauma more in the schizophrenic than the nonschizophrenic co-twin. For instance, the schizophrenic co-twins

222 OPPORTUNITIES IN BIOLOGY generally have a lower birth weight and a greater likelihood of neonatal infection. More detailed studies of this type in the next decade should tease out specific factors that might be crucial in the transformation of a genetic predisposition into frank schizophrenia. Perhaps the most exciting possibility is the potential identi- f~cation of the molecular genetic abnormality associated with schizophrenia, as we have discussed in relation to neurological disease. In families with an extremely high incidence of schizophrenia, one might be able to search for specific genetic markers differentiating schizophrenics from the general popula- tion by using strategies that have been successful in conditions such as Hunting- ton's disease. Being depressed is such a common experience that establishing diagnostic criteria for depressive "disease" is difficult. It is thought that as many as 5 percent of the population suffer from major affective disturbance. Both depression and mania are episodic and have been differentiated by the nature of the episodes. Bipolar disorder is characterized by episodes of both depression and mania, and unipolar illness by recurrent depression but no episodes of mania As with schizophrenia, affective disorders have a strong genetic component. Genetic studies support a fundamental distinction between bipolar and unipolar illness, although there is much overlap. Adoptive and twin studies reveal that affective disturbances and schizophrenia share similar genetic predispositions. Also, as with schizophrenia, drugs that influence neurotransmitters have provided strong hints as to a possible pathophysiology. Hypertensive patients treated with reserpine evince roughly a 15 percent incidence of severe depression, clinically indistinguishable from endogenous major depression. Reserpine de- pletes the brain of its biogenic amines-dopamine, norepinephrine, and serotonin. Alpha-methyldopa, also used to treat high blood pressure, selectively depletes the brain of norepinephrine and causes depression in many patients. Beta-adrenocep- tor blockers used to treat hypertension also elicit depression, presumably by antagonizing endogenous catecholamines. The major antidepressant drugs all seem to act through biogenic amines. The monoamine oxidase inhibitors increase brain concentrations of norepinephrine, dopamine, and serotonin. Tricyclic antidepressants inhibit the inactivation by reuptake of these three amines. Since some of the most effective antidepressants do not inhibit dopamine uptake, it is less likely that dopamine is involved in their action, but norepinephrine and serotonin are important candidates. Recently, several antidepressants have been introduced that selectively inhibit serotonin uptake with no influence on norepinephrine or dopamine. Some psychiatrists feel that drugs more selective for norepinephrine relieve depression by enhancing "drive," whereas serotonin-selective drugs act by increasing a sense of well- being. They hypothesize that there may exist two distinct subtypes of depression, one associated with deficits in norepinephrine functioning and the other, with deficits in serotonin functioning.

THE NERVOUS SYSTEM AND BElIAVIOR 223 Lithium relieves both mania and depression and is prophylactic against recurrence of all affective episodes. Conceivably, knowledge of its action may shed light on fundamental aberrations that occur in both poles of affective illness. The interference of lithium in the phosphoinositide cycle may be a valuable clue. If lithium acts specifically by inhibiting a phosphatase in the phosphoinositide cycle, organic chemicals can be developed to mimic this effect. Such agents would not compete for intracellular sodium to cause the typical toxic effects of lithium. Direct studies of the seratoninergic biochemical system in depression have been particularly promising. Postmortem brains of suicides manifest abnormally low concentrations of serotonin. The spinal fluid of depressed patients consis- tently shows a bimodal distribution of the serotonin metabolite S-hydroxyindo- leacidic acid. One group of patients has concentrations similar to those of normal subjects, whereas another group of patients of approximately equal numbers has a markedly lower concentration. Several researchers have shown that depressives with the lower hydroxyindoleacidic acid levels are more impulsive and prone to attempt suicide. It is now possible to image serotonin receptors by PET scanning, which may permit an overall evaluation of serotoninergic neuronal function. Similar tech- niques will likely be feasible for noradrenergic and dopaminergic neurons. It is honed that molecular Genetic techniques can be aDDlied to the diagnosis and ---rip -A of ¢~ treatment of affective illness, with the ultimate view of identifying specific molecular aberrations that reflect the cause of the illness.

Next: 7. The Immune System and Infectious Diseases »
Opportunities in Biology Get This Book
×
Buy Paperback | $125.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Biology has entered an era in which interdisciplinary cooperation is at an all-time high, practical applications follow basic discoveries more quickly than ever before, and new technologies—recombinant DNA, scanning tunneling microscopes, and more—are revolutionizing the way science is conducted. The potential for scientific breakthroughs with significant implications for society has never been greater.

Opportunities in Biology reports on the state of the new biology, taking a detailed look at the disciplines of biology; examining the advances made in medicine, agriculture, and other fields; and pointing out promising research opportunities. Authored by an expert panel representing a variety of viewpoints, this volume also offers recommendations on how to meet the infrastructure needs—for funding, effective information systems, and other support—of future biology research.

Exploring what has been accomplished and what is on the horizon, Opportunities in Biology is an indispensable resource for students, teachers, and researchers in all subdisciplines of biology as well as for research administrators and those in funding agencies.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!