At the Cusp of the 21st Century

Eric R. Kandel and Darcy B. Kelley

An age or culture is characterized less by the extent of its knowledge than by the nature of the questions it puts forward.

François Jacob
The Logic of Life (1981)

When Thomas Hunt Morgan started his work on Drosophila at the beginning of the twentieth century, the problems of heredity, development, and evolution were complete mysteries. How does like beget like? How does a single cell, the egg, divide and differentiate to give rise to a complex organism—a whole person? How does a group of organisms evolve to give rise to a new species? These were the major questions that confronted biology in the year 1900.

Morgan and his students at Columbia pioneered the modern discipline of genetics, and then went on to harness its power to address each of these questions. As a result, we now understand heredity, development, and evolution in principle, even if we do not yet have all the details.

With the approach of the new millennium, what challenges remain for genetics? In a broader sense, what mysteries remain for biology in the twenty-first century?

Perhaps the last great remaining mystery for modern biology, and, in fact for all of science, is to understand the biological basis of mental experience. How do we sense the world around us, how do we remember that perception, and how do we color it with emotion? Because of the continuing fascination the brain holds for us, biologists predict that such studies will be, for the twenty-first century, what the exploration of the gene was for the twentieth century, and what investigations of the cell were for the nineteenth century. When Morgan developed genetics at Columbia, his goal was to determine how genes control development. One hundred years later, we ask: can genetics also contribute to an understanding of mind?

An effective approach to mind/brain studies must obviously reach beyond genetics to direct observation of mental activity in humans and other primates. Such a wide-ranging initiative, known as molecular cognition, is now underway at Columbia. This effort is combining the techniques and insights of genetics and molecular biology with those of cognitive psychology, to continue and extend the tradition of accomplishment that is the enduring legacy of Thomas Hunt Morgan, his students, and colleagues at the University.

Cognitive Biology: What Is the Mind and Where Is It Located?
Cognitive psychologists and biologists have come to share a belief that what we call "mind" is the range of functions carried out by a physical organ, the brain. Thus, the workings of the brain underlie not only relatively simple mental acts such as the motor behaviors of walking, running, or hitting a tennis ball, but also elaborate emotional and cognitive behaviors such as feeling, learning, thinking, or composing a symphony. As a corollary, the disorders that characterize neurotic and psychotic illness can be seen as specific disturbances of brain function.

Viewed in this way, the scientific challenge posed by the mind becomes that of understanding how the brain produces its mental functions, a subject we can now approach using the tools of modern biology. Of course, understanding how the brain gives rise to mental activity still poses an enormous scientific challenge: creating a larger unification of thought to demystify the mind. There is nevertheless a sense of optimism in the air—driven by the gradual convergence during the past decade of neural science (the science of the brain) and cognitive psychology (the science of the mind).

The central principle for understanding how mental functions relate to brain functions derives from the finding that the functions of mind have a distributed representation in the brain: various mental functions are mediated not by single brain regions, but by distributed sets of different regions connected in very specific ways. Each region analyzes a particular component of a mental process (parallel processing), with some sending the processed information on to other brain regions (serial processing). This awareness that most mental faculties are made possible by the serial and parallel interconnections of multiple brain regions has given rise to the related insight that almost all mental operations are divisible into subfunctions.

This insight is quite counterintuitive. Most of us have the sense that we store and recall our knowledge of representations of the people and objects we encounter—and of the events that occur in our environment—as unified images that can be evoked by appropriate sensory reminder cues or even by the imagination alone. Our knowledge about our grandmother, for example, seems to be stored in a single representation as "grandmother," one that is equally accessible whether we see her, hear her voice, or simply think about her. But we now know that this belief is not supported by the facts, and that knowledge of people, objects, and events, far from being stored as single images, is instead subdivided into distinct categories.

This remarkable insight first emerged from an examination of the fragmented nature of mental deficits following damage to different brain regions. Damage to a particular part of the cerebral cortex (a brain region essential for higher cognition) can block the ability to name a person—for example, one's grandmother—without interfering with the ability to recognize her. A lesion to a different cortical area will destroy the ability to recognize grandmothers and other friends and relations, but not inanimate objects.

The details of this subdivision of mental processes are best understood for visual perception. The brain does not simply replicate the external world like a kind of three-dimensional photograph; rather, the brain's representation of the world is an abstraction—and, even more surprising, the brain builds up this image of the outside world only after first breaking it down into its constituent parts. In scanning a visual scene, the brain analyzes the form of objects as distinct from their movements, and both apart from their color, all before reconstituting the full image again according to the brain's own rules. Each component is processed in a different region, explaining why—as in the case of recognizing one's grandmother—individuals with lesions to a certain part of the brain might have no difficulty in visual reception per se, even though their visual recognition or visual knowledge might be severely impaired.

It was Sigmund Freud who first proposed at the end of the nineteenth century that defects in visual perception might be caused not by a sensory deficit of the eye or its nerve, but by a cortical deficit that affects the brain's ability to combine components of the visual impression into a meaningful pattern. Freud called these defects agnosias, or loss of knowledge. A striking kind of agnosia is the inability to recognize faces, or prosopagnosia. Patients suffering from it can identify a face as a face, name its parts, and even recognize specific emotions expressed by it, but they are unable to identify a particular face as belonging to a specific person—such as our grandmother.

Perhaps the most astonishing example of the combinatory structure of mental processes is the finding that our very sense of ourselves as a self—a coherent being—depends on neural connections between the two hemispheres of the cerebral cortex. In some epileptic patients the fiber tract connecting the two hemispheres, the corpus callosum, is severed to prevent seizures from spreading from one side of the brain to the other. As a result, each hemisphere carries an independent awareness of the self—for example, each responds to tactile stimuli applied to the opposite hand, but not to stimuli applied to the same-side hand. When identical objects are placed in both hands of such patients, the object in the left cannot be compared with the one in the right hand, because the two hemispheres are no longer in communication with each other. Even more dramatic, in most of these patients the right hemisphere cannot understand language that is well understood by the isolated left hemisphere. As a result, conflicting commands can be given selectively to each side of the brain!

The brain accomplishes its remarkable computational feats—the initial deconstruction and the subsequent reconstruction, or binding together, of perceptions and other mental phenomena—because the nerve cells that form its many components are wired together in very precise ways. Yet, equally remarkable, this wiring is not immutable. Connections between nerve cells can be altered by learning. We remember events because the structure and function of connections within the brain can be modified through experience. Thus, if you remember anything tomorrow about this essay that you have read today, it is because your brain has been altered by this particular reading experience.

Molecular Cognition: Bridging Genes to Mind
A major factor behind Thomas Hunt Morgan's success in relating genes to chromosomes, and then in showing how both are connected to the organism's overt characteristics, was his decision to use the fruit fly as the subject for his experiments. But most biologists believe that for processes as complex as the functioning of the brain, no single organism will suffice—there is no Drosophila to serve as a model organism for molecular cognition. So, in addition to studying simple organisms such as worms, flies, and snails, we will also need to study mice and monkeys and ultimately to examine the human mind directly.

Of the more complex animals, the mouse has a long history as a subject for the study of behavioral genetics. At the turn of the century, "waltzing" strains of mice produced by spontaneous mutations made popular household pets. These spontaneous mutations, changes in the DNA, were induced by events in the mouse's environment such as natural radiation. It was, however, the pioneering work in the 1940s by L.C. Dunn, a chairman of Columbia's zoology department who as an undergraduate had been inspired by Morgan's writings, that introduced the study of spontaneous mutations as a new technique for understanding vertebrate development. Determining exactly which genes were affected by spontaneous mutations was very difficult in mice, so scientists initially turned to flies and worms, where methods for gene discovery are much easier. In the 1980s, a revolutionary technical breakthrough allowed researchers to replace any mouse gene at will with altered or non-functional versions (the latter are often called "knockouts"), which could in turn be transmitted to the mouse's offspring. The new knockout methodology catapulted mice to the forefront of genetics. The opportunity to modify the genome of mice has given rise to important questions about the role of genetics in solving the remaining problems that confront biology, and more specifically about the role that genetics can play in helping to address the problems of the mind.

Several of the first gene knockouts in mice were achieved at Columbia by Elizabeth Robertson working with Argiris Efstratiadis and Steven Goff on studies of development and cancer biology. Influenced by their efforts, Eric Kandel used these and other knockout mice to examine how specific genes affect neuronal communication in the brain on the one hand, and learning and memory storage on the other.

Like humans, mice have two memory systems: an implicit system that stores information about how to do things (perceptual and motor strategies), and an explicit one that stores information about the things themselves (facts or events involving places, objects, and other living things). The explicit memory system is located in the medial temporal lobe of the cerebral cortex and its focus is a specific brain region, the hippocampus. Lesions of the hippocampus disrupt memories of space. Kandel and his colleagues found that altering the expression of certain genes in the hippocampus interferes with changes in the signaling between neurons—signaling that is crucial for memory storage, especially memory of place. Moreover, mice carry in their hippocampus an internal representation of space—a kind of cognitive map—and Kandel and his colleagues at Columbia together with Susumu Tonegawa and his colleagues at MIT found that genes that block the storage of spatial memory do so by interfering with the normal stability of the animal's internal representation. In this way, the use of mouse knockouts has provided a preliminary understanding of how connections between nerve cells change as a result of experience (learning) and how these changes are maintained in memory.

It soon became clear that genetics could be used to study not only what an animal knows and remembers—its cognitive capabilities—but also what it feels—its emotional or affective sensibilities. The latter includes aggression and mood, which provide the focus of René Hen's studies of genetically modified mice. Many of the drugs used to treat disorders of mood and emotion in humans act via a brain chemical called serotonin or the different kinds of proteins that this neurotransmitter binds on the surfaces of brain cells. Hen knocked out a number of the proteins through which serotonin acts, one at a time, in each case revealing specific effects on the mouse's behavior. He is currently identifying the serotonin receptors that mediate the antidepressant effects of drugs such as Prozac.

Richard Axel '67C has used molecular cloning to open up the study of smell, one of our most elaborate senses. Humans are able to perceive perhaps 10,000 different odors, all of which tap into our mental processes and our memory storage in complex ways. Until recently it has been difficult to study smell because we knew nothing about the molecules that detect smells (receptors)—not even how many there are. Does smell work like vision, in which we perceive the richness of the visual word using three receptors for colors and mixing their outputs? Or is olfaction more elaborate?

Axel succeeded in identifying the olfactory receptors and found that there are not three or four, but 1,000; fully one percent of all human genes are devoted to the perception of odors. He demonstrated that the molecular decoding of smells begins at the olfactory periphery, in the sensitive tissue lining the nasal cavity, and that any olfactory neuron expresses only one receptor. Each olfactory neuron in the nose sends a projection to a particular spot in the brain, with all the members of one olfactory receptor class converging on a few specific sites in the first olfactory way station (glomeruli). As the brain "reads" the activity of individual glomeruli we construct a world of smells.

Yet another Columbia scientist, Thomas Jessell, uses genetics to establish how the neurons of the brain develop their specific identities and how they form the correct connections essential for all brain functions. How is an entire neural circuit for a behavior, a complete reflex from sensory input to motor response, constructed? Jessell's research has shown that neurons that wire up to each other to form a particular circuit all share a common master regulatory factor, one that may be responsible for coordinating their patterns of genetic activity and permit the formation of specific connections between them.

Genes and the Human Brain
Human beings are far from ideal as experimental organisms for the study of genetics, but they provide fertile ground for exploring cognition. One reason for this, of course, is that they are the organisms whose cognition we ultimately hope to understand. In addition, not only do humans have remarkable cognitive capabilities, but they are also verbally insightful, making it easy to test them for a variety of cognitive skills.

Despite experimental limitations, we have learned a great deal recently about human genes, thanks to the introduction of DNA markers which facilitate gene mapping. Once mapped onto different chromosome regions, the genes can be cloned and identified—a key thrust of the ongoing Human Genome Initiative. We can identify genes that contribute to cognition and other higher mental functions, and may gain insight into the genetic factors that influence the attributes of personality and intelligence, a possibility that is exciting and at the same time daunting. It is a prospect that raises ethical issues for the University and for society as a whole, difficult ethical questions that must be faced with each new advance in molecular biology.

One of the most instructive examples to date of how genes affect human behavior has come from an analysis of Huntington's disease, a degenerative disease of the nervous system and the first complex human behavioral abnormality to be traced to a single gene. This important feat of genetic detective work has been spearheaded by Columbia's Nancy Wexler and her colleagues throughout the world. Huntington's disease affects both men and women with a frequency of about five per 100,000. Over the long course of the disease, neurons in the basal ganglia, a region of the brain involved in regulating voluntary movement and aspects of cognitive function, misfunction and then die. Huntington's is characterized by incessant, rapid, jerky movements and cognitive impairment (hence, its old name, Saint Vitus' dance, and its common name, Huntington's chorea), also by changes in memory and emotion. Everyone who inherits even one copy of the gene gets the disease; it strikes people in their late forties and early fifties, after they have married and had children, and progresses ruthlessly toward death.

Wexler and her colleagues identified the Huntington's disease gene on chromosome 4. This gene encodes a large protein called Huntingtin; we don’t know yet what the normal protein does. The mutated form of the Huntingtin gene revealed a fascinating feature now known to be shared with some other degenerative diseases of the brain: it encodes a protein containing a repeated stretch of one particular amino acid, glutamine. In individuals with the disease, the stretch is abnormally long, and this DNA repeat (CAG for glutamine) kills certain nerve cells in ways that are not yet understood. The use of mouse models of Huntington's promises to reveal key links between the gene, its effects on nerve cells, and the behavioral impact—both emotional and cognitive—of this devastating disease.

Even at this early stage in the study of human genes, research findings are beginning to reveal the scope of the issues that lie ahead. Long-held distinctions between the effects of genes (nature) and the environment (nurture) can be viewed in new ways. Both genetics and environmental factors may influence the same biological substrates; accumulating evidence suggests, for example, that the adverse impact of environmental stressors can be reversed by pharmacological manipulations. Researchers have discovered that learning gives rise to long-term memory, leading to changes in the expression of genes and structural alteration of the brain—in short, that the activity of genes themselves can be influenced by environmental factors.

Genetic studies of personality and mood suggest, for example, that natural variations among personality traits and pathological mood disorders actually lie along a continuum, effectively blurring any distinction between normal and abnormal. We have long recognized that some of the most creative members of our species are subject to extremes of temperament. Genetic knowledge provides us not only with a molecular explanation for various behaviors, but also with the realization that we all carry vulnerabilities to disorders of gene function, disorders that contribute to cancer or psychiatric breakdowns. An understanding of these molecular mechanisms promises to bring new cures for the diseases, along with new challenges on issues involving individual accountability and freedom. Future advances in understanding the biological basis of thought and mood will have significant repercussions for our medical and legal systems.

Cognitive Neuroscience of Human and Non-human Primates
Millions of items . . . present to my senses which never properly enter my experience. Why? Because they have no interest for me. My experience is what I agree to attend to. . . . Everyone knows what attention is. It is taking possession by the mind in clear and vivid form of one out of what seem several simultaneous objects or trains of thought. Focalization, concentration of consciousness are of its essence. It implies withdrawal from some things in order to deal effectively with others.

William James
Principles of Psychology (1890)

What we would ultimately like to understand are elements of consciousness—and, if we are to come to grips with the biological underpinnings of consciousness, we will have to look beyond the gene to brain systems that have evolved over the centuries in primates. One approach to exploring consciousness has been to examine the biological basis of selective attention, a component of conscious awareness, via a rigorous analysis of complex behavior in intact, awake, behaving primates as they carry out highly controlled perceptual or motor tasks. Monkeys have proven to be a valuable subject for such research, especially since studies with humans performing similar tasks while undergoing brain imaging have shown strong similarities in the elementary mechanisms of perception and movement control.

Initial hints about the physiological substrate of selective attention date back to the nineteenth century and the work of the English neurophysiologist David Ferrier. In 1876, Ferrier described what happened when an electrical stimulus was applied to the prefrontal cortex of a rhesus monkey: "The eyes open widely, the pupils dilate, and head and eyes turn towards the opposite side." In short, the monkey behaved in a way consistent with shifting attention to a new location in space. But it was not for another hundred years, in the 1970s, that neurophysiologists started to investigate systematically the neural basis of selective attention. They found that neurons in the prefrontal cerebral cortex were activated when monkeys moved their eyes to look at visual stimuli (overt attention), while neurons in the parietal cortex were instead activated when monkeys shifted attention to a new location, even if their eyes didn’t move (covert attention). The response of visual neurons is affected by attention in a way that suggests that the area of visual space driving their neurons (receptive fields) "shrinks" to focus on attended stimuli. In research now underway at Columbia, Vincent Ferrera is studying how spatial attention and feature-selective attention are integrated in the prefrontal cortex of monkeys. In the process, he is developing computational models to link cellular and behavioral correlates of attention.

Technology now allows us to visualize the activity of millions of neurons in specific regions of the human cortex while engaging in solving complex problems via PET (positron emission tomography) and functional NMR (nuclear magnetic resonance). At Columbia, these brain imaging methods are being applied by Yaakov Stern and Lynn Cooper to study normal implicit and explicit memory storage, by Richard Mayeux '91SPH and Scott Small to examine changes in memory storage during aging, and by John Mann to examine changes in brain activity associated with depression.

Other research builds on the finding that attention and emotion not only color our memories, but are integral to their creation, providing the contextual cues later used to bind the fragments together into recollected wholes. In this vein, Janet Metcalfe and Walter Mischel '87C are exploring a monitoring-and-control system for human explicit memories that generates a "feeling-of-knowing." Their analysis focuses on the way the control system assesses incoming events and adjusts the attention or effort assigned to them on the basis of novelty: little energy goes to old and already well-known events, but considerably more attention is devoted to novel events.

In the long run, we hope to be able to study similar tasks in monkeys and humans and then carry out intervention experiments in monkeys in order to distinguish brain activity patterns that are actually responsible for (or cause) components of conscious experience. But because of the brain's extraordinary complexity and the sheer number of nerve cells it contains—a million-million—we will never be able to achieve a complete understanding of its information processing functions without some sort of a computational theory, just as we cannot comprehend how a computer works without concepts such as an operating system and data structure. Columbia's Ning Qian has been applying the techniques of mathematical analyses and computer simulations to his investigations of how a population of cells with realistic physiological properties could act in concert to solve perceptual problems such as depth perception and motion detection. In the process, he, like his colleagues in a range of scientific disciplines throughout the University, is laying the essential groundwork for advances still to come.

Columbia and the Mind/Brain Initiative
The scope and complexity of initiatives now underway at Columbia highlight the formidable challenges posed by mind/brain studies, which have now been taken up by the University's combined neuroscience and cognitive psychology faculty, including scientists on the Health Sciences and Morningside Heights campuses. The product of their shared commitment is the new Mind/Brain Initiative, launched with the enthusiastic endorsement of President George Rupp and representing the most extensive and far-reaching collaboration to date between these two units of the University. This effort builds on the University’s well-established base in molecular neural science, which was recognized by the Howard Hughes Medical Institute when it selected Columbia as the site of one of four national centers for neuroscience in 1984. The mission of the Mind/Brain Initiative is to focus the strengths of psychology and neural science at Columbia on mental processes, including perception, action, memory, thought, emotion, attention, consciousness, and self-awareness.

Such an ambitious undertaking will have consequences for many areas of academic life at the University. To begin with, the importance of neurobiological insights into the human brain will most likely lead to a merger between the disciplines of Psychology and Neurobiology. Psychology will continue to provide the detailed analysis of mental processes, but biology will provide the tools to discover how these mental processes are generated. From cellular recordings of the primate cortex on the one hand to the imaging of human brain activity on the other, we will be able to determine how the brain reconstructs the most complex memories and the role of emotion and attention in re-creating the composite whole. In a corollary development, Psychiatry and Neurology will come together as two aspects of a new science of brain function, with disorders of mind being perceived as a distinctive set of disorders of the brain, much along the lines of other neurological disorders.

A focus on understanding the biology of mind will also have a profound effect on the computational sciences. Studies of artificial intelligence and of pattern recognition by computers have made us realize that the brain recognizes movement, form, and color using strategies that no existing computer begins to approach. Simply to look out into the world and recognize a face or enjoy a landscape entails an amazing computational achievement, one that is much more difficult than the ones required for solving logic problems or playing chess. Understanding how the brain sees, hears, thinks, and feels will have a major and instantaneous impact on the design of computers and robots alike.

In the field of education, the Mind/Brain Initiative will inspire a new level of empirical investigation into the kinds of questions that generations of students have assumed are purely philosophical, answerable only by introspection and logic: What aspects of the mind are innate? How does experience interact with the mind’s innate organization? How do we perceive the world, learn about each other, and remember what we experience? The biological investigation of mental activity might even serve as an intellectual bridge in Columbia's Core Curriculum, forging a new synthesis between the humanities, which have traditionally been concerned with the place of the individual in society, and the biological and physical sciences, with their traditional focus on nature and the universe. Understanding the biology of the mind would thus come to represent not only a scientific goal of great promise, but also one of the ultimate aspirations of humanistic scholarship, part of the continuous attempt of each generation of scholars to understand human thought and human action in new and more complex terms.

On a broad societal level, many of the most pressing problems that confront us today—addiction, aggression, and war, to name just three—revolve around the biological nature of human behavior. Our understanding of how humans bond to one another, whether in the mother-child interaction or larger group cohesion, can also be enriched by understanding the underlying biology. Without question, these issues are on the distant horizon, but most neural scientists believe they will be within reach during the next century. Work undertaken today will allow the biology of mind to have a major impact on sociology and on social psychology.

Genetics in Perspective: Columbia’s Contribution to Modern Thought
In anticipation of the University's upcoming 250th anniversary, we have looked both backwards and forwards, focusing on genetics to illustrate how biology has matured as a science during the twentieth century and to highlight the distinctive role Columbia has played at key points in this process. As we approach the twenty-first century, it is clear that genetics in particular and biology in general will exert an even greater influence on modern thought.

Already the pace of achievement is breathtaking. We now have in hand the complete genome of yeast, all 7,000 genes of a single organism possessing a nucleus. This year, the first genome was obtained for a multicellular organism, the 17,000 genes of the worm C. elegans. By the year 2001, we will be in possession of all of the 100,000-odd genes that make up the human genome (as well as the 20,000 genes of the fly genome). We then will have in hand, for the first time, the equivalent in biology of a periodic table. We will know all the genetic elements that make up life in all of its forms.

As a result, the very nature of biology—including human biology—will change dramatically. These developments will challenge all aspects of the University, raising ethical, political, social, and scientific issues in biology, medicine, philosophy, psychology, and religion. We shall have tasted once again from the tree of knowledge, and we shall be confronted with all the benefits and challenges that this exposure entails. The true challenge for a great University such as ours is to savor that knowledge and to debate its consequences for the betterment of society.

We have benefited from comments on this essay by Thomas Jessell.


Jacob, François (1981) The Logic of Life, Betty E. Spillman (trans.), Princeton: Princeton University Press.

James, William (1890) Principles of Psychology, New York: Henry Holt.

Kandel, Eric R., Thomas M. Jessell, James H. Schwartz (2000) Principles of Neural Science, 4th ed. New York: McGraw-Hill.

Living Legacies Home