Miller, Neal Elgar

Updated About content Print Article Share Article
views updated


(b. Milwaukee, Wisconsin, 3 August 1909; d. Hamden, Connecticut, 23 March 2002);

learning, motivation, psychotherapy, hypothalamus, biofeedback, behavioral medicine.

The newsletter of the American Psychology Association (APA) has ranked Miller among the ten most eminent psychologists of the twentieth century. Highly influential as a learning theorist, neuroscientist, science statesman, educator, and, above all, a consummate experimentalist, Miller authored eight books and more than 276 articles.

The great variety of areas in which Miller made important conceptual and research contributions mainly concerned reward and motivation mechanisms: 1) underlying thought processes and behaviors relevant to problem solving in psychotherapy, 2) as mediated by the nervous system, and 3) involved in learning control over voluntary (conscious) skeletal-muscle and autonomic (normally unconscious) internal-organ response systems for minimizing stress, treating disease, and promoting health. Furthermore, his career shows psychology’s evolution from a largely theory-driven but relatively data-impoverished experimental discipline in the 1930s to one that in the early twenty-first century integrates a vast body of clinical, social, and physiological knowledge.

Promising Early Years . Miller was the only child of Irving E. and Lily Rose Miller. His father obtained his PhD at the University of Chicago, where he studied with John Dewey and James Rowland Angell (later president of Yale University), and after several teaching positions elsewhere became professor of educational psychology at what later became Western Washington State College in Bellingham. His son Neal was a gifted child and avid reader. He gave his father credit for surreptitiously guiding his education by bringing home books and articles and leaving them lying casually about. After Miller read them, he noticed they disappeared to be replaced by others.

He earned his bachelor’s degree in 1931 at the University of Washington, where he had a major learning theorist, Edwin R. Guthrie, as a teacher. Only in his senior year did he decide on psychology as a career because one could be near its frontier and get into research without first having to go through so much specialized work and also because it combined his interests in writing, people, and science in general. He worked with the famed IQ researcher Louis M. Terman at Stanford University, and obtained his master’s degree there in 1932. But beforehand, he took advanced experimental psychology with Walter R. Miles and, because of his proficiency in physics and chemistry, helped Miles to put all the apparatuses in the Stanford Laboratory in good working order. Therefore, when Miles was invited to join the psychology faculty at Yale, he took Miller with him for doctoral work.

Laws of Learning and Foresight: Seeking a Unified Theory . At Yale, Miller came under the influence of Clark L. Hull, the most prominent learning theorist of his time. Hull’s program aimed at showing how the principles of classical conditioning, discovered in Pavlov’s lab, could be applied to understanding Edward Thorndike’s trial-and-error learning, human verbal learning, and higher mental processes such as purposeful, goal-oriented, and foresightful behavior, now described as cognitive. For example, in 1935, in his PhD dissertation Miller demonstrated that, by using the letter T as the critical cue (stimulus) predicting a shock and the number 4 as the neutral one predicting no shock, a learned change in the electrical conductivity of a person’s skin—now elicited by the critical cue but not by the neutral one—could be transferred from the cues’ overt presentations to the person’s merely thinking about them. From this emerged the idea that the mental acts of thinking, remembering, and imagining are themselves responses that can then function as cues (response-produced cues) to which other responses can be made and are subject to the same laws of learning as are external responses and cues. These mental responses, unconstrained by the real-time sequencing of cues in the physical world, permit the playback of events in reverse of their actual happenings. Thereby, one can work backward in one’s mind’s eye from a hoped-for goal along a route that better illuminates how to reach it than groping blindly forward from the start. It is such use of response-produced cues that Miller identified as the basis for much foresightful behavior in problem solving.

Extending the Quest to Freudian and Social Phenomena . As a result of an insight Miller had about the similarity between Freud’s conception of repression and Pavlov’s conception of inhibition, he resolved to extend Hull’s program to an examination of Freudian theory and practice in terms of the laws of learning. Accordingly, he obtained a postdoctoral fellowship to study in Vienna at Freud’s Psychoanalytic Institute, where he underwent a didactic analysis with Heinz Hartmann. He long regretted he had turned down at least one analytic session with the great Freud himself because an hourly fee of $20 seemed more than he could afford.

In 1936 Miller returned to Yale as instructor in psychology and research assistant psychologist in the multi-disciplinary Institute of Human Relations. Observing a dominant male monkey self-mutilate when prohibited from attacking a competitor given his harem, Miller thought of the Freudian concept of aggression-turned-inward. The result, in 1939, was his book, Frustration and Aggression, coauthored, among others, with the sociologist John Dollard. Its major hypothesis, as later reformulated in the journal Psychological Review, claimed that when a segment of society is frustrated from attaining its goals it tends—depending on what avenues are open or closed for expressing that frustration—to relieve it as angry persecutions against an innocent, less powerful segment. Examples included Miller’s European encounters with anti-Semitism inflamed by the still-lingering economic privations from reparations required of Germany after World War I.

Hull originally hypothesized that a response is reinforced (strengthened) if immediately followed by a reduction in a need. However, the assumption was that all needs drive (motivate) responses to reduce them. But some needs cannot do so because they cannot be detected (for example, the need to escape carbon monoxide). Thus, in Frustration and Aggression, Hull’s hypothesis was restated by Hobart Mowrer and Miller as the drive-reduction hypothesis, which now dealt only with detected needs, henceforth, defined as drives. In its strong form, the drive-reduction hypothesis asserts that the only events that can act as reinforcers of a response are those that immediately follow it and are themselves soon followed by a reduction in the drive motivating it or are highly associated with that drive’s later reduction. Assessing this strong form’s validity was the concern of much of Miller’s later research.

While testing in rats the Freudian idea of reaction formation, Miller noted the following: Hungry rats trained to go down a short alley for food and then given an electric shock at the goal tended on subsequent trials to approach part way and then stop, stopping further away the stronger the shock had been. From such observations emerged Miller’s theoretical-experimental analyses of

approach-avoidance conflict behavior. Published as “Experimental Studies of Conflict Behavior,” in 1944, these stated that if a goal is something an organism both wants and fears, there is an approach tendency to it, called an approach gradient, that grows stronger the nearer one gets to it but there is also an avoidance gradient that does the same. However, the avoidance gradient increases more rapidly with nearness than does the approach. The gradients often will cross each other—at that intersection the organism’s approach will stop. But just how close to the goal that will be depends on the relative strengths of the approach gradient versus the avoidance one.

In 1939 Mowrer hypothesized that any fear—or anxiety, as it is called when its source is vague or unconscious—that is induced by a noxious situation is then acquired by the cues associated with that situation. Thus, the fear subsequently can be induced by those cues alone and motivate responding to them. Most importantly, a reduction of this acquired fear by escaping or avoiding those cues will reinforce any specific response doing so. Miller rigorously tested this hypothesis of fear as an acquirable—that is, learnable—drive by first shocking rats in the white side of a two-compartment box until they learned to run rapidly through a door into the black side to escape shock. Afterward, when put in the white side with the door closed but without shock administered, they defecated and showed other signs of having learned to fear the cues there to which shock had become associated. If, by trial and error, they then rotated a small wheel, which opened the door and allowed them to escape the white side’s cues and, thus, the fear itself, they quickly learned to rotate the wheel for escape on subsequent non-shock trials. This confirmed Mowrer’s hypothesis.

Principles of Learning in the Acquisition of Social Behavior . In 1941, the year Miller became an associate professor, he and Dollard wrote another book, Social Learning and Imitation, to show how a wide range of human behavior can be understood by knowing a few important principles of learning plus the social-condition contexts in which the learning takes place. In the book they listed four fundamentals necessary for instrumental learning, that is, for the remembering of which behaviors—guided by what signposts—have proven, for future reference, to be instruments of success in achieving one’s goals:

  1. DRIVE (or Motivation); a person must want something. A drive may be innate, as with hunger, or it may be learned, as with fear or the desire for money.
  2. CUE (or Stimulus); a person must notice something. A cue may be response-produced, as with the thought of the letter T in the experiment already described.
  3. RESPONSE; a person must do something. A response may be an overt act or a central nervous system event, such as a thought, a perception, or paying attention.
  4. REWARD (or Reinforcement); a person must get something that is wanted. A reward following a response to a cue strengthens the tendency for the cue subsequently to elicit the response. A reward may be a learned one, such as getting money. Pain is a stimulus that elicits fear, and a reduction in pain or fear strengthens (rewards) any response immediately followed by that reduction.

An Interruption: Contributions in the War Years . In World War II, Miller served in the U.S. Army Air Corp (1942–1946). As a captain, he helped develop tests to select cadets likely to succeed in pilot training. He also initiated a study of factors contributing to fear and courage in combat. After being promoted to major, he helped identify behavioral and perceptual areas where improvements could be made in pilot training and in hitting targets via fixed gunnery. John C. Flanagan, one of his Army Air Corps colleagues, years later commented that Miller also instituted a pilot flight-check evaluation list “which provided the basis for today’s procedures, making my flights on commercial airlines much more pleasant and giving me more confidence that the airline pilots will do the right things (Flanagan to Miller, 3 September 1980, unpublished letter).

Retrospectively, in 1987, Miller wrote that his program’s success resulted from three factors: (1) finding something that the Air Force needed and that psychologists could deliver—initially, selection of personnel, (2) delivering it, and (3) then providing data proving it had been delivered. These wisdoms he adapted to achieving success in the many other missions for which he later served as a statesman in psychology and other behavioral sciences.

Learned Basis of Freudian Phenomena Revisited . When the war ended in 1946, Miller returned to Yale, where he attained tenure in 1947. He married Marion Edwards—a social worker there—in 1948. He was awarded full professorship in 1950 and appointed the first recipient of the James Rowland Angell Chair of Psychology in 1952. Again he collaborated with Dollard, taking psychoanalysis as a point of departure for analyzing psychotherapy as learning. As part of that effort Miller published an article, Theory and Experiment Relating Psychoanalytic Displacement to Stimulus-Response Generalization, in which, harkening back to his “Experimental Studies of Conflict Behavior,” he posited the following: When the approach to a stimulus is inhibited by conflict with an avoidance of that same stimulus, responses tend to displace to other stimuli that are still similar enough to motivate the prospect of a successful approach but are dissimilar enough to minimize the interfering avoidance. For example, given the Freudian Oedipal conflicts between a young son’s erotic love of his mother and retaliatory fear of his father, one can understand the displacement implied in the old vaudeville song, “I want a girl just like the girl [but not the same one, God forbid] that married dear old dad.” This and other predictions were borne out by three other studies on displacement, all in 1952.

In 1950 appeared another Miller book with Dollard, Personality and Psychotherapy: An Analysis in Terms of Learning, Thinking, and Culture. It was immensely influential in training the first post–World War II generation of clinical psychologists in the treatment of the neuroses and was, for years, widely used as a text in learning theory. It paid special attention to how in therapy the appropriate use of response-produced cues, particularly verbal ones, can facilitate generalizations between likenesses that should be perceived in one’s life but maladaptively are not and distinctions between differences that, likewise, should be perceived but, again, are not.

While writing Personality and Psychotherapy, Dollard and Miller submitted a proposal, encouraged by the Ford Foundation, to study coping behavior in normal people. But the foundation responded that, because of a policy change, it would take more than a year to decide whether to fund studies in that area. This unendurable delay forced the two researchers apart to formulate separate projects to support themselves. From the National Institute of Mental Health, Dollard found funding for the analysis of

psychotherapeutic interviews. From the same source, Miller found funding for studies of the mechanisms of reinforcement.

Into the Gut and the Brain . Around 1950–1952, Miller started turning to physiological interventions because they offered unique opportunities to test the strong form of the drive-reduction hypothesis of reinforcement against competing possibilities. For example, instead of the reinforcing value of food for a hungry animal residing in the food’s ability to reduce hunger, might it instead be either the pleasures of taste or of the swallowing of the food that is reinforcing? But, if one could reward behavior by reducing hunger while bypassing both taste and swallowing, that would clearly support the drive-reduction hypothesis. Indeed, delivery of food to a hungry rat via a tube directly into its stomach rewarded the learning of correct choices in a T-maze for that delivery. This supported the drive-reduction hypothesis of reward but did not discount that taste and swallowing could also be rewarding.

A preliminary step to yet another plan for testing the drive-reduction hypothesis was to lesion (destroy) the ventromedial nucleus in the hypothalamus of a rat’s brain, which then causes overeating and obesity. If this overeating had all the aspects of normal hunger, Miller could then proceed to the test proper. However, contrary to hunger motivation, these lesioned rats, while eating a larger amount of highly palatable foods than normal rats, worked less hard for food and were less tolerant of less palatable foods. This result spoiled his plan but taught him the importance of taking a variety of measures before inferring the nature of an underlying state—a lesson which he in 1961 strongly communicated in print as a cautionary tale for psychologists working in the brain.

The salience of the brain approach was heightened by two dramatic findings in the mid-1950s. One was the discovery by James Olds and Peter Milner of sites in the lateral hypothalamus that rats find rewarding to self-stimulate with volleys of brief electrical pulses by pressing a lever. The other was a reverse discovery by Jose Delgado, Warren Roberts, and Miller of sites where electrical stimulation would motivate cats to learn a response to escape or avoid the stimulation. But it was puzzling that at some sites cats would learn a response to terminate stimulation but not a response to avoid it—an observation leading to the discovery of the reward-escape effect in which Gordon Bower was involved. Implanted rats showing this effect cycled repeatedly between pressing a lever to turn on the stimulation and rotating a wheel to turn it off. In 1957 Miller seized this opportunity for an unusual test of some drugs. He showed that methamphetamine enhanced and chlorpromazine reduced the rewarding aspects of the cycle while leaving the punishing aspects unaffected. This was a first evidence of what later was recognized as the involvement of the neurotransmitter dopamine in promoting reward. Miller presented these data and others with Herbert Barry to drug companies to advertise the potential benefits of behaviorally evaluating pharmacological agents. With his encouragement, this approach was to become the field of behavioral psychopharmacology.

Another advantage of implanting electrodes in the lateral hypothalamus was to search there in rats for where W. R. Hess in cats had found that stimulation could induce them to eat. But as Miller much later reported in his autobiographical article, “Behavior to the Brain to Health,” the search took two years before yielding success. Edgar Coons discovered the site where rats, even thoroughly satiated, would eat ravenously while the current was on but stop immediately when it was turned off. Behavioral tests confirmed that the electrically elicited eating had all the earmarks of normally motivated hunger. Then why, contrary to the drive-reduction hypothesis, would these animals not press a lever to turn the hunger off but would press to turn it on? Miller noted in a Federation Proceedings article that Coons had found that amphetamine raises the threshold required to elicit feeding and lowers that required to sustain self-stimulation, showing that a single system does not subserve both self-stimulation and feeding. Also Coons later found that, at the lowest current required to elicit eating, the rat would NOT press a lever for it unless food was available to eat while the current was on—just as the drive-reduction hypothesis would predict.

The phenomenon of stimulation-elicited hunger lent itself to yet another question. Was hunger a learnable drive like fear? If so, would not a satiated animal then eat in the presence of a cue to which the experience of being hungry had in the past been closely associated? Other attempts with Arlo K. Myers to demonstrate such learning had proved a failure, but that could be because the slowness of onsets and offsets of normal hunger attenuated its association with the cue too much. The rapidity, however, with which stimulation-elicited hunger could be turned on whenever a designated cue was presented made for a much better test. However, as Miller reported in 1964, unstimulated eating in satiated animals was never observed to occur in the presence of the cue, even after numerous trials of pairing stimulation-elicited eating with that cue.

Major Recognitions and High Public Service . By 1962 his growing reputation and impact as a scientist was such that Miller was asked to chair a panel that made a published report to the President’s Science Advisory Committee on the needs of the behavioral sciences and how to meet them. It was a report over which he said he “sweated blood.” In further recognition of this impact, Miller was extended in 1964 the country’s highest scientific award, the National Medal of Science, handed to him in person by President Johnson. The citation accompanying it reads, “For sustained and imaginative research on principles of learning and motivation and illuminating behavioral analysis of the effects of direct electrical stimulation of the brain.”

At the time Miller moved to Rockefeller University in 1966 a great urgency was being felt by the many disparate disciplines conducting research on the nervous system that some kind of coordinating network be set up, as detailed in a 1983 oral history interview with Miller sponsored by the Brain Research Institute of UCLA. By 1969, while Miller was chair of the Committee on Brain Science within the National Research Council, this urgency had become a clear mandate. “Like a crystal dropped into a supersaturated solution,” as Miller characterized it, a “committee motion” proceeded quickly to the action of forming the Society for Neuroscience, with a membership of fifty thousand in the early twenty-first century. Miller, as always, exercised his uncanny ability to see to the heart of what needed to be done in organizations as well as in research. Unasked, he took the initiative to secure from the Sloan Foundation a grant of $20,000 to help cover the start-up costs of forming the society: legal fees of incorporation, membership recruiting, and public relations for educating legislators and the public as to the missions of the society. As a result, Miller is considered a key founding member. At the Society for Neuroscience’s first meeting in 1970, he was voted president elect. It constituted for him one of the pinnacles of his career, along with being elected to the National Academy of Science in 1958, holding the office of APA president in 1960–1961, and receiving the National Medal of Science.

Learnable Voluntary Control of Autonomic Functions? . In 1957 W. H. Gantt’s translation of K. M. Bykov’s book, The Cerebral Cortex and the Internal Organs, reported that autonomic responses in a wide variety of internal organ (visceral) systems, when elicited by their innately triggering stimuli, can then become elicitable by other stimuli that routinely closely precede and, thus, strongly predict these triggers. The well-known Pavlovian prototype for this classical conditioning is the learning of a dog to anticipatively salivate to the sound of a tone that he has come to associate as being followed immediately by meat powder to which he automatically salivates. The book’s publication stimulated Miller to follow up on a long-standing hunch first recorded in print in 1951. He had entertained the possibility, against popular opinion, that autonomic responses in visceral motor systems are not limited to becoming learned reactions to stimuli but, if properly rewarded, can be trained—like ordinary “voluntary” responses—to become “intended” behaviors to obtain those rewards. For example, if one could learn how to consciously control internal body processes, the medical benefits would be enormous, and success would fulfill Hull’s and his overarching hope to show an underlying relatedness of all laws of learning, spanning across voluntary, cognitive, and—now—autonomic domains of behavior.

Indeed, Alfredo Carmona with Miller showed in 1967 that thirsty dogs were able to increase or to decrease their autonomic response of salivation in order to obtain water rewards but, puzzlingly, displayed different postures during increases compared to decreases. Maybe just the postures were learned but somehow triggered the salivations. To rule this out, rats were treated with curare, which completely paralyzes the voluntary muscle system but leaves the autonomically controlled visceral muscle system unaffected. Then, the autonomic response of increasing (or, alternately, decreasing) the rats’ heart rates was designated the specific response basis for their obtaining very rewarding brain stimulation.

Just as predicted, and reported in two companion articles to the Carmona study, the autonomic changes in heart-rate responding required for rewards did seem dramatically to occur. Over many studies from 1965 to 1972, even the general public, vis-à-vis The New Yorker and other media, became aware of the medical benefits this promised. But then Miller and Barry Dworkin in his laboratory began finding that these results mysteriously diminished until they could no longer be replicated even after repeated and varied attempts. When finally convinced of failure, Miller, though heartbroken, courageously took great pains to publicize it widely.

Despite disappointments in this line of research, it led to advances in technology to measure otherwise impossible-to-detect subtle changes in heart rate and other physiological responses. After finding that paralyzed rats failed to learn autonomic control, Miller shifted this technology to seeing whether people who had been paralyzed by gunshot wounds that severed their spinal cords could gain that control. They differed from the rats in being better candidates in terms of Miller and Dollard’s four fundamentals necessary for effective instrumental learning: These patients had a high drive to try gaining control because their blood pressure was so low that whenever they sat or stood up they fainted. Unlike the rats, they were shown their own amplified heart rate and blood pressure readings, thus providing them biofeedback informational cues about their own performance, as Miller was invited to report in 1973. To this biofeedback information, the response they initially reported using— to try to change their readings—was to think emotional, often sexy, thoughts to which the desired blood pressure changes are normally reflexly connected. As these paralyzed patients became successful, they were gradually able to command these changes “directly.” Whenever there was a desired response, even if too small an increment initially to be clinically relevant, the mere detectable fact of it was a reward, given the paralytics’ high achievement motivation. But the summing of such increments mounted to clinically significant levels, and as they did, the rewards became enormous because not only could the paralytics now sit up without fainting but, as a result, they could now also attend plays and ball games. Nevertheless, how the patients achieved voluntary control of blood pressure via the autonomic nervous system was surprising; with a severed spinal cord the usual route of elevating blood pressure via the sympathetic component of the autonomic nervous system is also cut off. Parasympathetic (vagal) or blood-borne humoral factors are suspected or perhaps some subtle respiratory mediation still surviving paralysis, as reported in 1976 regarding vasomotor responses.

This and other studies using biofeedback also convincingly suggested the ability to bring autonomic responses under voluntary control, whether directly or indirectly. And as Miller loved to point out, toilet training, particularly the learning of control over the autonomic bladder sphincters, is a well-known—and rewarded—universal fact of life. Certainly, by 1985 the application of biofeedback methodology promoted by Miller and his associates had proved highly beneficial medically in treating a wide variety of problems, such as idiopathic scoliosis, enuresis, and migraine, problems involving both voluntary and autonomic response systems.

From the mid-1970s until a few years before his death in 2002, Miller’s inquiries into biofeedback and learned behavior took on a new emphasis, that of their use to maintain homeostasis and minimize stress. This emphasis contributed substantially to the establishment not only of biofeedback as a discipline but also the fields of behavioral medicine and health psychology, all of which consider Miller a founding father.

Among Miller’s last research contributions, 1993–1994, were his collaborations with Dr. Patricia Cowings and with Dr. Edward Taub. In particular, he took great pleasure in working with Taub in the development of constraint-induced movement therapy, a very effective treatment to rehabilitate stroke victims with motor impairment by overcoming their learned nonuse, which in turn also promotes neuroplastic changes in the brain that further enhance motor recovery. As a result, a fitting research epitaph to his entire career, devoted to understanding to what the laws of learning apply, can be the following statement: “The brain controls learned behavior but in turn learned behavior also controls the brain—a biofeedback cooperation.”

Research Mentor and Educator . There were a number of research ventures in Miller’s lab that do not fit nicely into the trajectory of his life characterizing this portraiture but which he strongly encouraged, supported, and included in his Selected Papers. For recommended readings there are his collaborations, among others, with: (1) David Egger on findings premonitory of the Rescorla-Wagner model, which now dominates studies in learning; (2) David Quartermain concerning memory consolidation as growing out of a study by Coons and Miller; and (3) studies devoted to understanding the signals for thirst with Don Novin, salt appetite with George Wolf and Edward Stricker, and hunger with Jack Davis, Eleanor Adair, Stan Tenen, David Booth, and Sarah Leibowitz. Among others whose research in his lab he fostered, were Sebastian P. Grossman for chemical coding of behavior discoveries published in 1960, E. E. Krieckhaus and George Wolf on latent learning in 1968, and Jay M. Weiss and Bruce S. McEwen for studies on stress, as cited in 1976.

Miller was considered a master of research design and communicator of the conceptual basis of scientific inquiry—so much so that in some respects that may be his greatest achievement, namely, helping psychology grow into a mature science. He always sought for parsimonious explanations of cause and effect but required that the hypotheses involved be rigorously defined in empirically testable ways that allow them to be confirmed or discon-firmed, ideally by a variety of measures. In his 1960 Federation Proceedings article he cautioned against “stopping as soon as a hypothesis is confirmed by a single test. [Especially in a] new field of investigation [such as the brain was in the 1950–1960s], it is essential to design careful behavioral tests of all conceivable alternatives.” In an interview taped in 2000, he said an experiment should be designed not only to discover something but also to communicate it and that it would be best to design experiments so that the results would be rather obvious and would not demand elaborate data analysis. He warned investigators, while running experiments, to keep an eye out for unexpected findings because sometimes these are more important than the findings sought for. “And regarding things in an experiment that give you a lot of difficulty, it may be something fairly important or it wouldn't be an important difficulty. So, perhaps you may want to change your goal and decide that the difficulty is a more important variable to study than what you originally started out with.” Indeed, as consulting editor of the Journal of Experimental Psychology for seven years, Miller passed on his research wisdoms widely. When he resigned, its chief editor, Arthur W. Melton, wrote: “It will be difficult, if not impossible, to replace you with another so keen at picking the flaws in logic or design” (Melton to Miller, 29 September 1956, unpublished letter).

During his professor emeritus years at Rockefeller, which began in 1980, he became quite alarmed by the dangers posed by the animal rights movement to research on treating illness and promoting health. Then, and after his return to Yale in 1985 as a research affiliate, he conducted vigorous efforts to educate the scientific and lay communities about the benefits of behavioral research on animals. He was an important figure in mobilizing the opposition of these communities to this threat.

Finally, as summarized in his last publication, How to Prepare for Our Future of Totally Unexpected Opportunities, he conducted vigorous efforts to communicate to the scientific and lay communities a basic understanding of the scientific method and the enormous benefits it has yielded and will continue to do, if well fostered. He served, thus, as an exemplary model for the more than 150 students he trained in research, many of whom became distinguished researchers themselves. In the year 2000 this was the theme of the final of his countless honors, the Award in Neuroscience Education, bestowed by the Association of Neuroscience Departments and Programs.

Miller was survived by a son, York, and daughter, Sarah Rose Mauch, children of his first wife, Marion Edwards Miller, who died 13 October 1997. He was also survived by his second wife, Jean Shepler Miller, whom he married on 21 July 1998 and who was a friend of the family and former music teacher to his children. To these partners he gave enormous credit and thanks: the first for protecting him from so much flotsam and jetsam of daily life—such as paying bills—so that he could concentrate on his science, and the second for providing him so much comfort and tolerance in his few remaining years.


Materials are held in the Neal E. Miller Papers collection at Manuscripts and Archives, Yale University Library.


“The Influence of Past Experience upon the Transfer of Subsequent Training.” PhD diss., Yale University, 1935.

With John Dollard, Leonard W. Doob, et al. Frustration and Aggression. New Haven, CT: Yale University Press, 1939.

With R. R. Sears, O. H. Mowrer, L. W. Doob, and J. Dollard. “I. The Frustration-Aggression Hypothesis.” Psychological Review 48 (1941): 337–342.

With John Dollard. Social Learning and Imitation. New Haven, CT: Yale University Press, 1941.

“Experimental Studies of Conflict Behavior.” In Personality and Behavior Disorders: A Handbook Based on Experimental and Clinical Research, edited by Joseph McVicker Hunt, 431–465. New York: Ronald, 1944.

Staff, Psychological Research Project (Pilot) [Miller, N. E. (Ed.)]. “Psychological Research on Pilot Training in the AAF.” American Psychologist 1 (1946): 7–16.

“Studies of Fear as an Acquirable Drive: I. Fear as Motivation and Fear-Reduction as Reinforcement in the Learning of New Responses.” Journal of Experimental Psychology 38 (1948): 89–101.

“Theory and Experiment Relating Psychoanalytic Displacement to Stimulus-Response Generalization.” Journal of Abnormal and Social Psychology 43 (1948): 155–178.

With John Dillard. Personality and Psychotherapy: An Analysis in Terms of Learning, Thinking, and Culture. New York: McGraw-Hill, 1950.

With Clark J. Bailey and James A. F. Stevenson. “Decreased ‘Hunger’ but Increased Food Intake Resulting from Hypothalamic Lesions.” Science 112 (1950): 256–259.

“Comments on Multiple-Process Conceptions of Learning.” Psychological Review 58 (1951): 375–381.

With D. Kraeling. “Displacement: Greater Generalization of Approach than Avoidance in a Generalized Approach-Avoidance Conflict.” Journal of Experimental Psychology 43 (1952): 217–221.

With E. J. Murray. “Displacement: Steeper Gradient of Generalization of Avoidance than of Approach with Age of Habit Controlled.” Journal of Experimental Psychology 43 (1952): 222–226.

With E. J. Murray. “Displacement and Conflict: Learnable Drive as a Basis for the Steeper Gradient of Avoidance Than of Approach.” Journal of Experimental Psychology 43 (1952): 227–231.

With M. L. Kessen. “Reward Effects of Food via Stomach Fistula Compared with Those of Food Via Mouth.” Journal of Comparative and Physiological Psychology 45 (1952): 555–564.

With A. K. Myers. “Failure to Find a Learned Drive Based on Hunger; Evidence for Learning Motivated by ‘Exploration.’” Journal of Comparative and Physiological Psychology 47 (1954): 428–436.

With J. M. R. Delgado and W. W. Roberts. “Learning Motivated by Electrical Stimulation of the Brain.” American Journal of Physiology 179 (1954): 587–593.

“Experiments on Motivation; Studies Combining Psychological, Physiological, and Pharmacological Techniques.” Science 126 (1957): 1271–1278.

“Objective Techniques for Studying Motivational Effects of Drugs on Animals.” In Psychotropic Drugs, edited by Silvio Garattini and Vittorio Ghetti. Amsterdam: Elsevier, 1957.

With G. H. Bower. “Rewarding and Punishing Effects from Stimulating the Same Place in the Rat’s Brain.” Journal of Comparative and Physiological Psychology 51 (1958): 669–674.

“Motivational Effects of Brain Stimulation and Drugs.” Federation Proceedings 19 (1960): 846–854.See pp. 850–851.

With H. Barry III. “Motivational Effects of Drugs: Methods which Illustrate Some General Problems in Psychopharmacology.” Psychopharmacologia 1 (1960): 169–199.

“Learning and Performance Motivated by Direct Stimulation of the Brain.” In Electrical Stimulation of the Brain, edited by Daniel E. Sheer. Austin: University of Texas Press, 1961. Contains advice as to the importance of taking a variety of measures before inferring the nature of an underlying state, particularly in brain research.

“Strengthening the Behavioral Sciences.” Science 136 (1962): 233–241. Miller chaired the Behavioral Science subpanel of the President’s Science Advisory Committee, which produced this report.

“Some Psychophysiological Studies of Motivation and of the Behavioural Effects of Illness.” Bulletin of the British Psychological Society 17 (1964): 1–20.

With L. DiCara. “Instrumental Learning of Heart-Rate Changes in Curarized Rats: Shaping, and Specificity to Discriminative Stimulus.” Journal of Comparative and Physiological Psychology 63 (1967): 12–19.

With A. Carmona. “Modification of a Visceral Response, Salivation in Thirsty Dogs, by Instrumental Training with Water Reward.” Journal of Comparative and Physiological Psychology 63 (1967): 1–6.

Neal E. Miller: Selected Papers. Chicago: Aldine-Atherton, 1971.

“Biofeedback: Evaluation of a New Technique” [Invited editorial]. New England Journal of Medicine 290 (1974): 684–685.

With B. R. Dworkin. “Visceral Learning: Recent Difficulties with Curarized Rats and Significant Problems for Human Research.” In Cardiovascular Psychophysiology: Current Issues in Response Mechanisms, Biofeedback, and Methodology, edited by Paul A. Obrist, A. H. Black, Jasper Brener, et al. Chicago: Aldine, 1974.

“Behavioral Medicine as a New Frontier: Opportunities and Dangers.” In Proceedings of the National Heart and Lung Institute Working Conference on Health Behavior, Basye, Virginia, May 12–15, 1975, edited by Stephen M. Weiss. Bethesda, MD: U.S. Dept. of Health, Education, and Welfare, Public Health Service, National Institutes of Health, 1975.

Evans, Richard I. The Making of Psychology: Discussions with Creative Contributors. New York: Knopf, 1976. See chapter 14, pp. 169–183.

“Behavioral Medicine: Symbiosis between Laboratory and Clinic.” Annual Review of Psychology 34 (1983): 1–31.

“Learning, Stress, and Psychosomatic Symptoms” [memorial paper in honor of Jerzy Konorski]. Acta Neurobiological Experimentalis 36 (1976): 141–156. Contains citations concerning Jay Weiss’s and Bruce McEwen’s research on stress.

With Wesley C. Lynch, Haruyo Hama, et al. “Instrumental Control of Peripheral Vasomotor Responses in Children.” Psychophysiology 13 (1976): 219–221.

With T. G. Pickering, B. Brucker, et al. “Mechanisms of Learned Voluntary Control of Blood Pressure in Patients with Generalized Bodily Paralysis.” In Biofeedback and Behavior, edited by Jackson Beatty and Heiner Legewie. New York: Plenum, 1977.

With B. S. Brucker. “A Learned Visceral Response Apparently Independent of Skeletal Ones in Patients Paralyzed by Spinal Lesions.” In Biofeedback and Self-Regulation, edited by Niels Birbaumer and H. D. Kimmel. Hillside, NJ: Erlbaum, 1979.

With B. R. Dworkin. “Different Ways in which Learning Is Involved in Homeostasis.” In Neural Mechanisms of Goal-Directed Behavior and Learning, edited by Richard F. Thompson, Leslie H. Hicks, and V. B. Shvyrkov. New York: Academic Press, 1980.

“Some Main Themes and Highlights of the Conference.” Health Psychology Suppl. 2 (1983): 11–14. Supplement to the Proceedings of the National Working Conference on Education and Training in Health Psychology, May 23–27, 1983, Arden House, Harriman, New York; Sponsored by the Division of Health Psychology of the American Psychological Association, edited by George D. Stone. Hillsdale, NJ: Erlbaum, 1983.

Brain Research Institute. “Neal Elgar Miller, PhD—Interviewed 8 November 1983.” Brain Research Institute, Neuroscience History Archive, Oral History Project Series CON Code MIL.

With D. Caroline Coile. “How Radical Animal Activists Try to Mislead Humane People.” American Psychologist 39 (1984): 700–701.

“The Value of Behavioral Research on Animals.” American Psychologist 40 (1985): 423–440.

With B. Dworkin, S. Dworkin, et al. “Behavioral Method for the Treatment of Idiopathic Scoliosis.” Proceedings of the National Academy of Sciences of the United States of America 82 (1985): 2493–2497.

With B. R. Dworkin. “Failure to Replicate Visceral Learning in the Acute Curarized Rat Preparation.” Behavioral Neuroscience 100 (1986): 299–314.

“Education for a Lifetime of Learning.” In Health Psychology: A Discipline and a Profession, edited by George C. Stone, S. M. Weiss, J. D. Matarazzo, et al. Chicago: University of Chicago Press, 1987. Contains reflections on important statesmanly lessons learned by Miller in the course of doing research on pilot training in the Army Air Force in World War II. Note p. 11.

With G. C. Stone, S. M. Weiss, et al., eds. “Health Psychology in the Twenty-First Century.” In Health Psychology: A Discipline and a Profession, edited by George. C. Stone, S. M. Weiss, J. D. Matarazzo, et al. Chicago: University of Chicago Press, 1987.

“Behavior to the Brain to Health.” In The Neurosciences: Paths of Discovery II, edited by Fred Samson and George Adelman. Boston, MA: Birkhauser, 1992.

With Edward Taub, T. A. Novack, et al. “Technique to Improve Chronic Motor Deficit after Stroke.” Archives of Physical Medicine and Rehabilitation 74 (1993): 347–354.

With Edward Taub, J. E. Crago, et al. “An Operant Approach to Rehabilitation Medicine: Overcoming Learned Nonuse by Shaping.” Journal of Experimental Analysis of Behavior 61 no. 2 (1994): 281–293.

With Patricia S. Cowings, W. B. Toscano, et al. Autogenic-Feedback Training as a Treatment for Airsickness in High Performance Military Aircraft: Two Case Studies. NASA Technical Memorandum 108810. Moffett Field, CA: National Aeronautics and Space Administration, Ames Research Center, 1994.

“How to Prepare for Our Future of Totally Unexpected Opportunities.” In Mind and Brain Sciences in the 21st Century, edited by Robert L. Solso. Cambridge, MA: MIT Press, 1997.

“Interview with Dr. Neal Miller.” Yale University School of Medicine: MedMedia Services. Taped interview for showing at the annual meeting of the Association of Neuroscience Departments and Programs held in 2000 in New Orleans.


Bykov, Konstantin M. The Cerebral Cortex and the Internal Organs. Translated by W. H. Gantt. New York: Chemical Publishing, 1959.

Coons, Edgar E. “Motivational Correlates of Eating Elicited by Electrical Stimulation in the Hypothalamic Feeding Area.” PhD diss., Yale University, 1964.

_____, and J. A. F. Cruce. “Lateral Hypothalamus: Food Current Intensity in Maintaining Self-Stimulation of Hunger.” Science 159 (1968): 1117–1119.

_____. “Obituaries—Neal Elgar Miller (1909–2002).” American Psychologist 57 (2002): 784–786.

Grossman, S. P. “Eating or Drinking Elicited by Direct Adrenergic or Cholinergic Stimulation of Hypothalamus.” Science 132 (1960): 301–302.

Hess, Walter Rudolf. Das Zwischenhirn: Syndrome, Lokalisationen, Funktionen. 2nd ed. Basel: Schwabe, 1954.

Jonas, Gerald. “Visceral Learning.” New Yorker, 19 August 1973 and 26 August 1973.

Krieckhaus, E. E., and G. Wolf. “Acquisition of Sodium by Rats: Interaction of Innate Mechanisms and Latent Learning.” Journal of Comparative and Physiological Psychology 65 (1968): 197–201.

McEwen, B. S., C. J. Denef, J. L. Gerlach, et al. “Chemical Studies of the Brain as a Steroid Hormone Target Tissue.” In The Neurosciences: Third Study Program, edited by Francis Otto Schmitt and Frederic G. Worden. Cambridge, MA: MIT Press, 1974.

Mowrer, O. H. “A Stimulus-Response Analysis of Anxiety and Its Role as a Reinforcing Agent.” Psychological Reviews 46 (1939): 553–566.

Olds, James, and Peter Milner. “Positive Reinforcement Produced by Electrical Stimulation of Septal Area and Other Regions of Rat Brain.” Journal of Comparative and Physiological Psychology 47 (1954): 419–427.

Trowill, J. A. “Instrumental Conditioning of the Heart Rate in the Curarized Rat.” Journal of Comparative and Physiological Psychology 63 (1967): 7–11.

Weiss, J. M. “Psychological Factors in Stress and Disease.” Scientific American 226, no. 6 (1972): 104–113.

Edgar E. Coons