Placebos and Psychsomatic Disease Part 1

An interview with Dr. Bob Ader


Many people, including physicians, may not know who Robert Ader is, but almost everyone is familiar with psychoneuroimmunology, a tongue twisting term he coined over three decades ago to describe his ability to condition the immune system, just as Ivan Pavlov had done with the gut. Pavlov’s classical conditioning study was based on his observation that if a dog saw a piece of meat, it would sniff it and immediately start to salivate. If someone rang a bell, the animal would simply turn around to see where the sound was coming from. However, if he repeatedly rang the bell first and immediately followed this by giving the dog some meat to eat, after repeating this several times, simply ringing the bell was sufficient to promote salivation. And since this sound was a signal that the meat would soon be coming, the dog’s body reacted as if it were already there, with an increase in gastrointestinal secretions and motility.


Similar conditioned responses were subsequently demonstrated in other animals as well as humans, and Ader, a psychologist, wondered how  long this conditioning effect would last. He injected rats with Cytoxan, which causes nausea, and simultaneously fed them water containing saccharin. The association of nausea with saccharin’s sugary taste resulted in subsequent avoidance of the sweetened water, a conditioned response. But the rats had to drink it when this aversion was overcome by severe thirst. Some avoided drinking longer than others, and a few died, not from dehydration, but infections.

In point of fact, deaths seemed to occur in those animals that drank the most saccharin-laced water on the single conditioning trial. Cytoxan was used in such taste aversion experiments because it predictably made the  animals feel sick. Cytoxan is used to treat certain cancers since it suppresses specific immune system components, which improves results. Unfortunately, like other chemotherapy drugs, patients who are sensitive or receive too much may be at increased risk for infections due to lowered immune system defenses. Although these influences did not seem to apply to this study since the rats were no longer receiving Cytoxan, was it possible that the sweetened water somehow continued to suppress their immune systems?

The Birth Of Psychoneuroimmunology And Its Skyrocketing Growth

Along with Nicholas Cohen, an immunologist, Ader did subsequent studies that confirmed this by measuring the amount of antibody that was produced in conditioned and unconditioned littermates. Others had previously suspected that the brain could influence the immune system, and George Solomon had actually established a “psychoimmunology” laboratory. However, this was first scientific proof that a nervous system signal (taste) could dramatically affect the immune system. The subsequent explosion of interest in this was unprecedented in my opinion. Some indication of this is evident in the first edition of Ader’s Psychoneuroimmunology in 1981, which had no references to AIDS. The second edition 10 years later, co-authored with Felten and Cohen, was dedicated to George Solomon, who was the senior author of the concluding chapter. This was entitled “Psychoneuroimmunologic Aspects of Human Immunodeficiency Virus Infection”, and its 32 pages included 150 references. The 2001 third edition was so huge that it required 2 volumes and the last in 2006 was even larger.

I had been planning to do a Newsletter interview with Bob Ader for some time to discuss the above and recent advances in this field as well as his views on psychosomatic disease, stress, placebos and other mind/body issues. This was delayed because of various scheduling and health  glitches, so without further ado, let me begin as follows:

 PJR: I would like to make sure that my brief description of what led up to your groundbreaking discovery is accurate. There are undoubtedly other important details that were omitted and would be grateful if you could fill in the blanks. I suspect many of our readers would be curious as to when you coined the term psychoneuroimmunology. How can you explain the dramatic escalation of interest in this discipline and its  continuing upward spiral?

RA: Notwithstanding its corrupted use by some clinical practitioners, psychoneuroimmunology, simply stated, is the study of the interactions among behavior, neural and endocrine function, and immune system processes. The term was first used in 1980 in my presidential address to the American Psychosomatic Society. Its most conspicuous use was as the title of my edited volume in 1981, reviewed and prophetically described as “The signature volume of a new field of research.” The central premise of this interdisciplinary field is that adaptation is the product of a single, integrated network of defenses. Each component of this network evolved to serve specialized functions. These are the parochial interests of the “disciplines” into which we have divided the biological sciences. At the same time, though, each  component of this defensive network monitors and responds to information derived from the others. Thus, we cannot fully understand immunoregulatory processes without considering the organism and the internal and external milieu in which immune responses take place.

As you know, the immune system was once considered a selfregulating, autonomous agency of defense, critical in defending the organism against the invasion of foreign material. At one time, the immune system was defined as that agency of defense that was independent of the nervous system. Research, most of which has been conducted over the past 30 years, however, has revealed that immunoregulatory processes are, in reality, influenced by the brain and, conversely, that neural and endocrine functions and behavior are influenced by the immune system.

With respect to filling in some blanks about the background of psychoneuroimmunology, and others who made significant contributions, were not aware of it at the time, but Russian scientists had conducted studies on the classical conditioning of immune responses in the 1920s. Indeed, that was the first sustained program of research on brain-immune system interactions. A conditioned stimulus (e.g., heat, tactile stimulation) was repeatedly paired with injections of foreign proteins. Subsequent exposure to the conditioned stimulus alone was thought to have induced antibody production. Although it was reviewed in English language journals by the eminent Clark Hull in 1934, it attracted little attention outside the Soviet Union. Within the Soviet Union, it provoked heated arguments since some investigators believed (but the scientific community rejected the notion) that an antibody response was the direct result of neural activity, i.e., that the nervous system, by itself, could stimulate antibody production. Other early indications of CNS influences on immunity came from Szentiványi’s studies in the late 50s showing that hypothalamic lesions could prevent anaphylactic shock in animals. Similar lines of research were initiated sporadically following this.

One of the earliest pioneers in the study of behavioral influences on immunity was Fred Rasmussen, a microbiologist at UCLA. Intrigued by the possibility that emotional states could influence the course of infectious illness, Rasmussen  teamed up with Norman Brill, a psychiatrist—probably the first such collaborative team—to start a program of research on stress and infectious disease. During the 1950s and 60s, Rasmussen and his colleagues examined the effects of various stressors on mice inoculated with different viruses. Susceptibility to infections was increased or  decreased, depending on the nature of the stressor. These studies, with obvious implications for the neuroendocrine modulation of immunity, also failed to attract much attention, although they were forerunners of some of the research on early life experiences and disease susceptibility initiated by Stan Friedman, a pediatrician, and myself and by George Solomon and Alfred Amkraut in the mid 60s. George Solomon was one of the real pioneers in the development of psychoneuroimmunology. His initial research examined the life histories and personality characteristics of patients with autoimmune disease. In the best known of their studies, Solomon and Moos compared rheumatoid arthritis patients with their “at risk,” but healthy, relatives. Their analysis also included the presence or absence of rheumatoid factor, an anti IgG antibody characteristic of rheumatoid arthritis. Compared to the patients, rheumatoid factor positive relatives were psychologically “healthy,” lacked anxiety, depression, or alienation and reported good relationships with spouses, relatives and friends. Psychological well being seemed to have had a salutary effect in the face of a genetic predisposition to autoimmune disease. George was convinced that experimental research would be more persuasive, so, as you indicated, he established a “psychoimmunology” laboratory and studied the effects of behavioral, social and endocrine manipulations in animals on immune function and responses to a bacterial antigen, virus-induced tumors, and adjuvant-induced arthritis. As in other such studies, the results varied depending on the stressor and the outcome measure. Solomon thought that “nobody was listening,” and, in the early 1970s, he had to discontinue this line of research—temporarily. Ten years later, he returned to it and adopted a psychoneuroimmunologic  erspective in his clinical research program on AIDS.

During the 1970s, Hugo Besedovsky, another very prominent figure in the development of what would later be called “psychoneuroimmunology,” was beginning to construct a neuroendocrine-immune system network with his studies of the effects of immune responses on neural and  ndocrine function. If, as he believed, immune function was integrated with other physiological processes, exposure to an antigen should evoke changes in neuroendocrine activity that, in turn, should have feedback effects on immunoregulatory processes and host defenses. There followed an innovative program of research that provided dramatic demonstrations that the nervous and endocrine systems could perceive and respond to signals emitted by an activated immune system.

The novel studies of several other figures played critical roles in the growing acceptance of this new discipline. There was the research of Ed Blalock who found lymphocytes could be a source of brain peptides and pituitary hormones. Now, it’s accepted that brain peptides and their receptors exist within the immune system and that the products of an activated immune system can function as neurotransmitters. Another critical link was forged by investigators such as Karen Bulloch and David Felten who described “hard-wired” connections from the nervous system to the immune system. At a behavioral level, Roger Bartrop described immunologic changes associated with the bereavement that followed the sudden death of a spouse and several other laboratories launched studies of the immune changes associated with stressful life experiences and emotional states. Marvin Stein, for example, who had studied the effects of hypothalamic lesions and stimulation on anaphylactic reactions in guinea pigs during the 1960s, returned to psychoneuroimmunology in the 1980s with a program of animal research on the immunologic effects of stressful experiences as well as human studies of the immunologic changes associated with loss and depression. Another interdisciplinary collaboration between Ronald Glaser, a virologist, and Janice Kiecolt-Glaser, a psychologist, developed an extremely productive research program beginning with studies of stress-induced immune function and the reactivation of latent viruses.

The research initiated in the 1970s and early 1980s was apparently “the right stuff at the right time!” It is likely that no one research program would have had quite the same impact had it not been for the converging evidence of brain-immune system interactions that was appearing in the literature at the same time. These initial studies legitimized questions that had not been asked before. And if the questions—and, sometimes, the questioners—were disparaged, a common experience, the data were compelling and then, undeniable. Thus, the coalescence of research initiated during the 1970s—and the identity provided by the label, psychoneuroimmunology—reactivated latent interests and attracted new investigators to this hybrid field.

In 1976, our research on behaviorally conditioned immunosuppression was the only NIH grant in this area. Today, there are hundreds of NIH grants from different Institutes that deal with psychoneuroimmunology. In 1984, Academic Press approached me about editing a new journal in the field. Initially, I thought it might be too early and could give other journals an additional excuse not to publish our material. I relented, however, and, in 1987, I became the Editor-in-Chief of Brain, Behavior and Immunity. I remained Editor-in-Chief until 2002 when I turned 70. According to 2009 figures, the Citations Index Impact Factor for this journal now places it in the top 16% of all immunology journals and the top 17% of all neuroscience journals. During this time, our informal meetings also included discussions of forming a scientific society representing psychoneuroimmunology. In 1993, I was elected founding President of the Psychoneuroimmunology Research Society and in 2003, Brain, Behavior and Immunity became its official journal. I am frequently asked, “Did you have any idea of what you had started or where your studies would lead?” I did know it could be very important, but never anticipated how rapidly or how large the field would grow, and I continue to be amazed by the number of scientists working in various psychoneuroimmunology laboratories—named as such—all over the world.

PJR:  I presented a brief description of your initial discovery, but you and Nick went on to conduct other experiments that contributed greatly to our understanding of relationships between the brain and immune system and the effects of conditioning. As you look back, what do you think was your greatest contribution and was their much opposition to your theories?

RA: My role in developing this new field will always be associated with a very controversial paper that Nicholas Cohen and I published in Psychosomatic Medicine in 1975. Our research had demonstrated a functional link between the brain and the immune system, which shouldn’t happen, because there were no connections between the brain and the immune system. In the early 1970s, I was studying taste aversion conditioning in rats. This is a variant of classical Pavlovian conditioning. It is an extremely robust one-trial, passive avoidance learning situation in which a novel, distinctively flavored drinking solution, the conditioned stimulus (CS), is paired with the unconditioned effects of a drug with noxious gastrointestinal consequences, the unconditioned stimulus (UCS). Under these circumstances, the rat will learn, after a single CS-UCS avoid consumption of the CS solution. In our study, rats drank different volumes of a saccharin solution and were then injected with a constant dose of Cytoxan, an immunosuppressive drug used in studies of taste aversion learning because it induces the desired gastrointestinal upset. As expected, the magnitude of the conditioned aversive response was directly related to the volume of saccharin consumed on the single conditioning trial. Also, repeated CS presentations without the drug extinguished the avoidance behavior, and the rate of extinction was inversely related to the magnitude of the CS. Unexpectedly, animals began to die during the course of these extinction trials—a troublesome but not particularly interesting observation.

It became evident, however, that, like the magnitude of the conditioned response, mortality rate varied directly with the amount of saccharin the rats consumed on the one conditioning trial—a troublesome but very interesting observation. As a psychologist, I was unaware that there were no connections between the brain and the immune system so I was free to consider any possibility that might explain this orderly relationship between the magnitude of the conditioned response and the rate of mortality. A hypothesis that seemed reasonable to me was that, in addition to conditioning the avoidance response, we were conditioning the immunosuppressive effects of Cytoxan. If reexposure to a CS previously paired with an immunosuppressive drug evoked a conditioned immunosuppressive response, and if the strength of the conditioned response was related to the magnitude of the CS, these animals might have been more susceptible to otherwise subthreshold levels of pathogenic stimulations in the laboratory environment. Thus, the serendipitous observation of mortality in a simple conditioning study and the need to explain an orderly relationship between mortality and the conditioned avoidance behavior prompted the hypothesis that immune responses could be modified by classical conditioning. Colleagues persuaded me to write a letter to Psychosomatic Medicine describing these observations and the hypothesis that immune responses were subject to conditioning. I asked George Engel to read a draft of the letter (from which I had deleted the title). Engel, who usually criticized the Discussion sections in my research papers for being overly cautious, predicted that my conservative reputation was now going to pay off: people were going to believe this just because I was the one who said it. Although it was meant as a compliment, I found the prospect somewhat unnerving. I wanted my ideas to be considered, of course, but I also wanted to retain my right to be wrong. I learned, however, that if you say something that’s not especially important, it doesn’t really matter whether you’re right or wrong; but, if you say something that could be important, you had better be right! The Letter to the Editor in Psychosomatic Medicine did not, as far as I know, attract much attention or generate any interest in testing the hypothesis. The exception was Nick Cohen, an immunologist, who thought these preliminary observations should be pursued. Using the taste aversion conditioning model, he and I designed a study to determine if immune responses could be modified by classical Pavlovian conditioning. The results: conditioned animals that were reexposed to a CS, saccharin, previously paired with the immunosuppressive effects of Cytoxan showed an attenuated antibody response to sheep red blood cells compared with (a) conditioned animals that were not reexposed to the CS, (b) nonconditioned animals that were exposed to saccharin and (c) a vehicle-treated control group. With some evident apprehension on the part of the Program Committee as well as the Editor of Psychosomatic Medicine, the manuscript entitled “Behaviorally Conditioned Immunosuppression” was presented at the 1975 meeting of the American Psychosomatic Society and published in the journal that year. This initial experiment demonstrated that, like other physiological processes, the immune system was subject to classical conditioning, thereby documenting a functional relationship between the brain and the immune system. In that paper, we wrote that …there may be an intimate and virtually unexplored relationship between the central nervous system and immunologic processes and that the application of behavioral conditioning techniques provides a means for studying this relationship in the intact animal. Confirmation of the capacity of behavioral conditioning procedures to suppress (or elicit) immune responses would raise innumerable issues regarding the normal operation of and modifiability of the immune system in particular and the mediation of individual differences in the body’s natural armamentarium for adaptation and survival in general. Such data also suggest a mechanism that may be involved in the complex pathogenesis of disease and bear eloquent witness to the principle of a very basic integration of biologic and psychologic function. Over the next several years, there were replications and extensions of the work on conditioned alterations of immune function resulting in an extensive literature documenting the acquisition and/or extinction of conditioned nonspecific host defense responses and different antibody- and cell8 mediated responses using different conditioned and unconditioned stimuli— and this includes the conditioned enhancement of antibody production using antigen, itself, as the UCS.

Part 2