Peter Herman

Introduction

C. Peter Herman is an American-Canadian psychologist, professor emeritus at the University of Toronto, who created the modern psychology of dieting. Working from the mid-1970s with Janet Polivy, he established that people who restrict their eating do not simply eat less: given the right trigger they eat more than people who are not dieting at all. That counterintuitive finding — counterregulation or disinhibition — became the empirical foundation of restraint theory, and the phenomenon he and Polivy named the "what-the-hell effect" has passed into ordinary language. His work supplied the first credible psychological account of why dieting so often fails, and it reframed overeating as a problem of self-regulation rather than of appetite.

Early Life and Education

Herman took his B.A. at Yale in 1968 and his PhD at Columbia in 1972, and taught at Northwestern University from 1972 to 1976. It was at Northwestern that the eating research began: Polivy, then a graduate student, had noticed that fellow students dieted all day and then "ate everything in sight" at night, and the two set out to see whether the pattern could be produced in the laboratory. Herman moved to the University of Toronto in 1977, spent the rest of his career there, became a full professor in 1983, and served as director of undergraduate and graduate studies and as department chair before retiring in 2012. He edited the Journal of Personality and was executive co-editor of Appetite, published more than two hundred papers, and has been funded continuously by Canada's natural sciences and engineering research council for four decades.

Academic and Professional Career

Herman's research programme has had two phases. The first, and the one for which he is known, established restraint theory: the development of the Restraint Scale, the preload experiments, the boundary model, and the demonstration that dieters' eating is governed by cognition rather than by internal hunger signals. The second, beginning in the 1990s and continuing after his retirement, turned to normal eating — why people eat what they eat in ordinary circumstances — with a focus on social and personal norms as determinants of intake, including modelling, social facilitation of eating, and the effect of portion size. Across both phases the unifying interest is self-regulation and its failures, and the same question recurs: under what conditions does the intention to control one's own behaviour collapse?

Major Contributions to Psychology

Restrained Eating and the Preload Experiment

The paradigm that made restraint theory famous was disarmingly simple. In the 1975 study with Deborah Mack, participants were asked to taste and rate ice creams, but some were first given a milkshake to drink. Non-dieters behaved as physiology predicts: the more milkshake they had, the less ice cream they ate. Dieters did the opposite — the larger the preload, the more they ate. Herman and Mack's interpretation was that a forced preload destroys the rationale for restraint; once the day's diet is broken, the cognitive brake is released and eating is driven by internal and external cues alone. A companion study the same year showed that anxiety did not reliably reduce eating, undermining the psychosomatic account of obesity then in circulation.

Restraint itself was operationalised by the Restraint Scale, later revised with Polivy into a ten-item instrument with two subscales, Concern for Dieting and Weight Fluctuation. Its central empirical claim is that what distinguishes dieters from non-dieters is not physiology but an intention, and that the intention is fragile.

The Boundary Model

Herman and Polivy's boundary model (1983–84) put the findings into a physiological frame. Eating is normally regulated inside a "zone of biological indifference" bounded by hunger below and satiety above. Unrestrained eaters stay inside that zone automatically. Restrained eaters impose an additional, cognitively maintained diet boundary well below the satiety boundary, and holding it requires attention and motivation. Two classes of event can breach it: strong emotional distress, and a perceived dietary violation — eating a forbidden food, or simply believing one has. Once the boundary is breached, eating continues to the satiety boundary, which in chronic dieters is set higher than normal. The model thus predicts both the emotional eating and the post-transgression eating that clinicians had long reported.

The What-the-Hell Effect

The name Herman and Polivy gave to the second route is the what-the-hell effect: having violated their own rule, restrained eaters abandon the attempt altogether and keep eating. The classic demonstrations manipulated belief as well as calories — participants told they were eating a high-calorie pudding subsequently ate more than those told the same pudding was low-calorie, regardless of what they had actually consumed, and the same pattern held for milkshake preloads. The effect is not confined to food; the literature on the abstinence-violation effect in addiction describes the same structure, and restraint theory has been applied to smoking, drinking, and spending.

Herman's later work asked what governs intake when nobody is dieting. With Deborah Roth and Polivy he reviewed, in a 2003 Psychological Bulletin article, the apparently contradictory literatures on social facilitation of eating and on modelling, and reconciled them with the idea of an inhibitory norm: people eat less in front of others not because of evaluation anxiety alone but because a norm against eating too much is activated, and they calibrate their intake against what a companion eats. This work made the case that ordinary eating is socially regulated to a degree that laboratory studies of individuals miss.

Dieting and Eating Disorders

With Polivy, Herman argued that dieting is a cause rather than merely a symptom of eating disorders, notably bulimia nervosa, by way of the restraint–disinhibition cycle. The claim was contested when it was made and remains debated, but it shifted clinical thinking away from the assumption that restraint is simply the appropriate response to overeating.

Major Publications

"Restrained and unrestrained eating" (with Deborah Mack, 1975), the preload experiment that founded restraint theory.
"Anxiety, restraint, and eating behavior" (with Janet Polivy, 1975), which failed to confirm the psychosomatic hypothesis of obesity.
Breaking the Diet Habit (with Janet Polivy, 1983), the trade book that carried the boundary model to a general readership.
"A boundary model for the regulation of eating" (with Janet Polivy, 1983), the formal statement of the theory.
"Causes of eating disorders" (with Janet Polivy, 2002), an Annual Review of Psychology synthesis of sociocultural, familial, and affective pathways.
"Effects of the presence of others on food intake: a normative interpretation" (with Deborah Roth and Janet Polivy, 2003), the inhibitory-norm account of social eating.

Impact and Legacy

Restraint theory changed the default assumption about dieting. Before it, overeating after restraint looked like weakness; after it, it looked like a predictable consequence of the restraint itself. Clinically, the work is a direct ancestor of the intuitive-eating approach — its developers have said the what-the-hell effect described exactly what they saw in patients — and conceptually it fed the broader literature on self-regulation failure and on why attempts at self-change collapse after a first lapse. The boundary model remains the standard account of the cognitive regulation of eating.

Three criticisms have proved durable. The first concerns the instrument. The Restraint Scale contains two separable components, concern for dieting and weight fluctuation, and high scorers turn out to be people who chronically diet and overeat — that is, unsuccessful dieters whose scores are elevated by their history of weight cycling. Critics argue this makes the central finding partly circular: a scale that identifies people by their tendency to lose control will, unsurprisingly, find that they lose control. Alternative measures, notably the Dutch Eating Behaviour Questionnaire and the Three-Factor Eating Questionnaire, sometimes identify restrained eaters who eat less, and the discrepancy has never been fully resolved. Michael Lowe has pressed the point hardest, arguing that current dieting and chronic restrained eating are different states that should not be conflated; Herman, Polivy, and Jennifer Mills replied in 2021, and the exchange is still live.

The second criticism is that specific predictions have failed to replicate. Several studies have not found the what-the-hell cognitions the model requires, and some have failed to obtain disinhibition effects at all. Supporters note that the preload effect itself has been replicated many times and that failures tend to occur when the preload is not perceived as a violation; detractors note that a theory which can absorb both outcomes is hard to test.

The third is about scope and direction. Most of the evidence comes from small laboratory studies using undergraduate women, milkshakes, and taste tests, and the causal arrow is not settled: people who are prone to overeating may become dieters rather than the reverse, and the association between restraint and later weight gain is compatible with both readings. Herman's own later work, with its shift toward norms and ordinary eating, implicitly concedes the point — the laboratory dieter is a special case, not the whole of human eating. Supporters would put it differently: the theory was always about a specific mechanism operating in a specific population, and on that narrower claim it has held up.

Learn More

References

Polivy, J., & Herman, C. P. (1985). Dieting and binging: A causal analysis. American Psychologist, 40(2), 193-201.