# Rigid vs flexible dieting: why all-or-nothing thinking backfires

October 17, 2026 · Weight Management · 6 min read · https://burnweek.fit/blog/all-or-nothing-dieting/

> One slice of pizza shouldn't cost you the rest of the day. For rigid dieters, experiments suggest it can, and the rule may matter more than the pizza.

**Key takeaways**

- In a 1975 lab study, restrained eaters ate more after a larger food preload, while unrestrained eaters ate less.
- In a 2010 experiment, restrained eaters who merely believed their pizza slice was larger ate more cookies afterward.
- In a 54,517-person weight-loss program, flexible control was linked to lower BMI, less binge eating, and a higher chance of successful weight loss.
- In a 241-person survey, black-and-white beliefs about food statistically linked restrained eating to weight regain; the design is cross-sectional, so causation is unproven.
- Among 106 women maintaining a loss, flexible restraint predicted better maintenance and rigid restraint less weight lost.

## Rigid rules turn one lapse into a lost day; flexible control is linked to better results

All-or-nothing dieting works on a simple rule: you're either on the diet or off it. The trouble is what happens at the edge. In lab studies of restrained eaters, breaking a rule, even slightly or only in their own mind, was followed by *more* eating, not less. Dieters sometimes call it the "what-the-hell effect": the day is ruined, so there's no reason to hold back. Decades of research on rigid versus flexible restraint point the same way. **Rigid control is associated with more disinhibited eating, more binge episodes, and higher BMI. Flexible control is associated with lower BMI and more successful weight loss** ([Westenhoefer et al., 1999](https://pubmed.ncbi.nlm.nih.gov/10349584/)).

Most of this evidence is correlational, and it should be read that way. The lab experiments behind it suggest a mechanism, and they help explain why "one meal is one entry, not a verdict" is more than a slogan.

If you're coming from our [guide to calorie deficits](/blog/calorie-deficit-weight-loss): the deficit is the arithmetic, and this article is about the kind of thinking that decides whether you keep doing it.

## The preload experiments: restrained eaters ate more afterward

The pattern was first measured in 1975. Herman and Mack split normal-weight participants into high-restraint (dieting) and low-restraint groups and gave them food "preloads" of different sizes before a "taste test" that measured how much more they ate. Non-dieters did what appetite predicts: the bigger the preload, the less they ate afterward. Dieters did the reverse. Their intake rose with the size of the preload ([Herman & Mack, 1975](https://pubmed.ncbi.nlm.nih.gov/1206453/)).

A 2010 experiment showed that the trigger doesn't have to be real. Every participant was served the same slice of pizza. Some were led to believe their slice was bigger than another participant's, some that it was smaller, and some saw no comparison. Then everyone could eat cookies freely. When the slice *looked* large, restrained eaters tended to eat more cookies, while unrestrained eaters tended to eat fewer. The authors interpreted the pattern as driven by the *belief* that they had overeaten ([Polivy et al., 2010](https://pubmed.ncbi.nlm.nih.gov/20691231/)).

So in these experiments, the extra eating seemed to follow the perception of having broken the rule more than the calories themselves. A diet built on bright lines may produce that feeling often, because real days rarely stay inside them.

> In all-or-nothing dieting, the meal that broke the rule may matter less than what the dieter decides it means.

## Rigid vs flexible restraint: the 54,517-person validation

If rigid rules cause trouble, is there another way to restrain eating? Westenhoefer and colleagues drew the distinction in a 1999 paper that developed two new measures:

- **Rigid control**: broadly, an all-or-nothing approach to restraint, where you are either on the diet or off it.
- **Flexible control**: broadly, a graded approach, where limits bend and a lapse is absorbed rather than treated as a breach.

They tested the measures in three samples. The largest included **54,517 participants** in a computer-assisted weight-reduction program who completed questionnaires and a 7-day food diary. A second sample was a random selection of 1,838 people from the West German population, and a third was a smaller sample of 85 participants ([Westenhoefer et al., 1999](https://pubmed.ncbi.nlm.nih.gov/10349584/)).

| Measure | Rigid control was associated with | Flexible control was associated with |
|---|---|---|
| Disinhibition (eating past intentions) | Higher | Lower |
| BMI | Higher | Lower |
| Binge eating episodes | More frequent and more severe | Less frequent and less severe |
| Self-reported energy intake | — | Lower |
| Weight loss in the 1-year program | — | Higher probability of success |

A later study followed 106 women who had already lost weight, about 10 kg on average, with a commercial slimming organization and were trying to keep it off. Over six months, **flexible restraint was associated with more weight lost and better maintenance, while rigid restraint was associated with less weight lost.** Rigid restraint also tracked with more attention bias toward food- and body-shape-related stimuli, a sign of preoccupation ([Westenhoefer et al., 2013](https://pubmed.ncbi.nlm.nih.gov/23265405/)).

These are associations. People who are already doing well may find flexibility easier. But the direction agrees across a lab mechanism, a very large cross-sectional sample, and a six-month follow-up of people maintaining a loss.

## Black-and-white thinking and regain

A 2015 study looked at the thinking style underneath rigid rules. Researchers surveyed 241 adults online, measuring *dichotomous thinking*, the tendency to see things in black-and-white terms, along with restrained eating, weight regain, BMI, and dieting history. The key result was that **eating-specific dichotomous thinking**, meaning black-and-white beliefs about food and eating, was a statistical *mediator* of the link between restrained eating and weight regain ([Palascha et al., 2015](https://pubmed.ncbi.nlm.nih.gov/25903250/)). In other words, in this survey the link between restraint and reported regain ran partly through all-or-nothing beliefs about food.

The caveat is essential. This was a one-time survey, so everything was measured at once and weight regain was self-reported. The analysis is *consistent with* black-and-white thinking contributing to regain, but it can't show that the thinking came first. The authors frame it as a possible link to rigid restraint, and that's how it should be read.

## One meal is one entry, not a verdict

This section is practical interpretation built on the studies above, not a tested program.

**Replace food bans with amounts.** "No chocolate ever" is a bright line that is easy to cross. "A couple of squares after dinner most days" is a flexible rule: you can go over it slightly without breaking it.

**Judge the week, not the meal.** A graded approach lets a large meal be absorbed over the week, without strict compensation. A single meal is a small part of a week's intake. Our article on [cheat meals and progress](/blog/cheat-meals-and-progress) shows the arithmetic, and [weekend overeating](/blog/weekend-overeating-undoing-deficit) shows where the real damage tends to build.

**Record the meal, not a judgment.** Logging a big meal as a plain entry, with no "bad day" label, keeps it as information. The rigid-dieting studies suggest that the sense of having broken a rule, more than the calories, can set off extra eating afterward.

**Change the next decision only.** The recovery steps are covered in [what to do after overeating](/blog/what-to-do-after-overeating). The thinking side matters too: in a lab study, strict dieters helped to respond self-compassionately after breaking a food rule ate less afterward, a finding covered in our piece on [self-compassion and weight loss](/blog/self-compassion-and-weight-loss).

**Watch for rigidity creeping into tracking.** Calorie tracking can support flexible control, since it shows the week's total. It can also become a new rulebook with a daily pass or fail. Our guide to [tracking without obsessing](/blog/tracking-without-obsessing) covers how to tell the difference.

## When rigid thinking is a warning sign

The research above covers ordinary dieters. If going over a rule regularly leads to eating that feels out of control, or to strict compensation such as fasting or purging, that goes beyond dieting style. It's worth discussing with a doctor or a specialist in eating disorders. Rigid control's links to more frequent and more severe binge episodes are one reason not to wave these patterns off ([Westenhoefer et al., 1999](https://pubmed.ncbi.nlm.nih.gov/10349584/)).

## FAQ

### What is the "what-the-hell effect" in dieting?

It's the pattern in which a dieter who feels they've broken their diet eats more, not less, afterward. In lab studies, restrained eaters ate more after a larger preload, and even after merely believing they had eaten more than another person ([Herman & Mack, 1975](https://pubmed.ncbi.nlm.nih.gov/1206453/); [Polivy et al., 2010](https://pubmed.ncbi.nlm.nih.gov/20691231/)).

### Is flexible dieting better than strict dieting for weight loss?

The evidence leans that way, though it's mostly observational. In a program with 54,517 participants, flexible control was associated with lower BMI and a higher probability of successful weight loss, while rigid control was associated with more binge eating and higher BMI ([Westenhoefer et al., 1999](https://pubmed.ncbi.nlm.nih.gov/10349584/)).

### Does black-and-white thinking cause weight regain?

It hasn't been shown to. A survey of 241 adults found that black-and-white beliefs about food statistically linked restrained eating to weight regain ([Palascha et al., 2015](https://pubmed.ncbi.nlm.nih.gov/25903250/)). Because everything was measured at once, the study can't establish cause.

## Sources

- [Herman CP, Mack D. Restrained and unrestrained eating. J Pers. 1975. doi:10.1111/j.1467-6494.1975.tb00727.x](https://pubmed.ncbi.nlm.nih.gov/1206453/)
- [Polivy J, Herman CP, Deo R. Getting a bigger slice of the pie. Effects on eating and emotion in restrained and unrestrained eaters. Appetite. 2010. doi:10.1016/j.appet.2010.07.015](https://pubmed.ncbi.nlm.nih.gov/20691231/)
- [Westenhoefer J, Stunkard AJ, Pudel V. Validation of the flexible and rigid control dimensions of dietary restraint. Int J Eat Disord. 1999;26(1):53ff.](https://pubmed.ncbi.nlm.nih.gov/10349584/)
- [Westenhoefer J, Engel D, Holst C, Lorenz J, et al. Cognitive and weight-related correlates of flexible and rigid restrained eating behaviour. Eat Behav. 2013. doi:10.1016/j.eatbeh.2012.10.015](https://pubmed.ncbi.nlm.nih.gov/23265405/)
- [Palascha A, van Kleef E, van Trijp HC. How does thinking in Black and White terms relate to eating behavior and weight regain? J Health Psychol. 2015. doi:10.1177/1359105315573440](https://pubmed.ncbi.nlm.nih.gov/25903250/)

Source: BurnWeek — "Rigid vs flexible dieting: why all-or-nothing thinking backfires", https://burnweek.fit/blog/all-or-nothing-dieting/. Licensed CC BY 4.0: free to quote or reuse with a link to this page.