The word is in general dictionaries, on sandwich packaging, and in the sort of half apology people make in a kitchen at twenty past seven. Hangry: bad tempered as a result of hunger. It travelled because it does two things at once. It names something recognisable, and it excuses it.
Underneath the joke sit three separate claims. They get said in the same breath and they do not have the same evidence behind them.
0.06
the ego depletion effect size across 36 laboratories and 3,531 participants
31
studies in the carbohydrate and mood meta-analysis, none finding a mood benefit
45%
of people told to fast for five hours arrived at the lab not feeling hungry
19%
of irritability variance explained by the fixed effects in the study that found the hunger link
The three things the word is doing
The first is a correlation in ordinary life: hunger and irritability show up together. The second is a mechanism: falling blood glucose shortens the temper, because self-control runs on fuel and the fuel ran out. The third is an attribution: that remark, in that kitchen, at twenty past seven, happened because of blood sugar.
They do not stand or fall together. The first is reasonably well supported and smaller than the phrase implies. The second is in poor shape. The third is not the kind of statement that can be checked.
| The claim | What is actually being asked | Where the evidence stands |
|---|---|---|
| Hunger and irritability travel together | Do the two co-occur in the same person at the same moment | Reasonably good within a moment, but hunger does not forecast a worse mood at the next check-in |
| Eating relieves irritation | Does the state lift when food arrives | The best supported of the four, from the largest dataset, and specific to irritation rather than mood in general |
| Falling blood glucose shortens the temper | Is there a metabolic chain from fuel to behaviour | Weak. The theory it rests on has failed repeated large replications, and the number rarely moves far in people not on glucose lowering medication |
| That outburst was caused by low blood sugar | Can one episode be traced to one input | Not testable. Nothing measured afterwards separates it from the other candidates |
What daily life studies found, and where they disagree
The study usually cited followed 64 adults in Austria, Germany and Switzerland for 21 days, five prompts a day, 9,142 reports in total. Hunger tracked anger and irritability within the same person, and the association survived adjustment for sex, age, body mass index, dieting behaviour and trait anger. That is a real result from a sensible design.
It is also modest in size, and the paper's own summary figure invites misreading. The headline is that hunger explained 56 percent of the variance in irritability. That number covers the whole model including person level random effects; the fixed effects alone account for 19 percent, and they include sex, age, body mass index, trait anger and each person's average hunger over three weeks. On the within-person coefficient, each point of hunger on a hundred point scale moved irritability by 0.16 of a point and anger by 0.10. Hunger at the top of the scale rather than the bottom is worth about sixteen points of irritability out of a hundred. Enough to notice. Not enough to carry an evening on its own.
Then there is the study pointing the other way. In Lausanne, 748 community adults with a mean age of 60 answered four times a day for a week. Hunger intensity predicted feeling more active and lively later, and less slowed thinking. It did not predict any shift toward sadness or anger. The authors read the difference as one of degree: naturalistic hunger is milder than deprivation arranged by a researcher.
The two results are compatible, and the way they are compatible is the interesting part. Hunger and irritability can sit in the same moment without hunger being able to tell you anything about the next one.
The largest of the three datasets is more encouraging about one narrow thing. Across nine studies, 801 participants and more than forty thousand questionnaire entries, researchers asked which emotional states preceded snacking and which followed it. Boredom preceded it. And of everything measured, irritation was the only state that reliably came down afterwards. Not sadness, not stress, not anxiety. Irritation.
So the defensible version of the first claim is narrower than the word: hunger and irritability co-occur, eating specifically relieves irritation, and hunger right now is a poor predictor of mood in an hour.
The blood sugar part is the weakest part
Almost every popular version of hangry contains a sentence about blood sugar crashing. That sentence is doing more work than the physiology supports.
Glucose is held in a narrow band, and the system that holds it there is redundant. Insulin falls, glucagon rises, the liver releases stored glucose and then manufactures more. The Endocrine Society's guideline on hypoglycaemic disorders in adults is blunt about the consequence: because those defences work, hypoglycaemia is an uncommon clinical event except in people who take drugs that lower plasma glucose, particularly insulin. The threshold commonly used for low blood glucose is 70 milligrams per decilitre, and the US National Institute of Diabetes and Digestive and Kidney Diseases does list irritability among its symptoms, in a document about people with diabetes who take medication for it.
So there are two conversations. For someone on insulin, the mechanism is documented and the symptom list is real. For most people who say they are hangry, the number is probably in range the whole time.
There is a more careful route to the same feeling, worth naming rather than overselling. The counterregulatory response to a falling glucose level involves adrenaline and cortisol, and a nervous system with more adrenaline in it has less tolerance for a third interruption. That is a plausible path to edginess without any hypoglycaemia at all, and it is a different claim from the crash story, with much less behind it.
The intervention everyone reaches for has been tested properly. A meta-analysis of 31 studies and 1,259 participants found no positive effect of carbohydrate on any aspect of mood, with alertness lower within sixty minutes and fatigue higher within thirty. Whatever happens when a meal helps, the sugar is not the active ingredient.
Where the fuel theory came from, and what happened to it
Blood sugar is attached to temper at all because of one research programme. In 2007 Gailliot, Baumeister and colleagues proposed that self-control draws on glucose as a limited energy source, and that willpower is therefore more than a metaphor. A clean idea with an obvious practical reading: a depleted person needs something sweet.
The decade that followed was unkind to it.
In 2012 Molden and colleagues had participants swish a carbohydrate solution and spit it out rather than swallow it. The effect appeared anyway, and not with a non-caloric sweetener. Nothing had been metabolised, so whatever the glucose was doing it was doing it as a reward signal rather than as fuel. In 2013 Job, Walton, Bernecker and Dweck showed the glucose effect depended on what participants believed about willpower, which is not how a metabolic constraint behaves. In 2016 Vadillo, Gold and Osman went through the glucose literature and found its evidential value weak and its effect sizes consistent with publication or reporting bias.
The parent effect fared no better. A 23 laboratory preregistered replication in 2016 concluded that if there is an ego depletion effect it is close to zero. A 36 laboratory paradigmatic test in 2021, with 3,531 participants, found a nonsignificant 0.06 and data four times more likely under the null than under an informed prior.
The study closest to hangry itself sits in this literature. In 2014 Bushman, DeWall, Pond and Hanus tracked glucose in 107 married couples for 21 days and reported that lower evening glucose went with more aggressive behaviour toward a spouse. Lange and Kurzban replied in print the same year: the study measured aggressive impulses rather than the inhibition of them, averaging glucose over weeks produces a trait measure rather than a momentary depletion, and a correlation of that shape does not validate the glucose model. Their closing line is the useful one. There is no reason to think that giving couples a sugary boost reduces partner violence.
The hangry experiment that went looking for the mechanism
The most careful laboratory attempt is MacCormack and Lindquist's, published in Emotion in 2019. Its results are more instructive than the headlines they generated.
Two hundred and thirty-six students were assigned either to fast for five or more hours or to eat a full meal within the hour before arriving. First finding, and it is about the whole literature rather than about this study: 54 of the 119 people in the fasting condition reported that they were not hungry. Nearly half. Satiety does not run on a clock.
Among those for whom the manipulation worked, hunger did raise self-reported negative high arousal emotion, but only where attention had not been directed at emotion beforehand. Participants who had just spent several minutes writing about anger or sadness showed no such effect. On the specific adjective anger there was no effect at all in any condition. Self-reports of hate moved; anger did not.
The study also tested the depletion account directly, by how long people persisted at a tedious mental rotation task. Hungry participants lasted 8.16 minutes on average, satiated participants 8.90, with a p value of .457. Entering perseverance as a covariate changed nothing anywhere.
Read plainly, that is a hangry experiment in which hunger did not deplete self-control, did not raise anger, and raised other unpleasant feelings only when people were not already attending to their own emotional state. The conceptualisation step is where the action was.
The parole study, which is not the support people think it is
Any discussion of this reaches the judges eventually. In 2011 Danziger, Levav and Avnaim-Pesso reported that the proportion of favourable parole rulings in Israeli boards fell across a session and jumped back up after a food break. It became one of the most repeated findings in popular psychology.
Later the same year, Weinshall-Margel and Shapard wrote to the journal to report that case ordering in those boards was not random: among other things, boards tended to finish all cases from one prison before taking a break. In 2016 Glöckner showed by simulation that an effect of similar magnitude is produced by a purely rational scheduler who avoids starting a case likely to run long just before a break, because favourable rulings take longer than unfavourable ones.
The finding may have something in it. It is no longer evidence anybody should be citing, and it appears here only as the clearest example of a story outrunning its data.
The claim nobody can check is the one people use most
Which leaves the third claim, the one that actually gets said out loud. Not hunger and irritability are associated. Not glucose constrains regulation. Simply: that was the blood sugar.
Nothing can be done with that sentence afterwards. The meal has happened, the mood has lifted, and no measurement remains that distinguishes low blood sugar from the long day, the conversation at four o'clock, the noise, the thing somebody said on Monday that went unanswered. Any of them predicts the same evening.
That is the real argument of this article. An explanation nobody audits stops the auditing, and whatever the anger was about does not get looked at, because the question was closed before it opened. Anger quite often arrives second, on top of something more exposing, and the case for that is made properly in the entry on anger as a secondary emotion. A convenient metabolic story is one of the more effective ways to never get there.
Thresholds are real even when the attribution is not
None of this means the experience is invented, and that would be a cheap conclusion. Not eating, not sleeping, pain: these lower tolerance, and people who found that eating at fixed times changed their evenings are not mistaken about their evenings. What they have is a result. What they do not have, and do not need, is a mechanism.
The honest form of the claim is a personal hypothesis rather than a general law, and it is one of the few here that can be settled cheaply.
That log is worth more than a laboratory for one reason. Laboratories have to average across people, and nearly half the participants in the best hangry experiment did not become hungry when instructed to. Individual variation is not noise here. It is the main finding.
If the shorter fuse is recent rather than lifelong, the test called Why Am I So Irritable Lately sorts the change into four sources rather than one, and the body is only one of them. Worth taking before committing to a favourite explanation, because the four want different things and one of them gets worse if treated as a scheduling problem.
The word will survive all of this. It is too useful not to. What is worth keeping separate is that hangry makes an excellent description and a poor diagnosis, and the gap between the two is where most of the evening's real information is sitting.
Sources
- Swami, V., Hochstöger, S., Kargl, E., Stieger, S. (2022). Hangry in the field: an experience sampling study on the impact of hunger on anger, irritability, and affect. PLOS ONE, 17(7), e0269629.
- de Rivaz, R., Swendsen, J., Berthoz, S., Husky, M., Merikangas, K., Marques-Vidal, P. (2022). Associations between hunger and psychological outcomes: a large-scale ecological momentary assessment study. Nutrients, 14(23), 5167.
- Aulbach, M. B., Bamberg, C., Reichenberger, J., Arend, A.-K., Blechert, J. (2025). Taming hanger and falling prey to boredom-emotional and stress-eating in 801 healthy individuals using ecological momentary assessment. Appetite, 207, 107897.
- MacCormack, J. K., Lindquist, K. A. (2019). Feeling hangry? When hunger is conceptualized as emotion. Emotion, 19(2), 301-319.
- Cryer, P. E., Axelrod, L., Grossman, A. B., Heller, S. R., Montori, V. M., Seaquist, E. R., Service, F. J. (2009). Evaluation and management of adult hypoglycemic disorders: an Endocrine Society clinical practice guideline. Journal of Clinical Endocrinology and Metabolism, 94(3), 709-728.
- National Institute of Diabetes and Digestive and Kidney Diseases. Low blood glucose, hypoglycemia.
- Mantantzis, K., Schlaghecken, F., Sünram-Lea, S. I., Maylor, E. A. (2019). Sugar rush or sugar crash? A meta-analysis of carbohydrate effects on mood. Neuroscience and Biobehavioral Reviews, 101, 45-67.
- Gailliot, M. T., Baumeister, R. F., DeWall, C. N., Maner, J. K., Plant, E. A., Tice, D. M., Brewer, L. E., Schmeichel, B. J. (2007). Self-control relies on glucose as a limited energy source: willpower is more than a metaphor. Journal of Personality and Social Psychology, 92(2), 325-336.
- Molden, D. C., Hui, C. M., Scholer, A. A., Meier, B. P., Noreen, E. E., D'Agostino, P. R., Martin, V. (2012). Motivational versus metabolic effects of carbohydrates on self-control. Psychological Science, 23(10), 1137-1144.
- Job, V., Walton, G. M., Bernecker, K., Dweck, C. S. (2013). Beliefs about willpower determine the impact of glucose on self-control. PNAS, 110(37), 14837-14842.
- Vadillo, M. A., Gold, N., Osman, M. (2016). The bitter truth about sugar and willpower: the limited evidential value of the glucose model of ego depletion. Psychological Science, 27(9), 1207-1214.
- Hagger, M. S., Chatzisarantis, N. L. D., et al. (2016). A multilab preregistered replication of the ego-depletion effect. Perspectives on Psychological Science, 11(4), 546-573.
- Vohs, K. D., Schmeichel, B. J., et al. (2021). A multisite preregistered paradigmatic test of the ego-depletion effect. Psychological Science, 32(10), 1566-1581.
- Bushman, B. J., DeWall, C. N., Pond, R. S., Hanus, M. D. (2014). Low glucose relates to greater aggression in married couples. PNAS, 111(17), 6254-6257.
- Lange, F., Kurzban, R. (2014). Sugar levels relate to aggression in couples without supporting the glucose model of self-control. Frontiers in Psychology, 5, 572.
- Danziger, S., Levav, J., Avnaim-Pesso, L. (2011). Extraneous factors in judicial decisions. PNAS, 108(17), 6889-6892.
- Weinshall-Margel, K., Shapard, J. (2011). Overlooked factors in the analysis of parole decisions. PNAS, 108(42), E833.
- Glöckner, A. (2016). The irrational hungry judge effect revisited: simulations reveal that the magnitude of the effect is overestimated. Judgment and Decision Making, 11(6), 601-610.