CoChat

Food Coma is Real: The Effect of Digestive Fatigue on Adolescents' Cognitive Performance

Open access

The DOI matches its Crossref record; 91 of the 91 references it cites were found in a registry or an index; 5 of the 20 sources its claims cite were checked for support.

Abstract

Food coma, also known as postprandial somnolence, is a commonly cited reason for experiencing reduced alertness during mid-afternoon worldwide. By using exogenous variation in the timing of tests and, hence, by extension, plausibly exogenous variation in the temporal distance between an individual's last meal and the time of test, we examine the causal impact of postprandial somnolence on cognitive capacities. Analyzing novel time use data on ~ 4,600 Indian adolescents and young adults, we find that testing within an hour after a meal reduces test-takers' scores on English, native language, math, and Raven's tests by 8, 8, 8, and 16 percent, respectively, compared to test-takers who took the tests more than an hour after their meal. We further find that the negative effect of postprandial somnolence on cognition operates through increased feelings of fatigue and depletion of cognitive resources that become more pronounced while dealing with more challenging test questions.

Overview

CoChat Summary

## TL;DR

Using quasi-random variation in test timing among about 4,600 Indian adolescents and young adults, the paper finds that taking cognitive tests within an hour of a meal lowers performance across several domains. The effects are sizable, operate mainly through fatigue, and are stronger for harder tasks, with some heterogeneity by SES and age.

## Key findings

  • Testing within an hour after a meal reduced scores by 0.27 SD in native-language reading, 0.21 SD in English reading, 0.25 SD in math, 0.42 SD in Raven’s matrices, and 0.15 SD in oral comprehension, relative to testing more than an hour after a meal.
  • In percentage terms, the paper reports score declines of about 8% for native-language reading, English reading, and math, 16%–17% for Raven’s, and 5%–6% for oral comprehension.
  • The treatment increased self-reported fatigue by 0.39 SD; the paper interprets this as evidence that postprandial somnolence works through fatigue and depleted cognitive resources.
  • The negative effect is larger for more difficult items: e.g., in native-language reading, effects were -18% for paragraphs, -7% for sentences, -4% for words, and not significant for letters; a similar pattern appears for English and math.
  • Heterogeneity: effects were larger in higher SES terciles, with gaps such as -0.08/-0.07/-0.12/-0.28 SD in tercile 1 versus -0.45/-0.38/-0.34/-0.55 SD in tercile 3 for native reading, English, math, and Raven’s, respectively.
  • The paper reports no evidence of meaningful gender differences, and mixed age-patterns: effects on Raven’s and oral comprehension are said to diminish with age.

## Methods

  • The study uses endline individual surveys from a larger project in 158 rural villages across two districts in Andhra Pradesh and Maharashtra, India, with respondents aged 12–22 in 2022; the paper says the endline sample is about 4,600 adolescents and young adults.
  • Identification comes from plausibly exogenous variation in test timing: the timing of household visits/tests was determined by field logistics, and the key treatment is whether testing occurred within 1 hour of the last meal versus more than 1 hour after.
  • Outcomes are standardized cognitive test scores in native-language reading, English reading, math, Raven’s matrices, and oral comprehension; the main analysis is OLS regressions with baseline covariates, village fixed effects, and age-group fixed effects, plus robustness checks and heterogeneity analyses.

## Limitations

  • The paper explicitly notes that the self-reported fatigue measure and the test-question completion-time measures are only proxies and are “not perfect indicators” of motivation.
  • It states that more research is needed to determine whether the findings replicate in other countries and age groups, and whether food coma affects learning generally or other high-stakes settings.
  • Evident from the text: the study is based on one endline survey context within a specific rural Indian program, so generalizability beyond that setting is not established in the paper.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.2139/ssrn.4788223

Cites

91 of 91 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Cognitive attainment by time distance from last meal
Figure 1 Cognitive attainment by time distance from last mealFrom the PDF
Figure A1: Coefficient estimates from the regression of SES on time dummies
Figure A1: Coefficient estimates from the regression of SES on time dummiesFrom the PDF
Figure A2: Distribution of survey times
Figure A2: Distribution of survey timesFrom the PDF

Verification

Claims the paper makes

2 Valid · 6 Not valid · 4 Not checked · 118 Own results and interpretations

83 more claims

Create a free account to see all 130 claims and to verify the 2 references still open.

See all claims — free account
  • Probably not a substantial claim. 14
  • It may state two separate claims as one. 21
  • The quote only partly states this claim. 11

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

4 of 4

Citations

Cites

Show all 91 references

Research works printed but not linked: 6

  1. Aser, 2018, Annual Status of Education Report
  2. Y Benjamini, 2006, Adaptive linear step-up procedures that control the false discovery rate, Biometrika, 93, 491, 10.1093/biomet/93.3.491 · DOI · Paper page
  3. S Bradley, 2020, The economics of education: A comprehensive overview
  4. S W Kim, 2009, Metabolic state, neurohormones, and vagal stimulation, not increased serotonin, orchestrate postprandial drowsiness, Bioscience Hypotheses, 2, 422, 10.1016/j.bihy.2009.07.008 · DOI
  5. J Park, 2018, Hot temperature, high stakes exams, and avoidance behavior: Evidence from new york city public schools
  6. M Walker, 2017, Why we sleep: Unlocking the power of sleep and dreams

Metadata

Authors
Hervé, J., Mani, S., Behrman, J. R., Laxminarayan, R., & Nandi, A.
Affiliations
  • Stevens Institute of Technology (US)
  • Fordham University (US)
  • University of Pennsylvania (US)
  • Population Council (IN)
Corresponding author
Hervé, Justine
Venue
Social Science Research Network
Publisher
Elsevier BV
Published
2024
ISSN
1556-5068 (ISSN-L)
OpenAlex
W4394697463
Cited by
0
References
75
Type
posted content
Language
en
Open access
Locations
  • SSRN Electronic Journal (acceptedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • Econstor (Econstor) (submittedVersion)
  • ScholarlyCommons (University of Pennsylvania) (submittedVersion, cc-by)
Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Sleep and Work-Related Fatigue
  • Social Sciences › Education › Educational Outcomes and Influences
  • Health Sciences › Nursing › Nutrition and Dietetics › Nutrition, Health and Food Behavior
Keywords
cognitive performance, adolescents, young adults, test scores, time use, meal timing, cognitive depletion, Raven's test

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.