Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 64 of the 64 references it cites were found in a registry or an index.

A neuro-computational account of procrastination behavior

Open access
Le Bouc, Raphaël, Pessiglione, Mathias · 2022 · Nature Communications

Abstract

Humans procrastinate despite being aware of potential adverse consequences. Yet, the neuro-computational mechanisms underlying procrastination remain poorly understood. Here, we use fMRI during intertemporal choice to inform a computational model that predicts procrastination behavior in independent tests. Procrastination is assessed in the laboratory as the preference for performing an effortful task on the next day as opposed to immediately, and at home as the delay taken in returning completed administrative forms. These procrastination behaviors are respectively modeled as unitary and repeated decisions to postpone a task until the next time step, based on a net expected value that integrates reward and effort attributes, both discounted with delay. The key feature that is associated with procrastination behavior across individuals (both in-lab and at-home) is the extent to which the expected effort cost (signaled by the dorsomedial prefrontal cortex) is attenuated by the delay before task completion. Thus, procrastination might stem from a cognitive bias that would make doing a task later (compared to now) appear as much less effortful but not much less rewarding.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-022-33119-w

Cites

64 of 64 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Behavioral tasks. Successive screens displayed in one trial are shown from left to right, with durations in ms. a Rating task. For each reward, effort and punishment (not shown), participants indicated on a keyboard the quantity that had the same subjective benefit (or subjective cost) than earning (or loosing) 1€ and 5€. b Intertemporal choice task. Participants first observed the two options shown successively and then indicated their preference by pressing one of two buttons with the left or right hand. The presentation order of sooner and later options was counterbalanced across trials. The task was divided into blocks of intertemporal choices between two rewards, two efforts, or two punishments (not shown). c ‘Now/tomorrow’ choice task. Participants were presented with an option combining reward and effort items. Then they indicated whether they preferred to exert the effort “Now”, and obtain the reward immediately, or “Tomorrow”, and obtain the reward the next day. The side of presentation of the “Now” and “Tomorrow” options was counterbalanced across trials. d ‘Form-filling’ home task. Participants were given 10 administrative forms, such as a passport renewal form. They had to fill in the forms and send a numeric copy via email within a time limit of 30 days, in order to receive their financial compensation for participating in the study. They were told that no compensation would be transferred after the deadline. e Experimental schedule. Tasks were performed in the alphabetic order. Only the intertemporal choice task was performed in the MRI scanner.
Fig. 1 Behavioral tasks. Successive screens displayed in one trial are shown from left to right, with durations in ms. a Rating task. For each reward, effort and punishment (not shown), participants indicated on a keyboard the quantity that had the same subjective benefit (or subjective cost) than earning (or loosing) 1€ and 5€. b Intertemporal choice task. Participants first observed the two options shown successively and then indicated their preference by pressing one of two buttons with the left or right hand. The presentation order of sooner and later options was counterbalanced across trials. The task was divided into blocks of intertemporal choices between two rewards, two efforts, or two punishments (not shown). c ‘Now/tomorrow’ choice task. Participants were presented with an option combining reward and effort items. Then they indicated whether they preferred to exert the effort “Now”, and obtain the reward immediately, or “Tomorrow”, and obtain the reward the next day. The side of presentation of the “Now” and “Tomorrow” options was counterbalanced across trials. d ‘Form-filling’ home task. Participants were given 10 administrative forms, such as a passport renewal form. They had to fill in the forms and send a numeric copy via email within a time limit of 30 days, in order to receive their financial compensation for participating in the study. They were told that no compensation would be transferred after the deadline. e Experimental schedule. Tasks were performed in the alphabetic order. Only the intertemporal choice task was performed in the MRI scanner.CC BY · PubMed Central
Fig. 2 Time preferences. a Intertemporal choices. Plots show patient choice rate (preference for the delayed option), as a function of the difference between option values /costs (top row) or option delays (bottom row). Choices were made between two rewards (blue), two efforts (red) or two punishments (yellow). The two options of a choice differed only in delay and value / cost (quantity converted into euros based on subjective ratings). Note that value represents a gain for reward and cost a loss for effort and punishment. In Exp. 1 (n = 8 participants), rewards (dark blue) were to be received and efforts (dark red) were to be exerted at the specified date. In Exp. 2 (n = 43 participants), rewards (light blue) were to be received at the specified date, but efforts (light red) and punishments (yellow) could be exerted or endured at any time before the specified deadline. On each dot plot, the color mark indicates the mean, the horizonal line indicates the median, the thick whiskers indicate the range from 25th to 75th percentiles, and the thin whiskers indicate the range from 5th to 95th percentiles. b Temporal discounting of reward (left), effort (middle), and punishment (right). The plots display individual discount curves for reward (thin blue lines), effort (thin red lines), and punishment (thin orange lines), as well as the population means (bold lines) and medians (dashed lines). c Accuracy of model fitting. Plots show correlations between modeled and observed choice probability for delayed rewards, delayed efforts, and delayed punishments. Individual choices were divided into 8 bins of increasing modeled probability; each dot represents modeled and observed choice probability averaged within one bin. The within-subject (trial-by-trial) fit can be assessed with the distribution of balanced accuracy across participants. Balanced accuracy is the average of prediction accuracy calculated separately for the two types of choices (now or later). d Temporal discount rates for reward (kR), effort (kE) and punishment (kP). Significance values are based on two-tailed paired t tests (n = 43). e Proportion of patient choices for the different categories of rewards, efforts and punishments. Error bars are inter-subject standard error of the mean (n = 43). Source data are provided as a Source Data file.
Fig. 2 Time preferences. a Intertemporal choices. Plots show patient choice rate (preference for the delayed option), as a function of the difference between option values /costs (top row) or option delays (bottom row). Choices were made between two rewards (blue), two efforts (red) or two punishments (yellow). The two options of a choice differed only in delay and value / cost (quantity converted into euros based on subjective ratings). Note that value represents a gain for reward and cost a loss for effort and punishment. In Exp. 1 (n = 8 participants), rewards (dark blue) were to be received and efforts (dark red) were to be exerted at the specified date. In Exp. 2 (n = 43 participants), rewards (light blue) were to be received at the specified date, but efforts (light red) and punishments (yellow) could be exerted or endured at any time before the specified deadline. On each dot plot, the color mark indicates the mean, the horizonal line indicates the median, the thick whiskers indicate the range from 25th to 75th percentiles, and the thin whiskers indicate the range from 5th to 95th percentiles. b Temporal discounting of reward (left), effort (middle), and punishment (right). The plots display individual discount curves for reward (thin blue lines), effort (thin red lines), and punishment (thin orange lines), as well as the population means (bold lines) and medians (dashed lines). c Accuracy of model fitting. Plots show correlations between modeled and observed choice probability for delayed rewards, delayed efforts, and delayed punishments. Individual choices were divided into 8 bins of increasing modeled probability; each dot represents modeled and observed choice probability averaged within one bin. The within-subject (trial-by-trial) fit can be assessed with the distribution of balanced accuracy across participants. Balanced accuracy is the average of prediction accuracy calculated separately for the two types of choices (now or later). d Temporal discount rates for reward (kR), effort (kE) and punishment (kP). Significance values are based on two-tailed paired t tests (n = 43). e Proportion of patient choices for the different categories of rewards, efforts and punishments. Error bars are inter-subject standard error of the mean (n = 43). Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 3 Neural correlates of time preferences. a Statistical parametric map (SPM) of discounted reward (chosen minus unchosen option discounted value), discounted effort and discounted punishment (chosen minus unchosen option discounted cost) during intertemporal choice, at the time of deliberation. Positive and negative effects are shown with yellow-red and green-blue color codes. b Regression coefficients (betas) of discounted reward (blue), discounted effort (red), and discounted punishment (orange) extracted from anatomically-defined regions of interest (see methods). Error bars are inter-subject standard error of the mean; significance values are based on two-tailed one-sample t tests (n = 27). Source data are provided as a Source Data file c Conjunction between the negative contrast for discounted reward and positive contrast for discounted effort, at the time of deliberation.
Fig. 3 Neural correlates of time preferences. a Statistical parametric map (SPM) of discounted reward (chosen minus unchosen option discounted value), discounted effort and discounted punishment (chosen minus unchosen option discounted cost) during intertemporal choice, at the time of deliberation. Positive and negative effects are shown with yellow-red and green-blue color codes. b Regression coefficients (betas) of discounted reward (blue), discounted effort (red), and discounted punishment (orange) extracted from anatomically-defined regions of interest (see methods). Error bars are inter-subject standard error of the mean; significance values are based on two-tailed one-sample t tests (n = 27). Source data are provided as a Source Data file c Conjunction between the negative contrast for discounted reward and positive contrast for discounted effort, at the time of deliberation.CC BY · PubMed Central
Fig. 4 Explaining procrastination with time preferences. a Procrastination behavior in the lab ‘now/tomorrow’ choice task. Plots show choice rates for the ‘tomorrow’ option as a function of reward value (left), effort cost (middle), or the task net value (right). On each dot plot, the color mark indicates the mean, the horizonal line indicates the median, the thick whiskers indicate the range from 25th to 75th percentiles, and the thin whiskers indicate the range from 5th to 95th percentiles. (n = 43). Source data are provided as a Source Data file b Procrastination behavior in the at-home ‘form-filling’ task. Histograms show the distribution of the delay before participants completed and returned administrative forms. c Regression estimates of the temporal discount rates for reward (kR), effort (kE) and punishment (kP) obtained from fitting a linear model (also including age and gender) to procrastination level in the ‘now/tomorrow’ choice task. Error bars represent SD; significance values are based on two-tailed one-sample t tests (n = 43). d Regression estimates of reward, effort and punishment discount rates obtained from fitting a linear model (also including age and gender) to procrastination level in the ‘form-filling’ home task. Error bars represent SD; significance values are based on two-tailed one-sample t tests (n = 37). Note that in both cases (in lab or at home), discount rates are inferred from intertemporal choices, which were observed independently from the procrastination behavior that they contribute to explain. e Accuracy of model fitting. The plot shows the correlation between modeled and observed choice frequency for the ‘tomorrow’ option. Individual choices were divided into 8 bins of increasing modeled frequency; each dot represents modeled and observed choice frequency averaged within one bin for one participant. The within-subject (trial-by-trial) fit can be assessed with the distribution of individual balanced accuracy (inset). Balanced accuracy is the average of prediction accuracy calculated separately for the two types of choices (now or later).
Fig. 4 Explaining procrastination with time preferences. a Procrastination behavior in the lab ‘now/tomorrow’ choice task. Plots show choice rates for the ‘tomorrow’ option as a function of reward value (left), effort cost (middle), or the task net value (right). On each dot plot, the color mark indicates the mean, the horizonal line indicates the median, the thick whiskers indicate the range from 25th to 75th percentiles, and the thin whiskers indicate the range from 5th to 95th percentiles. (n = 43). Source data are provided as a Source Data file b Procrastination behavior in the at-home ‘form-filling’ task. Histograms show the distribution of the delay before participants completed and returned administrative forms. c Regression estimates of the temporal discount rates for reward (kR), effort (kE) and punishment (kP) obtained from fitting a linear model (also including age and gender) to procrastination level in the ‘now/tomorrow’ choice task. Error bars represent SD; significance values are based on two-tailed one-sample t tests (n = 43). d Regression estimates of reward, effort and punishment discount rates obtained from fitting a linear model (also including age and gender) to procrastination level in the ‘form-filling’ home task. Error bars represent SD; significance values are based on two-tailed one-sample t tests (n = 37). Note that in both cases (in lab or at home), discount rates are inferred from intertemporal choices, which were observed independently from the procrastination behavior that they contribute to explain. e Accuracy of model fitting. The plot shows the correlation between modeled and observed choice frequency for the ‘tomorrow’ option. Individual choices were divided into 8 bins of increasing modeled frequency; each dot represents modeled and observed choice frequency averaged within one bin for one participant. The within-subject (trial-by-trial) fit can be assessed with the distribution of individual balanced accuracy (inset). Balanced accuracy is the average of prediction accuracy calculated separately for the two types of choices (now or later).CC BY · PubMed Central
Fig. 5 Explaining procrastination with neural correlates of time preferences. a Statistical parametric maps (SPM) of delay (chosen minus unchosen delay), – (undiscounted) reward value and + (undiscounted) effort cost, and the conjunction between the two regressors, at the time of deliberation. b Correlation between behavioral and neural estimates of the weight on delay (temporal discount rates in choice model versus delay regression estimates in dmPFC activity). Source data are provided as a Source Data file. c Regression estimates of beta weights on dmPFC activity for delay and (undiscounted) value or cost of reward, effort, and punishment, obtained from fitting a linear model to procrastination level observed either in the lab choice task (preference for ‘tomorrow’ options) (n = 27) or at home (delay in returning administrative forms) (n = 23). Note that neural activity was recorded during intertemporal choices, independently from the procrastination behavior assessed in the lab or at home. Error bars represent SD; significance values are based on two-tailed one-sample t tests.
Fig. 5 Explaining procrastination with neural correlates of time preferences. a Statistical parametric maps (SPM) of delay (chosen minus unchosen delay), – (undiscounted) reward value and + (undiscounted) effort cost, and the conjunction between the two regressors, at the time of deliberation. b Correlation between behavioral and neural estimates of the weight on delay (temporal discount rates in choice model versus delay regression estimates in dmPFC activity). Source data are provided as a Source Data file. c Regression estimates of beta weights on dmPFC activity for delay and (undiscounted) value or cost of reward, effort, and punishment, obtained from fitting a linear model to procrastination level observed either in the lab choice task (preference for ‘tomorrow’ options) (n = 27) or at home (delay in returning administrative forms) (n = 23). Note that neural activity was recorded during intertemporal choices, independently from the procrastination behavior assessed in the lab or at home. Error bars represent SD; significance values are based on two-tailed one-sample t tests.CC BY · PubMed Central
Fig. 6 From static to dynamic computational model of procrastination. a Static model. Effort cost and reward value are both discounted with time, and reward value vanishes after the deadline. The predicted delay for task completion (i.e., procrastination duration) is the one that maximizes the net value function (green dot). Simulations of procrastination duration under the static model shows that it mostly occurs with low temporal discount rates, even lower for reward than for effort b Dynamic model. Every day, the net value of completing the task now is compared to all other remaining available dates (white dots) through a softmax function that provides choice probabilities. As time goes by and the deadline gets closer, the comparison includes fewer dates, which increases the probability of performing the task. This probability obtained for each day is added to the probability of having completed the task on every past day, to generate a cumulative probability. The predicted delay of task completion is the expected value under this cumulative distribution. Simulations of procrastination duration under the dynamic model shows that it occurs with a wide range of temporal discount rates, provided that they are lower for reward than for effort. c Daily and cumulative probability of task completion across time, under the dynamic model. Each curve represents one participant. d Observed procrastination duration plotted against temporal discount rates for reward (kR) and effort (kE). Each dot is a participant. Red bars represent the means, and whiskers the 25th and 75th percentiles, in three bins of equal size. e Inter-participant one-tailed Pearson’s correlation between observed and modeled procrastination duration, under the dynamic model, based on behavioral data or neural data. Each dot is a participant. The number of dots corresponds to the number of participants who did send back the completed forms (n = 37 for behavioral data and n = 23 for fMRI data). Source data are provided as a Source Data file.
Fig. 6 From static to dynamic computational model of procrastination. a Static model. Effort cost and reward value are both discounted with time, and reward value vanishes after the deadline. The predicted delay for task completion (i.e., procrastination duration) is the one that maximizes the net value function (green dot). Simulations of procrastination duration under the static model shows that it mostly occurs with low temporal discount rates, even lower for reward than for effort b Dynamic model. Every day, the net value of completing the task now is compared to all other remaining available dates (white dots) through a softmax function that provides choice probabilities. As time goes by and the deadline gets closer, the comparison includes fewer dates, which increases the probability of performing the task. This probability obtained for each day is added to the probability of having completed the task on every past day, to generate a cumulative probability. The predicted delay of task completion is the expected value under this cumulative distribution. Simulations of procrastination duration under the dynamic model shows that it occurs with a wide range of temporal discount rates, provided that they are lower for reward than for effort. c Daily and cumulative probability of task completion across time, under the dynamic model. Each curve represents one participant. d Observed procrastination duration plotted against temporal discount rates for reward (kR) and effort (kE). Each dot is a participant. Red bars represent the means, and whiskers the 25th and 75th percentiles, in three bins of equal size. e Inter-participant one-tailed Pearson’s correlation between observed and modeled procrastination duration, under the dynamic model, based on behavioral data or neural data. Each dot is a participant. The number of dots corresponds to the number of participants who did send back the completed forms (n = 37 for behavioral data and n = 23 for fMRI data). Source data are provided as a Source Data file.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

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

The self-check needs the paper’s full text.

Citations

Cites

Show all 64 references

Research works printed but not linked: 8

  1. Akerlof, G. A. Procrastination and obedience. Am. Econ. Rev. 81, 1–19 (1991).
  2. O’Donoghue, T. & Rabin, M. in Behavioral Dimensions of Retirement Economics (ed. Henry Aaron) (Brookings Institution Press, 1999).
  3. Gustavson, D., Miyake, A., Hewitt, J. & Friedman, N. Genetic relations among procrastination, impulsivity, and goal-management ability. Behav. Genet. 44, 661–661 (2014).
  4. Ainslie, G. Picoeconomics: The strategic interaction of successive motivational states within the person. (Cambridge University Press, 1992).
  5. Mazur, J. E. An adjusting procedure for studying delayed reinforcement. In: Commons M.L., Mazur J.E., Nevin J.A. (eds) 55–73 (Psychology Press, 1987).
  6. Soman, D. The illusion of delayed incentives: evaluating future effort–money transactions. J. Mark. Res. 35, 427–437 (1998).
  7. Murphy, J. G., Vuchinich, R. E. & Simpson, C. A. Delayed reward and cost discounting. Psychol. Rec. 51, 5 (2001).
  8. Ferrari, J. R. & Emmons, R. A. Methods of procrastination and their relation to self-control and self-reinforcement - an exploratory-study. J. Soc. Behav. Pers. 10, 135–142 (1995).

Cited by

See all 44

Metadata

Authors
Le Bouc, R., & Pessiglione, M.
Affiliations
Centre National de la Recherche Scientifique (FR)
Venue
Nature Communications
Published
2022-09-26
Volume
13
Issue
1
Pages
5639-5639
ISSN
2041-1723 (ISSN-L)
OpenAlex
W4297222346
Cited by
44
References
64
Citations by year
2026: 18 · 2025: 9 · 2024: 8 · 2023: 9
FWCI
5.93
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • HAL (submittedVersion, other-oa)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Fondation pour la Recherche Médicale
Topics
  • Social Sciences › Psychology › Clinical Psychology › Perfectionism, Procrastination, Anxiety Studies
  • Social Sciences › Decision Sciences › General Decision Sciences › Decision-Making and Behavioral Economics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural and Behavioral Psychology Studies
Keywords
procrastination, intertemporal choice, computational model, effort cost, reward discounting, delay discounting, dorsomedial prefrontal cortex, cognitive biases, expected value
MeSH terms
Procrastination, Choice Behavior, Humans, Reward, Time, Prefrontal Cortex

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.

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: A neuro-computational account of procrastination behavior. https://app.cochat.ai/papers/10.1038/s41467-022-33119-w
The paper
Le Bouc, R., & Pessiglione, M. (2022). A neuro-computational account of procrastination behavior. Nature Communications. https://doi.org/10.1038/s41467-022-33119-w

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.