Bastien Blain

Ongoing / under review

• 2026

  PsyArXiv

  Controllable progress as an intrinsic value signal in cognitive control

  Machado Sanchez, S., de Gardelle, V.*, Blain, B.*

  Shows that people keep exerting effort and find progress rewarding mainly when they feel they caused it — controllability gates the motivational value of improvement.

  AbstractPreprint

  Why do people persist in demanding activities even when immediate rewards are modest? We propose that the answer lies in the interaction between progress (improvement in performance) and controllability (how much one's efforts and actions shape events): controllability amplifies the intrinsic value of progress by letting individuals attribute improvement to their own effort. Across two experiments (a pilot and a preregistered replication), progress increased task enjoyment, led to more effort, and drove task preferences exclusively when controllability was high. Controllability therefore gates the motivational value of progress, providing a computational principle for why people remain engaged in challenging activities.

• 2026

  Research Square

  Risk escalation is amplified by stakes, not by a sense of control

  Haj-Ali, H., Glickman, M., Blain, B., Sharot, T.

  Shows that merely repeating risky financial choices gradually increases risk-taking, and that higher stakes — not a sense of control — drive how fast this escalation happens.

  AbstractPreprint

  People are often faced with the same risky choices repeatedly. Isolating the effect of mere repetition from outcome learning, across three experiments participants made repeated financial wagers on positive-expected-value lotteries. Initially they wagered far below the risk-neutral optimum, but with repetition risk-taking gradually escalated. Higher stakes and agency both boosted risk-taking, yet only stakes shaped the trajectory of escalation, with steeper escalation emerging exclusively at higher reward potential.

Published

• 2023

  Nature Mental Health

  Sensitivity to intrinsic rewards is domain general and related to mental health

  Blain, B., Pinhorn, I., Sharot, T.

  Demonstrates that how strongly people value intrinsic rewards (curiosity, competence, social contact) is a single general trait that is linked to mental health.

  AbstractArticle

  Humans frequently engage in intrinsically rewarding activities (for example, consuming art, reading). We show that sensitivity to intrinsic rewards is domain general and associated with mental health. Participants online (N = 483) were presented with visual, cognitive and social intrinsic rewards as well as monetary rewards and neutral stimuli; all rewards were liked, wanted and reinforcing. Factor analysis revealed that ~40% of response variance was explained by a general sensitivity to all rewards, and affective aspects of mental health were associated with sensitivity to intrinsic, but not monetary, rewards.

• 2021

  J. Neurosci.

  A neurocomputational model for intrinsic reward

  Chew, B.*, Blain, B.*, Dolan, R. J., Rutledge, R. B.

  Introduces a computational 'hedonometer' showing that intrinsic rewards from skilled performance shape momentary happiness and engage the brain's valuation system (vmPFC).

  AbstractArticle

  We introduce a computational tool that measures the affective value of experiences, validated with fMRI while subjects performed a reinforcement learning task with periodic ratings of subjective affective state. Learning performance determined payment (extrinsic reward), while a skilled-performance component (intrinsic reward) did not. Both influenced affective dynamics, and individuals for whom intrinsic rewards mattered more had greater ventromedial prefrontal activity for intrinsic than extrinsic rewards — a 'computational hedonometer' indexing the subjective value of intrinsic relative to extrinsic rewards.

• 2021

  Curr. Opin. Behav. Sci.

  Potential mechanism of intrinsic rewards

  Blain, B., Sharot, T.

  Proposes that activities feel intrinsically rewarding because they raise our sense of self-efficacy, giving a single mechanism behind curiosity, play, art and mastery.

  AbstractArticle

  From child play to scientific discovery, many activities are rewarding in and of themselves. We propose the answer is an increased sense of self-efficacy: an intrinsically rewarding activity strengthens a person's belief that they can execute the actions required to deal with prospective situations. This can explain the rewarding nature of activities from crosswords to helping others, art and sport; processes that increase self-efficacy, such as executing agency and learning, activate the neural reward system.

• 2020

  eLife

  Momentary subjective well-being depends on learning and not reward

  Blain, B., Rutledge, R.

  Demonstrates that moment-to-moment happiness tracks how much we learn about our environment rather than the rewards we actually receive.

  AbstractArticle

  Momentary happiness is associated with reward prediction error. We tested subjects in a reinforcement learning task in which reward size and probability were uncorrelated, dissociating reward from learning. Happiness was sensitive to learning-relevant variables (probability prediction error) but not to learning-irrelevant ones (reward prediction error); depressive symptoms reduced happiness more in volatile than stable environments. How we learn about our world may be more important for how we feel than the rewards we actually receive.

Ongoing / under review

• 2024

  Under review

  Opposed mood dynamics of depression and anxiety are related to reward prediction error

  Xu, P., Wang, Z., Wang, T., Nan, T., Xu, J., Aleman, A., Luo, Y., Blain, B., Liu, Y.

  Shows that depression and anxiety are linked to opposite ways in which mood reacts to reward prediction errors.

  AbstractPreprint

  Using computational models of mood, we find that depression and anxiety are associated with opposite relationships between momentary mood and reward prediction errors.

• 2024

  Under review

  A computational model for the impact of altruism on happiness depending on social preference

  Wang, Z., Shimura, Y., Vaghi, M., Rutledge*, R., Blain, B.*

  Shows that giving boosts happiness only when it matches a person's own social preferences, and that several small donations increase happiness more than one large gift.

  AbstractPreprint

  Helping others is widely believed to be a route to happiness, but the relevance of individual social preferences remains unclear. Across two experiments (total N = 235), greater generosity was not associated with greater happiness; however, the outcomes of altruistic actions had a greater hedonic impact in more generous individuals. Computational modelling showed happiness was predicted by prediction errors weighted toward actions aligned with a person's social preferences, and outcome size did not matter — a prediction confirmed in a third experiment (N = 198): three separate small donations increased happiness more than one equivalent gift.

Published

• 2025

  eLife

  Mood computational mechanisms underlying increased risk behavior in adolescent suicidal patients

  Wang, Z., Nan, T., Wang, T., Liu, Y., Lu, F., Yu, Y., Cai, X., He, Z., Luo, Y., Blain, B.

  Identifies computational mood and decision markers that distinguish adolescent patients with suicidal behaviour and predict symptom severity.

  AbstractArticle

  Among 83 adolescent inpatients with affective disorders (58 with suicidal thoughts and behaviours, S+; 25 without, S−) and 118 healthy controls, S+ showed greater risk-taking. Computational modelling attributed this to an elevated approach parameter, while mood-model analyses revealed reduced sensitivity to certain rewards in S+. These signatures predicted suicidal symptom severity and generalized to an independent sample (n = 747).

• 2025

  J. Risk & Uncertainty

  Adapting temporal preference to scarcity: a role for emotion?

  Blain, B., Globig, L., Sharot, T.

  Demonstrates, using a real income shock, that scarcity makes people prefer immediate rewards independently of their emotional state.

  AbstractArticle

  Taking advantage of one of the largest global 'income shocks' in history, we tested 1,145 individuals as the market crashed in March 2020 and retested 200 as it recovered in June 2020. Income shock was strongly related to an increase in delay discounting in both cross-sectional and longitudinal data, and this relationship was independent of the negative impact on affect — suggesting people adapt discounting directly to environmental constraints, without input from the affective system.

• 2024

  Scientific Reports

  “How” web searches change under stress

  Kelly, C., Blain, B., Sharot, T.

  Shows that stress changes the questions people ask online — increasing action-oriented 'How' searches — so search patterns can track population stress levels.

  AbstractArticle

  In response to stressful public and private events, the high-level features of information people seek online change. People selectively ask 'How' questions when they want information to guide action; 'How' searches on Google increased during the pandemic and their proportion predicted weekly self-reported stress of ~17K individuals, an effect reproduced by experimentally manipulating stress.

• 2022

  J. Risk & Uncertainty

  Perceptions of personal and public risk: dissociable effects on behavior and well-being

  Globig, L., Blain, B., Sharot, T.

  Shows that personal risk perception mainly affects happiness, whereas public risk perception mainly drives protective behaviour.

  AbstractDOI

  Surveying a large representative sample of Americans during the COVID-19 pandemic at two times (N1 = 1145, N2 = 683), people perceived their own risk as relatively low while estimating risk to others as high. Perceived personal but not public risk was associated with happiness, while both predicted anxiety; protective behaviours were predicted by estimated risk to the population, not to oneself, suggesting people acted mainly for the benefit of others.

• 2022

  PLOS Comput. Biol.

  Observing others give & take: a computational account of bystanders' feelings and actions

  Blain, B.*, Marks, J.*, Czech, P., Sharot, T.

  Provides a computational account of how merely watching others act selfishly or unfairly shapes bystanders' emotions and their willingness to punish.

  AbstractArticleCode

  A person's well-being is influenced not only by interactions they experience but also by those they observe. We develop computational models relating others' (un)selfish acts to observers' emotional reactions and non-costly punishment decisions; these predict reactions in out-of-sample participants and highlight two social values — 'selfishness aversion' and 'inequality aversion' — with even small violations from equality having a disproportionately large impact.

• 2021

  J. Neurosci.

  A neurocomputational model for intrinsic reward

  Chew, B.*, Blain, B.*, Dolan, R. J., Rutledge, R. B.

  Introduces a computational 'hedonometer' showing that intrinsic rewards from skilled performance shape momentary happiness and engage the brain's valuation system (vmPFC).

  AbstractArticle

  We introduce a computational tool that measures the affective value of experiences, validated with fMRI while subjects performed a reinforcement learning task with periodic ratings of subjective affective state. Learning performance determined payment (extrinsic reward), while a skilled-performance component (intrinsic reward) did not. Both influenced affective dynamics, and individuals for whom intrinsic rewards mattered more had greater ventromedial prefrontal activity for intrinsic than extrinsic rewards — a 'computational hedonometer' indexing the subjective value of intrinsic relative to extrinsic rewards.

• 2020

  eLife

  Momentary subjective well-being depends on learning and not reward

  Blain, B., Rutledge, R.

  Demonstrates that moment-to-moment happiness tracks how much we learn about our environment rather than the rewards we actually receive.

  AbstractArticle

  Momentary happiness is associated with reward prediction error. We tested subjects in a reinforcement learning task in which reward size and probability were uncorrelated, dissociating reward from learning. Happiness was sensitive to learning-relevant variables (probability prediction error) but not to learning-irrelevant ones (reward prediction error); depressive symptoms reduced happiness more in volatile than stable environments. How we learn about our world may be more important for how we feel than the rewards we actually receive.

Published

• 2025

  Trends Cogn. Sci.

  Origins and consequences of cognitive fatigue

  Pessiglione, M., Blain, B., Wiehler, A., Naik, S.

  Argues that cognitive fatigue is best measured not by self-report but by a shift toward low-effort, immediate-reward choices, and links it to brain metabolism via the MetaMotiF model.

  AbstractArticle

  A review of the origins and consequences of cognitive fatigue. We argue that fatigue is poorly captured by introspection, self-report or performance decline, and that a preference for low-cost options in economic choice is a more reliable marker. We propose the MetaMotiF model, in which cognitive fatigue emerges from metabolic alterations in cognitive-control brain regions following their excessive mobilization, and in turn biases motivation.

• 2019

  Current Biology

  Physical overtraining and decision-making: evidence for cognitive control depletion

  Blain, B., Schmit, C., Le Meur, Y., Hausswirth, C., Pessiglione, M.

  Shows that endurance overtraining, like prolonged mental work, makes economic choices more impulsive and lowers lateral-prefrontal activity — evidence for a shared cognitive-control fatigue.

  AbstractArticle

  We induced a mild form of overtraining in endurance athletes and compared them to normally trained athletes during fMRI. Training overload enhanced impulsivity in economic choice — a bias favouring immediate over delayed rewards — and diminished lateral prefrontal activation during choice, providing causal evidence for a functional link between physical exercise and cognitive control.

• 2016

  PNAS

  Neural mechanisms underlying the impact of daylong cognitive work on economic decisions

  Blain, B., Hollard, G., Pessiglione, M.

  Demonstrates that a full day of demanding cognitive work makes people more impulsive by depleting activity in a specific lateral-prefrontal brain region.

  AbstractArticle

  Targeting the lateral prefrontal cortex (LPFC) as a substrate for cognitive control, we prolonged fatigue induction to more than 6 h (a workday). Choice impulsivity in intertemporal choices increased in participants who performed hard executive tasks, but not in controls who did easy versions or enjoyed leisure; fMRI linked this to a specific decrease in activity of a left middle-frontal-gyrus LPFC region recruited by both executive and choice tasks.

Published

• 2023

  Decision

  Discounting future reward in an uncertain world

  Story, G. W., Kurth-Nelson, Z., Moutoussis, M., Iigaya, K., Will, G.-J., Hauser, T. U., Blain, B., Vlaev, I., Dolan, R. J.

  Demonstrates that people discount future rewards more when those rewards are volatile, linking impatience to uncertainty via reduced hippocampal–prefrontal coupling.

  AbstractArticleDOI

  We propose that reward is discounted proportional to the rate of random change in its magnitude across time, termed volatility. Across three experiments (total N = 158) discounting increased with volatility, held over delays of up to 4 months, and fMRI showed a volatility-dependent decrease in hippocampal–prefrontal coupling during intertemporal choice.

• 2018

  Psychopharmacology

  Dual contributions of noradrenaline to behavioural flexibility and motivation

  Jahn, C. I., Gilardeau, S., Varazzani, C., Blain, B., Sallet, J., Walton, M. E., Bouret, S.

  Shows, in monkeys, that noradrenaline separately controls behavioural flexibility and the mobilization of physical effort.

  AbstractPubMed

  Using a novel sequential cost/benefit decision task in rhesus monkeys, we manipulated noradrenaline with clonidine. It had two distinct effects: it decreased choice variability without affecting the cost/benefit trade-off, and reduced force production without modulating the willingness to work — supporting an overarching role for noradrenaline in facing challenging situations.