About

Fiery Cushman is a Professor of Psychology at Harvard University and the Director of the Moral Psychology Research Laboratory. His research aims to organize the complexity of moral judgment around basic functional principles, motivated by the idea that morality serves to teach others how to behave and to learn appropriate patterns of behavior through punishments and rewards. His laboratory investigates these issues using a variety of methods, including surveys, laboratory behavioral studies, psychophysiology, infant and child research, functional neuroimaging, economic games, and formal modeling. His ultimate goal is to use the moral domain to understand broader phenomena such as the balance between learned and innate contributions to cognition, the human capacity to explain, predict, and evaluate others’ behavior, the relationship between automaticity and control, and the architecture of learning and decision-making in social contexts. Cushman received his BA and PhD from Harvard University, where he also completed a post-doctoral fellowship. He served as an Assistant Professor at Brown University from 2011 to 2014. His research has been recognized with awards including the Stanton Prize by the Society for Philosophy and Psychology and the Daniel M. Wegner Theoretical Innovation Prize by the Society for Personality and Social Psychology. His teaching excellence was acknowledged with the Henry Merrit Wriston Fellowship at Brown University. His research has been supported by the National Science Foundation, the Office of Naval Research, and the John Templeton Foundation.

Research topics

• Computer Science
• Social psychology
• Cognitive psychology
• Artificial Intelligence
• Psychology
• Epistemology
• Cognitive science
• Microeconomics

Selected publications

• Balancing precedent and mutual benefit in tacit coordination

  PsyArXiv (OSF Preprints) · 2026-03-16

  preprintOpen access

  Human coordination depends on two complementary mechanisms: forward-looking strategies that enable flexible adaptation to new circumstances, and backward-looking mechanisms that rely on precedent, convention, and rule-following. Most cognitive and computational models of coordination emphasize one mechanism or the other—either explaining how equilibria emerge and persist when agents adapt their behavior based on past experience, or how agents creatively generate novel solutions and strategies to achieve anticipated mutual benefit in the challenges of the moment—but not how the two interact. Here we introduce a cognitive model and experimental paradigm to capture the dynamics of both processes and, crucially, the arbitration between them. In two preregistered experiments (n = 510; 30,420 choices), participants repeatedly solve coordination problems that can be addressed either by generalizing past solutions or by adopting novel ones when precedent becomes inefficient. This design allows us to examine the conditions under which individuals or dyads decide to abandon entrenched equilibria and transition to novel coordination solutions by arbitrating between mutual benefit and precedent. By formally modeling both forward- and backward-looking mechanisms, and the process of arbitration between them, we provide a unified framework for understanding how human coordination can be both stable and adaptable—a property that underlies everyday cooperative behavior, social norms, and institutional evolution.

  Publisher

• Balancing precedent and mutual benefit in human coordination

  2026-04-07

  articleOpen accessSenior author

  Human societies rely on shared conventions that allow people to coordinate efficiently, from the conversational etiquette and the “rules of the road” to workplace roles and the handshake economy. To be effective, the conventions must be reliable, and yet also adaptable: when circumstances change, people must decide whether to follow established precedents or shift to new ways of coordinating. Most cognitive and computational models explain either how coordination equilibria emerge and persist through learning from past experience, or how agents generate novel strategies to maximize anticipated mutual benefit. How people arbitrate between these two modes of coordination remains poorly understood. Here we introduce a computational model and experimental paradigm to study this arbitration. In two preregistered experiments (n = 510; 30,420 choices), participants repeatedly solved coordination problems in which established precedents could be followed or abandoned as their efficiency declined. Our results quantify the conditions under which individuals and pairs transition from entrenched coordination equilibria to new shared solutions, revealing a dynamic trade-off between the reliability of precedent and the potential gains from innovation. These findings provide a unified account of how coordination remains both stable and adaptable, offering new insights into the evolution of conventions, norms, and institutions.

  PublisherDOI

• Mutual observability determines when precedent is moralized in multi-agent coordination games

  2026-03-12

  articleOpen access

  Moral cognition evolved, in part, to facilitate cooperation. Since coordination problems constitute a large subclass of cooperation problems, we should expect coordination behavior to be subject to moralization. But which of several coordination strategies is preferable depends on their respective probability of success. This, in turn, often depends on whether other people’s actions are mutually observable. Past research has focused on the moralization of coordination behavior in dyadic contexts. In two preregistered experiments (n = 472) based on incentivized real-time coordination games, we extend investigations of the moralization of coordination behavior to small groups, and study how it depends on the mutual observability of others’ actions. In Experiment 1, we show that moving away from the current coordination solution (thus inducing miscoordination) is viewed as morally worse than sticking to it. In Experiment 2, we show that this is not the case when actions are mutually observable, making switching to a more mutually beneficial alternative achievable. In such contexts, abandoning precedent to coordinate on a more efficient equilibrium is viewed as morally superior than sticking to it. Crucially, the same coordination rule is viewed as morally superior in one informational context, and morally inferior in the other. Understanding the cognitive processes involved in successful coordination can shed light on which conventions, norms, and rules are morally appropriate and which aren’t, when this is so, and why.

  PublisherDOI

• Philosophical arguments can boost charitable giving

  PsyArXiv (OSF Preprints) · 2026-02-26

  preprintOpen access1st authorCorresponding

  Do reasoned arguments increase charitable giving? Evidence is mixed. One possible explanation is that arguments vary in effectiveness. We crowdsourced 90 philosophical arguments in favor of charitable giving, coding them for a variety of features. In Study 1 (N=2161, MTurk), participants read either a control text or one of five arguments preselected as especially promising. Participants who read the philosophical arguments expressed more positive attitudes towards donating and donated about $1 more on average of a potential $10 bonus than participants who read the control text. Study 2 (N=8982, MTurk) employed all 90 arguments, with each participant reading one argument. We analyzed the relationship between the coded argument features and donation behavior, and found arguments to be more effective when they mentioned children, mentioned the specific impacts of donating, and/or emphasized that large benefits can be achieved at low cost. Among the features less predictive of donation: addressing counterarguments, mentioning the equivalence of persons or the participant’s own assumed good fortune, appealing to religion or expert authorities, or appealing to the participant’s self-interest. Together, these studies provide evidence that reasoned arguments can indeed shift moral behavior such as charitable giving, but that some arguments are simply more effective than others. Further, we identify specific argument features that are most and least effective in eliciting donations.

  Publisher

• Mutual observability determines when precedent is moralized in multi-agent coordination games

  PsyArXiv (OSF Preprints) · 2026-03-11

  preprintOpen accessSenior author

  Moral cognition evolved, in part, to facilitate cooperation. Since coordination problems constitute a large subclass of cooperation problems, we should expect coordination behavior to be subject to moralization. But which of several coordination strategies is preferable depends on their respective probability of success. This, in turn, often depends on whether other people’s actions are mutually observable. Past research has focused on the moralization of coordination behavior in dyadic contexts. In two preregistered experiments (n = 472) based on incentivized real-time coordination games, we extend investigations of the moralization of coordination behavior to small groups, and study how it depends on the mutual observability of others’ actions. In Experiment 1, we show that moving away from the current coordination solution (thus inducing miscoordination) is viewed as morally worse than sticking to it. In Experiment 2, we show that this is not the case when actions are mutually observable, making switching to a more mutually beneficial alternative achievable. In such contexts, abandoning precedent to coordinate on a more efficient equilibrium is viewed as morally superior than sticking to it. Crucially, the same coordination rule is viewed as morally superior in one informational context, and morally inferior in the other. Understanding the cognitive processes involved in successful coordination can shed light on which conventions, norms, and rules are morally appropriate and which aren’t, when this is so, and why.

  Publisher

• Balancing precedent and mutual benefit in human coordination

  PsyArXiv (OSF Preprints) · 2026-04-06

  preprintOpen access

  Human societies rely on shared conventions that allow people to coordinate efficiently, from the conversational etiquette and the “rules of the road” to workplace roles and the handshake economy. To be effective, the conventions must be reliable, and yet also adaptable: when circumstances change, people must decide whether to follow established precedents or shift to new ways of coordinating. Most cognitive and computational models explain either how coordination equilibria emerge and persist through learning from past experience, or how agents generate novel strategies to maximize anticipated mutual benefit. How people arbitrate between these two modes of coordination remains poorly understood. Here we introduce a computational model and experimental paradigm to study this arbitration. In two preregistered experiments (n = 510; 30,420 choices), participants repeatedly solved coordination problems in which established precedents could be followed or abandoned as their efficiency declined. Our results quantify the conditions under which individuals and pairs transition from entrenched coordination equilibria to new shared solutions, revealing a dynamic trade-off between the reliability of precedent and the potential gains from innovation. These findings provide a unified account of how coordination remains both stable and adaptable, offering new insights into the evolution of conventions, norms, and institutions.

  Publisher

• A Counterfactual Pathway to Responsibility Judgments in Children

  2026-02-27

  articleOpen accessSenior author

  Judgments of responsibility are foundational to human social life. Adults judge responsibility by considering not just what someone chose to do, but also the alternative choices available to them. Thus, the development of adult-like responsibility judgments in children depends on their capacity to reason about counterfactual choices. We distinguish three key elements of this capacity: first, understanding that counterfactual choices are possible, second, successfully reasoning about counterfactual worlds, and third, identifying the most relevant counterfactuals in any given situation. This framework clarifies the relationship between counterfactual reasoning and responsibility judgments in childhood, identifies conceptually distinct elements of children’s understanding of responsibly, and opens new avenues for inquiry into an essential element of children’s social reasoning.

  PublisherDOI

• Modeling fairness judgments of splits between multiple parties

  2026-03-12

  articleOpen accessSenior author

  Fair divisions are a fundamental problem for moral cognition. Past experimental work has provided evidence in support of three principles of fairness in divisions between two parties: proportional splits, equal splits, and splits that equalize net gains (the Nash bargaining solution). When the number of parties increases, so does the complexity of such decisions, potentially influencing people’s cognitive strategies and their reliance on precise explicit heuristics vs intuitive approximations. We design a novel task in which participants can easily and intuitively sample and select among various distributions of resources between multiple people (2 to 18) by adjusting a continuous slider that updates divisions in real time. In two preregistered experiments (n = 378; 2,268 choices), we quantitatively model participants’ fairness judgments at the individual level. We find that participants can be categorized into three main groups. Overall, around 50% are best fitted by proportionality, 40% by the Nash bargaining solution, and fewer than 10% by equality. At the aggregate level, proportionality and the Nash bargaining solution perform best. Fairness judgments remain stable as the number of involved parties increases. When they only have access to the slider (Experiment 1), participants best fitted by the Nash bargaining solution seem to intuitively approximate it, but in an imprecise way. By contrast, they tend to precisely select it when three buttons (one for each model, Experiment 2) are available to automatically adjust the slider. These results suggest that a substantial proportion of participants rely on intuitive approximations for the Nash bargaining solution, consistent with bargaining-based (contractualist) theories of fairness.

  PublisherDOI

• Balancing precedent and mutual benefit in tacit coordination

  2026-01-14

  articleOpen accessSenior author

  Human coordination depends on two complementary mechanisms: forward-looking strategies that enable flexible adaptation to new circumstances, and backward-looking mechanisms that rely on precedent, convention, and rule-following. Most cognitive and computational models of coordination emphasize one mechanism or the other—either explaining how equilibria emerge and persist when agents adapt their behavior based on past experience, or how agents creatively generate novel solutions and strategies to achieve anticipated mutual benefit in the challenges of the moment—but not how the two interact. Here we introduce a cognitive model and experimental paradigm to capture the dynamics of both processes and, crucially, the arbitration between them. In two preregistered experiments (n = 510; 30,420 choices), participants repeatedly solve coordination problems that can be addressed either by generalizing past solutions or by adopting novel ones when precedent becomes inefficient. This design allows us to examine the conditions under which individuals or dyads decide to abandon entrenched equilibria and transition to novel coordination solutions by arbitrating between mutual benefit and precedent. By formally modeling both forward- and backward-looking mechanisms, and the process of arbitration between them, we provide a unified framework for understanding how human coordination can be both stable and adaptable—a property that underlies everyday cooperative behavior, social norms, and institutional evolution.

  PublisherDOI

• Environmental variability shapes the representational format of cultural learning

  Proceedings of the National Academy of Sciences · 2025-07-07

  articleOpen accessSenior author

  Cumulative culture requires learning mechanisms that are both efficient and flexible in the face of environmental change. We examine models of this learning mechanism that emphasize teaching what to do (causally opaque procedures) and those that foreground what to aim for and why (goals and causal reasoning). Learning procedures is cheap but inflexible; learning goals is more flexible to changing circumstance, but requires expensive individual learning about how to achieve them. In an iterated learning experiment, we demonstrate that cultural learning adapts in precisely this way: Microcultures more often instruct future generations to follow procedures when the world is stable, but they tend to share information about valuable outcomes and causal relations when the world is variable.

  PublisherOA PDFDOI

Recent grants

• Investigating the functional match between punishment and learning.

  NSF · $85k · 2012–2014

Frequent coauthors

• Sydney Levine

  Allen Institute

  55 shared

• Nick Chater

  University of Warwick

  46 shared

• Joshua B. Tenenbaum

  Massachusetts Institute of Technology

  46 shared

• Max Kleiman‐Weiner

  Seattle University

  39 shared

• Jonathan Phillips

  30 shared

• Adam Morris

  28 shared

• Mark K. Ho

  New York University

  27 shared

• Samuel J. Gershman

  Harvard University

  24 shared

Labs

• Cushman LabPI

Awards & honors

• Stanton Prize by the Society for Philosophy and Psychology
• Daniel M. Wegner Theoretical Innovation Prize by the Society…
• Henry Merrit Wriston Fellowship