News

Maria Pykälä will receive a doctoral prize at HEC Lausanne’s upcoming Annual Doctoral Event in November, recognizing outstanding work in the social sciences and humanities. Maria completed her PhD in the Department of Organizational Behavior at HEC Lausanne, University of Lausanne, under the supervision of Prof. Charles Efferson. She defended her dissertation, Social Learning in Complex Systems: Experiments and Simulations, in late 2025.

People often look to what most others are doing for guidance, yet the same majority can be informative in one setting and misleading in another. We examined how people interpret social frequency information when group membership and payoff alignment vary. In an incentivized experiment, demonstrators first learned through private feedback which option maximized their own payoff. Social learners received no outcome feedback and decided only after observing demonstrators’ choices. This separation ensured that the social signal reflected demonstrators’ private learning rather than mutual influence among simultaneously observing learners. We varied whether demonstrators were ingroup or outgroup members and whether their incentives were aligned or misaligned with the social learner’s. The same majority could therefore point to the best option in one context and to the worst option in another. The design thus tested whether people interpreted the same signal differently across contexts. Performance was highest when demonstrators were aligned with the learner’s incentives, especially when those demonstrators were also ingroup members, even though group membership was induced by minimal cues. Cognitive load reduced memory performance but did not measurably alter the use of social information. We repeated the design in Switzerland and Kenya to study how findings extend across distinct participant pools. Individual strategies were heterogeneous in both sites but more dispersed in Kenya. These findings show that people do not treat the same social frequency signal as interchangeable across contexts defined by group membership and incentive alignment, and they highlight the value of studying social learning across distinct participant pools.

von Flüe, L., Vogt, S., & Efferson, C. (2026). The (anti-)conformity strategy space: How flexible is social learning when group membership and incentive alignment vary? Journal of Experimental Psychology: General. Advance online publication.

Efferson, C.
Arbitrary Cooperation
Can we use behavioral experiments to draw inferences about the evolution of human social behavior, and if so how? To illustrate the complexities, I examine the evolution of strategies for the repeated play of social dilemmas. Repeated interactions enjoy favorable status as an explanation for the evolution of human cooperation. So favorable, in fact, that one influential hypothesis paradoxically invokes repeated interactions to explain human cooperation in anonymous one-shot experiments. Some studies, however, suggest that repeated interactions in isolation, given a suitably flexible strategy space, do not reliably support cooperation. If generally true, this latter finding would contradict the status of repeated play as a basis for the evolution of both repeated and one-shot cooperation. Using a mix of analytical results and a large number of simulations, I consider this question with a social dilemma involving a continuous action space. Continuous action spaces arguably capture the social dilemmas people play in their daily lives better than games in which players can only choose to cooperate fully or defect fully. They also allow us to study the effects of extending the strategy space without positing super-human players who can remember long histories of past play. Under repeated play of such a game, I examine ten strategy spaces and seven approaches to implementation errors in all combinations. Results show an overall pattern of limited cooperation. In particular, given a combination of strategy space and implementation error that supports the evolution of cooperative strategies, some suitable generalization of the model typically destabilizes cooperation. Although these results do not constitute a general proof, they are consistent with the notion that repeated interactions only support the evolution of cooperation given arbitrary, biologically groundless restrictions on the set of strategies that can arise via mutation or innovation. This finding further implies that an important evolutionary explanation for one-shot cooperation enjoys a privileged status it does not deserve.
View abstract ↗

Moorthy, A.
Learning under different information acquisition modes: Experimental evidence
We study how the mode of information acquisition shapes belief updating and retention in the context of macroeconomic expectations. We conduct a randomised online experiment to study how beliefs about the risk of a recession or future inflation respond when the same informative signal is acquired in different ways. Participants are assigned to receive information in a passive mode, or one of three active modes – directed online search under low or high search frictions, and undirected online search. We elicit prior and incentivized posterior beliefs about recession risk and inflation and record detailed browsing behavior throughout the experiment. Comparisons across acquisition regimes provide causal evidence on how active retrieval and search frictions affect belief updating relative to the widely-used information provision paradigm, and identify the interaction between acquisition mode and search costs. The web trace data offers a descriptive benchmark for endogenous information acquisition and rich information on people’s search pathways. A follow-up survey provides evidence of differences in the retention of acquired information across the different modes.
View abstract ↗

Can we use contemporary empirical results to draw inferences about the evolution of human social behavior, and if so how? To illustrate the complexities, I examine the evolution of strategies for the repeated play of social dilemmas. Repeated interactions enjoy favorable status as an explanation for the evolution of human cooperation. So favorable, in fact, that one influential hypothesis paradoxically invokes repeated interactions to explain human cooperation in anonymous one-shot experiments. Some studies, however, suggest that repeated interactions in isolation, given a suitably flexible strategy space, do not reliably support cooperation. If generally true, this latter finding would contradict the status of repeated play as a basis for the evolution of both repeated and one-shot cooperation. Using a mix of analytical results and a large number of simulations, I consider this question with a social dilemma involving a continuous action space. Continuous action spaces arguably capture the social dilemmas people play in their daily lives better than games in which players can only choose to cooperate fully or defect fully. Continuous action spaces also allow us to study the effects of extending the strategy space without positing super-human players who can remember long histories of past play. Under repeated play of such a game, I examine eleven strategy spaces and seven approaches to implementation errors in all combinations. Results show an overall pattern of limited cooperation. In particular, given a combination of strategy space and implementation error that supports the evolution of cooperative strategies, some suitable generalization of the model typically destabilizes cooperation. Although these results do not constitute a general proof, they are consistent with the notion that repeated interactions only support the evolution of cooperation given arbitrary, biologically groundless restrictions on the set of strategies that can arise via mutation or innovation. This finding further implies that an important evolutionary explanation for one-shot cooperation enjoys a privileged status it does not deserve.

Efferson, C., Lalive, R.
Irreducible cultural variation in Switzerland
A prominent hypothesis in gene-culture coevolution says that cultural evolutionary processes stabilize group-level differences that would otherwise not be possible. This kind of variation is a necessary but not sufficient condition for the cultural group to be a meaningful unit of organization - and by extension selection - in human populations. While many of us may be comfortable with the notion that cultures vary when regaling party goers with tales from our latest vacation, isolating cultural variation from other forms of group-level variation is a formidable empirical task. The challenge is that cultural boundaries often correlate with others boundaries linked, for example, to the environment or political institutions. The French-German language border in Switzerland offers an unusual opportunity to sidestep these confounds. Using a pre-registered regression discontinuity design, we analyzed several potential discontinuities related to health, reproduction, and well-being. Results show that people in the two parts of Switzerland lead different lifestyles in terms of sleep, social support, and physical activity. They do not differ, however, in terms of diverse outcomes related to fertility, physiological disease, and physiological function. The two groups do differ psychologically, with significant discontinuities associated with major depressive episodes, psychological distress, and suicide attempts. Given that the daily flow of people and ideas across the language border is extreme, Switzerland presents conditions especially unfavorable for the maintenance of cultural variation. Observed variation thus supports the claim that cultural transmission, as a distinct evolutionary process, can and does sustain grouplevel differences.

Bonner, K.
The Effects of the Social Learning Strategy Space in Gene-Culture Coevolutionary Modeling
Understanding how individuals learn socially is central to explaining aggregatelevel cultural evolution. Prior research has shown that even small heterogeneities in individual social learning strategy use can have outsized impacts on population-level cultural dynamics. Yet, most theoretical research has relied on assumptions that preclude the majority of possible strategies from ever arising in evolutionary models, potentially distorting our understanding of cultural evolution. This study aims to address this potential issue by comparing the evolutionary dynamics and stable strategies that emerge under two distinct modeling approaches. The first, Bayesian updating, imposes structured constraints on the social learning process, thereby limiting the strategy space under consideration. The second approach, using Finite State Machines, allows for an unrestricted exploration of all feasible strategies. By comparing these methods, we aim to uncover how strategy space restrictions may bias our understanding and to identify methods for safely restricting the strategy space under consideration without being subject to this bias.

Von Flüe, L., Efferson, C.
(Anti-)conformity in three dimensions: social learning in the big space between simple heuristics and general-purpose reasoning
Do people flexibly adjust to social information or rely on rigid heuristics that weight others’ behaviour by who they are and whether their goals align with their own? The answer matters because the ways individuals learn from others shape how culture changes. We ran an incentivised behavioural experiment with two roles. Demonstrators made initial choices based only on private information. Social learners then made the same choice after observing several demonstrators. Social learners encountered four informational contexts defined by two factors. One was whether the demonstrators belonged to the learner’s own group or to another group. The other was whether this group relation implied that the learner and the demonstrators shared a similar incentive structure or were likely to benefit from different options. This combination of group membership and incentive alignment dictated whether copying the majority or the minority was optimal and success therefore required flexible adjustment rather than blind conformity. To probe the role of deliberation, half of the learners completed the task under cognitive load. Cognitive load had no measurable effect, which points to a deeply rooted learning bias that is too robust for load to alter, namely a tendency to process incentive aligned information more effectively, especially when it comes from ingroup members. We repeated the design with participant pools in Switzerland and Kenya to assess cultural generality. Both samples were sensitive to group identity and incentive alignment, yet they followed markedly different overall strategies, underscoring the importance of cultural diversity in research on social learning.

Scalable behavioural interventions often struggle to engage the cognitive and psychological mechanisms that underlie durable changes in preferences and habits. This study provides a proof of concept for an underexplored intervention format: edutainment through video games. Partnering with a large video game company, we develop a game that embeds educational content on sustainable food consumption into an entertaining storyline. In a pre-registered field experiment (N = 4,034 UK adults), participants are randomly assigned to play either one of three treatment versions of the game or a control version without environmental content. Real-world food choice behaviour is measured through incentivised online supermarket tasks. Relative to the control group, treated participants select grocery baskets with 20% lower environmental impact immediately after gameplay, an effect that remains at 8–10% in a follow-up 2–3 weeks later. Behavioural change results from a combination of knowledge gains, short-term salience and preference change. Strikingly, effects were particularly persistent among subjects with low baseline sustainability. Further evidence suggests that the intervention was effective partly because it provided an enjoyable experience and affected a rich set of beliefs and attitudes, including personal norms, efficacy, and perceived social norms.

Credible leadership signals are observable pieces of information emitted by potential leaders that credibly convey the state of unobservable leadership qualities. Credible signals bridge information asymmetries between leaders and followers and play a crucial coordination role in the leadership process by assisting followers in identifying competent leaders. Yet, despite empirical work on leadership signaling, its theoretical foundation remains underdeveloped. We thus integrate Spence’s signaling model into a theoretical framework for better understanding signaling in the leadership process. Specifically, we propose three conditions for credible leadership signals: 1) an unobservable leadership quality that followers care about, 2) an observable signal, and 3) signaling costs that are negatively correlated with the quality. We then systematically review the literature on leadership signaling, showing that conditions for credible signals are not respected in a majority of cases. We conclude by deriving theoretical implications and providing practical guidelines for future research on credible leadership signals. This manuscript advances the discourse on leadership signaling by clarifying when signals effectively guide followers to discern capable leaders.