Social Influence and Perceptual Decision Making

• Applicant: Dr. Markus Germar
• Subject Area: Social Psychology, Industrial and Organisational Psychology
• Term: from 2018 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 411073522

The seminal studies by Sherif (1935) and Asch (1956) showed that social influence can alter decision-making in simple perceptual tasks (i.e. perceptual decision-making). Yet, it remains unclear which cognitive mechanisms explain these effects. As a consequence, the fundamental question of whether social influence can alter basic perceptual and attentional processes is left unanswered. While current re-search suggests that social influence can change stimulus processing, the precise cognitive underpinnings still remain to be uncovered. Furthermore, there is no comprehensive theoretical frame-work for understanding these processes.The aim of the proposed research project is to close this important theoretical and empirical gap. To this end, we have developed a new model that is based on the social reinforcement learning approach, and integrates hitherto separate lines of research from social and cognitive psychology. Based on a few simple but well-established assumptions (i.e. applying the principles of social reinforcement learning on cognitive processes), innovative and untested hypotheses can be derived from our model. Our model postu-lates that repeated exposure to social influence (e.g. a social norm or a majority opinion) leads to a reinforcement learning process that alters the allocation of (visu-al) attention. As a result, the processing of stimulus information that is consistent with the social consensus is prioritized which, in turn, increases the tendency to de-cide in line with the social consensus. Since the changes in stimulus processing are learned, they persist even when perceptual decisions are made alone.We have designed five experiments to test the central implications of our model. We aim to employ a paradigm that will allow us to overcome the methodological limita-tions of previous research. Experiments 1-3 are designed to test the postulated me-diating mechanisms of our model by employing three complementary approaches: a diffusion model analysis to disentangle important decision parameters, a visual search task to measure the allocation of attention, and event-related potentials to uncover the temporal dynamics of the mediating attentional processes. Finally, Ex-periments 4 and 5 aim to investigate whether the postulated effects are robust against objective feedback and robust over time, and whether the confrontation with a consensus of a disliked group reverses the direction of the postulated effects, ultimately leading to anti-conformity.

• DFG Programme: Research Grants
• Co-Investigator: Professor Dr. Andreas Mojzisch