Experiments show that irrelevant facial trustworthiness affects learning dynamics in a probability task, suggesting wider implications.
Although social cues such as perceived facial trustworthiness influence people’s ongoing decision-making progress in a dynamic world, the ecological validity of facial appearance is limited or nonexistent. We conducted two experiments (N total = 258) to investigate the impact of the valence-defining but task-irrelevant social dimension of facial trustworthiness on learning dynamics. In a multiple-cue probability learning task, (in) congruence was manipulated by associating (less) trustworthy faces with the binary valenced outcome. In the Incongruence conditions, participants performed worse and showed a slower learning rate initially. As demonstrated by a rolling regression approach, they utilized the facial trustworthiness information across a greater number of trials, as opposed to basing probabilistic predictions on learning experience. The effect was driven by valence matching rather than by a prevailing semantic relationship between cues and outcome. We discuss the perception of cues based on preformed associations versus associated behavior and the generalizability of the congruence-incongruence effect.
No takes yet. Share an insight, caveat, or question.
Scholten et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: