Using evolutionary algorithms to uncover individual differences in how humans represent facial emotion
How humans perceive and represent emotional facial expressions critically guides social interactions and has a profound impact on social cognition. However, much emotion research to date is potentially flawed by relying on the assumption that people represent categorical emotions in the same way, using standardised stimulus sets and therefore overlooking important individual differences in emotion processing. To resolve this problem, we developed a task using Genetic Algorithms and derived participant-generated emotional expressions from 105 individuals. A separate group (N=108) was then asked to identify the expression on these faces. Taken together, results showed relative population-level consistency in the representation of happy faces alongside a high degree of variability in the representation of fear and sadness. High test-retest reliability was observed. This novel method enables the efficient quantification of variation in how people represent emotional expressions, presenting promising advances for the study of individual differences in social cognition and emotion processing.