Eliz Shimshek¹, Marco A. Sama¹, Jonathan S. Cant¹; ¹University of Toronto Scarborough

The visual system efficiently processes multiple sources of information by leveraging ensemble encoding, the ability to extract statistical summaries (e.g., the average size of circles) from sets of similar objects. This also occurs for high-level stimuli such as faces, and summary statistics such as the average expression in a crowd of faces provides critical cues about the intentions of others. Emotional expressions are pervasive drivers of decision-making, and similarly, affective images, such as the valence of photographs, have been shown to significantly influence approach-avoidance decisions. While ensemble processing has been extensively studied in visual perceptual tasks, its influence on approach-avoidance decisions, a fundamental aspect of human behavior, remains insufficiently explored. To address this, we examined how ensemble processing influences approach-avoidance decisions in real-world scenarios, namely, when deciding whether or not to watch a movie. To investigate this, we examined the relationship between implicit ensemble processing (i.e., passive viewing of ensemble stimuli) and approach-avoidance behavior. Participants viewed ensembles of six faces expressing a positive, negative, or neutral average emotion. Following the presentation of each face ensemble, participants completed an approach-avoidance task using a social decision-making paradigm. Specifically, participants were presented with a positive, negative or neutral movie poster and then quickly decided whether or not they would prefer to watch the depicted film. We found that average expression did not influence approach-avoidance behaviors towards affective movie posters. Instead, viewing decisions depended only on the affective content of the movie poster, independent of the face ensemble it was paired with. Critically, this research deepens our understanding of the intersection between ensemble processing and decision-making. As a next step, we will test whether explicit ensemble processing influences approach-avoidance decisions, as explicit judgments of an average feature are known to create more precise ensemble representations compared with implicit processing.

Acknowledgements: This research was funded by an NSERC Discovery Grant to J.S.C.