Associative learning of social interaction alters attention to human faces

Read the full article

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Humans excel at leveraging relational knowledge (e.g., relative features or locations) to facilitate object processing. This knowledge is thought to emerge gradually through associative learning over the course of experience. While extensive research has shown how experience shapes attentional biases toward non-social regularities, far less is known about whether the learning of social relationships – the most consequential form of relational knowledge for humans – tunes attention toward social entities. Across four experiments (N = 144), we examined this question by training participants to associate face pairs with either interacting (face-to-face) or non-interacting (back-to-back) configurations. In a subsequent cueing task, direct-gaze face pairs from interacting configurations captured attention more strongly than those from non-interacting configurations, an effect abolished for upside-down faces. This bias persisted even when individual faces from learned dyads were tested in isolation. In contrast, this bias did not occur for arrows, ruling out domain-general cueing, and for animal faces, even when training promoted identity-level processing, indicating its specificity to human faces. Our findings demonstrate that social relationships can be internalized through experience as emergent attributes of face representations, highlighting a key role of social learning in shaping attention to social information.

Article activity feed