Abstract
Viewing art is generally assumed to be a subjective experience driven by an interaction between external input from the world and internal associations which give the art its personal meaning. In art theory, this personal contribution of meaning to a painting is termed the Beholder’s Share. The cognitive and neural mechanisms that underlie the Beholder’s Share, and their link to how individuals experience art, remain unknown. Here we attempt to fill this gap. We focus on a key tenet of the Beholder’s Share - that a viewer brings more personal meaning to abstract art than to representational art - and test this tenet by comparing the variability in responses to abstract and representational paintings, both behaviorally and in the brain. 30 participants were scanned with fMRI while viewing and making decisions about a series of real-world paintings that were either abstract or representational. In a separate online experiment, 30 independent subjects generated written captions for the same paintings. We hypothesized that abstract paintings would elicit more personal contributions than representational art, and that this would be reflected in cross-subject variability in both behavioral and neural responses. Specifically, at the neural level we predicted that abstract art would be associated with greater cross-subject variability in patterns of BOLD activity, particularly in areas involved in narrative interpretation (Default Mode Network). Behaviorally, we hypothesized that abstract paintings would be associated with more cross-subject variability in the semantic meaning of captions. As expected, we found that abstract paintings elicited more cross-subject variability in BOLD responses than representational paintings, specifically in the DMN. Additionally, semantic meaning of captions was more variable across subjects for abstract than representational paintings. Taken together, these findings suggest that abstract paintings evoke more individual contributions than representational paintings, reflected in variability in both brain responses and semantic meaning generation.