Purchase this article with an account.
Edward Vessel, Jonathan Stahl, Isaac Purton, Gabrielle Starr; Domain-General Representation of Visual Aesthetic Appreciation in the Medial Prefrontal Cortex. Journal of Vision 2015;15(12):124. doi: 10.1167/15.12.124.
Download citation file:
© 2017 Association for Research in Vision and Ophthalmology.
Are there “domain-general” neural processes that support aesthetic appreciation regardless of stimulus type, similar to the encoding of abstract expected value of goods? Recent work examining face and place attractiveness reported a common representation of attractiveness in medial prefrontal cortex (MPFC; Pegors et al., in press). We sought to test a) whether moving aesthetic experiences with artwork and architecture, artifacts of human culture that show much more individualistic preferences than landscapes or faces, rely on similar domain-general patterns of activation in MPFC, and b) whether such putative domain-general mechanisms overlap with the default-mode network (DMN), previously shown to be engaged by highly moving artworks (Vessel et al., 2012). Thirteen observers made aesthetic judgments (“how much does this image move you?”) about images of artworks, natural landscapes, or architecture on a continuous scale while being scanned using fMRI. Classifiers were trained to distinguish most vs. least moving trials using patterns of trialwise BOLD responses. When provided data from the entire MPFC, classifiers trained on one category and tested on another performed better than chance for all train/test combinations (63-67%, p< 0.01), providing evidence for a domain-general aesthetic mechanism in the MPFC. Furthermore, classifiers trained on patterns from a spatially restricted ROI corresponding to the left anterior MPFC portion of the DMN (derived from subject-specific “rest” scans), also performed better than chance (55-62%, p< 0.02), providing strong evidence that this portion of the DMN contains domain-general information about aesthetic appreciation. We also found that when classifiers were trained and tested on images of the same category (halfwise-split; using the entire MPFC), performance was best for artworks (72%, p< 0.01), followed by landscapes (60%, p< 0.05), and lastly architecture (53%, n.s.), which suggests that MPFC as a whole contained more information about the aesthetic appeal of artworks than for the other categories.
Meeting abstract presented at VSS 2015
This PDF is available to Subscribers Only