Charting Decodability of Dynamic Facial Expressions in Young and Old Adults: Similarities and Differences

Dynamic facial expressions carry a wide range of signals, encompassing emotional but also more conversational content important for social interaction, for which the dynamic aspect is crucial. Likewise, we know from previous behavioral and neuroimaging studies that processing of emotional stimuli changes across aging - little, however, is known about how age may impact brain activity for dynamic facial expressions. To address this open issue, here we used two cohorts of older and younger adults (total N=77) within a whole-brain MVPA decoding paradigm in fMRI. We used a range of dynamic and conversational expressions as stimuli shown with a foil task in the scanner and had participants rate these post-scanning in terms of their affective content along 12 dimensions (including valence and arousal). The behavioral ratings were used to cluster the facial expressions and the resulting similarity matrix was used in a searchlight decoding paradigm to identify common areas. Using robust bootstrap analyses, we identified the insula as a common brain region able to decode the wide range of emotional and conversational dynamic facial expressions for both participants groups. We also discuss additional brain areas specific to the younger group. Our study adds to the growing literature on neural processing of dynamic expressions in the context of aging.