Semantic encoding improves face recognition in prosopagnosia

Prosopagnosic individuals suffer from severe difficulties in face perception and recognition. Interventions that improve their face recognition abilities are scarce and mostly involve extended training on perceptual information. Given that most individuals with developmental prosopagnosia appear to make trait judgments about faces normally, in the current study we took advantage of recent reports on a semantic encoding benefit in face recognition and examined whether semantic encoding will also improve face recognition in developmental prosopagnosia (DP). Recent studies show that making trait inferences about faces (e.g., how intelligent does the face look like?) during encoding improves face recognition relative to making perceptual inferences about facial features (e.g., how round is the face?) or no evaluations. This semantic manipulation during encoding was also shown to improve face recognition for other race faces, suggesting that it can enhance poor recognition abilities. Thus, we ran a group of DPs (N = 17) on a face recognition task in which participants evaluated faces socially, perceptually, or made no evaluations during the study phase. During the test phase they were presented with different images of the learned identities and were asked to decide whether the face was old or new. Results show a robust semantic encoding benefit that was comparable in prosopagnosics and controls. Surprisingly, the overall performance of prosopagnosics across all conditions of this task was not worse than controls. These findings are the first to show that a non-perceptual, social manipulation can enhance face recognition in prosopagnosia. This may be consistent with a previous report that oxytocin enhances face recognition in prosopagnosia, suggesting that social processing mechanisms may help DPs to alleviate their face recognition difficulties.