Item

Abstract

Face and voice are two preeminent physical cues describing a person. In unimodal face studies, faces are believed to be represented in a psychological multidimensional space built around a norm face. This norm-based framework is purely visual and can account for the recognition advantage of distinctive faces. Could this visual framework be extended to include voices? First, we investigated whether face distinctiveness can be of multi-modal nature. Participants saw faces paired to distinctive or typical auditory stimuli during a learning phase. Thereafter their recognition performance was tested with the learned faces presented alone among new faces. We assessed whether their recognition was affected by the distinctiveness of the auditory stimuli. Further, we asked whether the type of auditory stimuli paired to the faces (voices or other sounds) mattered. We found that recognition performance was better for faces previously paired with distinctive than with typical voices while no such distinctiveness effect was observed after face-sound pairing. These findings suggest that facial distinctiveness can be indeed of a multi-sensory nature as the memory for faces can be modified by the perceptual quality of related vocal information. Second, we investigated whether faces and voices are integrated early on to form a multisensory representation of a person in memory. We used a similar learning phase followed by a crossmodal priming paradigm to test whether a face familiarity decision could be primed by voices. We found that voice primes facilitated the recognition of their paired faces. Our results suggest a strong early association between voices and faces in memory. Furthermore, we found no evidence of priming with paired sounds, thus confirming the special status of the face-voice pairing for person identification.