Abstract
Can people correctly match an unfamiliar voice to a static image of a face? There’re conflicting reports: Kamachi et al. (2003) showed participant at chance level performance, whereas Mavica & Barenholtz (2013) observed above chance. We hypothesized that this contradiction may be due to the variability of difficulties in face-voice matching among the models, and examined whether the variability would be explained by similarity between the personality impressions formed from a face image and voice. We conducted a matching task with a static face image and voice, and measured the personality impressions received from the face image and voice. The result demonstrated that overall accuracy of the matching task was significantly above chance. More importantly, face-voice matching performance was correlated with the similarity between the face and voice ratings for the model, suggesting that the personality impression from the model’s face and voice plays an important role in face-voice matching.
