Previous research has shown that facial impressions can change when paired with vocal cues of opposing valence. However, it remains unclear whether voices can also influence mental representations of faces—an effect that would indicate deep cross-modal integration in impression formation. Here, we tested that possibility by using reverse correlation (RC) across two preregistered experiments. In Experiment 1, participants (N=61) read negative behavioral statements about a target and completed an RC task to generate mental face representations (Classification Images, CIs). After hearing a pretested positive or negative voice associated with the target, they repeated the RC task. In Experiment 2, we tested whether voice-face integration occurs spontaneously and without explicit evaluation. Participants (N=120) viewed positive or negative faces at Time 1 without making judgments; after hearing a voice with congruent or incongruent valence at Time 2, they completed the RC task. In both experiments, the resulting CIs were rated by independent judges (N1+2 = 140). Results showed that face representations were judged differently after hearing a voice with opposing valence, suggesting that voices can alter mental images of faces. These findings indicate that vocal information shapes face representations, when elaborated as social information both intentionally and spontaneously, extending evidence of cross-modal integration in impression formation and updating.
Brambilla, M., Masi, M. (2026). Facial impressions are not set in stone: Voice-driven changes in facial representations obtained via reverse correlation. In Proceedins of the General Meeting of the European Association of Social Psychology (EASP).
Facial impressions are not set in stone: Voice-driven changes in facial representations obtained via reverse correlation
Brambilla, M
;Masi, M
2026
Abstract
Previous research has shown that facial impressions can change when paired with vocal cues of opposing valence. However, it remains unclear whether voices can also influence mental representations of faces—an effect that would indicate deep cross-modal integration in impression formation. Here, we tested that possibility by using reverse correlation (RC) across two preregistered experiments. In Experiment 1, participants (N=61) read negative behavioral statements about a target and completed an RC task to generate mental face representations (Classification Images, CIs). After hearing a pretested positive or negative voice associated with the target, they repeated the RC task. In Experiment 2, we tested whether voice-face integration occurs spontaneously and without explicit evaluation. Participants (N=120) viewed positive or negative faces at Time 1 without making judgments; after hearing a voice with congruent or incongruent valence at Time 2, they completed the RC task. In both experiments, the resulting CIs were rated by independent judges (N1+2 = 140). Results showed that face representations were judged differently after hearing a voice with opposing valence, suggesting that voices can alter mental images of faces. These findings indicate that vocal information shapes face representations, when elaborated as social information both intentionally and spontaneously, extending evidence of cross-modal integration in impression formation and updating.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


