McGurk effects according to sex of talker and listener

The McGurk effect has been shown to be more robust in female listeners. It has also been shown that the perceived similarity in talker-listener dyads may influence speech perception and attention. It is unknown if the sex of the talker, vis-à-vis the listener, influences the McGurk effect. It is thus hypothesized that there is a greater McGurk effect for female listeners when perceiving same-sex talkers. Using a web-based paradigm, 200 native English-speaking participants, half men, reported their perception of 78 target syllables spoken by both male and female talkers in a classic McGurk paradigm. Consistent with some reports in the literature, a small percentage of participants failed to perceive the McGurk effect. Confirming prior reports, men were less likely than women to report the McGurk effect. Unexpectedly contrary to the hypothesis, no listener-talker sex concordance effect was observed. This lack of effect may be the result of an attentional bias preferring the opposite sex, predictive coding in the perceiver, or the relative lack of effect permuted across many participants. Findings have implications for audio-visual integration, real-time perception of speech, and offer a proof-of-concept of web-based perception studies.