Abstract

Contemporary literature suggests that listeners use auditory and visual gender cues during /s/-/ʃ/ categorization. This study builds upon previous research on sibilant categorization by investigating whether self-perceived power, “an individual’s relative capacity to modify others’ states by providing or withholding resources or administering punishments” (Keltner et al. 2003), can serve as a mediating factor in gender and linguistic perception. Social processing seems to show some sensitivity to one's degree of self-perceived power - high-power individuals tend to attend less to information in conflict with their expectation of another according to the other’s social category. (Goodwin et al. 2000). I report whether gender cue congruity and the participant's primed degree of self-perceived power shapes the outcome of /s/-/ʃ/ categorization.

Participants were primed for a high or low degree of self-perceived power and completed a forced-choice identification task. During each trial, participants saw an image of a face and heard one of a continuum of words ranging from "shy" to "sigh"; they indicated whether they heard "shy" or "sigh". A mixed logistic regression revealed that participants overall were significantly more likely to respond “sigh” for a male voice (p<0.001) and male face (p=0.01). Participants primed for low-power were also significantly more likely to respond “sigh” when a given voice was paired with a male face, relative to when it was paired with a female face (p<0.001) . Participants primed for high-power, however, were not significantly more likely to respond “sigh” for a given voice according to what face was paired with it (p=0.4). As hypothesized, the responses of high-power participants showed less sensitivity to gender cue congruity than did those of low-power participants.