Gender and cerebral lateralization of audio-visual perception of emotion

Presentation of brief (120ms, 160ms, 520ms) audio prosody-, video- and audio-visual clips containing congruent emotion (anger, fear, happiness, neutral, sadness/ positive- and negative valence) was used in a divided visual field technique/dichotic listening behavioral experiment consisting of 17 males and 17 females to investigate a possible gender difference in cerebral lateralization in perception of emotion. Clips were created from Montréal affective voices and the Montréal Pain and Affective Face Clips. Accuracy percentages of correct recognition of emotion were recorded. Findings showed no support for either the right-hemisphere- or the valence hypothesis. Gender as a between subject factor was non significant. Clips containing both audio and video had the highest accuracy score of all modalities. Audio-only prosody had significant lower accuracy score compared to video-only and audio-visual clips. Positive valence in the short length may have an early accuracy advantaged compared to negative valence in the audio-visual modality that dissipates in 120ms-160ms range, with the accuracy difference disappearing between the categories. The same advantage can be found in anger, while happiness, fear and neutral have no significant differences in accuracy in lengths in the audio-visual modality