Vivid Face Perception as a Constructive Component of Multimodal Affective States

<p>The features of expression and perception of vivid facial expressions in the system of multimodal affective states of a person are studied. The study is based on the Russian-language version of the Geneva Emotion Recognition Test (GERT) and consists of two series. The first serie of experiment was devoted to demonstration of short audio—video clips of 14 affective states expressed by specially trained actors, the second —to the same videos without sound accompaniment (intonation of voice in pseudolinguistic utterances). The subjects — 72 women in each series of the experiment, age: 18-45 years (M1=22.4 g, SD1=5.4; M2=27.47 l, SD2=5.7) - identified the states of the actors—sitters (five women and five men), using an interactive image of the Geneva Wheel of Emotions on the display screen for evaluation. The analysis of the characteristics of the accuracy and semantic structure of the expression and identification of emotional expressions, depending on the type of modality and exposure conditions, is carried out. The main expression and identification metrics of facial, prosodic and multimodal states are correlated. According to experimental data, within the framework of bimodal (face+voice) expressions of emotions, a mobile face has a higher expressive potential and is the basis of cross-modal integration.</p>