VISUAL PERCEPTION OF FACIAL EXPRESSIONS

Psychophysical method for quantitative specification of facial expressions is presented. The method is based on subjective estimations of similarity between pair of stimuli – face expression and emotional name. Each pair of stimuli was randomly chosen from banks of 25 face pictures and 25 Russian names of emotion, and simultaneously presented on the monitor’s screen. Duration of presentation was equal 1 c., and during the 2 following seconds subject should estimate similarity between the face expression and the name of emotion using integer numbers from 9 (maximal similarity) till 1 (minimal similarity). The estimates were averaged across ten presentations of each pair. In general the results for each subject was presented in the form of a matrix whose rows (faces) and columns (names) represent the two sets of stimuli while the entries contain all pairwise similarities among them. Each row of the matrix can be considered as semantic emotional function of face stimulus as well as spectral distribution of light intensity is considered The visual perception of the emotional characteristic of stimuli is one of main information channels during human interpersonal relationship. It stipulates importance of measurement technique in this area. It is obvious that perception of facial expression are determined by personal experience as well as by detection of face configuration (Ekman & Friesan, 1978, Paramey, 1996; Bimler & Kirkland, 2001). Therefore, research into the role of the subjective estimates of facial expression and its recognition is of prime interest. The principal characteristics of facial configuration are the outline of the mouth, and eye and brow form. Psychophysical investigations using photographs and schematically presented faces as stimuli have shown that facial expression can be defined as a multidimensional function of the curvature of the lips, angle of eyebrows, and other configuration characteristics (Ekman &amp