Purchase this article with an account.
Hanako Ikeda, Katsumi Watanabe; Recognizing emotional states from biological motion within noise. Journal of Vision 2008;8(6):915. doi: 10.1167/8.6.915.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Perceiving the activities of other people is an important social skill of obvious survival value. Human vision has highly sensitive mechanisms for recognizing activities performed by others (Johansson, 1973). We previously showed that foveal and peripheral performance of biological motion detection within noise could not be equated by any magnitude of size scaling (Ikeda et al., 2005), suggesting that the resources for keen biological motion perception are concentrated on the central region of the visual field. Biological motion provides information not only for recognizing others' activities but also for inferring others' emotional states (Dittrich et al., 1996). In the present study, we examined whether a similar central limitation would be observed for the recognition of emotion from biological motion. Neutral and emotional (sad, angry, or happy) biological motion sequences were embedded in motion noise and presented successively (2AFC). Participants indicated which of the two intervals contained emotional biological motion. A staircase procedure varied the number of noise dots to produce a criterion level of performance. The poorer spatial resolution within the periphery was compensated by spatially magnified the entire stimulus. As in the previous detection task, foveal performance regularly increased with stimulus size until it reached the saturation level. Performance at the far most periphery (more than 8 deg) could not be equated by size scaling. The saturation level and the saturation size appeared to differ among different emotions. In addition, performance in the moderate eccentricity (e.g., 4 deg) showed inconsistent tendencies; for example, performance with some of emotional biological motion increased steadily with stimulus size, and in some cases, performance increased irregularly. These results indicate that the eccentricity dependency (and underlying processes) may differ, at least partially, between biological motion detection and emotion inferences from biological motion.
This PDF is available to Subscribers Only