Individual differences in the use of form and motion in the perception of sex in biological motion displays.

Past research on individual differences in biological motion perception has used different tasks as a way of determining the relative use of form and motion information. However, a single task, the identification of the sex of a biological motion actor, can be based on form information, motion information, or both. We collected data from 76 observers completing a sex identification task to determine what information observers used to complete the task. In intermixed trials, participants viewed biological motion displays that varied in 1) both motion and form information, 2) just motion information, and 3) just form information. In addition to basic demographic information, participants also completed ADHD measures and surveys on autism, empathy, social anxiety, and the vividness of movement imagery. We analyzed the biological motion data by examining the slopes of logistic regression functions for the three types of biological motion displays. Observers were classified as using motion and form equally (18 observers), using motion more (31 observers), or using form more (27 observers). Performance on ADHD measures and survey scores for autism, empathy, and social anxiety were not correlated with performance on the biological motion tasks as measured by the slope of the logistic regression functions. There was a significant negative correlation between performance on the form only biological motion displays and vividness of movement imagery. Analysis of demographic data showed that females were more likely to be classified as "using form information more" or "using both motion and form about equally" while males were more likely to be classified as using "motion information more." These data show that there are individual differences on what information observers use to complete a sex identification task and these differences may in part be related to the sex of the observer and their visual movement imagery ability. Meeting abstract presented at VSS 2018