Akademik Veri Yönetim Sistemi
International Symposium on Brain and Cognitive Sciences (ISBCS), Ankara, Türkiye, 29 Mayıs 2022
Biological motion carries complex and dynamic information created by the motion kinematics of the actor. The complexity of biological motion allows animals to extract different types of information that are used to predict the intent of others. Humans excel at extracting relevant information from biological motion. We can infer the action, emotional state, and gender of others from a set of dots known as point-light displays. However, the majority of studies with point-light displays investigate bottom-up perception of biological motion and it remains unknown how top-down knowledge (priors) affects biological motion perception. In order to address this question, we conducted three experiments in which we manipulated the type of prior. In experiment 1 (N=20) we provided the action category as the prior, in experiment 2 (N=19) we used the emotional state as the prior, and in experiment 3 (N=20) we used gender information as the prior. In all these experiments, each trial started with a cue (a symbol of kick or walk in experiment 1, happy or sad in experiment 2, and male or female in experiment 3). Next, an intact and a scrambled version of point-light biological motion were presented on different sides of the screen relative to a central fixation (left or right). Participants were asked to locate the biological motion stimulus using the arrow keys. In order to increase the probability of the cue being used, we increased the difficulty of the task by using additional noise dots. We controlled the number of noise dots with an adaptive procedure (QUEST). The cue predicted the type of information in the biological motion stimulus. For each experiment, we also manipulated the validity of the cue in four sessions, (neutral, 50%, 75%, 100%). Our results showed that participants were able to locate biological motion faster in the congruent (valid) trials than in the incongruent trials (p<0.01) when the cue validity was high (at 75%) and when the cue was about action category. There were no significant differences between congruent and incongruent trials in the other experiments and validity conditions. Overall, our results suggest that biological motion perception is affected by top-down knowledge in line with predictive processing accounts of perception.