Abstract
Purpose. A region on the human posterior superior temporal sulcus (posterior STS) has been implicated in the perception of human activity, including biological motion as portrayed by Johansson's point-light animations. This area is more active when observers view biological motion than when they view the same motion vectors scrambled such that the perceptual coherence is destroyed (Grossman et al, 2000, Grossman & Blake, 2001). To test whether the neural response in this area is specific to human body movement, we measured the BOLD signal changes when observers viewed point-light animals, human hand and face movements versus the scrambled counterparts. Methods. Point-light human body motion animations were created by digitizing video of an actor with reflective tape on the joints performing various activities. Point-light animals were created by encoding the joint positions of various animals from the published Muybridge collection, and displaying the frames in rapid sequence to give the appearance of natural movement. Point-light hand animations were created by digitally recording hand movements in a dark room with glow paint applied to the joints. Because faces have fewer joint positions, point-light facial expressions were created by placing glow paint in scattered positions on the entire face. Scrambled versions of all types of animations were created by randomizing the starting positions of the dots. Results. Posterior STS was strongly modulated by the human whole body animations, but less so by any of the other animations. Still, viewing animal movements and facial expressions resulted in more neural activity in posterior STS than scrambled motion. The results indicate a strong selectivity for human movement over these other types of natural, biological motion patterns.