Learned bias for 3-D shape perception without object motion

Abstract

Was the highlighted surface closer or farther? During “cue recruitment”, uninformative visual signals such as retinotopic location and object translation direction can come to bias the perceived rotation direction of a Necker cube (a bistable stimulus) after training [e.g. Haijiang et al., 2006; Harrison & Backus, 2010]. Studies to date used 3D rotating stimuli, so it is possible that the learning was peculiar to neurons in the motion sensitive areas, such as MT, that are jointly tuned to motion direction and disparity [DeAngelis, Cumming & Newsome, 1998]. We examined whether cue recruitment occurs when stimuli do not contain motion signals. Learned Bias for 3-D Shape Perception without Object Motion