Representational momentum using complex, continuous motion

Representational momentum (RM) refers to a memory distortion in which the final position or configuration of a moving object is misremembered as being too far forward along its actual or implied trajectory. RM has been well established when motion or change is implied, however, some debate still exists as to whether it can be observed with smooth, continuous motion. Typically, smooth motion studies have involved simple translation or rotation in the picture plane. Under these conditions, when eye movements are controlled, little if any forward shift is obtained. This suggests that overtracking may account for the errors observed with smooth motion. However, it is also possible that the addition of a salient central reference point (fixation cross) also influences localization accuracy given such simple displays. The purpose of the current work was to measure RM with continuous displays where eye movement artifacts would be unlikely to occur. In one series of studies, video sequences of human crowd scenes were used as stimuli. In another, complex, non-rigidly deforming novel shapes were expanded and contracted around a central fixation point. In all studies we found memory distortions consistent with the errors observed in previous RM studies. These findings suggest that Freyd's (1987) “dynamic mental representations” may underlie our perception of continuous as well as implied motion sequences