Nicolaas Prins¹ (); ¹University of Mississippi

Apparent motion refers to the phenomenon in which motion is perceived between two or more stationary visual tokens that are presented in successive frames. When frames contain multiple tokens, the visual system must solve the motion correspondence problem: Which token in one frame corresponds to which token in the alternate frame? Generally, the visual system prefers to assign correspondence between tokens that are separated by short distances (the nearest neighbor principle). However, when motion tokens positioned on diagonally opposed corners of a rectangle are alternated with tokens on the other two corners, horizontal and vertical motion are equally likely to be perceived when the vertical sides of the rectangle are longer than the horizontal sides. Substantial individual differences exist in the magnitude of this bias. Here I investigate whether the magnitude of this bias also varies within observers across different testing occasions. Each of eight observers tested in five identical sessions that were conducted on separate days. In each session, six independent adaptive trial sequences were conducted to relate the probability of perceiving horizontal motion to the ratio of vertical to horizontal distances between potential motion matches and to determine the equivalence ratio: The aspect ratio at which horizontal and vertical motion are equally likely to be perceived. While equivalence ratios were highly consistent within individual sessions, they varied substantially not only between observers but also between the sessions of individual observers. This within-observer, between-session variance was comparable in magnitude to the between-observer variance, exceeding the within-session variance of equivalence ratios by approximately half an order of magnitude. The mechanism behind the observed within-observer variance across sessions is speculative at present, but its existence carries important implications for the design, analysis, and interpretation of research into visual apparent motion.