Individual differences in long-range visual apparent motion

The correspondence problem in apparent motion occurs when motion tokens in one motion frame have multiple possible matches in the subsequent frame. Previously, Prins (2023, https://doi.org/10.3758/s13428-023-02061-0) observed a single outlier among six observers who, relative to the other observers, displayed a quantitatively and qualitatively different pattern of the effects of three variables on perceived motion correspondence. These variables were: the relative distances between tokens, whether motion paths preserved or violated the figural integrity of the tokens, and the attentional load of a secondary task. Here I test 33 observers to explore individual differences among observers further. Motion tokens positioned on diagonally opposed corners of an imaginary rectangle were alternated with tokens positioned on the other diagonal. Each frame contained a square and a round token such that only one of the possible solutions to the correspondence problem would preserve token integrity. During the motion sequence, observers performed one of two secondary attention-demanding tasks that required fixation at center of motion display. Observers can be divided into three categories: (1) those who tend to report motion along the shorter of the two possible motion paths (the nearest-neighbor principle), with an average bias towards reporting vertical motion and towards motion that preserves token integrity (n = 8), (2) those who (almost) exclusively report motion that preserves token integrity, regardless of relative distances between tokens, (n = 7), (3) those who (almost) exclusively report vertical (n = 14) or horizontal (n = 1) motion, or respond independent of any of the variables (n = 3). Strikingly, while all of (1) displayed strong trial-to-trial hysteresis, none of (2) did. Overall, results are consistent with individual differences being mediated by different relative roles of a low-level, competitive mechanism acting on the nearest-neighbor principle on the one hand and top-down cognitive control on the other.