Abstract
Ambiguous apparent motion stimuli have long been used to study the effects of motion priming. It is well established that, at certain frame and prime durations, attending to a unidirectional drifting motion-pattern can constrain the correspondence solution of subsequently viewed ambiguous frames. In a recently published study (Davidenko et al., 2017) we showed that priming with a bidirectional rebounding motion-pattern at 2.5Hz produces a similar effect in sufficiently ambiguous stimuli: persistent rebounding illusory apparent motion was reported as lasting for M = 6.2 frames (2.48 seconds) on average. The ability to prime rebounding motion could be interpreted as evidence for a high-level mechanism that operates by 1) encoding a two-step motion-pattern (i.e. left-right-left-right), 2) maintaining the temporal order of each step during ambiguous frame transitions, and 3) deploying selective attention during each frame transition such that the encoded pattern is reinstantiated. However, there is also evidence that illusory rebounding motion is a kind of default percept (Verstraten, Cavanagh, & Labianca, 2000; Hock, Park, & Schöner, 2002; Hsieh, Caplovitz, & Tse, 2005). For example, if neural populations that are tuned to opposing directions along the same axis (i.e. left and right) are related through inhibitory interactions, any consistent activation of these populations may result in a type of rivalry effect that maintains the rebounding pattern. In the present study, we dissociate these two explanations by priming with a non-rebounding two-step motion-pattern: staircase motion (e.g. up-right-up-right). We show that, as with drifting and rebounding patterns, exposure to a slowly moving staircase motion-pattern (1.5Hz) constrains subsequently viewed frames of refreshing random dot arrays to maintain the illusory staircase motion-pattern (M = 5.9 frames, 3.87 seconds). These results provide further evidence for the existence of a priming mechanism operating at the level of motion-pattern encoding.
Meeting abstract presented at VSS 2018