Abstract
In two experiments, we investigated the processing of rotational motion in a same/different task. In Experiment 1, participants viewed eight 33 ms frames (20° apart) or four 67 ms frames (40° apart) of a rotating (right-to-left and left-to-right) Greeble (courtesy Michael Tarr), followed by a pre- and post-masked, briefly presented (167 ms) test probe. Test probes were presented in differing orientations; in an orientation prior to the first exposed rotation frame, at all four exposed 40° rotations, and in an orientation beyond the exposed rotational sequence. We found that same judgment accuracy of probes beyond the exposed rotational sequence was less accurate than for probes in the final exposed position in the 40° rotation condition [consistent with Vuong and Tarr (Vision Sciences 2002)] but not in the “smoother” 20° rotation condition. In Experiment 2, we varied rotation duration; participants viewed four 67 ms frames (40° apart) or four 133 ms frames (40° apart) of a rotating Greeble followed by a pre- and post-masked, briefly presented (100 ms) test probe. We found that same judgment accuracy decreased for test probes beyond the exposed rotational sequence, and that accuracy was lower at the first exposed position than at later exposed positions in the sequence. A consistent rotational representation facilitates responding to test probes further along in the rotational sequence (Experiment 2), but the effect is quite specific; accuracy for test probes beyond the exposed rotational sequence were not facilitated (Experiments 1 and 2). Further research using test probes at interpolated positions will be discussed.