Department of Psychology and Key Laboratory of Machine Perception (Ministry of Education), Peking University, Beijing, P. R. China.
PLoS One. 2011 Feb 23;6(2):e17130. doi: 10.1371/journal.pone.0017130.
Previous studies have shown that in tasks requiring participants to report the direction of apparent motion, task-irrelevant mono-beeps can "capture" visual motion perception when the beeps occur temporally close to the visual stimuli. However, the contributions of the relative timing of multimodal events and the event structure, modulating uni- and/or crossmodal perceptual grouping, remain unclear. To examine this question and extend the investigation to the tactile modality, the current experiments presented tactile two-tap apparent-motion streams, with an SOA of 400 ms between successive, left-/right-hand middle-finger taps, accompanied by task-irrelevant, non-spatial auditory stimuli. The streams were shown for 90 seconds, and participants' task was to continuously report the perceived (left- or rightward) direction of tactile motion. In Experiment 1, each tactile stimulus was paired with an auditory beep, though odd-numbered taps were paired with an asynchronous beep, with audiotactile SOAs ranging from -75 ms to 75 ms. Perceived direction of tactile motion varied systematically with audiotactile SOA, indicative of a temporal-capture effect. In Experiment 2, two audiotactile SOAs--one short (75 ms), one long (325 ms)--were compared. The long-SOA condition preserved the crossmodal event structure (so the temporal-capture dynamics should have been similar to that in Experiment 1), but both beeps now occurred temporally close to the taps on one side (even-numbered taps). The two SOAs were found to produce opposite modulations of apparent motion, indicative of an influence of crossmodal grouping. In Experiment 3, only odd-numbered, but not even-numbered, taps were paired with auditory beeps. This abolished the temporal-capture effect and, instead, a dominant percept of apparent motion from the audiotactile side to the tactile-only side was observed independently of the SOA variation. These findings suggest that asymmetric crossmodal grouping leads to an attentional modulation of apparent motion, which inhibits crossmodal temporal-capture effects.
先前的研究表明，在要求参与者报告视错觉运动方向的任务中，当听觉刺激在时间上接近视觉刺激时，与任务无关的单一声响可以“捕获”视觉运动感知。然而，多模态事件的相对时间和事件结构，调节单一和/或跨模态感知分组的作用仍然不清楚。为了研究这个问题，并将研究扩展到触觉模态，当前的实验呈现了触觉双点击视错觉运动流，在连续的左手/右手中指点击之间的 SOA 为 400 毫秒，同时伴随着与任务无关的非空间听觉刺激。这些流显示了 90 秒，参与者的任务是连续报告触觉运动的感知（向左或向右）方向。在实验 1 中，每个触觉刺激都与一个听觉哔哔声配对，尽管奇数编号的点击与异步哔哔声配对，听觉触觉 SOA 从-75 毫秒到 75 毫秒不等。触觉运动的感知方向与听觉触觉 SOA 系统地变化，表明存在时间捕获效应。在实验 2 中，比较了两个听觉触觉 SOA--一个短（75 毫秒），一个长（325 毫秒）。长 SOA 条件保留了跨模态事件结构（因此时间捕获动力学应该与实验 1 相似），但现在两个哔哔声都出现在一侧的点击时间附近（偶数编号的点击）。发现这两个 SOA 对视错觉产生相反的调制，表明跨模态分组的影响。在实验 3 中，只有奇数编号的点击与听觉哔哔声配对，而不是偶数编号的点击。这消除了时间捕获效应，而是观察到来自听觉触觉侧到仅触觉侧的视错觉的主导感知，与 SOA 变化无关。这些发现表明，不对称的跨模态分组导致视错觉的注意力调制，抑制了跨模态的时间捕获效应。
