Competition between color and luminance in motion correspondence
Competition between color and luminance in motion correspondence
复制标题
DOI：
10.3758/s13414-018-1589-5
复制
发表时间：
2019-01-01
影响因子：
1.7
通讯作者：
Schuetz, Alexander C.
中科院分区：
文献类型：
--
作者：
Hein, Elisabeth;Schuetz, Alexander C.
视觉系统需要解决对应问题(即，哪些元素跨越空间和时间属于一起)，以允许对象的稳定表示。已有研究表明，时空信息和特征信息可以影响这种对应过程，但这些因素如何相互作用尚不清楚，特别是当它们由于对比度大小的变化而或多或少地突出时。我们使用Ternus显示的一种变体来研究这个问题，这是一种模棱两可的外观运动显示，其中三个元素可以被感知为一起移动(群体运动)或一个元素跳过其他元素(元素运动)。在第一个实验中，我们通过呈现一些具有相同特征(等光颜色或亮度)的元素来偏向感知，使得它们要么与群体运动兼容，要么与元素运动(简单的特征偏向)兼容。为了改变特征偏向的强度，我们操纵了特征的对比度大小。在另外三个实验中，我们引入了竞争性显示，其中一些元素显示基于颜色/亮度的不同对比度大小的元素偏差，而其他元素显示不同对比度大小的基于亮度/颜色的组偏差(竞争特征偏差)。我们发现，对于简单的特征偏向，对比度大小不影响偏向的强度。然而，对于相互竞争的特征偏差，对比度大小确实影响了一致性，因为偏差强度随着对比度的增加而增加。讨论了我们的结果对当前运动和基于特征的对应理论的影响。
The visual system needs to solve the correspondence problem (i.e., which elements belong together across space and time) to allow stable representations of objects. It has been shown that spatiotemporal and feature information can influence this correspondence process, but it is unclear how these factors interact with each other, especially when they are more or less prominent due to changes in contrast magnitude. We investigated this question using a variation of the Ternus display, an ambiguous apparent motion display, in which three elements can either be perceived as moving together (group motion) or as one element jumping across the others (element motion). In the first experiment, we biased the percept by presenting some of the elements with the same feature (isoluminant color or luminance), such that they were either compatible with group motion or with element motion (simple feature biases). To change the strength of the feature bias, we manipulated the contrast magnitude of the feature. In three more experiments we introduced competitive displays, in which some of the elements showed a color/luminance based element bias of varying contrast magnitude, while other elements showed a luminance/color based group bias of varying contrast magnitude (competing feature bias). We found that for a simple feature bias the contrast magnitude did not affect the strength of the bias. For competing feature biases, however, the contrast magnitude did influence correspondence, as the bias strength increased with contrast. The implications of our results for current motion and feature-based theories of correspondence are discussed.