Visual search for dimensionally redundant pop-out targets: parallel-coactive processing of dimensions is location specific.
Visual search for dimensionally redundant pop-out targets: parallel-coactive processing of dimensions is location specific.
复制标题
DOI：
10.1037/0096-1523.28.6.1303
复制
发表时间：
2002-10
期刊：
影响因子：
--
通讯作者：
Joseph Krummenacher;H. Müller;D. Heller
中科院分区：
文献类型：
--
作者：
Joseph Krummenacher;H. Müller;D. Heller
三个视觉搜索实验研究了单个和双奇一出特征目标的冗余增益，这些目标与干扰物在方向、颜色或两者上不同。在实验1中，冗余目标显示包括(a)在2个维度上定义的单一目标，(b)在不同维度上定义的双重目标，或(c)在同一维度上定义的双重目标。相对于单个非冗余目标，冗余增益从第一个条件开始下降，仅在前两个条件中出现了对J. Miller（1982）种族模型不平等（RMI）的违反。实验2系统地改变了在不同维度上定义的双目标之间的空间间隔。只有当两个目标占领附近地点时，才明显违反了RMI。实验3提供了在预先（可能）和未提示（不可能）显示位置的维度冗余目标违反RMI的证据。综上所述，这些结果表明，不同维度的目标信号（而不是同一维度的信号）共同激活了一种共同机制，共同激活效应具有空间特异性，共同激活机制位于加工的前注意、知觉阶段。
Three visual search experiments investigated redundancy gains for single and dual odd-one-out feature targets that differed from distractors in orientation, color, or both. In Experiment 1, redundant-target displays contained (a) a single target defined in 2 dimensions, (b) dual targets each defined in a different dimension, or (c) dual targets both defined in the same dimension. The redundancy gains, relative to single nonredundant targets, decreased from the first condition on, with violations of J. Miller's (1982) race model inequality (RMI) manifested only in the first 2 conditions. Experiment 2 systematically varied the spatial separation between dual targets each defined in a different dimension. Violations of the RMI were evident only when the 2 targets occupied nearby locations. Experiment 3 provided evidence of RMI violations by dimensionally redundant targets at both precued (likely) and noncued (unlikely) display locations. Taken together, these results suggest that there is coactivation of a common mechanism by target signals in different dimensions (not by signals in the same dimension), that the coactivation effects are spatially specific, and that the coactivated mechanisms are located at a preattentive, perceptual stage of processing.