Zheng Yujie, Wang Tengfei, Li Zhi
Department of Psychology and Behavioral Sciences, Zhejiang University.
J Exp Psychol Hum Percept Perform. 2025 Aug;51(8):1025-1043. doi: 10.1037/xhp0001337. Epub 2025 May 8.
In the present study, we tested the validity of an assumption of the guided search model that visual selection is a serial process. This assumption predicts that a top-down search, such as the distinct color search (DCS; e.g., finding a target color among a set of distractors each of a unique color), is a serial search. We examined the search mode of DCS with the multiple-target search paradigm, which uses target redundancy gains to distinguish parallel from serial searches. The data suggest that DCS is normally a parallel search (Experiment 1) but becomes a serial search if its start time is delayed (Experiment 2). One plausible account to reconcile these results is that visual selection is serial by default but, right after the presentation of the search array, there is a short-lived bottom-up mechanism that may enable parallel visual selection. Our data further show that this short-lived mechanism has a capacity limit between 4 and 8 (Experiment 3). The present findings suggest that the current mainstream theories of visual search need to be modified and the effect of temporal factors on the search mode should be taken into account. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
在本研究中，我们检验了引导搜索模型的一个假设的有效性，即视觉选择是一个串行过程。该假设预测，自上而下的搜索，如独特颜色搜索（DCS；例如，在一组颜色各异的干扰项中找到目标颜色），是一种串行搜索。我们使用多目标搜索范式来研究DCS的搜索模式，该范式利用目标冗余增益来区分并行搜索和串行搜索。数据表明，DCS通常是并行搜索（实验1），但如果其开始时间延迟，就会变成串行搜索（实验2）。调和这些结果的一个合理的解释是，视觉选择默认是串行的，但在搜索阵列呈现后，会有一个短暂的自下而上机制，该机制可能会实现并行视觉选择。我们的数据进一步表明，这种短暂的机制的容量限制在4到8之间（实验3）。目前的研究结果表明，当前视觉搜索的主流理论需要修正，并且应该考虑时间因素对搜索模式的影响。（PsycInfo数据库记录（c）2025美国心理学会，保留所有权利）
