工作记忆表征引导视觉注意选择的眼动研究

Working Memory Representation Does Guide Visual Attention:Evidence from Eye Movements

工作记忆表征能否引导视觉注意选择?目前实验结果尚不一致。有研究者认为能否观察到注意引导效应取决于视觉搜索类型。研究采用工作记忆任务与视觉搜索任务相结合的双任务范式,结合眼动追踪技术,对不同视觉搜索类型下的注意引导效应进行验证。实验1结果发现,不管视觉搜索任务的靶子是否变化,在早期的眼动指标上都发现了显著的注意引导效应,但注意引导效应在靶子固定的视觉搜索任务下表现得更强。实验2在平衡两种视觉搜索任务中的工作记忆负载后发现,两种视觉搜索任务下都出现了显著的注意引导效应,但没有发现实验1中所出现的任务间差异。实验结果否定了视觉搜索类型对注意引导效应的决定性影响,同时也提示工作记忆负载可能在注意引导效应中起重要作用。

Whether the working memory representations could guide visual attention to select the matching stimuli in visual search is still controversial. By requiring the participants to perform a visual search task while online keeping some objects in working memory, some researchers have observed a stronger interference from the distractor when it was identical or related to the object held in memory. But other researchers did not observe such attentional guidance effect even using similar procedures. Olivers （2009） examined several possible influencing factors through a series of experiments and finally attributed the discrepancy to the search type whether the search target was varied or not across trials throughout the experiment. However, according to our analysis, there were several factors might confound the results in the critical experiment of Olivers （2009）. So here, we used the classic dual task combined with eye movement tracking technology to reexamine and evaluate the effect of the search type on the top-down guiding process of visual attention from working memory representations. Experiment 1 aimed to reexamine the effect of search type on attentional guidance via counterbalancing the perceptual difficulty of the search array in two types of the visual search task. The experimental procedure was similar to that used in the Experiment 5 of Olivers （2009） except making the perceptual difficulty of search array to be equal between visual search tasks. The eye movement data showed that, both in fixed- and varied-target visual search task, the distractor which was identical to the working memory representation was more easily to capture visual attention than the control distractor which was irrelevant to the working memory representations. It is suggested that the attentional guidance would appear no matter which visual search type used. However, we also found that the magnitude of attentional guidance was greater in the fixed-target visual search task than in the varied-target visual search task. Experiment 2 aimed to examine whether search type or working memory load was the real fact which caused the difference of guidance effect between visual search tasks in experiment 1. In the present experiment, we counterbalanced the working memory load between two types of visual search task. Consistent to experiment 1, the results also showed significant attentional guidance in both visual search task, however, the magnitude of the attentional guidance between visual search tasks became equivalent. In conclusion, these results disobeyed the explanation that the visual search type was the determinant factor for whether the attentional guidance has been observed or not in the previous studies, and suggested that the attentional guidance from working memory was to some degree affected by the factors of working memory load.