4.4 Article

Luminance Contrast Drives Interactions between Perception and Working Memory

期刊

JOURNAL OF COGNITIVE NEUROSCIENCE
卷 34, 期 7, 页码 1128-1147

出版社

MIT PRESS
DOI: 10.1162/jocn_a_01852

关键词

-

资金

  1. Brain and Behavior Research Foundation
  2. Biotechnology and Biological Sciences Research Council [BB/H019731/1, BB/R009287/1]

向作者/读者索取更多资源

This study investigates the differences in the role of luminance and chromaticity in visual working memory and finds that luminance plays a crucial role in the representation of shape in WM.
Visual working memory (WM) enables the use of past sensory experience in guiding behavior. Yet, laboratory tasks commonly evaluate WM in a way that separates it from its sensory bottleneck. To understand how perception interacts with visual memory, we used a delayed shape recognition task to probe how WM may differ for stimuli that bias processing toward different visual pathways. Luminance compared with chromatic signals are more efficient in driving the processing of shapes and may thus also lead to better WM encoding, maintenance, and memory recognition. To evaluate this prediction, we conducted two experiments. In the first psychophysical experiment, we measured contrast thresholds for different WM loads. Luminance contrast was encoded into WM more efficiently than chromatic contrast, even when both sets of stimuli were equated for discriminability. In the second experiment, which also equated stimuli for discriminability, early sensory responses in the EEG that are specific to luminance pathways were modulated by WM load and thus likely reflect the neural substrate of the increased efficiency. Our results cannot be accounted for by simple saliency differences between luminance and color. Rather, they provide evidence for a direct connection between low-level perceptual mechanisms and WM by showing a crucial role of luminance for forming WM representations of shape.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.4
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据