Distractors induce space-specific neural biases in visual working memory

Author:

Gupta SanchitORCID,Sridharan DevarajanORCID

Abstract

AbstractInformation in working memory is remarkably resilient to distraction. Yet, recent evidence suggests that distractors containing task-relevant features can disrupt working memory by inducing subtle biases in mnemonic representations. With multivariate decoding of human electroencephalography recordings, we show that temporally unpredictable distractors produce spatially-antagonistic mnemonic biases, across the visual hemifields. Grating distractors produced either an attractive or a repulsive mnemonic bias – a shift in the neural representation of the memorandum toward or away from the distractor’s orientation – depending, respectively, on whether the distractor appeared in the same hemifield as the memorandum, or opposite to it. Behavioral biases closely tracked these neural effects. We devised a two-tier ring attractor model with cross-hemifield inhibition, which comprehensively explains how the distractor’s timing, encoding strength, and input gating control these mnemonic biases. Our results provide a mechanistic account of distractor-induced biases, across space and time, in visual working memory.Lay SummaryWorking memory – the capacity to momentarily store important items or events – is remarkably robust to distraction. Yet, when a salient distractor with features resembling the memorized items appears, it can subtly alter (bias) working memory. Applying state-of-the-art neural decoding to human electrophysiological data, we show the such distractor biases are space-specific. The neural memory representation begins to resemble the distractor (attractive bias) when the distractor appears at the same location as the memory items, but becomes less like the distractor (repulsive bias) when the distractor appears further away. We tie together these experimental findings with a simple, computational model. The findings enhance our understanding of neural mechanisms underlying working memory and may be relevant for cognitive therapies addressing memory deficits.

Publisher

Cold Spring Harbor Laboratory

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3