Abstract

Visuospatial working memory (WM) is a fundamental but severely limited ability to temporarily remember selected stimuli. Several studies have investigated the underlying neural mechanisms of maintaining various visuospatial stimuli simultaneously (i.e., WM load, the number of representations that need to be maintained in WM). However, two confounding factors, namely verbal representation and encoding load (the number of items that need to be encoded into WM), have not been well controlled in previous studies. In this study, we developed a novel delayed-match-to-sample task (DMST) controlling for these two confounding factors and recorded scalp EEG signals during the task. We found that behavioral performance deteriorated severely as memory load increased. Neural activity was modulated by WM load in a dynamic manner. Specifically, higher memory load induced stronger amplitude in occipital and central channel-clusters during the early delay period, while the inverse trend was observed in central and frontal channel-clusters during late delay. In addition, the same inverse memory load effect, that was lower memory load induced stronger amplitude, was observed in occipital channel-cluster alpha power during late delay. Finally, significant correlations between neural activity and individual reaction time showed a role of late-delay central and frontal channel-cluster amplitude in predicting behavioral performance. Because the occipital cortex is important for visual information maintenance, the decrease in alpha oscillation was consistent with the cognitive role that is “gating by inhibition.” Together, our results from a well-controlled DMST suggest that WM load not exerted constant but dynamic effect on neural activity during maintenance of visuospatial objects.

Highlights

  • When immersed in a cluttered environment filled with objects, the limited mind can only remember items perceived to be relevant for later behavioral purposes

  • While in this kind of paradigm, it means the number of stimuli displayed in sample (Luck and Vogel, 1997; Todd and Marois, 2004; Palva et al, 2011; Roux et al, 2012), which confuses with the encoding load

  • This result suggested that individual working memory (WM) capacity was severely limited and could maintain only about three objects in the mind at the same time, consistent with the finding of a previous functional magnetic resonance imaging study (Xu and Chun, 2006)

Read more

Summary

Introduction

When immersed in a cluttered environment filled with objects, the limited mind can only remember items perceived to be relevant for later behavioral purposes This cognitive ability is known as working memory (WM), which consists of verbal and visual forms (Baddeley, 2012). Many researchers used the N-back task to determine the underlying mechanism (Cohen et al, 1997; Gevins et al, 1997) This task confuses the three stages of visual WM, namely encoding, maintenance, and retrieval. Increasing WM load by adding more items in sample display will inevitably induce greater encoding load Encoding load is another confounding factor that needs to be controlled, especially when investigating the time course of WM load effects on representation maintenance, since the exact end time of sample encoding is not known

Objectives
Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call