Lateralized alpha oscillations are irrelevant for the behavioral retro-cueing benefit in visual working memory.

Wanja A Mössing,Niko A Busch

doi:10.7717/peerj.9398

Abstract

The limited capacity of visual working memory (vWM) necessitates the efficient allocation of available resources by prioritizing relevant over irrelevant items. Retro-cues, which inform about the future relevance of items after encoding has already finished, can improve the quality of memory representations of the relevant items. A candidate mechanism of this retro-cueing benefit is lateralization of neural oscillations in the alpha-band, but its precise role is still debated. The relative decrease of alpha power contralateral to the relevant items has been interpreted as supporting inhibition of irrelevant distractors or as supporting maintenance of relevant items. Here, we aimed at resolving this debate by testing how the magnitude of alpha-band lateralization affects behavioral performance: does stronger lateralization improve the precision of the relevant memory or does it reduce the biasing influence of the irrelevant distractor? We found that it does neither: while the data showed a clear retro-cue benefit and a biasing influence of non-target items as well as clear cue-induced alpha-band lateralization, the magnitude of this lateralization was not correlated with any performance parameter. This finding may indicate that alpha-band lateralization, which is typically observed in response to mnemonic cues, indicates an automatic shift of attention that only coincides with, but is not directly involved in mnemonic prioritization.

Highlights

Working memory is the ability to maintain and process information that is no longer physically present
The goal of this study was to test whether alpha lateralization during the maintenance interval is a neuronal mechanism supporting the retro-cueing effect in visual working memory (vWM), and whether it accomplishes this via facilitation or inhibition
The maximum likelihood estimates (MLEs) of the mixture model parameters (Fig. 3B) showed that precision κ was significantly higher for retro-cued trials (Mean difference (Md) = 5.40, 95% CI [3.13, 7.66], t(38) = 4.83, p < 0.001, d = 0.35)

Summary

Introduction

Working memory is the ability to maintain and process information that is no longer physically present. It acts as the mind’s main work space and is a core element of human cognition underlying many critical functions (cf., Baddeley, 2015) such as general intelligence (Conway, Kane & Engle, 2003), attention (Oberauer, 2019), or reading comprehension (Tighe & Schatschneider, 2016). The more items are maintained, the less resources are available per item, reducing the quality of the items’ representations While these models disagree on the architecture of vWM and on the exact cause of capacity limitations, they both highlight that without prioritization of relevant over irrelevant information, the available capacity may be insufficient for maintaining all information with the desired quantity or quality

Objectives

Results

Discussion

Conclusion