Flexible representations in visual working memory and interactions with long‐term learning: Commentary on the special issue

Abstract

This special issue of the British Journal of Psychology brings together cutting edge research on a range of topics in visual working memory (VWM). In this commentary, we attempt to summarize common themes in current VWM research exemplified in this issue. The articles include several reviews of important topics as well as empirical papers covering three main themes. The first concerns the nature of mental representations of memoranda in the commonly used delayed estimation task, where both fine-grained and broad categorical details appear to be represented, and their susceptibility to interference. The second concerns interactions between VWM representations, both those that produce individuation of representations and those that create an overarching ensemble structure. Finally, the third main topic concerns the use of VWM during visual search and in the learning of repeated configurations in search displays. The work presented here, and other work in the field, points to a rich interplay between representations in VWM but also between VWM and information in long-term memory. Opportunities for further investigation are highlighted throughout.

Dependence of Working Memory on Coordinated Activity Across Brain Areas

“…The role of long-term memory in WM ( Jensen and Lisman, 2005 ) remains controversial. The observation of prefrontal-hippocampal synchrony during WM maintenance ( Brincat and Miller, 2015 ) seems consistent with the suggestion that the hippocampus supports WM by activating long-term memory representations ( Eriksson et al, 2015 ; Loaiza and Halse, 2019 ; Rhodes and Cowan, 2019 ). In support of recruitment of the hippocampus during WM, there is evidence for an increase in the γ band power within PFC and the hippocampus during WM ( Brzezicka et al, 2019 ), and imaging studies have shown an interaction between PFC and the hippocampus during WM tasks ( Gluth et al, 2015 ; Calabro et al, 2020 ).…”

“…Distributed network models of WM posit that interactions between association areas (such as parietal and prefrontal cortex) are necessary for memory maintenance ( Leavitt et al, 2017 ). In the activation of long-term memory theory, PFC-hippocampal interactions maintain WM via activation of long-term memory representations ( Eriksson et al, 2015 ; Loaiza and Halse, 2019 ; Rhodes and Cowan, 2019 ); it has also been suggested that information must pass through WM before entering long-term memory ( Eriksson et al, 2015 ; Loaiza and Halse, 2019 ; Rhodes and Cowan, 2019 ). Synchrony between the PFC and areas associated with long-term memory (such as the hippocampus) could reflect either of these processes.…”

Current directions in visual working memory research: An introduction and emerging insights

“…The collection of articles is rounded off by an integrative commentary by Rhodes and Cowan (). Although some current topics, such as the active manipulation of VWM content (e.g., Christophel, Cichy, Hebart, & Haynes, ; Liesefeld & Zimmer, ) and the relationship of LTM and VWM (e.g., Brady, Störmer, & Alvarez, ; Schurgin, Cunningham, Egeth, & Brady, ) are underrepresented here, we believe that reading through the articles in this issue provides a relatively comprehensive overview on the current state of VWM research.…”