Working memory maintenance is sufficient to reduce state anxiety

Abstract

According to the attentional control theory (ACT) proposed by Eysenck and colleagues, anxiety interferes with cognitive processing by prioritizing bottom-up attentional processes over top-down attentional processes, leading to competition for access to limited resources in working memory, particularly the central executive (Eysenck, Derakshan, Santos, & Calvo, ). However, previous research using the n-back working memory task suggests that working memory load also reduces state anxiety. Assuming that similar mechanisms underlie the effect of anxiety on cognition, and the effect of cognition on anxiety, one possible implication of the ACT would suggest that the reduction of state anxiety with increasing working memory load is driven by activation of central executive attentional control processes. We tested this hypothesis using the Sternberg working memory paradigm, where maintenance processes can be isolated from central executive processes (Altamura et al., ; Sternberg, ). Consistent with the n-back results, subjects showed decreased state anxiety during the maintenance period of high-load trials relative to low-load trials, suggesting that maintenance processes alone are sufficient to achieve this state anxiety reduction. Given that the Sternberg task does not require central executive engagement, these results are not consistent with an implication of the ACT where the cognition/anxiety relationship and anxiety/cognition relationship are mediated by similar central executive mechanisms. Instead, we propose an extension of the ACT such that engaging working memory maintenance suppresses state anxiety in a load-dependent manner. Furthermore, we hypothesize that the efficacy of this effect may moderate the effect of trait anxiety on cognition.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 37 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.