Induced-anxiety differentially disrupts working memory in generalized anxiety disorder

Abstract

BACKGROUND: Anxiety is characterized by a bias towards threatening information, anxious apprehension, and disrupted concentration. Previous research in healthy subjects suggests that working memory (WM) is disrupted by induced anxiety, but that increased task-demand reduces anxiety and WM is preserved. However, it is unknown if patients with generalized anxiety disorder (GAD) can similarly normalize their performance on difficult WM tasks while reducing their anxiety. Increased threat-related bias and impoverished top-down control in trait anxiety suggests that patients may not reap the same cognitive and emotional benefits from demanding tasks that those low in anxiety. Here we examine this possibility using a WM task of varying difficulty. METHODS: GAD patients (N = 30) and healthy controls (N = 30) performed an n-back task (no-load, 1-back, 2-back, and 3-back) while at risk for shock (threat) or safe from shock (safe). Anxiety was measured via startle reflex and self-report. RESULTS: As predicted, healthy controls' performance was impaired under threat during low-load tasks and facilitated during high-load tasks. In contrast, GAD patients' performance was impaired under threat regardless of WM load. Anxiety was reduced as cognitive load increased in both groups. CONCLUSIONS: The divergence of emotion regulation (reduction) and performance (persistent impairment) in the patient but not the control group, suggests that different top-down mechanisms may be operating to reduce anxiety. Continued WM disruption in patients indicates that attentional resources are allocated to emotion regulation instead of goal-directed behavior. Implications for our understanding of cognitive disruption in patients, and related therapeutic interventions are discussed.

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 — 38 citations exist upstream.

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