When visual and verbal memories compete: Evidence of cross-domain limits in working memory

Cowan, Nelson · 2004 · OpenAlex

DOI: 10.3758/bf03196573

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper investigates whether visual and verbal working memory rely on separate, domain-specific resources or share a central, cross-domain limitation. Previous research, such as Luck and Vogel (1997), suggested that visual working memory operates independently of verbal retention, implying distinct capacity limits for different sensory modalities. However, the authors argue that these earlier studies may have used insufficient verbal loads to tax a shared central faculty, such as the focus of attention or central executive processes. The study aims to resolve this controversy by examining interference between a visuospatial task and varying levels of verbal memory load, specifically distinguishing between passive verbal rehearsal and active maintenance of novel information. The researchers employed a dual-task experimental design with 24 undergraduate participants. The primary task was a visual array comparison, where participants viewed a sample array of colored squares (four, six, or eight items) and then determined if a specific square in a subsequent test array had changed color. Concurrently, participants performed a secondary auditory task involving the vocal recitation of one of four conditions: no load, a two-digit list, a seven-digit random list, or their own memorized seven-digit telephone number. Vocal recitation was used to suppress phonological rehearsal, ensuring that any interference resulted from working memory demands rather than verbal coding of the visual stimuli. The design allowed for the analysis of performance based on whether the digit load was recalled correctly or incorrectly. The results demonstrated that visual working memory performance was unaffected by the no-load, two-digit, and telephone number conditions, with accuracy rates around 90–92%. However, performance decreased significantly when participants maintained a load of seven random digits, dropping to 85% accuracy. Crucially, the interference was not uniform; trials where the seven-digit load was recalled incorrectly showed markedly lower visual accuracy (74%) compared to trials where the digits were recalled correctly (84%). Furthermore, errors on the visual task and the digit task tended to co-occur, indicating a shared resource constraint rather than a simple tradeoff. The study also replicated the finding that a two-digit load does not impair visual performance, confirming that the interference effect depends on the magnitude of the verbal load. These findings challenge the hypothesis that visual and verbal working memory are entirely independent systems. The authors conclude that both domains are subject to at least one shared limit, likely related to the capacity of the focus of attention or a central executive resource. The data suggest that while domain-specific mechanisms may exist, they are constrained by a central bottleneck when information exceeds the capacity of passive rehearsal. The study implies that previous claims of domain specificity may have been artifacts of using loads too small to engage central resources. The results support theoretical models proposing a limited central workspace for working memory, though the authors note that further research is needed to distinguish whether this limit applies to storage capacity or the transfer of information into working memory.

Key finding

Visual working memory performance decreased markedly when paired with a load of seven random digits, whereas it was unaffected by concurrent recitation of a two-digit list or a known seven-digit sequence.

Methodology

lab_experiment

Sample size: 24

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via openalex_abstract on 2026-05-08 (13 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• working memory