Measuring Working Memory Capacity With Automated Complex Span Tasks

Engle, Randall W · 2012 · OpenAlex

DOI: 10.1027/1015-5759/a000123

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 addresses the need for standardized, reliable, and valid measures of working memory capacity (WMC), a construct critical to understanding cognitive functioning, intelligence, and executive attention. While traditional complex span tasks (CSTs) have been widely used, their administration often varies across studies, hindering comparability. The authors present an evaluation of automated versions of three widely used CSTs—Operation Span, Symmetry Span, and Reading Span—which are computerized, mouse-driven, and freely available. The study aims to provide normative data, assess psychometric properties, and demonstrate the utility of these automated tasks for researchers and clinicians. The methods involved aggregating data from over 6,000 young adult participants (ages 17–35) across three testing locations: the University of Georgia, the University of North Carolina at Greensboro, and the Georgia Institute of Technology. This diverse sample included students of varying academic selectivity and nonstudents, allowing for a broad representation of cognitive abilities. Participants completed at least one of the automated CSTs, which require interleaved processing and storage tasks. The authors analyzed test-retest reliability, internal consistency, convergent validity, and criterion-related validity. They also examined performance differences based on sex and college status, comparing partial-credit scoring (sum of correctly recalled items) against absolute scoring (all-or-nothing trial success). The results indicate that automated CSTs possess strong psychometric properties. Test-retest reliabilities were high, with partial scores showing greater stability than absolute scores. Internal consistency coefficients for partial scores exceeded recommended thresholds, whereas absolute scores yielded lower reliability. Convergent validity was demonstrated through strong correlations among the three tasks (r ≥ .52) and significant associations with fluid intelligence measures like Raven Progressive Matrices, though correlations with crystallized intelligence were weaker. Gender differences were minimal; males slightly outperformed females on storage tasks, but effect sizes were small. Performance varied by college status, with students from more selective institutions scoring higher. Crucially, the analysis confirmed that partial-credit scoring is superior to absolute scoring in terms of reliability and sensitivity to individual differences. The significance of this work lies in establishing automated CSTs as robust, standardized tools for measuring WMC. By providing normative data and demonstrating high validity and reliability, the authors facilitate cross-study comparisons and replication. The findings support the view that WMC reflects executive attention rather than mere storage capacity. The availability of these standardized tasks encourages broader research into how individual differences in WMC relate to various behaviors, including clinical conditions like schizophrenia and aging-related cognitive decline. The paper concludes that while multiple measures should be used to avoid method variance, the automated CSTs offer a reliable, accessible, and psychometrically sound approach to assessing working memory.

Key finding

Automated complex span tasks demonstrate high test-retest reliability and internal consistency, with partial scoring methods yielding superior psychometric properties compared to absolute scoring.

Methodology

dataset

Sample size: 6274

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 (6 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.

• cognitive capacity variation
• working memory