Correlative comparison of visual working memory paradigms and associated models

Hojjati, Fatemeh; Motahharynia, Ali; Adibi, Armin; Adibi, Iman; Sanayei, Mehdi · 2024 · Crossref

DOI: 10.1038/s41598-024-72035-5

archive: archived pipeline: cataloged verified

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

Summary

This study addresses the ongoing theoretical debate regarding the mechanisms of visual working memory (WM), specifically contrasting the "slot model," which posits a fixed number of discrete storage units, and the "resource model," which suggests a continuous pool of resources distributed among items. The authors sought to determine the relationship between two widely used experimental paradigms associated with these theories: the delayed match-to-sample (DMS) task, often linked to slot-based capacity, and the analog recall paradigm, associated with resource-based precision. By correlating performance across these tasks, the study aimed to clarify whether they measure distinct aspects of WM or share underlying cognitive processes. The researchers recruited 30 healthy volunteers to perform two psychophysical tasks. In the sequential analog recall task, participants memorized the orientations of three colored bars presented sequentially and adjusted a probe bar to match the target orientation. In the DMS task, participants memorized a checkerboard pattern and identified the correct pattern from two options after delays ranging from 0.5 to 8 seconds. Statistical analysis employed Spearman’s correlation coefficients to compare DMS performance (accuracy) with recall error, precision, and error sources (target, non-target/swap, and uniform/guessing) derived from a three-component mixture model in the analog task. The results revealed significant correlations between the two paradigms. DMS performance was negatively correlated with mean recall error ($r = -0.60$) and positively correlated with precision ($r = 0.60$) in the sequential task. Furthermore, DMS accuracy correlated positively with the proportion of target errors ($r = 0.59$) and negatively with swap errors ($r = -0.55$), but showed no significant correlation with random guessing. These correlations persisted even when controlling for task difficulty using a single-bar analog condition. The strength of correlation varied by stimulus order, with the third bar in the sequence showing the strongest association with DMS performance, suggesting that lower memory load conditions yield more comparable results across paradigms. The findings indicate that DMS and analog recall tasks are not independent measures but reflect overlapping elements of WM processing. This supports hybrid models that integrate features of both slot and resource theories, suggesting that WM capacity and precision are intrinsically linked. The study concludes that binary paradigms like DMS cannot fully capture all error types, such as random guessing, which are better assessed via analog tasks. These insights have practical implications for clinical settings, suggesting that simpler, content-based tasks like DMS can provide assessments comparable to more complex precision tasks for monitoring cognitive decline in neurological disorders.

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.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

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

• dual task multitasking