Disruption of visual feature binding in working memory

Ueno, Taiji; Allen, Richard J.; Baddeley, Alan; Hitch, Graham J.; Saito, Satoru · 2010 · Memory & Cognition

DOI: 10.3758/s13421-010-0013-8

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 investigates the fragility of visual feature binding in working memory, specifically examining how subsequent visual stimuli (suffixes) disrupt the retention of bound object representations compared to individual features. Motivated by prior findings that bound representations are susceptible to retroactive interference, the authors sought to determine if this vulnerability persists when the interfering stimulus is to be ignored rather than intentionally encoded. The research aims to differentiate between general attentional costs of filtering and specific interference with bound memory traces. The researchers conducted five experiments using a visual suffix paradigm. Participants viewed arrays of colored shapes and were tested on their memory for color, shape, or the binding of both features. A visual suffix was presented during the retention interval, which participants were instructed to ignore. Experiments 1A, 1B, and 2 utilized suffixes with features external to the study set, varying only in consistency and variability. Experiments 3A and 3B used suffixes drawn from the same feature pool as the study items, creating potential feature overlap. All experiments employed a single-probe recognition task with articulatory suppression to prevent verbal coding. Results from Experiments 1A, 1B, and 2 showed a modest but equivalent disruption of memory for individual features and bound features when suffixes were external to the study set. No significant interaction between suffix presence and stimulus type was found, indicating that variable, non-overlapping suffixes did not selectively impair binding. However, Experiments 3A and 3B, where suffixes shared features with the study set, revealed significantly greater disruption for bound features than for individual features. This selective interference occurred despite participants being instructed to ignore the suffix. The findings suggest that visual suffix interference comprises two components: a general component affecting all memory types, attributed to the attentional cost of filtering out the suffix, and a specific component affecting bound representations. The specific disruption arises when the suffix features could potentially form part of the to-be-remembered set, causing the filtering mechanism to fail and allowing interference with the fragile bound object representations. This supports models of visual working memory where feature and binding information are stored at different levels, with binding being particularly vulnerable to interference from stimuli sharing feature dimensions.

Key finding

Visual suffixes composed of features from the same pool as the target stimuli disrupt memory for bound features significantly more than memory for individual features, whereas suffixes with non-overlapping features disrupt both equally.

Methodology

lab_experiment

Sample size: 204

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 author_sweep_intake on 2026-05-28.

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.

• inattentional change blindness