Selective storage and maintenance of an object's features in visual working memory

Woodman, Geoffrey F.; Vogel, Edward K. · 2008 · OpenAlex-citations

DOI: 10.3758/pbr.15.1.223

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 whether visual working memory (VWM) obligatorily stores all features of an attended object or if individuals can selectively store only task-relevant attributes. While VWM capacity is limited to approximately three or four objects, it remains unclear if encoding one feature (e.g., color) necessitates the storage of other visible features (e.g., orientation). The authors contrast the "obligatory storage" hypothesis, which suggests all features are encoded together, with the "selective storage" hypothesis, which posits that VWM can flexibly encode and maintain only relevant information. To test these hypotheses, Woodman and Vogel employed two distinct methodologies: psychophysical masking experiments to measure encoding efficiency and electroencephalography (EEG) to measure maintenance. In the masking experiments, participants viewed multifeature objects (colored/oriented bars or colored/shaped stimuli) and were instructed to remember either a single feature or the conjunction of features. Performance was measured using change-detection tasks with varying stimulus-onset asynchronies (SOAs) to determine the time required to consolidate information into VWM. In the electrophysiological experiment, researchers recorded the contralateral delay activity (CDA), a neural marker of VWM maintenance, while participants remembered color, orientation, or both features of objects. The results supported the selective storage hypothesis. Psychophysical data revealed that consolidation rates varied systematically based on the task-relevant feature. Encoding color was significantly faster (approximately 68–84 msec/item) than encoding orientation or shape (approximately 118–170 msec/item). Crucially, when participants were required to remember both features, the consolidation rate matched that of the slower-to-encode feature, indicating parallel processing rather than additive serial processing. Electrophysiological results confirmed that maintenance was also selective; the CDA amplitude was significantly larger when orientation was task-relevant compared to when only color was relevant, despite identical stimuli. This indicates that irrelevant features were not actively maintained in VWM. Additionally, selective storage conferred processing advantages, as reaction times were faster in single-feature conditions than in conjunction conditions. The findings demonstrate that VWM is not a rigid system that obligatorily binds all object features. Instead, individuals can exert top-down control to selectively encode and maintain only task-relevant information. This selective mechanism offers efficiency benefits, allowing for faster encoding and reduced response competition from irrelevant information. The study concludes that the ability to select specific features for storage is a common capability across individuals, distinct from general VWM capacity limits, and suggests that VWM representations are flexible and dependent on current task goals.

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.

• working memory