Attention rhythmically samples multi-feature objects in working memory

Chota, Samson; Leto, Carlo; van Zantwijk, Laura; Van der Stigchel, Stefan · 2022 · DOAJ

DOI: 10.1038/s41598-022-18819-z

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 attention directed toward internal visual working memory (VWM) representations operates rhythmically, similar to attention directed at external stimuli. While previous research established that attention samples single-feature objects in VWM in a theta-rhythmic manner (3–8 Hz), it remained unclear if this mechanism extends to multi-feature objects. The authors hypothesized that non-spatial retro-cues would initiate comparable theta-rhythmic sampling of multi-feature items, suggesting a shared attentional mechanism for internal and external visual processing. The experiment involved 26 participants performing a two-alternative forced-choice working memory task. Participants encoded two circular gratings, each defined by a unique color and orientation. A retro-cue indicated which item would likely be probed, followed by a variable delay (200–700 ms) and a probe stimulus. Participants judged whether the probe’s orientation was tilted clockwise or counterclockwise relative to the memorized item. Reaction times (RTs) and accuracy were measured across densely sampled time points to detect rhythmic fluctuations. The authors analyzed the data using two statistical pipelines: a traditional time-shuffling method and a more robust autoregressive AR(1) model recommended by Brookshire (2021) to control for autocorrelation and Type I errors. The results demonstrated that participants responded significantly faster and more accurately to cued probes than uncued ones. Crucially, spectral analysis of RTs revealed significant rhythmic modulation in the theta range (~4.4–4.6 Hz) for validly cued trials. Phase analysis confirmed that RT fluctuations for cued and uncued items oscillated in antiphase, indicating that attention alternates sequentially between the two memory representations. The AR(1) analysis provided stronger evidence, showing significant rhythmic peaks in both RTs and accuracy for the valid condition, whereas the traditional method only detected significant RT rhythms. Accuracy fluctuations did not show a consistent antiphasic relationship, suggesting that while speed is rhythmically modulated, accuracy may be influenced by other factors. These findings extend the understanding of rhythmic attention to multi-feature objects in working memory, supporting the hypothesis that internal and external visual representations are accessed via similar oscillatory mechanisms. The antiphasic modulation suggests that attention does not maintain a static focus but rhythmically samples competing representations. This provides a potential solution to the binding problem in working memory, implying that rhythmic sampling helps integrate multiple features of an object. Furthermore, the study highlights the importance of using appropriate statistical methods, such as AR(1) models, to accurately detect behavioral rhythms and avoid false positives in time-series analysis.

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.

• attention selective divided