Just-in-Time Encoding Into Visual Working Memory Is Contingent Upon Constant Availability of External Information

Hoogerbrugge, Alex J.; Strauch, Christoph; Böing, Sanne; Nijboer, Tanja C. W.; Van der Stigchel, Stefan · 2024 · DOAJ

DOI: 10.5334/joc.364

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 trade-off between maintaining information in visual working memory (VWM) and sampling external information just-in-time. While previous research established that predictable costs, such as fixed delays or increased distance, shift reliance toward VWM, it remained unclear how unpredictable access to external information affects this balance. The authors hypothesized that any disruption to the constant availability of external information would increase VWM usage, regardless of whether that disruption was predictable or variable. The research employed two experiments using a copying task where participants recreated a stimulus layout from an example grid to a working grid. In Experiment 1, the example grid was intermittently occluded for varying durations (High, Medium, and Low availability conditions) compared to a baseline where it was always visible. In Experiment 2, the example grid was occluded by default and only appeared after a gaze-contingent delay. This delay was either constant (2 seconds) or variable (drawn from Gaussian distributions with low or high variance), allowing the authors to isolate the effects of predictability while controlling for average delay duration. Eye-tracking data measured inspection frequency, fixations per inspection, and correct placements. Results from Experiment 1 showed that intermittent occlusion reduced the frequency of external inspections compared to the baseline, indicating increased reliance on memory. Participants attempted to encode more information per inspection by increasing fixations, though this did not consistently result in more correct placements. Completion times increased, and participants spent a significant proportion of time unproductively waiting for the grid to reappear. Experiment 2 confirmed that introducing any delay, whether constant or variable, significantly decreased external inspections and increased VWM load compared to the no-delay baseline. Crucially, the variability of the delay did not alter behavior; participants responded similarly to constant and variable delays. The findings demonstrate that the decision to engage VWM is highly sensitive to the continuity of external information access. Any disruption to constant availability drives increased VWM usage, irrespective of whether the cost of sampling is predictable or variable. This suggests that the primary factor influencing the storage-sampling trade-off is the mere presence of a barrier to immediate access, rather than the uncertainty or magnitude of that barrier. The study highlights the dynamic nature of VWM, showing that humans prioritize just-in-time sampling only when external information is continuously and immediately accessible.

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.

StageOutcomeToolModelPromptAttemptsCompleted
discover success DOAJ 1 2026-06-17
archive success unpaywall 1 2026-06-25
extract success cached 2 2026-06-26
clean success clean 1 2026-06-18
chunk success chunk 1 2026-06-18
embed success embed Qwen/Qwen3-Embedding-8B 1 2026-06-18
promote success 1 2026-06-17
summarize success llm qwen3.6-27b-prismaquant summ-v5 1 2026-06-26
tag success vector_similarity 6 2026-06-18
verify success 1 2026-06-26

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.