The load to distribute retrospective attention to multiple items within visual working memory

Takeno, Masae; Kitagami, Shinji · 2016 · DOAJ

DOI: 10.4189/shes.14.23

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 limitations of retrospective attention in visual working memory, specifically addressing whether attention can be effectively distributed to multiple items after encoding. While retrospective attention (retro-cuing) is known to enhance memory accuracy for a single item, previous research indicated a null effect when attention was distributed to multiple items. The authors sought to determine if this limitation arises from a strict capacity constraint (i.e., attention can only focus on one item) or from the cognitive load and effort required to maintain split attention across multiple items. To test these hypotheses, the researchers conducted two experiments using a probed-recall paradigm, which offers higher sensitivity than previous methods. In Experiment 1, 20 participants memorized four colored shapes. After a delay, they received retro-cues indicating either no items, one item, or two items to attend to, followed by a probe requiring oral recall of the missing feature. Results showed a large effect size for single-cueing ($d = 1.14$), significantly improving accuracy compared to the control condition. However, double-cueing did not significantly improve accuracy over the control condition, though a weak effect ($d = 0.35$) was detected, suggesting that distributing attention imposes a cognitive load. Experiment 2 aimed to reduce this load by eliminating the inter-stimulus interval (ISI) between item offset and cue presentation, thereby leveraging iconic memory to support attention distribution. Twenty-one participants performed the same task with an ISI of 0 ms. While single-cueing accuracy improved significantly compared to Experiment 1, double-cueing accuracy did not improve relative to the control condition or Experiment 1. The lack of improvement in the double-cue condition, despite the reduction in ISI, indicated that the cognitive load of splitting attention persisted and hindered performance. The findings suggest that the inability to enhance memory for multiple retro-cued items is not due to a hard limit on the number of items attention can target, but rather the significant cognitive effort required to distribute and maintain split attention. The study concludes that retrospective attention is limited by the load associated with separating attentional foci, rather than by a fundamental constraint on the number of items that can be attended. This implies that reducing the cognitive load of attention distribution could potentially allow for enhanced memory performance across multiple items.

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 2026-06-25; verification: verified.

Topics

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

• attention allocation