Investigating memory episodes in location probability learning: Can altering response features reset spatial bias?

Yu, Xinger; Woodman, Geoffrey F. · 2025 · Crossref

DOI: 10.3758/s13414-025-03106-6

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 altering motor response features can reset spatial attentional biases acquired through implicit statistical learning. The research is motivated by the episodic-retrieval account of statistical learning, which posits that individuals store task occurrences as memory episodes binding stimuli, responses, and outcomes. If this account holds, changing the motor response should disrupt the retrieval of these episodes, thereby negating the learned attentional bias toward high-probability locations. The authors tested this hypothesis using a probability cueing paradigm, where participants performed a visual search task to detect a target letter "T" among distractors "Ls." In two experiments, participants underwent a training phase where the target appeared in one specific quadrant 50% of the time (the "rich" quadrant) and equally across three other quadrants (the "sparse" quadrants). Participants were unaware of this distribution. During training, they used specific finger-key mappings based on target orientation. In the testing phase, Experiment 1 maintained the same spatial distribution but required half the participants (the "switch" group) to change their response keys and hands, while the other half (the "same" group) continued with their original configuration. Experiment 2 altered the spatial distribution so that a previously sparse quadrant became the new rich quadrant, while also manipulating response features for the switch group. The researchers measured reaction times (RTs), scan-path ratios, and fixation dwell times to assess attentional bias. The results from both experiments contradicted the initial hypothesis. In Experiment 1, changing response features did not affect the attentional bias; both groups showed significantly faster RTs and more efficient scan paths for targets in the rich quadrant compared to sparse quadrants, with no significant differences between the switch and same groups. In Experiment 2, despite the shift in target probability to a new quadrant, participants in both groups maintained a persistent attentional bias toward the original rich quadrant. The switch group did not demonstrate an enhanced ability to reallocate attention to the new rich quadrant or a reduced bias toward the old one. Furthermore, analyses of explicit awareness confirmed that the effects were implicit, as performance did not differ significantly between participants who were aware or unaware of the spatial regularities. These findings suggest that implicit location probability learning is robust and not easily disrupted by changes in task-irrelevant motor response features. The results challenge the notion that statistical learning relies solely on the episodic retrieval of bound stimulus-response-outcome episodes, at least regarding the motor component. Instead, the data imply that spatial attentional biases are formed independently of specific motor responses, highlighting a dissociation between spatial statistical learning and motor execution. This contributes to the understanding of how visual systems adapt to environmental regularities, indicating that such learning mechanisms are stable and resistant to minor changes in response demands.

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.

• inattentional change blindness