Sequential spatial regularities bias visual attention and guide anticipatory eye movements
Poster Presentation 36.470: Sunday, May 17, 2026, 2:45 – 6:45 pm, Pavilion
Session: Attention: Spatial
Schedule of Events | Search Abstracts | Symposia | Talk Sessions | Poster Sessions
Keara M. Ginell1, Ren E. Hentz1, David E. Osher1; 1The Ohio State University
Implicitly learned sequential spatial regularities bias visual attention during search. Previous research shows that memory-guided and top-down attentional biases reflect predictive eye movements. However, whether memory-guided attention, through sequential spatial regularities, also guides anticipatory eye movements is unclear. This question is imperative for understanding how attentional habits emerge through experience. We investigated whether implicitly learned spatial sequences guide pre-stimulus eye movements, using an attentional serial reaction time (attSRT) task. On each trial, a Gabor appeared briefly at one of four locations, and participants reported its tilt. Blocks alternated between learned sequences, where target locations followed a repeating spatial sequence across trials; and novel sequences, where locations were randomized. Participants were not informed of the sequence and could freely move their eyes. Eye movements were analyzed using time-to-first fixation on the target location, and the proportion of gaze samples in target vs. non-target locations (during both the pre-stimulus period and stimulus presentation). After learning the sequential regularities, fixations tended to reach the target location earlier than when the regularities were absent. This pattern, along with faster reaction times for the learned condition, suggests that sequential regularities may provide a spatial attentional bias via efficient eye movements. The proportion of gaze samples at stimulus onset was higher within the target location than non-target locations, confirming that eye movements were made to the Gabor target. During the pre-stimulus period, the proportion of gaze samples to the target location was greater for the learned condition compared to the novel condition. Collectively, the faster time-to-fixate and higher proportion of gaze samples to the pre-stimulus target location suggest that implicitly learned sequential spatial regularities can guide predictive eye movements while boosting behavioral performance (e.g., faster response times). We conclude that attentional habits may form efficient search performance, partially through guiding predictive eye movements.