Saccade trajectory curvature in visual search using natural scene images

Sogo, Hiroyuki; Takeda, Yuji · 2007 · Crossref

DOI: 10.4992/jjpsy.78.512

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 visual context influences saccade trajectory curvature during visual search, specifically examining if the inhibitory mechanism preventing re-examination of rejected distractors operates similarly in natural scenes as it does with abstract stimuli. Previous research established that saccade trajectories often curve away from previously fixated locations, a phenomenon termed inhibition-induced saccade trajectory deviation (ISTD). While ISTD had been documented using abstract stimuli (e.g., finding an "O" among "Cs"), it remained unclear if this effect persists in complex, naturalistic environments, given that other inhibitory effects like inhibition of return (IOR) are known to vary based on stimulus meaningfulness and context. To address this, the authors conducted an experiment with seven participants who performed a visual search task using computer-generated natural scene images. Participants were instructed to locate a toy car within the scenes or indicate its absence. Eye movements were recorded using an Eyelink II eye-tracker at a sampling rate of 250 Hz. The stimuli consisted of 84 unique indoor and outdoor scenes, half of which contained a toy car placed in physically plausible positions. The researchers analyzed saccade trajectories by calculating "Area Curvature," a metric representing the area enclosed between the saccade path and the straight line connecting its start and end points. They employed fixation maps and regression models to determine the influence of previous fixation locations on the curvature of subsequent saccades. The results demonstrated that saccade trajectories in natural scene search were significantly affected by previous fixations, consistent with findings from abstract stimulus studies. Specifically, saccades curved away from locations of the preceding three fixations, with the inhibitory effect decreasing exponentially as the number of intervening fixations increased. The influence of fixations older than three was negligible, corresponding to a temporal window of approximately 800 milliseconds. Statistical analysis confirmed that saccades deviated away from the side where previous fixations were more frequent. These findings indicate that ISTD occurs robustly in naturalistic contexts, unlike IOR, which can be modulated by scene semantics. The authors conclude that ISTD and IOR likely arise from distinct neural mechanisms. ISTD is attributed to motor control processes involving the superior colliculus, where inhibitory inputs from the frontal eye field prevent saccades toward recently fixated locations. In contrast, IOR involves cortical processing related to attentional planning and is sensitive to object information. The study suggests that ISTD serves as a simple, position-based inhibitory strategy that remains effective regardless of scene context, facilitating efficient visual search in dynamic environments by preventing inefficient re-examination of recently inspected areas.

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.

• visual search
• eye movements scanning