Coarse-to-fine eye movement behavior during visual search

Godwin, Hayward J.; Reichle, Erik D.; Menneer, Tamaryn · 2014 · Crossref

DOI: 10.3758/s13423-014-0613-6

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 the mechanisms underlying "coarse-to-fine" eye movement behavior during visual search, a pattern characterized by increasing fixation durations and initially increasing, then decreasing, saccade amplitudes as a trial progresses. While previous research interpreted this pattern as evidence of a deliberate strategic shift from global to local processing, Godwin, Reichle, and Menneer propose an alternative explanation: that these behavioral trends emerge from stochastic factors controlling eye movements rather than a conscious strategy. The authors aim to determine if a parsimonious computational model, incorporating well-established principles of saccadic programming and execution, can replicate these empirical findings without assuming a strategic shift. To test this hypothesis, the authors conducted an experiment with 16 participants who searched for a colored T-shape among offset L-shape distractors on a virtual grid. Eye movements were recorded using an Eyelink 1000 system. The experimental design included 640 trials, with participants searching for targets of varying colors to ensure generalizability. Data processing excluded outliers, incorrect responses, and fixations influenced by response preparation or saccadic inhibition. The authors then developed a Monte Carlo simulation model based on four principles: probabilistic saccade targeting biased toward proximal stimuli; inhibition of return to previously fixated locations maintained in a short-term visual buffer; fixation durations determined by stimulus encoding time and saccade programming latency (which increases with saccade length); and the inspection of only a subset of stimuli based on similarity and response criteria. The results replicated the coarse-to-fine pattern: fixation durations significantly increased across ordinal fixations, while saccade amplitudes peaked early in the trial and showed a trend toward decreasing later. Crucially, the computational model accurately simulated these observed means, with a mean absolute percentage deviation of only 1.74%. The model’s best-fitting parameters, such as a visual buffer capacity of four items and saccadic latencies between 110 and 260 ms, were consistent with prior estimates from visual-cognitive research. This demonstrated that the interaction of stochastic factors—specifically probabilistic selection, inhibition of return, and saccadic programming constraints—suffices to explain the observed eye movement patterns. The significance of this work lies in challenging the interpretation of coarse-to-fine behavior as a deliberate strategic shift. By showing that simple, well-established oculomotor principles can generate these patterns, the authors argue that future models of visual search do not need to incorporate complex strategic assumptions. This parsimonious framework provides a robust theoretical tool for simulating eye movements and highlights the importance of accounting for stochastic variability in experimental analyses to avoid inaccurate conclusions about search strategies.

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
• inattentional change blindness