Fractal fluctuations in gaze speed visual search

Stephen, Damian G.; Anastas, Jason · 2010 · Crossref

DOI: 10.3758/s13414-010-0069-3

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 role of fractal fluctuations in gaze speed during visual search tasks, proposing that these fluctuations facilitate perceptual exploration beyond what is explained by higher-order cognitive mechanisms like memory. The authors reanalyze eye-tracking data from Stephen and Mirman (2010), focusing on single-feature and conjunction search tasks. The research is motivated by findings that individuals with autism spectrum disorders (ASDs) perform visual search faster than typically developing individuals despite having comparable visual working memory. This suggests a lower-order perceptual mechanism, potentially involving hyperdiffusive gaze fluctuations, contributes to search efficiency. The authors hypothesize that gaze orientation changes exhibit fractal fluctuations (Hypothesis 1) and that greater fractality predicts faster reaction times (Hypothesis 2). The study involved six participants who completed visual search tasks while their eye movements were recorded at 500 Hz. The single-feature task required identifying the letter "O" among distractors, while the conjunction task required finding a green "N" among green "X"s and brown "N"s. Each task consisted of blocks with varying set sizes (1, 5, 15, and 30 items). The authors analyzed the angular change of gaze (arctangent of Δy/Δx) rather than gaze position to avoid contamination from saccades. They employed Detrended Fluctuation Analysis (DFA) to estimate the scaling exponent H, which indicates the degree of fractality and diffusion speed. To corroborate findings of long-range temporal correlations, they also used Autoregressive Fractionally Integrated Moving Average (ARFIMA) modeling and dispersion analysis. Growth Curve Modeling (GCM) was used to test whether trial-by-trial changes in H predicted changes in reaction time (RT), controlling for task type, set size, and practice effects. The results confirmed both hypotheses. DFA analysis showed that original gaze time series exhibited fractal fluctuations with a mean scaling exponent H of 0.627, significantly higher than shuffled data (H = 0.478), indicating hyperdiffusion. ARFIMA and dispersion analyses provided converging evidence of long-range temporal correlations. Crucially, GCM revealed that higher fractality (greater H) significantly predicted lower reaction times. While the standard interaction between task type and set size was replicated, the addition of H as a predictor significantly improved the model's fit. The negative coefficient for H indicates that as the fractality of gaze orientation increases, visual search reaction time decreases. The significance of these findings lies in demonstrating that fractal fluctuations in gaze speed play a key role in the efficacy of perceptual exploration. The study suggests that hyperdiffusive gaze movements allow for efficient sampling of the visual field across multiple time scales, providing an advantage in exploration that is distinct from memory-driven mechanisms like inhibition of return. This highlights the importance of lower-order perceptual dynamics in cognitive performance, offering a potential explanation for individual differences in visual search efficiency, such as those observed in ASD populations. The work distinguishes itself from cognitive diffusion models by focusing on overt, measured perceptual-motor time series rather than inferring internal cognitive processes.

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 2026-06-26; verification: verified.

Topics

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

• visual search