Estimation of mental effort in learning visual search by measuring pupil response.

Takeuchi, Tatsuto; Puntous, Théodore; Tuladhar, Anup; Yoshimoto, Sanae; Shirama, Aya · 2011 · DOAJ

DOI: 10.1371/journal.pone.0021973

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 relationship between perceptual learning and mental effort, specifically testing the hypothesis that task automatization leads to a reduction in cognitive load as performance improves. While behavioral metrics like reaction time typically decrease with training, it remains unclear whether this efficiency corresponds to a linear decrease in mental effort or if effort increases to maintain high performance. The authors utilized pupil diameter, a physiological marker governed by the autonomic nervous system and correlated with mental workload, to estimate effort during visual search training. The experimental design involved twelve subjects performing a conjunction visual search task over 16 consecutive days. Participants searched for a target Gabor patch, defined by a specific combination of orientation and spatial frequency, among 15 distractors. Reaction times were recorded via touch-pad responses, and pupil diameter was monitored using an infrared eye-tracking system. To isolate mental effort from the pupil light reflex, the researchers included a passive viewing condition where subjects observed the displays without performing the search task. Data analysis focused primarily on target-absent trials to ensure error-free conditions and consistent serial search strategies. Behavioral results showed a monotonic improvement in performance, with reaction times decreasing significantly from approximately 4 seconds on the first day to under 2 seconds by the final day. In contrast, pupil size followed an inverted U-shaped trajectory. Pupil diameter increased rapidly during the early phase of training, reaching a maximum on the fifth day, before decreasing to half of its peak value in the later stages. Crucially, the pupil size never returned to pre-training baseline levels. The passive viewing condition confirmed that these changes were task-dependent, as pupil size remained stable across days when no search effort was required. Statistical analyses confirmed significant main effects of training day on both reaction time and pupil size, with no interaction between task type and training day for behavioral performance. The findings challenge the simple automatization hypothesis, which posits that learning reduces mental effort. Instead, the divergence between behavioral performance and pupil response suggests that mental effort and behavioral efficiency reflect distinct aspects of perceptual learning. The initial increase in pupil size may indicate heightened cognitive engagement or subjective fatigue as subjects adapt to the task, while the subsequent decrease suggests a stabilization of effort rather than its elimination. The authors conclude that mental effort continues to be invested to maintain improved performance, implying that perceptual learning involves complex neural mechanisms where behavioral gains do not necessarily equate to reduced cognitive load. This work highlights pupil response as a valuable tool for uncovering the subjective and physiological dimensions of learning that are not captured by behavioral metrics alone.

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.

• automaticity skill acquisition