Can a Low-Cost Eye Tracker Assess the Impact of a Valent Stimulus? A Study Replicating the Visual Backward Masking Paradigm

Vlastos, Dimitris D.; Kyritsis, Markos; Varela, Vasiliki; Gulliver, Stephen R.; Papaioannou-Spiroulia, Afroditi · 2020 · OpenAlex-citations

DOI: 10.1093/iwc/iwaa010

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 low-cost eye-tracking devices can effectively detect physiological responses to emotionally charged (valent) stimuli, specifically addressing the barrier of high equipment costs in academic and commercial research. The authors aimed to replicate the visual backward masking paradigm to determine if a lower-end tracker could distinguish between unconscious physiological reactions to fearful versus neutral faces. The motivation stems from the growing interest in using eye-tracking for affective computing and user experience studies, where detecting emotional responses is crucial, yet high-end equipment remains inaccessible to many researchers and impractical for integration into consumer devices. The experiment employed a repeated-measures design with 44 participants aged 19–48. Participants completed a target detection task while being subliminally exposed to either fearful or neutral face distractors, which were masked to prevent conscious awareness. Data were collected using a low-cost USB eye-tracker (EyeTribe) with a 60 Hz sampling rate. The primary dependent variables were saccadic duration (approximated as the time between fixations) and pupil size. The study utilized parametric and non-parametric statistical tests, including a repeated measures MANOVA, paired samples t-tests, and Wilcoxon signed-rank tests, alongside Bayesian analysis to assess the strength of evidence. The results demonstrated significant main effects of stimulus valence on both measured variables. Pupil sizes were significantly larger during trials with fearful distractors (mean = 22.7) compared to neutral distractors (mean = 18.17), with a large effect size (d = 3.91). Similarly, saccadic durations were significantly longer for fearful trials (mean = 396 ms) than for neutral trials (mean = 331.99 ms), also showing a large effect size (d = 0.96). Bayesian analysis reinforced these findings, indicating that the data were 89 times more likely to occur under the alternative hypothesis for pupil size and 24 times more likely for saccadic duration, providing strong to very strong evidence that valence impacts physiological responses. The study concludes that low-cost eye-trackers, when properly calibrated and used under controlled conditions, can successfully detect relative physiological differences associated with affective processing. These findings suggest that such affordable devices are viable tools for capturing unconscious emotional responses in research settings, particularly for students or commercial applications where high-end equipment is prohibitive. While the authors note limitations regarding motion error and the lack of comparative accuracy testing against high-end devices, the results support the use of low-cost trackers for assessing relative variance in user emotion during interactive tasks.

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.

• workload measurement