Related Experiment Video
Updated: Jul 10, 2026
Loneliness Assuaged: Eye-Tracking an Audience Watching Barrage Videos
Published on: May 29, 2020
Integrating eye-tracking into psychometric research: computational approaches to explore response burden in
Monica Casella1, Francesca Borghesi2, Elena Lupi1
1Natural and Artificial Cognition Laboratory "Orazio Miglino", Department of Humanistic Studies, University of Naples "Federico II", Naples, Italy.
Introduction:
Long questionnaires can increase cognitive and motivational demands, causing response burden. This study investigates whether eye-movement behavior varies with instrument length.
Methods:
Twenty-five participants completed the BFI-44 and the longer BFQ-2 during gaze recording. Analysis included a 2 x 3 repeated-measures ANOVA (Test x Time) and the evaluation of seven machine-learning classifiers (including Random Forest and AdaBoost) trained on the final questionnaire segment.
Results:
Fixation, visit, glance, and pupil metrics decreased over time across both instruments. Significant Test x Time interactions revealed that the longer BFQ-2 produced more pronounced temporal variability in gaze behavior. The AdaBoost classifier achieved 73% accuracy in distinguishing between the two instruments.
Discussion:
Findings suggest that longer questionnaires elicit stronger temporal fluctuations in gaze, likely reflecting fatigue or strategic adaptation. Eye-tracking provides a valuable objective tool for assessing the stability of respondent engagement over time.
