Eye Feel You: A DenseNet-driven User State Prediction Approach

2026-01-28

AI Summary

This study addresses the challenge of reliably predicting users’ subjective states—such as fatigue, cognitive load, and perceived task difficulty—from objective eye-tracking data, thereby reducing reliance on subjective self-reports. To this end, the authors propose an end-to-end deep learning regression model based on DenseNet that directly learns predictive representations from raw eye movement velocity signals, eliminating the need for handcrafted features. Through cross-session and cross-participant generalization experiments, the work systematically demonstrates that eye movement dynamics contain robust, generalizable signals correlated with subjective states. The results show that the model effectively predicts these states across both longitudinal variations and individual differences, establishing eye movement behavior as a stable and objective indicator of internal cognitive and affective conditions.

Technology Category

• Computer Vision: Motion & Tracking
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Subjective self-reports, collected with eye-tracking data, reveal perceived states like fatigue, effort, and task difficulty. However, these reports are costly to collect and challenging to interpret consistently in longitudinal studies. In this work, we focus on determining whether objective gaze dynamics can reliably predict subjective reports across repeated recording rounds in the eye-tracking dataset. We formulate subjective-report prediction as a supervised regression problem and propose a DenseNet-based deep learning regressor that learns predictive representations from gaze velocity signals. We conduct two complementary experiments to clarify our aims. First, the cross-round generalization experiment tests whether models trained on earlier rounds transfer to later rounds, evaluating the models'ability to capture longitudinal changes. Second, cross-subject generalization tests models'robustness by predicting subjective outcomes for new individuals. These experiments aim to reduce reliance on hand-crafted feature designs and clarify which states of subjective experience systematically appear in oculomotor behavior over time.

Problem

• subjective state prediction
• eye-tracking
• gaze dynamics
• longitudinal studies
• user state

Innovation

• DenseNet
• gaze dynamics
• subjective state prediction
• cross-round generalization
• cross-subject generalization