Beyond Instructed Tasks: Recognizing In-the-Wild Reading Behaviors in the Classroom Using Eye Tracking

2025-01-30

AI Summary

This study addresses the limitation of conventional reading instruction assessment—its reliance on controlled, instruction-based tasks that fail to capture natural reading behaviors in authentic classroom settings. We propose an eye-tracking–driven reading behavior recognition method designed for *in-the-wild* contexts. For the first time, we formally define prototypical reading behaviors—including skimming, close reading, and scanning—integrating cognitive theory, statistical analysis, and a lightweight 2D CNN into a generalizable hybrid analytical framework. Feature extraction emphasizes oculomotor velocity, fixation density, and sequential patterns. Evaluated on real-world classroom eye-tracking data, our model achieves an F1-score of 0.80, enabling fine-grained and interpretable behavior classification. Our core contribution is the first theoretically grounded, data-cooperative (combining instructed and *in-the-wild* data), lightweight, and deployable framework for naturalistic reading behavior analysis—breaking free from laboratory constraints and providing empirical support for precision pedagogical intervention and formative assessment.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Computer Vision: Motion & Tracking
• Intelligent Robots: Behavior Learning & Control

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Understanding reader behaviors such as skimming, deep reading, and scanning is essential for improving educational instruction. While prior eye-tracking studies have trained models to recognize reading behaviors, they often rely on instructed reading tasks, which can alter natural behaviors and limit the applicability of these findings to in-the-wild settings. Additionally, there is a lack of clear definitions for reading behavior archetypes in the literature. We conducted a classroom study to address these issues by collecting instructed and in-the-wild reading data. We developed a mixed-method framework, including a human-driven theoretical model, statistical analyses, and an AI classifier, to differentiate reading behaviors based on their velocity, density, and sequentiality. Our lightweight 2D CNN achieved an F1 score of 0.8 for behavior recognition, providing a robust approach for understanding in-the-wild reading. This work advances our ability to provide detailed behavioral insights to educators, supporting more targeted and effective assessment and instruction.

Problem

• Reading Habits
• Educational Improvement
• Assessment Methods

Innovation

• AI-based reading habit recognition
• Natural reading state analysis
• Educational strategy optimization