Riding the Carousel: The First Extensive Eye Tracking Analysis of Browsing Behavior in Carousel Recommenders

2025-07-14

AI Summary

This study investigates users’ multi-list browsing behavior in carousel recommendation interfaces, focusing on how browsing initiation points, cross-list navigation patterns, and topic preferences influence user interaction. Leveraging eye-tracking and free-browsing experiments, we construct the first large-scale eye-movement dataset specifically for carousel scenarios, empirically identifying a “post-swipe position bias”—a systematic tendency for users to fixate disproportionately on items near their post-swipe location. Grounded in observed gaze distributions following swipes, we propose a topic-aware re-ranking strategy tailored to multi-theme carousels. Our contributions are threefold: (1) addressing a critical gap in behavioral modeling of carousel interfaces; (2) providing the first fine-grained eye-tracking evidence characterizing the dynamic evolution of visual attention during carousel browsing; and (3) advancing behaviorally grounded interface design and recommendation optimization, yielding statistically significant improvements in click-through rate and user experience.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Humans and AI: Intelligent User Interfaces
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Topic discovery and tracking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Carousels have become the de-facto interface in online services. However, there is a lack of research in carousels, particularly examining how recommender systems may be designed differently than the traditional single-list interfaces. One of the key elements for understanding how to design a system for a particular interface is understanding how users browse. For carousels, users may browse in a number of different ways due to the added complexity of multiple topic defined-lists and swiping to see more items. Eye tracking is the key to understanding user behavior by providing valuable, direct information on how users see and navigate. In this work, we provide the first extensive analysis of the eye tracking behavior in carousel recommenders under the free-browsing setting. To understand how users browse, we examine the following research questions : 1) where do users start browsing, 2) how do users transition from item to item within the same carousel and across carousels, and 3) how does genre preference impact transitions? This work addresses a gap in the field and provides the first extensive empirical results of eye tracked browsing behavior in carousels for improving recommenders. Taking into account the insights learned from the above questions, our final contribution is to provide suggestions to help carousel recommender system designers optimize their systems for user browsing behavior. The most important suggestion being to reorder the ranked item positions to account for browsing after swiping.These contributions aim not only to help improve current systems, but also to encourage and allow the design of new user models, systems, and metrics that are better suited to the complexity of carousel interfaces.

Problem

• Analyzes eye tracking behavior in carousel recommenders
• Examines user browsing patterns across multiple carousels
• Investigates genre preference impact on browsing transitions

Innovation

• Eye tracking analyzes carousel browsing behavior
• First extensive empirical study on carousel recommenders
• Reorder ranked items based on swiping behavior