Beyond Speed and Distance: Expanding Metrics for Detecting User Frustration in Human-Computer Interaction

Abstract

The detection of emotional states, particularly frustration, during human-computer interaction (HCI) is critical for optimizing user experience and system performance. While prior research has primarily focused on mouse cursor speed and distance as indicators of negative emotion, this study seeks to broaden the scope by introducing and evaluating novel behavioral metrics. We explore additional indicators such as hesitation patterns, interaction pauses, click dynamics, and trajectory irregularities to capture more nuanced emotional responses. By integrating these metrics with traditional tracking data, we aim to enhance the accuracy of frustration detection in real-time system interactions. Through a series of experiments conducted in both controlled environments and live user interfaces, our results demonstrate a significant improvement in detecting emotional variability and user frustration. This approach paves the way for the design of adaptive, emotion-aware systems that can dynamically adjust to users' emotional states, ultimately leading to more intuitive and satisfying user experiences. Our findings provide a foundation for the development of next-generation HCI designs that prioritize user emotion as a key component of interaction design.