The interplay of user preference and precision in different gaze-based interaction methods

TL;DR

This paper investigates how user preference and precision interact across four gaze-based interaction methods in a VR visual search game. The authors compare dwell-time selection, head-confirmation, nod gestures, and smooth pursuit, using a 52-participant study with subjective NASA-TLX and objective performance metrics. They find significant method-dependent differences in workload, search time, and scoring, with nod generally yielding the best scores and smooth pursuit showing higher mental demand and lower accuracy, while preferences differ by gender. The results underscore the value of customizable, context-aware gaze interfaces to boost accessibility, usability, and satisfaction, and point to potential trade-offs when designing gaze-driven systems for real-world scenarios.

Abstract

In this study, we investigated gaze-based interaction methods within a virtual reality game with a visual search task with 52 participants. We compared four different interaction techniques: Selection by dwell time or confirmation of selection by head orientation, nodding or smooth pursuit eye movements. We evaluated both subjective and objective performance metrics, including NASA-TLX for subjective task load as well as time to find the correct targets and points achieved for objective analysis. The results showed significant differences between the interaction methods in terms of NASA TLX dimensions, time to find the right targets, and overall performance scores, suggesting differential effectiveness of gaze-based approaches in improving intuitive system communication. Interestingly, the results revealed gender-specific differences, suggesting interesting implications for the design of gaze-based interaction paradigms that are optimized for different user needs and preferences. These findings could help to develop more customized and effective gaze interaction systems that can improve accessibility and user satisfaction.

Figures (7)

• Figure 1: Summary of the gaze-based selection methods, each method is distinguished by the purple colored boxed. All methods start by looking at a target. The gray boxes indicate the interface actions in response to the user input for each of the methods. Then, a confirmation is necessary (blue, yellow and orange), aside from the method in (a).
• Figure 2: Boxplots summarizing NASA-TLX scores across six dimensions for each of the interaction methods, showing the distribution of responses. Gaze on the y-axis is the shortened version of Gaze Dwell
• Figure 3: Distribution of the time required by the participants to find and select the target for each interaction method. The orange line indicates the mean and the red line the median.
• Figure 4: Boxplot summary of the points obtained with the different interaction methods. Significant differences between the methods are labelled with ‘***’ to indicate $p < 0.001$.
• Figure 5: Distribution of time spent on a task, broken down by gender. The vertical lines indicate the median value.