Augmented Reality Fitts' Law Input Comparison Between Touchpad, Pointing Gesture, and Raycast

Mifsud, Domenick; Williams, Adam; Ortega, Francisco R.; Teather, Robert J. · 2022 · 2022 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)

DOI: 10.1109/vrw55335.2022.00146

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the impact of target transparency on selection performance in augmented reality (AR) and compares three common input modalities: finger-based Pointing Gesture, controller Touchpad, and controller Raycast. Motivated by the increasing adoption of optical see-through AR head-mounted displays (HMDs) like the Magic Leap One, the authors sought to determine how varying the opacity of virtual targets affects user interaction, a factor previously underexplored using Fitts’ Law metrics. The research aims to provide systematic data on how transparency influences throughput, error rates, and movement time across different pointing techniques. The experiment employed a within-subjects design with 18 participants using a Magic Leap One HMD. Participants performed a 2D reciprocal Fitts’ Law pointing task involving 11 targets arranged in a ring. The study manipulated two independent variables: input modality (Touchpad, Pointing Gesture, Raycast) and target transparency level (0%, 30%, 60%, and 90%). To control for field-of-view limitations, target distance and width were held constant. Data collected included throughput (bits per second), movement time, error rate, target re-entry count, and incorrect click count. Statistical analysis utilized repeated-measure ANOVAs and Kruskal-Wallis tests, with post-hoc comparisons to identify significant differences between conditions. The results demonstrated that target transparency had no significant effect on throughput, movement time, error rate, or incorrect click counts. However, input modality significantly influenced performance. The Raycast method outperformed both Pointing Gesture and Touchpad inputs across all transparency levels. Specifically, Raycast yielded the highest throughput (M = 2.63 bits/s) and lowest movement time (M = 1068.77 ms), while also exhibiting significantly lower error rates (M = 0.01) compared to Touchpad (M = 0.05) and Pointing Gesture (M = 0.04). Pointing Gesture and Touchpad did not differ significantly in throughput or error rates, though Pointing Gesture resulted in significantly more target re-entries. Post-experiment questionnaires revealed that participants preferred Raycast for overall use, accuracy, and speed, while Touchpad was preferred for crowded environments. Participants reported high fatigue and tracking issues with the Pointing Gesture method. The study concludes that while transparency adjustments do not hinder selection performance in AR, the choice of input modality is critical. Raycast emerges as the superior technique for accuracy and efficiency in this context. These findings suggest that developers should prioritize ray-based interaction for precision tasks in AR interfaces. Future work is recommended to examine the impact of lost finger tracking on task re-acquisition and to test these results across different AR display types to assess generalizability.

Key finding

Raycast input significantly outperformed pointing gesture and touchpad inputs in throughput and error rates, while target transparency had no significant effect on selection performance.

Methodology

lab_experiment

Sample size: 18

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via author_sweep_intake on 2026-05-28.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• control interfaces