3D Interaction Study Interfaces for Additive Model Validation

As part of research into KLM-style additive models for predicting task completion time in 3D user interfaces, I developed two study applications for the Magic Leap 2 AR head-worn display: a menu selection interface and a tabletop manipulation interface.

The menu selection study had participants select buttons on a virtual two-page menu using six input modalities: controller ray casting, gaze with controller confirmation, gaze with airtap, gaze with dwell, blink confirmation, and direct hand touch. The manipulation study had participants translate and scale a virtual cylinder on a tabletop surface using the same modalities.

Both applications logged per-trial movement times and confirmation operator times, enabling validation of existing Fitts’ law-style predictive models against actual AR interaction performance across a wide range of input modalities.

Stack: Unity, C#, Magic Leap 2 SDK, Magic Leap 2

Publication: Evaluating the Viability of Additive Models to Predict Task Completion Time for 3D Interactions in Augmented Reality — IEEE VR 2026