Sensing the properties of virtual objects without physical feedback

This study investigates whether users can perceive the physical properties of virtual objects—specifically the rigidity of simulated C60 molecules—in the absence of physical haptic feedback. The researchers explore the effectiveness of 'Subtle Sensing,' a paradigm where users infer object properties by interacting with real-time molecular simulations in an immersive extended reality (XR) environment.

The researchers conducted two psychophysics experiments using a two-alternative forced-choice (2AFC) task. Participants were asked to identify the 'softer' of two buckyballs (C60 molecules) that varied in their bond-angle rigidity. The study compared two conditions: direct interaction, where participants manipulated the molecules using XR controllers, and observation-only, where participants watched a recording of an expert interacting with the same simulations. The researchers measured the just-noticeable difference (JND) to quantify the threshold at which participants could reliably distinguish between different levels of rigidity.

The results demonstrate that participants are capable of sensing molecular rigidity without physical feedback. In the direct interaction condition, participants achieved a higher sensitivity, detecting rigidity differences of 11.5%, compared to 18.5% in the observation-only condition. Furthermore, participants who engaged in direct interaction first performed better in the subsequent observation-only task, suggesting that active interaction serves as a form of training that enhances the ability to interpret visual and dynamic cues about virtual object properties.

These findings suggest that human perception can adapt to virtual environments by building predictive models of object behavior, even when those objects lack a physical counterpart. This 'Subtle Sensing' approach provides a flexible, low-cost alternative to traditional haptic hardware, offering significant potential for enhancing molecular research, education, and other XR applications where understanding the dynamic properties of complex, non-physical systems is essential.