Social Interaction
Solutions

Treatment through Virtual Self

Client: Stanford University, Virtual Human Interaction Lab VHIL

Application: Scientific research program examining the effectiveness of self-models in virtual worlds.

Equipment Used: WorldViz Vizard 3D software toolkit, WorldViz PPT X4 optical inertial hybrid wide-area tracking system, NVIS nVisor SX head-mounted display

Summary

Using WorldViz technology, VHIL developed virtual worlds in which participants could interact with avatars of their own selves. The researchers studied effects of self-modeling in contexts such as perceived aging and advertising, as well as exercising and weight management. In the future, their efforts could lead to treatments where the virtual self influences the real self, convincing us to alter our decisions and behavior.

Vision

Stanford University's Virtual Human Interaction Lab VHIL sought to develop several experiments which would examine participants' reactions to seeing themselves in a virtual world. In these immersive virtual environments, participants would view and interact with lifelike 3D digital humans resembling their own selves performing desirable or undesirable behaviors.

Solution

Building compelling, lifelike avatars was a top priority for the researchers at VHIL. The Vizard 3D software toolkit's built-in human face and body models and its accompanying morph designer empowered the researchers to instantly insert virtual humans into their virtual environments. With the high-level morph controller Stanford University's researchers were able to blend complex facial expressions, creating realistic social interaction between avatars and participants.

As a second component, WorldViz provided VHIL with Vizard Complete Characters, a collection of 100 carefully crafted bodies and faces optimized for real-time performance in virtual environments. Thirdly, to instantly create compelling virtual self-models of large groups of participants, the BioVirtual 3DMeNow software was chosen. It allows to build high-quality faces directly from two photographs per participant within very few steps, and to integrate dynamic facial expressions for ultimate realism.

Finally, participants were invited to meet their virtual counterparts by donning an NVIS nVisor SX stereo head-mounted display with a crisp high-resolution image. Users could roam around freely in the virtual environment and naturally interact with their virtual selves, tracked by the WorldViz PPT X4 wide-area tracking optical system.

Results

With the help of WorldViz technology, Stanford University's researchers have been able to build several virtual treatments that are unprecedented in the field. This work is funded by the National Science Foundation and in the past year, their research has been featured in Science Magazine, The New York Times, Business Week, Forbes, Time Magazine, and many more.

One study investigates the modeling of healthy and unhealthy behaviors. Participants are able to view themselves exercising on a treadmill or being inactive. The finding was that seeing the self-model exercise in the virtual environment causes significantly more exercise in the real world within the 24 hours following the experiment.

In another study, participants see their virtual self-model gain weight as the participant physically remains inactive, and the self-model lose weight as the participant physically exercises. The exposed users resulted exercising significantly more frequently than other participants whose self-model did not appear to lose or gain weight.

VHL is finding that self-models in virtual environments can be effective treatments that may be useful in encouraging fitness and curbing the obesity epidemic.