Simulation Trust Calibration using Virtual Humans

Simulation Trust Calibration using Virtual Humans


Funded by the National Science Foundation

VERG members: Jacob Stuart, Ileri Akinnola

Goals: The Goal of this project is to explore how trust in simulations and virtual humans can differ within the same experience and how this can be calibrated to an appropriate level.

Collaborators: Karen Aul, Michael Bumbach, Anita Stephens

Our role as virtual agent researchers:

  • Provide virtual agent domain knowledge for non-computer science experts. We helped the team understand best-practices for designing agent-based medical simulations based on prior literature on virtual agents and our prior experience of developing virtual agents.
  • Develop and iterate prototypes based on user feedback during iterative prototyping.
  • Develop a virtual experience delivered via Qualtrics to senior-level nursing students at UF who are practicing using SBAR tools.

Target Audience: Senior-level nursing students at the University of Florida

Our approach:

  • Kirkpatrick Evaluation Model: The Kirkpatrick evaluation model is being to identify evaluation metrics.