High-Fidelity Avatars for Behaviour and Cognition Research
Lead Research Organisation:
University of Kent
Department Name: Sch of Psychology
Abstract
Virtual Reality (VR) is becoming more and more sophisticated. It is now possible to build worlds inhabited by avatars representing real people, whose likenesses have been 'scanned in' to the system. Of course, these developments have been driven by the entertainment and gaming industry, but the development of this technology also opens up many new opportunities for social sciences researchers. In short, it is possible to carry out research and conduct experiments in 'virtual space' situations that would be very difficult to set up in real life.
Our research aims to develop an easy-to-use toolkit for researchers to build virtual worlds for the purpose of research. We will develop procedures for constructing realistic avatars, based on real people, which can interact with experimental participants in natural ways, including moment-by-moment changes in facial expression and eye-direction. These are cues that are very important to everyday interactions, but they are difficult to control in typical lab-based experiments - for example to ensure that all participants are exposed to the same experience. This becomes possible in virtual reality, where experimenters can manipulate the behaviour of the avatars relatively straightforwardly.
Our previous ESRC-funded work in a Virtual Reality airport has shown that certain tasks, for example checking passports, can be simulated in ways that are very close to real-life settings. Furthermore, experimental results are very similar in both settings. Here we will develop further the range of possibilities for interacting with avatars (i.e. beyond simple identity checking). We will also develop general tools for researchers to create rich virtual environments containing people who can interact dynamically and flexibly with users. We will create manuals and walk-through videos to support researchers building their experiments, without the need for complex programming skills. In this way, we aim to make available the recent advances in VR technology to social scientists, opening up exciting possibilities for new research.
Our research aims to develop an easy-to-use toolkit for researchers to build virtual worlds for the purpose of research. We will develop procedures for constructing realistic avatars, based on real people, which can interact with experimental participants in natural ways, including moment-by-moment changes in facial expression and eye-direction. These are cues that are very important to everyday interactions, but they are difficult to control in typical lab-based experiments - for example to ensure that all participants are exposed to the same experience. This becomes possible in virtual reality, where experimenters can manipulate the behaviour of the avatars relatively straightforwardly.
Our previous ESRC-funded work in a Virtual Reality airport has shown that certain tasks, for example checking passports, can be simulated in ways that are very close to real-life settings. Furthermore, experimental results are very similar in both settings. Here we will develop further the range of possibilities for interacting with avatars (i.e. beyond simple identity checking). We will also develop general tools for researchers to create rich virtual environments containing people who can interact dynamically and flexibly with users. We will create manuals and walk-through videos to support researchers building their experiments, without the need for complex programming skills. In this way, we aim to make available the recent advances in VR technology to social scientists, opening up exciting possibilities for new research.
Organisations
People |
ORCID iD |
| Markus Bindemann (Principal Investigator) | |
| A Burton (Co-Investigator) |