EVE: A framework for running virtual reality experiments in neuroscience

Experiments in neuroscience often rely on virtual reality (VR) to investigate navigation inside the MRI scanner. While these studies are limited by the lack of locomotion and proprioceptive feedback, researchers have successfully immersed participants using passive (e.g., videos, images) or active (e.g., movement mediated by a control interface) virtual environments in order to study different aspects of their spatial memory. However, conducting VR experiments inside the MRI scanner can be cumbersome and time consuming due to issues with the hardware, software and the extraction/synchronization of data. In
this talk, we will present some of the functionalities of the EVE (Experiments in Virtual Environments) framework that can facilitate the setup, implementation, and evaluation of VR experiments in neuroscience. EVE is designed to reduce repetitive and error-prone steps that often occur during the design of VR experiments. The framework is based on the popular platforms of Unity 3D and MiddleVR. This allows researchers, who do not have specialized training in computer science, to design virtual
environments and connect with physical (e.g., eye tracker, electrodermal activity) and virtual sensors (e.g. location marker, collectible items) for data collection and synchronization. EVE further assists researchers by providing data management and evaluation capabilities including database support, questionnaire
tools, visualization tools, and scripting for R. The framework includes a series of replay tools that researchers can use to run quick diagnostic analyses (e.g., video playback, path tracing) at the scanner or more in depth analyses at the laboratory using the custom made R package evertools. We will present some of these functionalities in the context of several experiments conducted at the Chair of Cognitive Science at ETH Zurich.