An immersive virtual reality task with physical movement for the assessment of spatial memory

Abstract

Virtual Reality (VR) has not yet been exploited in the assessment of spatial memory. Current VR systems for the assessment of spatial memory include interaction by using traditional devices (e.g., computer screens, mouses or keyboards). This classical type of interaction with the system is ineffective because the user's sense of presence is very poor. If the user does not have the feeling of being in the virtual environment, the spatial memory may not be assessed correctly. The user's physical movement can contribute to a high level of presence. Therefore, by developing a system that allows the participants to become active in a virtual environment, spatial memory can be assessed as the same way it could be evaluated in a real environment. The main objective of this thesis was to design, develop and validate an immersive VR system in which the user could interact by physical movements. The virtual environment was designed based on a city square. The cognitive task, based on this environment, comprised six levels. The goal of the levels for the main task was to assess children short-term memory for object location in the virtual environment. The VR system incorporated two types of interaction. 1) A physical active condition (physically walking on a Wii Balance Board and changing the direction by turning a wireless steering wheel). 2) An inactive condition (stand up and use a gamepad). For the visualization of the task, a large stereo screen was used. For the validation, two studies were carried out. The performance of our task was compared with traditional methods (the Corsi Block Tapping Test). We carried out two studies involving 212 children. Correlations were found between our task and traditional methods, indicating that our task has proven to be a valid tool for assessing spatial short-term memory in children. With regard to the interaction type, the results showed that there were no statistically significant differences regarding the score obtained in our task based on the interaction used (inactive condition vs. physical active condition). Although unexpected, this is a good result because it means that the task is well suited for the assessment of spatial memory and that the two interaction types can be used for this purpose. With regard to gender differences in the task score and the Corsi Block Tapping Test, the results indicated that there were no statistically significant differences for gender. With regard to usability and satisfaction, our studies have shown that the use of the inactive condition did not differ significantly from the physical active condition for the usability and satisfaction questions. Previously to the development of the mentioned VR system, we developed a VR system, with Natural User Interfaces (NUI) and an autostereoscopic screen, for dental learning. The system included two modes: neutral and real world background. This system was validated with 33 dentistry students. With this first development, the required knowledge for facing the second development, core of the thesis, was acquired. The following general conclusions were extracted from the two developments and the three studies: Learning - Autostereoscopic VR systems, with different background modes and NUI, have proven to be effective tools for learning teeth morphology - With this type of systems, children can learn and at the same time, they can have a good time - Stereoscopy and NUI are appropriated for developing educational games and they can be exploited in their development Spatial memory - VR systems, with stereoscopy and two different user interfaces (inactive and physical active conditions), have proven to be reliable and effective tools to assess spatial memory in children - With this type of systems, the children can be assessed meanwhile are having a good time - Our task and similar tasks could be used for assessment and training of spatial memory in children and adults