Showing papers by "Nick Tyler published in 2022"
TL;DR: While human spatial representation appears differentially influenced by environmental boundaries, the influence is similar across virtual and physical environments.
Abstract: Boundaries define regions of space and are integral to episodic memories. The impact of boundaries on spatial memory and neural representations of space has been extensively studied in freely-moving rodents. But less is known in humans and many prior studies have employed desktop virtual reality (VR) which lacks the body-based self-motion cues of the physical world, diminishing the potentially strong input from path integration to spatial memory. We replicated a desktop-VR study testing the impact of boundaries on spatial memory (Hartley et al., 2004) in a physical room (2.4m x 2.4m, 2m tall) by having participants (N = 27) learn the location of a circular stool and then after a short delay replace it where they thought they had found it. During the delay, the wall boundaries were either expanded or contracted. We compared performance to groups of participants undergoing the same procedure in a laser-scanned replica in both desktop VR (N = 44) and freely-walking head mounted display (HMD) VR (N = 39) environments. Performance was measured as goodness of fit between the spatial distributions of group responses and seven modelled distributions that prioritised different metrics based on boundary geometry or walking paths to estimate the stool location. The best fitting model was a weighted linear combination of all the geometric spatial models, but an individual model derived from place cell firing in Hartley et al. 2004 also fit well. High levels of disorientation in all three environments prevented detailed analysis on the contribution of path integration. We found identical model fits across the three environments, though desktop VR and HMD-VR appeared more consistent in spatial distributions of group responses than the physical environment and displayed known variations in virtual depth perception. Thus, while human spatial representation appears differentially influenced by environmental boundaries, the influence is similar across virtual and physical environments. Despite differences in body-based cue availability, desktop and HMD-VR allow a good and interchangeable approximation for examining human spatial memory in small-scale physical environments.
4 citations
TL;DR: This paper proposes low-level features constructed from the IMU data gathered from 35 participants, while they performed a stair climbing and descending task in a real-world simulated environment, and demonstrates that with these features the machine learning models predict dementia with 87.02% accuracy.
Abstract: Posterior Cortical Atrophy is a rare but significant form of dementia which affects people's visual ability before their memory. This is often misdiagnosed as an eyesight rather than brain sight problem. This paper aims to address the frequent, initial misdiagnosis of this disease as a vision problem through the use of an intelligent, cost-effective, wearable system, alongside diagnosis of the more typical Alzheimer's Disease. We propose low-level features constructed from the IMU data gathered from 35 participants, while they performed a stair climbing and descending task in a real-world simulated environment. We demonstrate that with these features the machine learning models predict dementia with 87.02% accuracy. Furthermore, we investigate how system parameters, such as number of sensors, affect the prediction accuracy. This lays the groundwork for a simple clinical test to enable detection of dementia which can be carried out in the wild.