Examining the Effects of Embodiment on Working Memory Performance in VR

Abstract

This study investigates the impact of immersive virtual reality (VR) environments on working memory task performance and the role of embodiment in VR. Sixteen participants engaged in N-back tasks presented either on a computer screen or in VR. Participants were assigned to one of four conditions: non-VR, VR without mimicking, VR with mimicking, or VR with mimicking and virtual mirror. Statistical analyses revealed no significant differences in sense of embodiment or task performance among the VR conditions. Additionally, there were no significant changes in working memory performance across conditions. While the findings suggest that variations in VR setups may not affect embodiment or task performance, the study’s preliminary nature emphasizes the need for larger sample sizes for more conclusive results. Further research is warranted to validate these findings and explore the effects of VR interventions on cognitive outcomes.

Introduction

Virtual reality (VR) technology has increasingly been utilized to create immersive environments for various applications, including cognitive training and rehabilitation (Lei et al., 2019; Lillo et al., 2014; Mej et al., 2021). Working memory is a cognitive function that facilitates the temporary availability of a limited amount of information while engaging in ongoing mental activities and actions (Forsberg et al., 2020; Williams & Sto, 2021). By immersing individuals in virtual environments where they embody avatars or interact with virtual objects, these experiences may engage working memory processes more effectively, leading to improved cognitive performance (Jang et al., 2010; Peck et al., 2018). The present work has practical significance as studies have shown that working memory capacity is an important predictor of academic achievement and has been linked to cognitive function, problem-solving skills, and academic performance (Peacock et al., 2022). The n-back task is a widely used measure of working memory and cognitive control because it requires the individual to maintain information in their minds while also inhibiting irrelevant information and focusing on the task at hand (Baddeley, 2003). Through sequential presentation of stimuli such as letters or numbers, individuals are required to discern whether the current stimulus corresponds to the stimulus presented n steps earlier (e.g., 2-back, 3-back), with the task’s difficulty adjustable by manipulating the value of n (Barnes et al., 2020). Advancements in technology, including VR, present a promising avenue with novel and engaging methods for enhancing working memory (Mej et al., 2021). N-back tasks in VR refer to the use of VR technology to present the traditional n-back task in an immersive environment where the stimuli are presented within a virtual environment (Aksoy et al., 2021). The sense of embodiment in VR is often enhanced through the creation of a virtual representation of the user’s body in the virtual environment, sometimes through a virtual mirror (Krogmeier & Mousas, 2020; Peck et al., 2013; Wu & Chen, 2021), and leading to heightened motivation, enhanced motor learning, and improved task performance (Johnson-glenberg et al., 2020; Petersen et al., 2022; Porssut et al., 2022). This study investigated the difference in working memory task performance between a computer screen presentation and an immersive virtual environment, and how the sense of embodiment impacts working memory performance in VR.

Methods

Sixteen participants (n = 16, age range: 19–36 years, mean age: 26.12 ± 4.73, ten female, five male, one non-binary) participated in the experiment with informed consent. Virtual scenes were presented through a Head-Mounted Display (HMD), with motion tracking facilitated by VIVE trackers. The experiment is a between-subjects design with four conditions: (a) non-VR, (b) VR without mimicking, (c) VR with mimicking, or (d) VR with mimicking and virtual mirror.

Firstly, all participants completed a pre-test consisting of the N-back tasks in a standard desktop computer setup (https://www.brainturk.com/dual-n-back), to assess participants’ baseline working memory capabilities.

Following that, participants were randomly and evenly assigned to one of the four experimental conditions to perform N-back tasks. In those N-back tasks, the sequence of stimulus was 3D human shapes with different poses. In the first condition, participants completed the tasks on a desktop computer, and in VR conditions participants wore the HMD and performed tasks in VR. Participants in the third condition were asked to mimic the human shapes with their own bodies, and in the fourth condition added a virtual mirror intending to enhance embodiment within the virtual environment.

After, all participants underwent a post-test in the same format as the pre-test. This allowed for the assessment of any changes in working memory capacities following the experimental manipulation.

Sense of embodiment and task performance were assessed for each participant. Statistical analyses including ANOVA, were conducted to compare the sense of embodiment and task performance among conditions. Significance level was set at 0.05.

Results

There was no statistically significant difference in the sense of embodiment among the three VR conditions. Furthermore, no statistically significant difference was observed in task performance across the four conditions, nor was there significant difference in the changes observed in the assessment of working memory performance among these conditions. Although this preliminary investigation involved only 16 participants (four participants for each condition), this study provides information on effect size, which can aid future power analysis and possible extension of the study.

Discussion

The findings suggested that variations in the VR setups may not significantly impact users’ sense of embodiment or working memory performance. However, the preliminary nature of the investigation underscores the importance of recruiting an adequate sample size to ensure the reliability of the results. Further research with a larger sample size is needed to validate these findings and draw more definitive conclusions regarding the effects of virtual reality interventions on perceptual and cognitive outcomes.