Predictors of improved attentional control after Virtual Reality-based cognitive rehabilitation in chronic traumatic brain injury.

Abstract

The purpose of this paper was to explore predictors of change in attentional control observed in a randomized controlled trial (RCT) investigating VR-based cognitive training, using the game BeatSaber. Data were collected at baseline of the RCT, including 100 participants in the chronic phase of TBI, defined as at least one year post-injury. The 51 participants randomized to the VR group were included in this analysis. Primary outcome measure was based on the main significant finding from the RCT, a ratio score between speed and accuracy, Inverse Efficiency Score (IES). To explore potential predictors of post-treatment change in IES, a multivariable regression analysis was performed. Based on previous research, age, years since injury, executive functioning, level of abstract thinking, and immersive tendencies were selected as predictors. The model explained 40% of the variance of the post treatment change in IES. More years since injury, lower baseline executive functioning and higher immersive tendencies towards games explained greater change in VR-training. However, after bootstrapping, only years since injury remained a significant predictor. In light of the bootstrapped analyses, the findings should be interpreted as exploratory and considered hypothesis-generating, warranting further investigation in larger samples.