Factor analysis of the virtual reality Stroop task.

Introduction: Virtual reality (VR) offers neuropsychologists high dimensional (3D) platforms for administering cognitive tasks that balance experimental control with simulations of naturalistic activities. A virtual reality version of the Stroop task, the Virtual Reality Stroop Task (VRST), was developed that leverages technological advances to enhance the ecological validity of neuropsychological assessments. The high mobility multipurpose wheeled vehicle (HMMWV) version of the VRST includes high arousal (ambush) and low arousal (safe) zones was employed in this study. This version of the VRST contains both cognitive (Stroop) and affective (arousal) components. While the VRST has been shown to have good construct validity, the factor structure has yet to be explored. This study aimed to examine the factor structure of the VRST and compare the results with a lower dimensional (2D) computer-automated Stroop task (i.e., the Automated Neuropsychological Assessment Metrics - ANAM).

Method: Data was drawn from college-aged students who completed the VRST and ANAM Stroop tasks (N = 115). Factor analyses utilized principal axis factoring (PAF), and output variables included percent of correct responses and response times the VRST and ANAM Stroop tasks.

Results: Results indicated that both Stroop tasks had two-factor solutions. Factor one measured response times and factor two measured accuracy. While factors respective of speed and accuracy factors were correlated across assessment modalities, within assessment factor correlations were low.

Conclusions: The factors possibly indicated participants response styles in that they either focus on responding accurately or responding quickly to stimuli. Therefore, including throughput in future research examining either the ANAM Stroop task or VRST may provide useful insight into participant performance. Finally, because similar factor structures were observed for both the VRST and ANAM Stroop task this study provided additional support for the construct validity of a higher dimensional Stroop task, the VRST.