Abstract
The use of videoconference neuropsychology increased during the SARS-CoV-2 pandemic; however, questions remain regarding the reliability of videoconference neurocognitive testing, particularly for visual and timed tasks. The purpose of this pilot study was to compare videoconference and in-person testing across a variety of visual tasks. The study utilized a counterbalanced, crossover design where veteran volunteers (N = 20) completed a test battery in both videoconference and in-person conditions. Functional videoconference perception was also measured. Nine common tests were included; alternate forms were leveraged. Wilcoxon comparisons and concordance correlations were used to compare across conditions; test visual characteristics were compared using Wilcoxon tests; and effect of functional perception was evaluated using linear regression. Test scores did not differ between conditions, with the possible exception of higher scores for Shape Learning Immediate and Delayed Recognition in the in-person condition (effect sizes r = .39 and r = .36, respectively; differences did not survive accounting for multiple comparisons). Functional videoconference perception was unrelated to scores. There were no differences in videoconference scores associated with color versus non-color and text versus non-text test attributes. This study failed to demonstrate that videoconference administration is associated with variability in complex visual neuropsychological tasks involving color and text, and suggested that videoconference administration may yield test–retest reliability comparable with that seen in in-person normative samples. It also did not find an association between functional perception and videoconference test performance. These findings require replication in larger samples and across a variety of clinical populations, and would not be expected to generalize to settings where location, technical, inter/intrapersonal, and comorbid clinical factors are inadequately managed.