Comparing web-based neuropsychological tests of verbal and spatial memory with standardized pen and paper versions

Abstract

In this study we studied if novel Internet versions of a verbal and a visuo-spatial test were comparable with the standard (i.e. pen and paper) versions of the test. On each of these two types of tests we wanted to see if the subjects had the same scores on both versions of the test, and if not, if there were any learning effects from session to session. In addition to this, we wanted to see if there was a correlation between different tests. Both intraclass correlation and Pearson’s correlation were used to investigate if the two versions of the tests were comparable. T-tests were performed to see if there was a difference between the scores on each test, and between the groups on each version. Subjects got significant higher scores in the iVLT compared to the CVLT-II on all trials. However this difference was constant across all learning and recall trials. We only observed a significant learning effect on the distraction list when having taken the iVLT before the CVLT-II. Subjects got significant higher scores on the pen and paper version of Objects in Grid on both immediate and late recall. No learning effects were observed. Performance on the web-based tests was might have been affected by the subject’s computer skills. This study shows that it is possible to use web-based memory tests and end obtain comparable results to traditional neuropsychological tests. Test scores are not directly comparable, but have to be adjusted to make comparisons. In addition, on tests where computer skills affect scores it is necessary to measure and adjust for this.