Construct validity of the Continuous Recognition Memory test

Abstract

A principal factor analysis was performed on variables derived from a neuropsychological battery administered to 100 healthy young adults in order to investigate the construct validity of the Continuous Recognition Memory test (CRM). It was hypothesized that CRM "hits" and "false alarms" would load on different factors. The factors that emerged in the analysis were labeled "Verbal Ability", "Divided Attention", "Attention to Visual Detail", "Visuomotor Integration and Planning", and "Learning and Memory". As expected, CRM hits had a significant loading on the Learning and Memory factor. However, CRM false alarms did not have a significant loading on the Divided Attention factor as expected and, instead, loaded significantly on the Attention to Visual Detail factor. A second analysis was performed using variables from the delayed condition of the memory measures. In this analysis, the CRM delayed recognition variable had significant loadings on both a "Nonverbal Memory" factor and a "Verbal Memory" factor. These analyses support the construct validity of CRM hits as a measure of learning and memory and suggest that false alarms provide a measure of attention to visual detail.