Abstract
Objective: Traditional neuropsychological testing with paper-and-pencil tests has been a mainstay of cognitive assessment. However, paper-and-pencil testing is prone to inter-examiner variability, insensitive to subtle cognitive changes, time-consuming, and often unavailable to rural and under-resourced communities. Computerized and web-based neuropsychological assessments such as the California Cognitive Assessment Battery (CCAB) have been developed to address these issues. Method: The CCAB is a fully automated battery with 30+ cognitive tests that parallel traditional paper-and-pencil measures. Participants were tested on the CCAB at home, with remote proctoring via web-based interface. In Experiment 1, we measured concurrent validity by comparing performance on CCAB measures to performance on corresponding paper-and-pencil measures. Test-retest reliability was evaluated with repeat CCAB administration. In Experiment 2, the CCAB was administered at baseline, 6, 18, and 30 months to evaluate cognitive performance longitudinally in healthy older adults. Results: The CCAB showed acceptable to good concurrent validity (rs = 0.44-0.79) and reliability (rs = 0.57-0.84). Factor analysis of data from Experiment 2 identified three primary factors: Memory, Semantics, and Visuospatial-Executive (variance explained: 6.9-11.7%). In the longitudinal analysis, age, gender, education, and ethnicity were significant predictors of cognitive performance. Conclusions: The CCAB has acceptable psychometric properties and can be used to measure cognitive performance longitudinally in an older cohort. Given the aging of the population and the need for early diagnosis of diseases such as MCI and dementia, computerized tools like the CCAB can play an important role in providing access to early cognitive assessments that help track and improve health outcomes.