Cognitive Performance in Geriatric Depression: Examining the Sensitivity of Clinical Neuropsychological and Experimental Working Memory Tasks

Abstract

Background: Traditional neuropsychological measurement tools have identified a proportion of individuals with late life Major Depressive Disorder (MDD) who experience modest to moderate cognitive impairments. These cognitive impairments are associated with impairments in everyday functioning. However, a large proportion of individuals with MDD report difficulties with cognition that are greater than those identified with traditional measurement tools. In this study, we explore cognitive performance using new methods that are different from those used in traditional cognitive testing. We aim to test if rather than using traditional measures of cognitive abilities, cognitive performance impairments in MDD reveal themselves as a function of increasing complexity of cognitive tasks, and to examine the relationship of cognitive performance on these tasks to everyday functioning. Method: Older participants with remitted MDD (MDD-R; n = 93), Mild Cognitive Impairment (MCI; n = 200), MCI with comorbid MDD-R (n = 80), and healthy comparison participants (n = 53) completed a baseline neuropsychological assessment battery for preventative intervention for Alzheimer’s Dementia for individuals at clinical risk were used in analyses. We examined group differences in cognitive performance accuracy across three experimental working memory tasks that allowed for titration of complexity (i.e., the Paced Auditory Serial Processing Task (PASAT), n-back, Continuous Performance Task – Identical Pairs (CPT-IP)). Changes in performance over the course of the cognitive tasks were entered into a regression equation to examine the variance accounted for above and beyond global cognition in predicting everyday functioning in all clinical groups. Results: Relative to healthy comparison participants, individuals with MDD-R had significantly worse performance on all levels of the PASAT and n-back tasks. The MDD-R group performed significantly better than the MCI groups on all levels of the PASAT and n-back. No significant group differences were found on the CPT-IP. Changes in PASAT performance were significantly associated with everyday functioning but did not add meaningful variance in understanding the relationship between cognition and everyday functioning. Conclusions: Our findings highlight the utility of using novel methods to probe cognition, such as the experimental cognitive tasks with the ability to titrate complexity used in this study, in late life MDD.