This paper validates the D-Cog, an iPad-based digital spatial working memory 'serious game' that is feasible in clinic, correlates with MMSE (r=0.65), and discriminates AD and DLB/PD from controls (AUCs up to 0.96).

Although it offers little mechanistic or therapeutic insight, a validated, repeatable digital cognitive test like D-Cog could improve remote cognitive monitoring and outcome measurement sensitivity in Parkinson's disease clinical trials and longitudinal care.

BackgroundCognitive testing is critical for screening, diagnosing, and monitoring neurocognitive disorders, but current tests are limited by long administration times and the need for trained personnel. Digital cognitive testing offers a convenient and highly repeatable alternative to track cognitive decline, which is essential for evaluating patient care and in clinical trials. To address these limitations, we developed a novel digital spatial working memory test called the D-Cog.ObjectiveThis study aimed to assess the feasibility, validity, and diagnostic utility of the D-Cog in a clinical population.MethodsPatients with Alzheimer's disease (AD), Lewy body dementia (DLB), Parkinson's disease (PD), and cognitively normal controls (CN) were included. During clinical visits, a medical assistant administered the D-Cog test, presented as a "serious game" on an iPad.ResultsThe D-Cog showed high participant engagement, although engagement decreased with advancing cognitive impairment. A significant difference in the mean D-Cog score was observed between CN and both AD and DLB/PD groups (p < 0.003). Additionally, a significant correlation between the D-Cog score and the Mini-Mental State Examination was identified (r = 0.65, p < 0.001, 95% CI: 0.52-0.75). The D-Cog demonstrated good discrimination between groups, with AUCs of 0.96 (95% CI: 0.91-1.00) for AD versus healthy controls and 0.88 (95% CI: 0.75-0.97) for MCI versus healthy controls.ConclusionsThe D-Cog is a feasible and valid digital cognitive test with potential clinical utility across neurocognitive disorders. Although engagement was lower in advanced dementia, these findings support further development and exploration of the D-Cog, particularly for potential in-home assessments.