Digital episodic memory tools for the identification of amnestic mild cognitive impairment: A systematic review and an exploratory meta-analysis

Within our ageing population, the prevalence of Amnestic Mild Cognitive Impairment (aMCI) is rising, underscoring a need for early detection strategies to mitigate potential progression to dementia. Traditional pen-and-paper cognitive tests remain standard but often fail to capture subtle early impairments. Digital episodic memory tasks have emerged as a promising alternative, offering automated scoring, increased accessibility, and potentially better sensitivity to cognitive decline. We systematically evaluated the diagnostic accuracy of digital episodic memory assessments for aMCI. Thirteen studies met inclusion criteria, with six (n = 510 participants) providing sufficient data for meta-analysis. Pooled estimates showed sensitivity of 0.78 (95% CI: 0.65-0.87), specificity of 0.79 (95% CI: 0.66-0.89), and AUC of 0.86, with low to moderate heterogeneity (I² = 36.1%), largely reflecting differences in task design and administration. Both quantitative and qualitative findings suggest digital episodic memory tasks can provide accessible, scalable, and effective tools for early aMCI detection. However, design variability and construct differences highlight the need for further validation across diverse populations and improved methodological standardisation before clinical implementation.