Recognition memory in individuals with mild cognitive impairment and Alzheimer's dementia : a systematic review

Abstract

Background: Episodic memory impairment is a hallmark of Alzheimer's disease (AD) and is already present in its prodromal stage. While recall deficits are well established as predictors of memory performance, the role of recognition memory-particularly the dissociation between familiarity and recollection-remains less explored. Objective: This systematic review aims to synthesize current evidence on recognition memory performance in healthy controls (HC), individuals with mild cognitive impairment (MCI), and those with AD, with a focus on the distinct contributions of familiarity and recollection processes. Methods: We searched PubMed (MEDLINE), Science Direct, PubPsych, and TRIP Medical databases for studies published until November 2025, which investigated recognition memory performance in individuals with MCI and AD including validated memory tools (such as scales, indices, scores, tests, and assessments). Forty-six studies (n = 3996) met the criteria and were included. Methodological quality was evaluated using the Newcastle-Ottawa scale. This study is registered with PROSPERO, CRD42022343750. Results: Most studies were cross-sectional and conducted in memory clinics or research centers, with considerable heterogeneity in sample size and assessment tools. The majority utilized experimental paradigms to differentiate familiarity and recollection, though traditional memory tests remain prevalent. Across studies, recollection was consistently impaired in MCI and AD, while familiarity showed a more variable pattern-often preserved in early MCI but impaired in advanced stages and AD. Structural and functional neuroimaging studies revealed that hippocampal atrophy is closely linked to recollection deficits, while alterations in entorhinal and parahippocampal cortices are associated with familiarity impairment. Combined deficits in recall and recognition, especially when recognition impairment reflects encoding failure, robustly predict conversion to dementia. Conclusions: Recognition memory assessment, particularly the dissociation between familiarity and recollection, provides valuable information for early detection and prognosis in the AD continuum. Incorporating nuanced recognition memory measures into clinical practice may improve diagnostic specificity and facilitate timely interventions. Further longitudinal research is needed to validate recognition memory as a predictor of dementia progression and to standardize its assessment in diverse populations.