A major challenge in research on cognitive decline and dementia is the identification of at-risk populations in the preclinical phase. In this context, there is growing interest in language markers as early indicators of cognitive impairment. Objectives: This study aims to identify early linguistic markers that may facilitate the detection of individuals at risk of cognitive decline and dementia during the preclinical stage. Additionally, it seeks to evaluate the effectiveness of various assessment techniques and instruments for detecting such language impairments. Methods: A systematic review was conducted in accordance with the PRISMA guidelines, encompassing studies published between 2014 and 2025. A total of 109 articles were included in the qualitative synthesis. Results: The findings indicate that syntactic–structural features—particularly complexity, discourse coherence, and global organization—together with acoustic parameters such as pause duration, exhibit a higher accuracy and predictive value for the early diagnosis of cognitive decline and its progression to dementia. Furthermore, narrative-based tasks analyzed through automated methods demonstrate significant advantages for the assessment of language impairments. Conclusions: The analysis of language markers—particularly through the examination of syntactic complexity, acoustic features, and automated narrative assessments—represents a promising and effective approach for the early identification of cognitive impairment and the prediction of subsequent dementia onset.
Espinoza et al. (2026) studied this question.