Related Experiment Video
Updated: Jun 2, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Neuropsychological predictors of transition from healthy cognitive aging to mild cognitive impairment: The PATH
Nicolas Cherbuin1, Perminder Sachdev, Kaarin J Anstey
1Centre for Mental Health Research, Australian National University, Canberra, Australia. nicolas.cherbuin@anu.edu.au
Objective:
To identify neuropsychological predictors of transition from healthy cognitive aging to mild cognitive impairment (MCI) or any mild cognitive disorder (any-MCD) in a community-based longitudinal study of aging.
Design:
Longitudinal
Participants:
Two thousand eighty-two individuals, aged 60–64 years and participating in a prospective epidemiologic study of mental health, and aging were assessed at two time points 4 years apart for MCI using the International Consensus Criteria, the clinical dementia rating scale (CDR, 0.5), or any of a suite of criteria sets for MCDs (any-MCD).
Measurements:
Logistic regression was used to assess the neuropsychological predictors of conversion to diagnosis including the Mini-Mental State Examination, immediate and delayed recall (IR and DR), Digit Backward, Spot-the-Word (STW), Symbol Digits Modalities Test (SDMT), simple and choice reaction time, and reaction time variability.
Results:
Of the 2,082 participants with no cognitive impairment in the first wave of data collection, 18 participants were diagnosed with MCI, 32 with CDR 0.5, and 64 participants presented with any-MCD 4 years later. The main neuropsychological predictors of conversion identified in multivariate analyses were measures of IR/DR, STW, Symbol Digit Modalities Task, and reaction time variability.
Conclusions:
Although most measures were significant predictors of conversion to MCI or any-MCD when assessed independently, four tests (IR/DR, STW, SDMT, and simple reaction time variability) accounted for the explained variance in diagnosis when all tests were assessed together. When predictive value, stability across clinical categories, and psychometric characteristics were considered together, the reaction time variability measure was the best predictor of future cognitive disorder.
More Related Videos
Related Concept Videos
Cognitive Development During Adulthood
Alzheimer Disease l: Introduction
Alzheimer Disease ll: Pathophysiology
Alzheimer's Disease: Overview
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ and tau...
Dementia l: Introduction
Information Processing Approach

