A new cognitive clock matching phenotypic and epigenetic ages.

Area of Science:

• Gerontology
• Cognitive Neuroscience
• Biomarkers

Background:

• Cognitive abilities naturally decline with advancing age, representing a key aspect of the aging process.
• Identifying reliable quantitative biomarkers for cognitive aging and their relationship with biological clocks remains a significant research challenge.

Purpose of the Study:

• To identify key age-related cognitive indices using specific tests.
• To develop a machine learning model, termed a Cognitive Clock, for predicting chronological and biological ages.
• To explore correlations between cognitive, phenotypic, and epigenetic aging.

Main Methods:

• Utilized cognitive tests including shade differentiation (campimetry), arithmetic correctness evaluation, and reversed letter detection.
• Constructed a machine learning-based Cognitive Clock using subsets of identified cognitive indices.
• Assessed the accuracy of the Cognitive Clock in predicting chronological age, epigenetic age, and phenotypic age.

Main Results:

• The Cognitive Clock achieved a mean absolute error of 8.62 years in predicting chronological age.
• The Cognitive Clock demonstrated higher accuracy in predicting epigenetic and phenotypic ages compared to chronological age.
• Demonstrated significant correlations between cognitive, phenotypic, and epigenetic age accelerations.

Conclusions:

• Cognitive tests can serve as effective biomarkers for aging.
• A machine learning Cognitive Clock can accurately predict various aging metrics.
• Cognitive performance is deeply interconnected with an individual's overall aging status.