# Similar minds age alike: an MRI similarity approach for predicting age-related cognitive decline

## TL;DR

This study shows that brain gray matter similarity networks can better predict age-related cognitive decline than traditional brain connectivity methods.

## Contribution

The study introduces gray matter similarity networks as a novel method for predicting cognitive aging.

## Key findings

- Gray matter similarity networks predict age and cognition better than anatomical and functional connectivity.
- Age-related changes in gray matter similarity are strongest in cortical layers II and III.
- These networks are more robust to individual differences in cognition, behavior, and sex.

## Abstract

As individuals age, cortical alterations in brain structure contribute to cognitive decline. However, the specific patterns of age-related changes and their impact on cognition remain poorly understood. This study assessed the effects of aging on individual gray matter similarity networks and compared them to anatomical and functional connectivity networks derived from diffusion-weighted imaging and resting-state fMRI, respectively. Our results showed that gray matter similarity networks outperformed anatomical and functional connectivity in predicting age and cognition, showing the earliest age-related changes across the adult lifespan. These networks also demonstrated greater robustness to individual differences in cognition, behavior, and sex. Notably, age-related changes in gray matter similarity were associated with the brain’s underlying cytoarchitecture, being strongest in brain regions from cortical layers II and III. These findings provide a new biological insight into the neural mechanisms of cognitive aging and highlight the potential of individual morphological similarity for capturing complex brain changes across the lifespan.