Distributed functional connectivity predicts neuropsychological test performance among older adults

Area of Science:

• Neuroscience
• Cognitive Science
• Medical Imaging

Background:

• Neuropsychological tests are crucial for evaluating cognitive changes in late-life neurocognitive disorders.
• The underlying brain mechanisms influencing neuropsychological test performance are not fully understood.

Purpose of the Study:

• To identify optimal functional brain connectivity patterns that predict neuropsychological test scores in individuals.
• To develop and validate predictive models based on brain connectivity.

Main Methods:

• Utilized resting-state functional connectivity data and neuropsychological test scores from the OASIS-3 dataset.
• Trained predictive models and validated them on internal and external test sets (OASIS-3, KSHAP).
• Employed a predictive modeling approach to link brain connectivity to cognitive performance.

Main Results:

• Connectivity-based predicted scores showed moderate correlation with actual behavioral test scores (r=0.08-0.44).
• Models using a broader range of connectivity features outperformed those using focal features.
• The neural basis of test performance appears to be distributed across multiple brain systems.

Conclusions:

• Late-life neuropsychological test performance can be characterized by distributed connectome-based predictive models.
• Further research is needed to develop theoretically valid and clinically useful predictive models for translational applications.