Task-induced brain state manipulation improves prediction of individual traits

Area of Science:

• Neuroscience
• Cognitive Science
• Brain Imaging

Background:

• Individual differences in brain functional organization link to behavior and cognition.
• The optimal brain state for revealing these brain-behavior relationships is not well-established.

Purpose of the Study:

• To determine if cognitive tasks enhance individual differences in functional connectivity compared to resting-state.
• To identify which tasks best predict fluid intelligence and if this varies by sex.

Main Methods:

• Analysis of two large, independent datasets using functional magnetic resonance imaging (fMRI).
• Comparison of predictive model performance for fluid intelligence using resting-state fMRI versus task-based fMRI data.
• Investigation of sex-specific differences in task-based predictive accuracy.

Main Results:

• Task-based fMRI data yielded significantly better predictive models of fluid intelligence than resting-state fMRI data.
• Specific cognitive tasks consistently outperformed others in predicting fluid intelligence.
• The optimal task for prediction demonstrated sex-specific effects.

Conclusions:

• Cognitive tasks amplify trait-relevant individual differences in functional connectivity patterns.
• Task-based functional connectivity analyses offer a more sensitive approach to understanding brain-behavior relationships than resting-state analyses.
• Considering task-induced brain states and sex is crucial for maximizing the explanatory power of brain-behavior models.