Brain network dynamics in schizophrenia: Reduced dynamism of the default mode network

Area of Science:

• Neuroscience
• Psychiatry
• Computational Biology

Background:

• Complex human behavior relies on synchronized neural activity across brain networks.
• Schizophrenia is associated with behavioral and cognitive deficits, potentially linked to altered neural dynamics.

Purpose of the Study:

• To model neural activity dynamics in schizophrenia using resting-state fMRI.
• To investigate the association between these dynamics and behavioral/cognitive deficits in schizophrenia.

Main Methods:

• Inferred a hidden Markov model (HMM) from resting-state fMRI data of schizophrenia patients and healthy controls.
• Described neural activity fluctuations using 12 brain states across 14 resting-state networks.
• Compared time spent and visit duration in each state between groups and correlated with symptom severity.

Main Results:

• Schizophrenia patients spent less time in default mode and executive networks compared to controls.
• Default mode network activations were less frequent but longer in duration in schizophrenia.
• Positive symptom severity correlated with prolonged inactive default mode/executive states and heightened sensory network activity.
• Classifiers achieved 76-85% accuracy in predicting diagnosis based on brain state descriptors.

Conclusions:

• Neural dynamics, characterized by altered brain state engagement and duration, differ significantly in schizophrenia.
• These dynamic alterations are linked to symptom severity and diagnostic status.
• HMM-based analysis of fMRI data offers a promising approach for understanding schizophrenia pathophysiology.