Maximum entropy model reveals frequent brain state switching in a multiversal brain function analysis in early...

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Psychiatric Neuroscience

Background:

• Traditional neuroimaging methods like regional activation and pairwise correlation have limitations in capturing dynamic brain network interactions.
• The maximum entropy model (MEM) offers a promising integrative approach to study evolving functional brain networks by integrating regional and inter-regional activity.

Purpose of the Study:

• To compare the efficacy of MEM with traditional methods in characterizing brain network alterations in psychosis.
• To investigate dynamic changes in the default mode network (DMN) and dorsal attention network (DAN) in individuals with psychosis.

Main Methods:

• Applied activation-only, pairwise-coactivation-only, and MEM analyses to the Human Connectome Project-Early Psychosis dataset.
• Utilized the HCP/Glasser atlas to define DMN and DAN regions, quantifying group differences in nodal frequency band power, functional connectivity (FC), and MEM features.
• Examined transition rates between energy landscape basins and their association with cognitive and psychopathological measures.

Main Results:

• Psychosis was associated with reduced BOLD frequency band power and FC graph metrics.
• MEM analysis revealed more frequent switching between DMN and DAN nodal states in psychosis, with higher basin transitions and reduced dwell times.
• MEM metrics showed significant negative correlations with working memory and positive correlations with positive symptom severity.

Conclusions:

• The MEM provides unique insights into nodal activation/deactivation patterns, basin transitions, and network energy, surpassing traditional methods.
• MEM analysis better characterized psychosis-related network alterations, including attenuated inter-network connectivity and reduced DMN/DAN stability.
• These network instabilities may underlie impaired working memory and increased psychopathology severity in psychosis.