Linking individual behavioral architecture to brain topological structure: A combined network analysis approach

Area of Science:

• Neuroscience
• Network Neuroscience
• Cognitive Neuroscience

Background:

• Understanding brain function necessitates a multimodal, network-level approach beyond isolated domains.
• Individual uniqueness arises from interactions between neural and behavioral facets.
• The Human Connectome Project provides a rich dataset for studying brain-behavior relationships.

Purpose of the Study:

• To investigate how large-scale brain network topology relates to individual mental architecture.
• To model nonlinear brain-behavior relationships using graph theory and Generalized Additive Models.
• To determine the contribution of resting-state network (RSN) topological properties to cognitive and behavioral performance.

Main Methods:

• Combined graph theory and exploratory graph analysis on Human Connectome Project data (N=379).
• Utilized Generalized Additive Models to analyze nonlinear brain-behavior associations.
• Examined topological properties of seven canonical resting-state networks (RSNs) across seven cognitive/behavioral domains.

Main Results:

• Identified domain-specific signatures of network integration and segregation correlating with cognitive/behavioral performance.
• Higher-level and core cognitive functions linked to fronto-parietal network integration and default mode network segregation.
• Maladaptive domains (substance abuse, impulsivity) associated with increased sensorimotor and limbic network integration.
• Single-node analysis revealed key hubs with topological features specific to distinct cognitive/behavioral profiles.

Conclusions:

• Network neuroscience offers a powerful approach to understanding the complexity of human cognition and behavior.
• Brain network topology plays a crucial role in characterizing individual differences in mental architecture.
• Domain-specific network configurations underlie both adaptive and maladaptive cognitive and behavioral outcomes.