Cortical high-density counterstream architectures.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Network Science

Background:

• Small-world network models offer insights into brain architecture, balancing integration and segregation.
• Previous studies suggested low-density interareal graphs in the cortex, implying efficient connectivity.
• However, these models may not fully capture the complexity of cortical organization.

Purpose of the Study:

• To challenge existing models of cortical architecture.
• To propose a new model for interareal connectivity based on empirical data.
• To investigate the role of weight heterogeneity and distance-weight correlations in cortical networks.

Main Methods:

• Analysis of high-density interareal graphs representing cortical connectivity.
• Modeling of network properties including weight heterogeneity and distance-weight correlations.
• Development of a bow-tie representation for interareal architecture.

Main Results:

• Cortical graphs are revealed to be high-density, not low-density as previously reported.
• Economy of connections is achieved through weight heterogeneity and distance-weight correlations.
• A core-periphery structure and dual counterstream organization of pathways were identified.
• A bow-tie model accurately predicts network features.

Conclusions:

• The proposed bow-tie model offers a more accurate representation of cortical interareal architecture.
• Understanding these network properties is crucial for comprehending cortical computation.
• The findings highlight the importance of weight heterogeneity and distance-weight correlations in brain networks.