Abstract
In this study, we aim to elucidate how intranasal oxytocin modulates brain network characteristics, especially over the frontal network. As an essential brain hub of social cognition and emotion regulation, we will also explore the association between graphic properties of the frontal network and individual personality traits under oxytocin (OT) administration. 59 male participants administered intranasal OT or placebo were followed by resting-state fMRI scanning. The Correlation-based network model was applied to study OT modulation effects. We performed community detection algorithms and conducted further network analyses, including clustering coefficient, average shortest path and eigenvector centrality. In addition, we conducted a correlation analysis between clustering coefficients and the self-assessed psychological scales. Modular organizations in the OT group reveal integrations of the frontoparietal network (FPN) and the default mode network (DMN) over frontal regions. Results show that frontal nodes within the FPN are characterized by lower clustering coefficients and higher average shortest path values compared to the placebo group. Notably, these modulation effects on frontal network property are associated with Interpersonal Reactivity Index (IRI) fantasy value. Our results suggest that OT elevates integrations between FPN, DMN and limbic system as well as reduces small-worldness within the FPN. Our results support graph theoretic analysis as a potential tool to assess OT induced effects on the information integration in the frontal network.
Introduction
Oxytocin (OT) is a critical neuropeptide involved in complex social behaviors, including trust, empathy, and bonding. While its effects on specific brain regions like the amygdala are well-documented, its impact on the brain's system-level functional architecture remains less understood. The frontal cortex, a central hub for social cognition and emotional regulation, is a primary target for investigating these system-wide changes. This study utilizes graph theoretic analysis to examine how intranasal OT modulates the functional connectivity and organizational properties of the frontal network, moving beyond regional activation to understand global information integration.
Methods
The study followed a double-blind, placebo-controlled design involving 59 healthy male participants. Participants were randomly assigned to receive either 24 IU of intranasal oxytocin (n=30) or a placebo (n=29). Following administration, resting-state fMRI (rs-fMRI) scans were conducted to capture baseline functional connectivity.
As illustrated in the workflow above, the brain was modeled as a network of 90 nodes based on the Automated Anatomical Labeling (AAL) atlas. Functional connectivity was defined using Pearson correlation. The analysis pipeline included:
- Community Detection: Identifying functional modules using the Louvain algorithm.
- Network Metrics: Calculating the Clustering Coefficient (CC) for local connectivity and the Average Shortest Path (ASP) for global efficiency.
- Outlier Management: For specific graph metrics (CC, ASP, and Eigenvector Centrality), one outlier was excluded from each group, resulting in final sample sizes of n=29 (OT) and n=28 (Placebo).
- Behavioral Association: Correlating network metrics with the Interpersonal Reactivity Index (IRI), specifically the "Fantasy" subscale, to assess how individual traits interact with OT-induced changes.
Results
Modular Reorganization and Integration
Community detection revealed a significant shift in the brain's functional architecture under oxytocin. In the placebo group, the Frontoparietal Network (FPN) and Default Mode Network (DMN) maintained distinct, segregated modular structures. In contrast, the OT group exhibited increased integration between these networks, particularly within the frontal regions. This suggests that OT facilitates cross-talk between systems typically involved in external task-switching (FPN) and internal self-referential processing (DMN).
Altered Network Efficiency
Node-wise analysis identified six specific frontal nodes—including the Middle Frontal Gyrus and Inferior Frontal Gyrus—that showed significantly lower clustering coefficients ($p < 0.01$) and higher average shortest paths in the OT group. In the context of graph theory, this indicates a reduction in "small-worldness" within the FPN, shifting the network toward a more integrated, globally-oriented communication style rather than localized, "cliquish" processing.
Interaction with Personality Traits
A key finding was the relationship between network changes and the IRI "Fantasy" subscale. In the placebo group, higher fantasy scores were associated with lower local clustering. However, oxytocin modulated this relationship: individuals with lower baseline fantasy scores showed the most pronounced network shifts. This suggests that the neurobiological effects of OT are contingent upon an individual's baseline psychological profile, specifically their capacity for social imagination.
Discussion
This research demonstrates that oxytocin acts as a powerful modulator of the frontal network's functional architecture, promoting integration between the FPN, DMN, and limbic systems. By reducing local clustering and increasing global integration, OT may enhance the brain's ability to process complex social information.
However, several limitations warrant further investigation. The study's exclusive focus on male participants limits generalizability, as OT effects are often sexually dimorphic. Additionally, while rs-fMRI provides a snapshot of the brain's "wiring," future studies using dynamic connectivity analysis or task-based fMRI could clarify how these network shifts translate into real-world social behavior. Despite these limitations, the results highlight graph theoretic analysis as a robust tool for assessing the systemic impact of neuropeptides on the social brain.
Summary updated