Understanding human behavior and cognition remains one of the most compelling challenges in neuroscience. Despite significant advances, non-invasive neuroimaging techniques still limit our ability to fully investigate brain function—particularly in the context of social cognition and behavior. Functional Magnetic Resonance Imaging (fMRI), recognized for its superior spatial resolution among non-invasive modalities, has played a central role in advancing our understanding of human brain organization and function. However, traditional activation-based approaches often fall short in capturing the complexity of social interactions and the underlying neural dynamics.

In response to these limitations, there has been growing interest in connectivity-based approaches as a means to more effectively characterize the brain's functional architecture. Dynamic Causal Modeling (DCM) is a powerful method for inferring the causal structure of dynamic neural systems by estimating effective connectivity from observed blood oxygenation level-dependent (BOLD) signals. In social neuroscience, DCM holds particular promise for disentangling complex neural interactions that underlie inter-individual variability in behavior and cognition. One of the major challenges in applying DCM to group-level studies is accounting for variability in connectivity across individuals. Such variability can obscure consistent patterns, complicating group-level inference. To address this, Parametric Empirical Bayes (PEB) provides a statistically rigorous approach that enables hierarchical modeling of subject-specific DCM parameters. By integrating individual estimates within a group-level framework, PEB facilitates the identification of both shared and divergent connectivity profiles.

Crucially, the DCM/PEB framework builds upon the results of General Linear Model (GLM) analyses, using identified activations and deactivations as priors to model the dynamic interactions among brain regions. This integration offers a coherent and biologically grounded means of linking observed neural responses to underlying network dynamics. However, a recurring issue in the literature involves the inappropriate application of DCM—particularly in relation to suboptimal experimental design and the selection of regions of interest. For example, volumes of interest (VOIs) that fail to meet statistical thresholds for activation in GLM analyses are frequently used, resulting in models that are underpowered or yield inconsistent findings. These shortcomings are not inherent flaws in the DCM framework itself, but rather reflect methodological oversights in experimental planning and data extraction.

To fully leverage the potential of DCM/PEB in social neuroscience, it is essential to ensure methodological rigor at multiple levels. This includes designing experiments that robustly elicit task-related activations, constructing models that comprise a necessary and sufficient set of interacting regions, and selecting VOIs based on reliable activation patterns. Only under such conditions can connectivity analyses yield interpretable and meaningful insights into the neural mechanisms underlying social cognition and behavior.

This Research Topic aims to showcase and promote best practices in the application of DCM/PEB to the study of social cognition and behavior. We invite contributions that present innovative experimental designs grounded in ecologically valid social tasks, rigorous statistical methodologies, and theoretically informed interpretations of connectivity. By advancing both methodological rigor and conceptual clarity, this collection seeks to highlight the unique strengths of DCM/PEB in uncovering the neural dynamics that support human social behavior. Potential themes may include, but are not limited to:

- Hierarchical modeling: Applications of Parametric Empirical Bayes (PEB) for group-level inference in dynamic causal modelling (DCM) of social cognition studies.

- Connectivity and behavior: Linking effective connectivity patterns with individual differences in social decision-making and behavior.

- Methodological rigor: Best practices in constructing and validating DCM/PEB models in social neuroscience.

- Experimental designs tailored for DCM/PEB: Creating ecologically valid social tasks that reliably elicit effective connectivity relevant to social cognition.

- Theoretical integration: Interpreting effective connectivity findings in light of cognitive, affective, and social neuroscience theories.

- Clinical and translational relevance: Using DCM/PEB to elucidate effective connectivity alterations in psychiatric or neurological disorders affecting social cognition.

Are you interested in contributing a manuscript to this Research Topic?

Dynamic Causal Modeling and Parametric Empirical Bayes in Social Neuroscience: Methods, Applications, and Insights