TL;DR
This study investigates the dynamic interplay between the anterior insula (AIC) and anterior cingulate cortex (ACC) during movie viewing using high-fidelity intracranial EEG. They found that these regions exhibit distinct but complementary roles in integrating sensory information with bodily signals (heart rate variability), highlighting their importance in emotional experiences.
What Problem Does This Paper Address?
The brain's ability to integrate external sensory information with internal bodily states is crucial for adaptive behavior and emotional processing. The anterior insula (AIC) and anterior cingulate cortex (ACC), key components of the brain's "salience network," are known to be involved in this process. However, the precise nature of their dynamic interactions and how they contribute to linking perceptual features with physiological responses, especially during complex, naturalistic stimuli like movies, remains unclear. This paper aims to unravel these context-dependent dynamic interactions.
Methods
The researchers employed a multi-pronged approach:
- High-Fidelity Intracranial EEG (iEEG): They recorded iEEG from epilepsy patients with electrodes implanted in both the AIC and ACC. This invasive technique offers excellent temporal and spatial resolution, allowing for the capture of rapid neural dynamics.
- Power Spectral Analysis: They analyzed the frequency content of the iEEG signals to identify characteristic brain rhythms associated with movie viewing.
- Dynamic Causal Modelling (DCM): This is a sophisticated computational framework used to infer effective connectivity (causal influence) between brain regions. They used a biophysically informed DCM (canonical microcircuit model) to model the interactions between AIC and ACC. This model allows for the investigation of both feedforward (bottom-up) and feedback (top-down) connections.
- Multilevel Parametric Empirical Bayes (PEB): This statistical approach was used to analyze the DCM results, allowing for the estimation of consistent connectivity patterns across subjects and over time, while accounting for individual differences and temporal fluctuations.
- Correlation Analysis: They correlated the estimated effective connectivity parameters with extracted movie features (visual, audio, emotional) and heart rate variability (HRV) to understand how these neural dynamics relate to external stimuli and internal bodily states.
Datasets
- Primary Dataset: iEEG recordings from five epilepsy patients with electrodes implanted in the AIC and ACC during movie viewing.
- Validation Dataset: An independent resting-state iEEG dataset from six participants with AIC and ACC implants was used for comparative analysis of power spectra.
Key Insights and Ideas
- Gamma Band Activity as a Marker of Integration: The study found a prominent gamma band power peak (30-35 Hz) in both AIC and ACC during movie viewing, which was absent during resting states. This suggests that this specific frequency band plays a role in the active integration of sensory information during engaging experiences.
- Dissociable Roles of AIC and ACC Connectivity: The research revealed that connectivity originating from the ACC was primarily associated with exteroceptive (external sensory) features of the movie (e.g., salience, audio, luminance). In contrast, AIC-driven connectivity was linked to both sensory features (audio) and internal bodily signals (HRV). This points to a division of labor, with ACC focusing on processing external information and AIC acting as a crucial bridge between external and internal states.
- Predictive Coding Framework: The findings are interpreted within a predictive coding framework, suggesting that feedforward connections carry prediction errors and feedback connections convey prediction signals. The positive correlation between these connections and movie features/HRV supports the idea that the AIC-ACC system dynamically refines predictions about the environment and the body's state.
- Emotional Experience as a Driver: The study highlights that these dynamic interactions are particularly pronounced during emotional experiences, as elicited by the movie, underscoring the role of the AIC-ACC circuit in processing affect.
Key Results
- Gamma Power Peak: A significant gamma power peak (30-35 Hz) was observed in both AIC and ACC during movie viewing, but only in AIC at rest.
- Correlations with Movie Features:
- Feedforward connectivity from ACC to AIC positively correlated with facial emotion and luminance.
- Feedback connectivity from AIC to ACC positively correlated with movie salience.
- Feedback connectivity from ACC to AIC positively correlated with facial emotion and audio.
- Correlations with HRV:
- Feedforward connectivity from AIC to ACC showed a negative correlation with the LF/HF ratio (a proxy for sympatho-vagal balance).
- AIC connectivity was related to HRV and audio.
- Interplay: The study found that connectivity parameters from the ACC were linked to stimulus-related sensory information, while AIC-driven connectivity was associated with both stimulus properties and physiological responses.
Limitations and Future Work
- Epilepsy Patient Cohort: The study utilized data from epilepsy patients, which, while providing high-quality iEEG, may introduce confounds due to underlying network pathology or residual artifacts. The authors acknowledge that epilepsy is a network disorder and that some patients had epileptic zones in structurally connected regions.
- Small Sample Size: The movie viewing data was collected from only five participants, limiting the generalizability of the findings. The authors emphasize the need for larger sample sizes and replication.
- Hemispheric Bias: Most contacts were located in the right hemisphere, with only one patient contributing left-hemisphere data, limiting the exploration of hemispheric asymmetries.
- Model Specificity: The use of a canonical microcircuit DCM model, while powerful, may not fully capture the specific cytoarchitecture of the AIC and ACC.
- Temporal Trajectories: The choice of temporal basis functions in the PEB analysis might not encompass all possible complex temporal dynamics.
- Subjective Engagement: The subjective engagement of participants during movie viewing could not be quantified due to surgical constraints.
- Limited Bodily Signals: The study primarily focused on HRV. Future work could incorporate a wider range of bodily signals (e.g., galvanic skin response, facial expressions) and explore more complex emotional stimuli.
- Future Directions: The authors suggest that these findings provide testable hypotheses for understanding deficits in brain disorders characterized by altered AIC-ACC function, potentially guiding the development of targeted interventions. Further research is needed to explore spatial specificity, hemispheric differences, and the impact of more complex stimuli and bodily signals.
Summary updated