Fronto - Parietal Regions Predict Transient Emotional States in Emotion Modulated Response Inhibition via Low Frequency and Beta Oscillations

Siddharth Nayak, Arthur C. TsaiSymmetry 2022 · DOI

TL;DR

This study investigates how emotions influence our ability to stop actions, finding that specific brainwave patterns in the frontal and parietal regions predict our success in inhibiting responses. The research uses advanced computational modeling and EEG to show that emotional states can enhance inhibitory control, likely by influencing how quickly our brains process information and set response thresholds.

What Problem Does This Paper Address?

In our daily lives, we constantly need to balance acting quickly with the ability to inhibit actions, a crucial aspect of executive function. This ability, known as response inhibition, is vital for safe and goal-directed behavior. While tasks like the stop-signal task (SST) are standard for measuring this, the role of transient emotional states in modulating this inhibitory control remains complex and sometimes contradictory in previous research. Some studies suggest emotions improve inhibition, while others indicate they impair it. This ambiguity arises because previous research often focused on overall brain activity or specific event-related potentials, missing the dynamic, trial-by-trial neural processes that might explain these discrepancies. This paper aims to bridge this gap by examining how the brain's electrical activity, specifically low-frequency (delta, theta) and beta oscillations, predicts inhibitory performance on a moment-to-moment basis, and how task-relevant emotions influence these neural dynamics.

Methods

The researchers employed a combination of behavioral tasks, electroencephalography (EEG), and computational modeling.

  • Emotional Stop-Signal Task (SST): Participants performed a modified SST where visual stimuli were either neutral, happy, or disgust faces. This allowed for the investigation of how different emotional valences affect response inhibition.
  • EEG Recording: High-density EEG (128 electrodes) was used to record brain activity at a high temporal resolution (1000 Hz sampling rate). This allowed for the analysis of event-related spectral perturbations (ERSPs) to examine power changes in specific frequency bands (delta, theta, beta) over time.
  • Hierarchical Drift-Diffusion Model (HDDM): This sophisticated computational model was used to analyze the reaction time data from the "Go" trials. The HDDM decomposes decision-making into key parameters: response threshold (how much evidence is needed to make a decision), drift rate (the speed of evidence accumulation), and non-decision time (time for sensory/motor processes). By using a hierarchical approach, the model accounts for individual differences while also allowing for trial-by-trial variations.
  • Trial-by-Trial Regression Analysis: The core of the study involved regressing single-trial EEG oscillatory power (delta, theta, beta) from specific brain regions (mid-frontal and mid-parietal) against the HDDM parameters. This allowed researchers to directly link neural activity at the time of stimulus presentation to the decision-making processes that underlie response inhibition.

Datasets

The study utilized a dataset collected from 33 healthy participants performing the emotional stop-signal task. The visual stimuli consisted of emotional faces (happy, disgust) and neutral faces, sourced from a Taiwanese face database to minimize cultural bias. No new benchmarks were introduced; rather, the study used existing computational modeling frameworks and standard EEG analysis techniques.

Key Insights and Ideas

  • Emotion Enhances Inhibition (When Task-Relevant): The behavioral results showed faster responses in both Go and Stop trials for emotional stimuli compared to neutral ones, suggesting that task-relevant emotions improve inhibitory control. This aligns with theories suggesting emotions can prioritize processing when relevant to the task.
  • Low-Frequency Oscillations Predict Response Threshold: Trial-by-trial variations in delta and theta power in the mid-parietal region (early processing) and mid-frontal region (later processing) were found to predict changes in the response threshold. This suggests that these brainwave patterns dynamically adjust how much evidence is needed to commit to an action or to inhibit it.
  • Beta Oscillations and Emotional Differentiation: Beta oscillations in the mid-frontal region were specifically implicated in differentiating between disgust and neutral emotions, both in early and later processing stages. This points to a role for beta activity in the specific processing of negative emotional valence during inhibitory control.
  • Frontoparietal Network for Emotional Inhibition: The findings highlight the importance of a frontoparietal network, with distinct roles for different frequency bands and processing stages, in mediating the influence of emotion on response inhibition.

Key Results

  • Behavioral: Participants responded faster to disgust and happy faces in both Go trials (mean RT: ~672 ms for disgust, ~662 ms for happy vs. ~680 ms for neutral) and Stop trials (mean SSRT: ~313 ms for disgust, ~309 ms for happy vs. ~327 ms for neutral) compared to neutral faces.
  • Neural (Early Window, 0-300 ms):
    • Mid-frontal beta power decreased the response threshold for disgust and happy emotions relative to neutral.
  • Mid-parietal delta power increased the response threshold for disgust and happy emotions relative to neutral.
  • Neural (Later Window, 300-600 ms):
    • Mid-frontal delta and theta power increased the response threshold for disgust and happy emotions relative to neutral.
  • Mid-frontal beta power affected the response threshold for disgust emotion relative to neutral.
  • Mid-parietal delta and theta power increased the response threshold for happy emotion relative to neutral.

Limitations and Future Work

The authors acknowledge several limitations:

  • Model Limitations: The Drift-Diffusion Model (DDM) is not the only framework for modeling response inhibition, and alternative models like Dynamic Models of Choice (DMC) might offer more flexibility.
  • Limited Emotion Scope: The study only included happy and disgusted faces, limiting the generalizability of findings to other emotional categories.
  • Region of Interest: Due to computational constraints, the analysis focused on only two midline regions (Fz and Pz). Future work could explore a broader range of brain regions and consider brain asymmetry.

Future research directions suggested include:

  • Investigating interactions between low-frequency oscillations and beta power in the mid-frontal region to better understand their combined role in cognitive control.
  • Extending the framework to the rest of the brain and using standard atlases to better understand emotion-modulated response inhibition in the context of brain asymmetry.
  • Exploring a wider range of emotional stimuli to assess the generalizability of the findings.

Summary updated

BibTeX
@article{nayak2022fronto,
  author = {Siddharth Nayak and Arthur C. Tsai},
  title = {{Fronto - Parietal Regions Predict Transient Emotional States in Emotion Modulated Response Inhibition via Low Frequency and Beta Oscillations}},
  journal = {Symmetry},
  year = {2022},
  doi = {10.3390/sym14061244},
}
Figure