Deconstructing the brain bases of emotion regulation: A Bayes factor system-identification approach
Poster No:
710
Submission Type:
Abstract Submission
Authors:
Ke Bo1, Thomas Kraynak2, Mijin Kwon1, Michael Sun1, Peter Gianaros2, Tor Wager1
Institutions:
1Dartmouth College, Hanover, NH, 2University of Pittsburgh, Pittsburgh, PA
First Author:
Co-Author(s):
Introduction:
Emotion regulation is fundamental to physical and mental health. Reappraisal is a particular emotion regulation tactic that involves reinterpreting the meaning of events to alter emotional responses to them. Reappraisal and emotion-generation processes may interact non-additively, and some brain regions may be uniquely or jointly involved in both emotion regulation and generation. At present, however, it is not clear whether some regions are uniquely engaged by reappraisal or emotion-generation states, which precludes developing brain measures of these processes.
Here, we applied a systems identification approach to two large community samples (n=182 and 178), who viewed and reappraised aversive images from the International Affective Picture System (IAPS) during fMRI scanning. We aimed to identify brain regions that correspond to four potential system components: (1) 'Reappraisal only' regions responding only to reappraisal demand, not negative images; (2) 'Common appraisal' regions activated by negative images and further increased during reappraisal; (3) 'Non-modifiable emotion generation' regions activated by negative images but unaffected by reappraisal; and (4) 'Modifiable emotion generation' regions activated by negative images and reduced by reappraisal.
Here, we applied a systems identification approach to two large community samples (n=182 and 178), who viewed and reappraised aversive images from the International Affective Picture System (IAPS) during fMRI scanning. We aimed to identify brain regions that correspond to four potential system components: (1) 'Reappraisal only' regions responding only to reappraisal demand, not negative images; (2) 'Common appraisal' regions activated by negative images and further increased during reappraisal; (3) 'Non-modifiable emotion generation' regions activated by negative images but unaffected by reappraisal; and (4) 'Modifiable emotion generation' regions activated by negative images and reduced by reappraisal.
·A.Task paradigm. B.Concepts of system components. C. Visualization of axiomatic methods. D. Specific reappraisal strategies in current study. E. Subjective negative rating of each condition
Methods:
We established an axiomatic method based on a Bayes Factor approach to identify specific voxels for each system component. This involved an exhaustive search of whole brain voxels. For a voxel to be classified into a particular system component, it must satisfy predefined axioms. These axioms are established based on the observed activation and null effects across two contrasts: [Look Negative – Look Neutral] and [Reappraise – Look Negative]. Bayes Factors are applied to quantify the evidence for and against activation or null effects. We calculated them using the JZS prior, based on T statistics and degrees of freedom.
The spatial similarity of the identified system components was compared with a series of neural transmitter receptor density maps. These analyses aim to understand the neural chemical associations of these system components
The spatial similarity of the identified system components was compared with a series of neural transmitter receptor density maps. These analyses aim to understand the neural chemical associations of these system components
Results:
Our data identified regions consistently associated with each component across both datasets. 'Reappraisal only' regions included anterior prefrontal cortex, temporal-parietal junction and temporal pole. 'Common appraisal' regions (the component with the largest number of associated brain voxels) included fronto-parietal regions, nucleus accumbens, and medial prefrontal cortex. Among emotion generation regions, most subcortical regions were not modified by reappraisal, including amygdala, brainstem, PAG, parabrachial complex, and thalamus, while visual and attention-related regions were modifiable by reappraisal. Brain activities in 'Reappraisal only', 'Common appraisal' and 'Modifiable emotion generation' regions correlated with successful regulation of negative emotions. These regions also coincided spatially with serotonin, GABA, and glutamate receptor-dense regions.
Conclusions:
Our findings indicate that the brain regions engaged by reappraisal highly overlap with regions that may generate emotion, but some activations patterns were spatially selective for reappraisal. Automatic appraisal was observed to be supported by subcortical structures and are not influenced by reappraisal, while regions pertaining to sensory representation were the main regulatory targets.
Emotion, Motivation and Social Neuroscience:
Emotion and Motivation Other 1
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
Bayesian Modeling
Univariate Modeling
Keywords:
Cortex
Emotions
Neurotransmitter
Sub-Cortical
Univariate
Other - Emotion regulation
1|2Indicates the priority used for review
Provide references using author date format
Rouder, Jeffrey N., et al. "Bayesian t tests for accepting and rejecting the null hypothesis." Psychonomic bulletin & review 16 (2009): 225-237.
Hansen, Justine Y., et al. "Mapping neurotransmitter systems to the structural and functional organization of the human neocortex." Nature neuroscience 25.11 (2022): 1569-1581.