Behavioral/Cognitive The VLPFC-Engaged Voluntary Emotion Regulation: Combined TMS-fMRI Evidence for the Neural Circuit of Cognitive Reappraisal

Introduction

• Emotion regulation (ER) is essential for mental health and adaptive social behaviors.
• ER involves neurobiological processes from cortical control systems (prefrontal regions) that modulate the activity of subcortical affective response systems (amygdala and insula).
• ER can be:
  • Automatic/implicit: Evoked unconsciously by affective stimuli, mainly engages the medial prefrontal cortex (MPFC).
  • Voluntary/explicit: Requires conscious effort, launched and maintained by the lateral PFC (LPFC).
• Cognitive Reappraisal:
  • Voluntary ER strategy that reduces negative experiences and emotion-related neural responses with long-lasting benefits.
  • Supported by the dorsolateral PFC (DLPFC), ventrolateral PFC (VLPFC), and ventromedial PFC (VMPFC).
  • DLPFC: Maintains multiple appraisals for the current affective situation in working memory.
  • VLPFC: Selects the appropriate appraisal and inhibiting others in accordance with the ER goal.

Background and Significance

• The neural circuit underlying emotion regulation (ER) is crucial for basic and translational research.
• There's a need for evidence combining neuroimaging and neuromodulation techniques.
• Key questions addressed:
  • Does prefrontal-subcortical activity causally contribute to the ER effect?
  • Does the prefrontal control system directly modulate the subcortical affective system?
• The study combines fMRI with transcranial magnetic stimulation (TMS) to map connections between the PFC and subcortical structures (amygdala, insula).
• TMS-induced ventrolateral PFC (VLPFC) facilitation enhances activity in the VLPFC and ventromedial PFC (VMPFC).
• It also attenuates activity in the amygdala and insula during reappraisal but not during nonreappraisal (baseline).
• Activated VLPFC intensifies prefrontal-subcortical couplings via the VMPFC during reappraisal only.
• The study provides evidence that downregulating negative emotion involves the prefrontal control system suppressing the subcortical affective system.
• The VMPFC serves as a crucial hub within the VLPFC-subcortical network, suggesting an indirect pathway model of the ER circuit.
• The findings outline potential protocols for improving ER ability by intensifying the VLPFC-VMPFC coupling in patients with mood and anxiety disorders.
• TMS amplifies neural changes within the ER circuit, highlighting the "bridge" role of the VMPFC under the reappraisal versus nonreappraisal contrast.
• TMS helps identify brain regions that causally support reappraisal (VLPFC and VMPFC) and those that are modulated by reappraisal (amygdala and insula).

Hypotheses

• VLPFC is an essential region in voluntary emotion regulation.
  • TMS facilitation would launch a chain reaction in the neural circuit of reappraisal: increased activity in the TMS target (i.e., VLPFC) and attenuated activity in subcortical regions of the amygdala and insula.
• Prefrontal-subcortical connectivity should increase as a result of VLPFC facilitation.
  • The study empirically tests whether VMPFC activity and connectivity were influenced by VLPFC TMS.
• ROIs:
  • right VLPFC
  • bilateral VMPFC
  • amygdala
  • insula. The bilateral amygdala and insula were chosen as they are representative hubs in the subcortical affective system and have been particularly associated with social pain processing.

Direct vs. Indirect Pathway Models

• Direct-pathway model:
  • Prefrontal regions directly modulate subcortical regions during ER implementation.
  • Evidence: VLPFC to amygdala and ventral striatum (Wager et al., 2008), DLPFC to ventral striatum in cognitive regulation of substance craving (Kober et al., 2010).
• Indirect-pathway model:
  • Voluntary ER is supported by prefrontal modulation of the subcortical responses via their mutual connections to the VMPFC
  • Evidence:
    • VMPFC mediated the connections between the VLPFC and amygdala during reappraisal (Johnstone et al., 2007).
    • Reappraising fearful situations engaged LPFC regions, which modulated amygdala activity via the VMPFC (Urry et al., 2006; Delgado et al., 2008).
    • The mediating role of the VMPFC in the LPFC input to the amygdala has also been highlighted in dynamic causal modeling (DCM) studies (Steward et al., 2021).

Methods

• Participants:
  • Included two TMS groups: active and sham stimulation.
  • A priori power analysis indicated that 26 participants would ensure 80% statistical power.
  • 120 healthy college students were recruited from Shenzhen University and randomly assigned into the TMS groups.
  • Participants were right-handed with normal or corrected-to-normal vision.
  • Final sample: 58 in the sham and 59 in the active TMS groups.
• Experimental design:
  • 2 (regulation type: no-reappraisal vs reappraisal) x 2 (TMS group: sham vs active) mixed design.
  • Regulation type was the within-subject factor, and TMS group was the between-subject factor.
  • The task was divided into two sessions, corresponding to the two regulation types.
  • 60 social exclusion pictures were selected from the Social Inclusion and Exclusion in Asian Young Adults image database (Zheng et al., 2022).
  • Each picture depicts a scenario of social exclusion, consisting of a rejected individual with sad or upset facial and body expressions and three or four rejecters who are talking and/or laughing together
• Task:
  • Each session began with a 3 s fixation cross.
  • A trial began with an image presentation for 7.5 s, during which participants were required to either watch passively (i.e., the no-reappraisal session) or downregulate their negative emotions using reappraisal strategies (i.e., the reappraisal session).
  • Participants reported their negative feelings on a 9 point scale.
  • Jitter duration between two trials was randomly set at 1.5, 3.0, 4.5, 6.0, or 7.5 s.
  • Two 10 min TMS sessions, each followed by one task session (fMRI session).
• Repetitive TMS (rTMS):
  • A figure-eight-shaped coil connected to the magnetic stimulator (M-100 Ultimate, Shenzhen Yingchi Technology).
  • TMS was applied over the right VLPFC in the active TMS group and over the vertex in the sham TMS group.
  • The coordinate of the left motor cortex (x = -38.3, y = -15.2, z = 67.9) was determined as the optimal coil position for motor cortex stimulation (Bungert et al., 2017).
  • rTMS was applied at 10 Hz; Each 10 min session contained 20 trains, each lasting 3.9 s with an intertrain interval of 26.1 s (Zhao et al., 2021).
• Image acquisition:
  • Brain images were collected using a 3-T MR scanner (Siemens Trio).
  • Functional images were collected using an EPI sequence (number of slices, 72; gap, 0.6 mm; slice thickness, 2.0 mm; TR, 1500 ms; TE, 30 ms; flip angle, 75°; voxel size, 2 mm x 2 mm x 2 mm; FOV, 192 mm x 192 mm).
  • Structural images were acquired through 3D sagittal T1-weighted MPRAGE (224 slices; TR, 1900 ms; TE, 2.23 ms; voxel size, 1.1 mm x 1.1 mm x 1.1 mm; flip angle, 8°; inversion time, 904 ms; FOV, 220 mm x 220 mm).
• Image processing and statistical analysis:
  • Images were preprocessed and analyzed using Statistical Parametric Mapping (SPM12).
  • Functional images were coregistered with the T1-weighted 3D images, normalized to MNI space, and smoothed with an 8 mm FWHM isotropic Gaussian kernel.

Results

• Ratings of negative emotion:
  • Significant interaction between TMS group * regulation type (F(1,115) = 5.57, p = 0.020, h^2_p = 0.046).
  • Active TMS group reported lower negative feelings than the sham TMS group in the reappraisal session (F(1,115) = 10.79, p = 0.001, h^2_p = 0.086).
  • Main effect of regulation type was highly significant (F(1,115) = 290.42, p < 0.001, h^2_p = 0.716).
  • Significant main effect of TMS group (F(1,115) = 4.97, p = 0.028, h^2_p = 0.041).
• Postscanning picture ratings:
  • Significant interaction between TMS group * regulation type (F(1,115) = 4.20, p = 0.043, h^2_p = 0.035).
  • Active TMS group reported higher valance of the pictures than the sham TMS group in the reappraisal session (F(1,115) = 14.40, p < 0.001, h^2_p = 0.111).
  • Main effect of regulation type was highly significant (F(1,115) = 9.03, p = 0.003, h^2_p = 0.073).
  • Significant main effect observed in TMS group (F(1,115) = 7.15, p = 0.009, h^2_p = 0.059).
• ROI analysis:
  • Significant interaction effects between TMS group * regulation type in the right VLPFC, right VMPFC, right amygdala, and left insula.
  • These interactions were all driven by a significant group difference during the reappraisal condition, but not during the no-reappraisal condition.
• Whole-brain analysis:
  • Main effect of the regulation type was observed in extensive brain regions in the frontal, temporal, parietal, and occipital lobes as well as the limbic system.
  • The interaction between TMS group * regulation type revealed significant brain clusters in the right VLPFC and left insula.
• DCM results:
  • The indirect model family, which had VLPFC-to-VMPFC-to-subcortical connections modulated by the ER effect, was a better explanation of the data with a total exceedance probability of 0.995 (active) and 0.992 (sham).
  • Model 22 outperformed the other models with an exceedance probability of 0.571 (active) and 0.432 (sham).
• Brain-behavioral correlations:
  • Higher regulation success correlated positively with activity enhancement from the no-reappraisal to the reappraisal condition in the VLPFC (r = 0.346, p(FDR) = 0.028) and VMPFC (r = 0.318, p(FDR) = 0.028) in the active TMS group.

Discussion

• The study aimed to answer two questions regarding voluntary ER.
  • Is voluntary ER dependent on the VLPFC, and is downregulating negative emotion implemented by the VLPFC’s suppression of the subcortical affective system?
  • How does the prefrontal top-down control system connect with the subcortical affect system?
• TMS-induced VLPFC facilitation led to enhanced activity in the prefrontal network and attenuated activity in the subcortical regions, and that the activated VLPFC intensified the prefrontal-subcortical couplings via the VMPFC.
• The study provides causal perturb-and-measure evidence regarding the voluntary ER being critically dependent on the VLPFC, and that downregulating negative emotion is a process that involves the prefrontal control system suppressing the subcortical affective system.
  • Voluntary ER is influenced by the VLPFC and its interactions with other brain regions, specifically the VMPFC and subcortical affective areas
  • The opposite neural changes in prefrontal (enhanced) and subcortical (attenuated) regions are not a byproduct of voluntary ER; instead, this prefrontal-subcortical activity intrinsically and causally contributes to the ER effect.
• Increased activity in three ER-modulated pathways (i.e., excitatory VLPFC-to-VMPFC, inhibitory VMPFC-to-amygdala, and inhibitory VMPFC-to-insula pathways) because of VLPFC facilitation has been found.

Conclusion

• This study revealed that the VLPFC serves as an essential brain region to support the voluntary ER process, while the downstream propagation of reappraisal unfolds as a “chain reaction” from the prefrontal control network into the emotion integrative area VMPFC and to the subcortical affective regions.
• The VLPFC-VMPFC coupling during voluntary ER provides novel avenues for clinical practice.
• Neuroimaging studies usually used threatening pictures from the International Affective Picture System (Lang et al., 2005) and revealed that the VMPFC is the “intermediate station” involved in reappraisal on the LPFC-to-amygdala pathway (e.g., Urry et al., 2006; Johnstone et al., 2007; Silvers et al., 2017; Steward et al., 2021).
• The inhibitory VMPFC-insula coupling is essential for the VLPFC’s modulation of the insula during reappraising (the insula is more sensitive to painful than fearful stimuli).
• The TMS-facilitated ER persisted for at least half an hour, as revealed by more positive rating of pictures 30 min following the ER implementation in the active compared with the sham TMS group.