nucleation

THEORY-878: Instructed reappraisal of aversive pictures engages anterior prefrontal, temporo-parietal and cerebellar regions that show evidence of no response to negative emotion generation, beside a larger set of regions engaged both by generation and by reappraisal: in this task part of regulation runs in regions apart from generation, and most of it in regions shared with it

4 min read

Properties4
StatusProposed
Tagsemotion-and-affect, neuroscience, cognition
SourceLIT-1359 — A systems identification approach using Bayes factors to deconstruct the brain bases of emotion regulation
Bears onQUESTION-054 — Is there a governing level among an agent's regimes, a system that specifies control apart from the systems that carry it out, or is governance the interaction of the regimes themselves?

Graph View

Source

Bo, Kraynak, Kwon, Sun, Gianaros & Wager, LIT-1359, read in full in NOTE-1061 from the bioRxiv preprint, version 2 (10 May 2023), pp. 5–12 and 17–20, with Figures 1, 2 and 5. The version of record, Nature Neuroscience 27(5), 975–987 (2024), is closed and was not read; its abstract matches the preprint’s.

What was actually shown

Participants in two samples looked at neutral pictures, looked at aversive pictures, or reappraised aversive pictures to feel less negative. Each voxel was tested on two contrasts, emotion generation [Look Negative − Look Neutral] and reappraisal [Reappraise Negative − Look Negative], with Bayes factors that can support a null (BF > 10 for an effect, BF < 1/10 for none).

  • A selective set. Sixteen regions met the axioms for “Reappraisal only” in both samples: an effect of reappraisal and evidence of no effect of negative pictures (p. 8). They are anterior DLPFC and VLPFC, TPJ, IPL, posterior middle and anterior inferior temporal gyri and superior cerebellum, mostly in default-mode and frontoparietal networks.
  • A shared set. Fourteen regions met the axioms for “Common appraisal”: engaged by negative pictures and further by reappraisal (p. 9). They are the larger set: DLPFC, VLPFC, SPL, pSTS, anterior insula, dMPFC/aMCC/SMA, precuneus and cerebellum. The authors: “the majority of the regions activated by reappraisal were also activated during negative emotion-generation without explicit reappraisal instructions” (p. 18).
  • Which predicts success. Reappraisal-related activity in the selective set correlated with the drop in reported negative affect (r = 0.2), more than the shared set (r = 0.14), and the best model joined it with the regions reappraisal reduces (r = 0.26, adjusted R² = 0.063) (pp. 11–12).

What could have come out otherwise: every region engaged by reappraisal could also have responded to negative pictures, which is what a view in which regulation is generation re-run would expect. A set did not.

What this does not say

  • That the selective regions regulate. They are engaged during regulation and predict its success weakly. No interference was tried, and the authors’ description of them as “in a position to regulate” sensory hierarchies (p. 17) rests on anatomy and on the literature.
  • That they are an emotion regulator in general. One deliberate strategy, instructed downregulation, on IAPS pictures. Effortless regulation, which is where THEORY-056 makes its claim, was not studied. Their best Neurosynth match is “response inhibition”, so they may be task control rather than anything particular to emotion.
  • That passive viewing is unregulated. The authors allow that the shared regions “may be involved in spontaneous self-regulation in the absence of explicit regulatory instructions” (p. 18). If so, some regulation is in the baseline, and the shared set mixes generation with regulation.
  • That the null is a true zero. BF < 1/10 at the voxel and region level is evidence against a mean difference of the size the prior expects. It does not exclude small or patterned responses.
  • The version of record. Only the preprint and the published abstract were read.

Connections

It answers QUESTION-054 in part. For one deliberate strategy there is a set of regions engaged only when regulating, high in the cortical hierarchy and predictive of success: a candidate governing level in the anatomical sense. It does not reach effortless regulation, where QUESTION-054’s accounts differ.

  • THEORY-056 (regulation is not separate from generation). The shared set agrees with it; the selective set is what a distinct regulating system would look like in imaging. THEORY-056’s own promote_when says that a difference in network activation “cannot settle it”. So this result does not refute it, but it does show that this exclusion keeps THEORY-056 out of reach of imaging data.
  • THEORY-680 (regulation is a set of processes distinct from generation). The selective set agrees with its first clause. Its stage claims are not tested, since only one strategy was used, and its demand for an independently estimated unregulated response is not met.
  • THEORY-070 (a distinct system that specifies control). The selective regions’ annotation (response inhibition, executive function) fits a control system that is not particular to emotion. That is the record’s reading; the paper does not discuss it.
  • THEORY-871. The same study’s other half: what reappraisal does and does not change.
  • CLAIM-412. Its grounds include THEORY-056 and THEORY-680 as the two readings of reappraisal. This result divides between them and does not choose.

Backlinks