Whole brain correlates of individual differences in skin conductance responses during discriminative fear conditioning to social cues
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
Vinbert et al. provide a conceptual replication on individual differences in conditioned skin conductance response during fear acquisition training and BOLD fMRI in a large sample (N=285) of healthy individuals (mono- and dizygotic twins). The authors report results that are in line with previous work and new results from a whole-brain analysis and suggest unique and shared contributions of individual brain regions. This is a timely and well-conducted replication study, the sample size is large for this area, and there are robustness (multiverse) analyses in place to ensure that findings are not driven by particular analysis choices.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their names with the authors.)
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (eLife)
Abstract
Understanding the neural basis for individual differences in the skin conductance response (SCR) during discriminative fear conditioning may inform on our understanding of autonomic regulation in fear-related psychopathology. Previous region-of-interest (ROI) analyses have implicated the amygdala in regulating conditioned SCR, but whole brain analyses are lacking. This study examined correlations between individual differences in SCR during discriminative fear conditioning to social stimuli and neural activity throughout the brain, by using data from a large functional magnetic resonance imaging study of twins ( N = 285 individuals). Results show that conditioned SCR correlates with activity in the dorsal anterior cingulate cortex/anterior midcingulate cortex, anterior insula, bilateral temporoparietal junction, right frontal operculum, bilateral dorsal premotor cortex, right superior parietal lobe, and midbrain. A ROI analysis additionally showed a positive correlation between amygdala activity and conditioned SCR in line with previous reports. We suggest that the observed whole brain correlates of SCR belong to a large-scale midcingulo-insular network related to salience detection and autonomic-interoceptive processing. Altered activity within this network may underlie individual differences in conditioned SCR and autonomic aspects of psychopathology.
Article activity feed
-
-
Evaluation Summary:
Vinbert et al. provide a conceptual replication on individual differences in conditioned skin conductance response during fear acquisition training and BOLD fMRI in a large sample (N=285) of healthy individuals (mono- and dizygotic twins). The authors report results that are in line with previous work and new results from a whole-brain analysis and suggest unique and shared contributions of individual brain regions. This is a timely and well-conducted replication study, the sample size is large for this area, and there are robustness (multiverse) analyses in place to ensure that findings are not driven by particular analysis choices.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. …
Evaluation Summary:
Vinbert et al. provide a conceptual replication on individual differences in conditioned skin conductance response during fear acquisition training and BOLD fMRI in a large sample (N=285) of healthy individuals (mono- and dizygotic twins). The authors report results that are in line with previous work and new results from a whole-brain analysis and suggest unique and shared contributions of individual brain regions. This is a timely and well-conducted replication study, the sample size is large for this area, and there are robustness (multiverse) analyses in place to ensure that findings are not driven by particular analysis choices.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their names with the authors.)
-
Reviewer #1 (Public Review):
Vinbert et al. provide a conceptual replication on individual differences in conditioned skin conductance response during fear acquisition training and BOLD fMRI in a large sample (N=285) of healthy individuals (mono- and dizygotic twins). The authors report results that are in line with previous work and new results from a whole-brain analysis and suggest unique and shared contributions of individual brain regions.
Strengths of the manuscript include the large sample size and the attempt to conceptually replicate previous work in a large sample as well as including an interesting extension in the scope of research beyond previous work (whole brain analyses, shared/unique contributions). The authors provide a number of robustness analyses in the supplementary material which is highly appreciated. Yet, …
Reviewer #1 (Public Review):
Vinbert et al. provide a conceptual replication on individual differences in conditioned skin conductance response during fear acquisition training and BOLD fMRI in a large sample (N=285) of healthy individuals (mono- and dizygotic twins). The authors report results that are in line with previous work and new results from a whole-brain analysis and suggest unique and shared contributions of individual brain regions.
Strengths of the manuscript include the large sample size and the attempt to conceptually replicate previous work in a large sample as well as including an interesting extension in the scope of research beyond previous work (whole brain analyses, shared/unique contributions). The authors provide a number of robustness analyses in the supplementary material which is highly appreciated. Yet, replication attempts are most useful when it is clearly outlined which effect is aimed to be replicated, a thorough and precise status quo of the literature is provided and in case of conceptual replications which procedural and analytical specifications differ from the previous, to-be-replicated work. It would be helpful for the reader if the exact results of previous work are, the employed procedures and analyses of previous work were described and discussed in relation to the present work in more detail.
The study sample is relatively large (N=285) - yet the sample is special in that participants were genetically related and as siblings (twins, mono- and dizygotic twins) shared environmental influences. This become only apparent in the method section (and discussion?) but needs to be mentioned upfront in the abstract, intro and included in the discussion as this may have an impact on the results. From the methods section it remained unclear how many pairs of di- and monozyotic twins were included in the study and more information on the sample (age range for instance) would be desirable.
The authors used 4 different trial sequences. Can they provide information on which CS+ trial was the first reinforced trial in these different sequences? The reason I am asking this is that if the first 5 CS+ presentations in sequence#1 were not reinforced but already the first one was reinforced in sequence #2 this would likely lead to differences in learning speed and ultimately average CS discrimination which may impact on the results. Are individual differences in discrimination related to trial sequences?
Can the authors elaborate on the advantages of using Z-transformed SCR in one set of analyses and square root transformed raw values in other sets of analyses? The reader would profit from a bit more detail to what extent Z-transformed values lead to confounding CS+ and CS- values with response magnitude (as indicated in section 4.3.3). I would also appreciate if the type of transformation used would be clearly indicated in the figures/figure captions in the main manuscript.
On a related note, have the authors investigated individual differences in SCRs to the US? The authors note that individuals that display stronger CS discrimination also had higher SCR magnitudes. Is it possible that the results provided here do reflect associations with physiological reactivity (SCR) rather than associative learning processes (CS discrimination)? I wonder if the findings they report here may be related to general physiological reactivity and would also be evident when looking at US responses. In the end, stronger responding to the US may provide a stronger "teaching signal". The work by Marin et al., (2019), which the authors cite in their manuscript relates to this question: https://doi.org/10.1111/psyp.13350
The robustness analyses provided in the supplementary material (excluded participants and SCR transformations) are highly appreciated. I suggest to include some basic information in table 1 which results were robust against these checks and which were not to facilitate "digestion" of this information.
In the discussion, the authors note that individual differences in SCRs are stable and provide 3 references for this. The authors may want to double check if these references really show demonstrate the stability of individual differences in CS discrimination. If I am not mistaken, neither Fredriksson (1993) nor Zeidan (2012) report stability measures for CS discrimination (but only for CS+ and CS- individually).
Did the authors record any other outcome measures than SCRs and BOLD fMRI? As the authors only report individual difference analyses with SCRs the question remains whether results can really be interpreted the way the authors do in the discussion (arousal/salience). It would be very interesting to see comparable analyses with ratings of fear or contingency awareness. If these are not available, I suggest to discuss this point in a bit more detail.
Can the authors provide a more information how exactly the eigenvariates were extracted as there are a number of different ways to do so (different tools, first-level, second level). I also suggest to add a little bit more information/explanation/ discussion what exactly is captured by the eigenvariate that was extracted. Given the level of details provided in the manuscript, I could not completely follow the procedure (i.e., are not 100% sure what was done) and hence interpretation.
The authors report that correlations to differential SCRs were mainly driven by increased responding to the CS+. Can the authors provide some information on the distribution of CS+ and CS- responses (I may have missed these if reported somewhere). Are individual differences/variances similar for SCRs to both CS types?
The possible link to psychopathology mentioned at a number positions in the manuscript could be elaborated on in somewhat more detail (direction of findings, which disorders, which experimental phase, CS discrimination effect of CS specific effects, which outcome measure was associated...). Also here, the authors report a rather unspecific association ("pathological fear and anxiety has been associated with altered SCR discrimination during fear conditioning", page 7). I suggest to check back on the original literature and specify this statement as the meta-analysis by Duits does not suggest an effect of CS discrimination during fear acquisition (but CS- responding).
-
Reviewer #2 (Public review):
The authors investigated the whole-brain neural correlates (as assessed with fMRI) of individual differences in SCR during fear conditioning in humans.
Strengths of the manuscript include:
Use of very big (in neuroimaging terms, and for a single study) sample size.
Use of sound methods, including whole-brain (instead of ROI) fMRI analyses and additional calculations using alternative methods to calculate SCR.
Analysis of the potential contribution of "each" brain activation to SCR response. Very few studies present this kind of analysis, and I found it very interesting.
Clear description of the methods/results and transparency in making most of the data available for the rest of the scientific community, therefore facilitating replication.
Weaknesses of the manuscript include:
Putting together whole-brain and …
Reviewer #2 (Public review):
The authors investigated the whole-brain neural correlates (as assessed with fMRI) of individual differences in SCR during fear conditioning in humans.
Strengths of the manuscript include:
Use of very big (in neuroimaging terms, and for a single study) sample size.
Use of sound methods, including whole-brain (instead of ROI) fMRI analyses and additional calculations using alternative methods to calculate SCR.
Analysis of the potential contribution of "each" brain activation to SCR response. Very few studies present this kind of analysis, and I found it very interesting.
Clear description of the methods/results and transparency in making most of the data available for the rest of the scientific community, therefore facilitating replication.
Weaknesses of the manuscript include:
Putting together whole-brain and ROI-based data in a regression analysis seems not "fair" to assess the contribution of different brain activations to SCR responses.
Some inclusion/exclusion criteria are not well defined (e.g., "current alcohol or drug-related problems") or unclear (e.g., why should someone receiving psychological treatment be excluded? were only psychotropic medications -and not other medications- excluded? )
The potential limitations of 1) using a twin sample and 2) asking the participant to press a button after each CS presentation are not discussed.
I think that overall the authors have achieved their aims, and their results support their conclusions.
I think the ms may impact the fear conditioning field and broader fields like the study of human emotion.
-
Reviewer #3 (Public Review):
This manuscript by Vinberg et al. reports brain-behaviour correlations between BOLD activity and fear-conditioned (differential) SCR in a delay fear conditioning task with social CS and a 6 s CS-US interval.
This is a timely and well-conducted study, the sample size is adequate, and there are robustness (multiverse) analyses in place to ensure that findings are not driven by particular analysis choices. The experimental paradigm and BOLD analysis appears appropriate.However, peak-scoring windows for the SCR analysis are unclear, and potentially problematic. This, together with the comparably large effect size for the CS+/CS- difference in SCR, suggests a potential risk that the authors may have inadvertently looked at outcome-driven (US- or omission-riven) SCR, rather than conditioned SCR. This would call …
Reviewer #3 (Public Review):
This manuscript by Vinberg et al. reports brain-behaviour correlations between BOLD activity and fear-conditioned (differential) SCR in a delay fear conditioning task with social CS and a 6 s CS-US interval.
This is a timely and well-conducted study, the sample size is adequate, and there are robustness (multiverse) analyses in place to ensure that findings are not driven by particular analysis choices. The experimental paradigm and BOLD analysis appears appropriate.However, peak-scoring windows for the SCR analysis are unclear, and potentially problematic. This, together with the comparably large effect size for the CS+/CS- difference in SCR, suggests a potential risk that the authors may have inadvertently looked at outcome-driven (US- or omission-riven) SCR, rather than conditioned SCR. This would call into question the brain-behaviour relationship results.
-
