Interpersonal physiological and psychological synchrony predict the social transmission of nocebo hyperalgesia between individuals

Witnessing another’s pain can heighten pain in the observer. However, research has focused on the observer’s intrapersonal experience. Here, a social transmission-chain explored the spread of socially-acquired nocebo hyperalgesia. Dyads of genuine participants were randomised to ‘Generations’ (G1–G3). G1-Demonstrators, observed by G2-Observers, experienced high/low thermal pain contingent on supposed activity/inactivity of a sham-treatment. G2 became Demonstrators, witnessed by G3-Observers. They experienced fixed low-temperature stimuli irrespective of sham-treatment ‘activity’. G3 then Demonstrated for G4-Observers (a confederate), also experiencing low-temperature stimuli only. Pain ratings, electrodermal activity, and facial action units were measured. G1’s treatment-related pain propagated throughout the chain. G2 and G3 participants showed heightened subjective and physiological response to sham-treatment, despite equivalent stimulus temperatures, and G3 never witnessing the initial pain-event. Dyadic interpersonal physiological synchrony (electrodermal activity) and psychological synchrony (Observer’s ability to predict the Demonstrator’s pain), predicted subsequent socially-acquired pain. Implications relate to the interpersonal spread of maladaptive pain experiences.


Summary
In general, the main question of the present manuscript was to investigate how pain perception and experiences with certain treatments (thus expectations and experiences) transmit through generations of participants (G1-G4).The authors created a sham nocebo treatment in G1 (high pain), which was then observed and demonstrated in the next generation without the treatment (low pain).The authors report that nocebo effects were transferred from G1 to the other generations and the degree of synchrony in skin conductance and pain prediction predicted the degree of this transfer.

Strengths
The research question is fascinating and relevant, I was really excited to read this paper and it did not disappoint, this paper was a pleasure to read!The manuscript is well structured and easy to follow.The authors provide a strong rationale for their work and clearly derived hypotheses.I highly appreciate that the authors employed a range of methods to tackle their research question, and a range of statistical techniques and analyses to gather multiple parts of evidence.I really like the fact that the authors nearly only used real participants and minimal confederates, thereby increasing the amount and usefulness of the collected data.Methods are very detailed and allow for replication.Results are well described, follow a logical story, and relevant statistics are reported.I appreciate the clear figures and especially Figure 5 that summarizes all findings.I also appreciate the transfer of the authors findings to real world examples in the discussion.The authors answered nearly every open question I had when reading the paper.Nevertheless, I have some reservations, as outlined below, which need to be addressed, before I can recommend the article to be of sufficient quality to be published.My comments are structured as they appear in the text.I hope the authors find my suggestions helpful to improve the manuscript.

Comments
Line 4 speaks of G1-G3, but later in the abstract there is also G4, maybe clarify to avoid confusion?In line 73 onwards, I would also mention G4 then?I understood only in line 88 that G4 is a confederate, so this needs to be made more explicit.Maybe also in the subtitle of Figure 1?
The authors investigated the effect of a negative treatment, which leads to nocebo hyperalgesia, yet the title and abstract never once mention the word nocebo.I think it would be beneficial to frame the work in light of negative expectations and nocebo effects for readers already early in the abstract to void misunderstanding.
A recent meta-analysis on social observational learning came out that the authors might want to incorporate (just a suggestion, only if it fits) https://journals.lww.com/pain/Fulltext/9900/Learning_pain_from_others__a_systematic_review_and.328.aspxAlong the same lines, the key Benedetti paper might also be interesting, as it also investigated transmission of negative information and symptoms among different people?(again, just a suggestion) https://www.sciencedirect.com/science/article/abs/pii/S0304395914000244 Were the heat stimuli individually calibrated, as line 77 indicates a fixed temperature for high and low stimuli?45 degrees Celsius is not a low intensity stimulus, it can actually be very painful for some people.Why did the authors choose to not use individuals pain levels and could this have influenced their results?
Please explain already at the start of the results, why "trial" was additionally included as a factor.
Line 95 psychological synchrony should be lower-case letters?
Out of curiosity and as a second manipulation check, I would also be interested whether the authors ran comparisons of their outcome measures with G1?I see comparisons between G2 and G3, but it would also be interesting how the nocebo effect differs from the experienced treatment effect in G1?
Figure 3c: Is there a typo?In the title it says Nocebo (G2) but the Figure shows Nocebo (G3) and also lines for Nocebo (G2).
Did the authors check whether there was a difference between generations in the matching of Responder and Nonresponder dyads?Is this high rate of > 50% non-responders common regarding previous own and others' work?It might be helpful to put these numbers into perspective.The fact that the authors preregistered the study only becomes apparent at the end, in the methods, I think this can already be highlighted much sooner Please also include the link to the preregistration directly in the paper.
Could the confusion of the cues have been due to color-blindness (Tritanomaly or Tritanopia)?Did the authors check for these issues/were the cues clearly separable?What role do the authors think that gender (and matching of gender between observer and demonstrator) played in their study and findings?
In the discussion, it might potentially be interesting to talk about bidirectional influences, so whether this chain can also reverse and influence future responses of the earlier chain members (considering the fact that synchrony is not unidirectional)?Just an idea :) This paper examined the effects of perceived pain on one's own pain experience, how this is passed down through generations, and modulated by synchrony (both physiological and behavioral) between the demonstrator and observer.I found the paper to be well-written, intriguing, and innovative.One of the most interesting findings was the evident transmission of nocebo pain up to the third generation.Below are several points that I believe could enhance this manuscript and connect it to more extensive work beyond the realm of nocebo effects.
Abstract -Upon my first reading, I must admit that I found the abstract rather puzzling, understanding it more clearly only after reading the entire paper.It begins with, "Witnessing another's pain can intensify pain in the observer.However, current research has concentrated on the observer's intrapersonal experience, overlooking social reciprocity."In discussing reciprocity, I expected the authors to examine the impact of the observer on the demonstrator (effects of G2 on G1) rather than how the effect disseminates to other members.The term "generations" in the abstract was also confusing, as I initially believed it referred to family generations or age variations.

Introduction:
The section on nocebo is clear and informative.Nonetheless, presenting this solely in the context of nocebo effects neglects a vast realm of research on the evolutionary facets of social contagion and learning, including those specific to pain, stress, or fear, which could be responsible for these "side-effects".For instance, observing another's pain and subsequently experiencing or simulating it is considered fundamental for empathy and for learning to avoid potentially painful or stressful situations.In extreme scenarios (e.g., chronic stress or nocebo effects), this might have adverse impacts on the learner, yet it is generally an adaptive (and widely studied) mechanism.
Here's a recent paper on rodents that also reviews other relevant studies (including human ones): Keysers, C., & Gazzola, V. (2023).Vicarious emotions of fear and pain in rodents.Affective Science, 1-10.
Discussion: Similar to the Introduction, while the paper predominantly focuses on nocebo effects, the impacts should be considered in the broader context of emotion contagion, especially concerning pain.Methods: Preregistration: I was unable to locate the preregistration on the OSF, making it impossible to compare the analyses.
How was the participant number determined?
The term Psychological Synchrony might be somewhat confusing, as synchrony often suggests continuous temporal dynamics (as the authors themselves concede).A more accurate term would be more fitting.Perhaps terms like "Expectation of Pain Accuracy" or "Prediction of Pain" would be more appropriate.
Reviewer #3 (Remarks to the Author): This paper investigates multi-generational social transmission of nocebo hyperalgesia.In particular it investigates if nocebo hyperalgesia can be transmitted across generational chains of participants in the presence of sham treatments and mechanistically if psychological or physiological synchrony between demonstrator and observer contributes to such transmission.The results support both these effects.The study is well-designed and analyses are appropriate.Interpretations of the results are in line with the data presented.I also appreciated the strong open science practices followed.In sum, this is a very good paper and I have only some minor comments, mostly suggestions for clarification.Thanks to the author for a fascinating read!The methods were very clear both with regard to procedure and hypotheses.Unfortunately, I didn't find the description in section 2.1 as clear.I think information about what dyad pairs are the target of analysis could come earlier, together with hypothesis.I also first found (both here and in abstract) the information that there are 3 generations but a G4 participant somewhat confusing.Consider revisiting this section.
Please clarify in Figure that G4 is confederate.
In Figure 3, please add "psychological" to synchrony label on the x-axis Maybe I missed it, but for psychological synchrony (correlation between expectancy ratings) what were the actual values here?How good were observers at predicting how much pain demonstrators thought they would be in?
For psychological synchrony, why is the demonstrator's pain expectancy rating used and not their actual pain?Unless I missed it the argument isn't spelled out in the paper and it would be useful to have it there.I don't think it is necessary to present the step-wise regression, and variable selection by step-wise methods are generally not recommended.Instead, just present the full model (synchrony, generation, sync X generation), for each of the three synchrony measures.
Lines 248-249 -is it possible to explain in greater detail why the analyses concern the expression of empathic pain?Additionally why this is significant to explore.
For Figure 5, please add a figure legend for the meaning of the filled and dashed lines (nonsignificant and significant?).Please note that we have re-formatted the manuscript to reflect the formatting of Communications Psychology.The primary change being a switch to Methods-Results-Discussion formatting, rather than Results-Discussion-Methods.
We have attempted to maintain consistency between the two versions in terms of overall content.We would like to thank all reviewers for taking the time to leave comments and suggestions that have improved the manuscript.

REVIEWERS' COMMENTS:
Reviewer #1 (Remarks to the Author): I am an expert on pain and empathy for pain, and I work on placebo and nocebo effects, but I have no experience with FAC and EDA data or synchrony analysis.

Summary
In general, the main question of the present manuscript was to investigate how pain perception and experiences with certain treatments (thus expectations and experiences) transmit through generations of participants (G1-G4).The authors created a sham nocebo treatment in G1 (high pain), which was then observed and demonstrated in the next generation without the treatment (low pain).The authors report that nocebo effects were transferred from G1 to the other generations and the degree of synchrony in skin conductance and pain prediction predicted the degree of this transfer.

Strengths
The research question is fascinating and relevant, I was really excited to read this paper and it did not disappoint, this paper was a pleasure to read!The manuscript is well structured and easy to follow.The authors provide a strong rationale for their work and clearly derived hypotheses.I highly appreciate that the authors employed a range of methods to tackle their research question, and a range of statistical techniques and analyses to gather multiple parts of evidence.I really like the fact that the authors nearly only used real participants and minimal confederates, thereby increasing the amount and usefulness of the collected data.Methods are very detailed and allow for replication.
Results are well described, follow a logical story, and relevant statistics are reported.I appreciate the clear figures and especially Figure 5 that summarizes all findings.I also appreciate the transfer of the authors findings to real world examples in the discussion.The authors answered nearly every open question I had when reading the paper.Nevertheless, I have some reservations, as outlined below, which need to be addressed, before I can recommend the article to be of sufficient quality to be published.My comments are structured as they appear in the text.I hope the authors find my suggestions helpful to improve the manuscript.

Comments
1. Line 4 speaks of G1-G3, but later in the abstract there is also G4, maybe clarify to avoid confusion?In line 73 onwards, I would also mention G4 then?I understood only in line 88 that G4 is a confederate, so this needs to be made more explicit.Maybe also in the subtitle of Figure 1?
We have reordered the abstract and have made clear that G4 is a confederate.We have also updated the caption of Figure 1 to reflect this.The description of the procedure in the introduction (Lines 68-93) has been updated to clarify the roles of each generation of participants and the purpose of the confederate.
2. The authors investigated the effect of a negative treatment, which leads to nocebo hyperalgesia, yet the title and abstract never once mention the word nocebo.I think it would be beneficial to frame the work in light of negative expectations and nocebo effects for readers already early in the abstract to void misunderstanding.
Thanks for this comment.We have updated the title to "You pain increases my pain: Exploring the effect of interpersonal synchrony on the transmission of nocebo hyperalgesia" to make this point clearer and now mention the nocebo effect in the abstract.Many thanks for these recommendations.We read the meta-analysis with interest and have included it in the first paragraph of the introduction (please see lines 34-35).We are aware of the Benedetti paper but decided not to include it due to limited word count.Our reasoning was that the study does not focus on experimentally induced pain per se, which was the focus of the literature review, but also involves transmission from one individual to many rather than along chains of individuals (social transmission, as studied here).

A recent meta
4. Were the heat stimuli individually calibrated, as line 77 indicates a fixed temperature for high and low stimuli?45 degrees Celsius is not a low intensity stimulus, it can actually be very painful for some people.

Why did the authors choose to not use individuals pain levels and could this have influenced their results?
We based the design of this study on a previously published paradigm (Barnes et al., 2021) that we are currently employing in multiple other studies.We have found that direct conditioning and social observation with the two temperatures selected produces a robust nocebo effect during test (i.e., at the lower temperature) in self-report ratings, but importantly, in facial action units and electrodermal activity, which the present study relies on.As such, we reasoned that the paradigm would sufficiently generate pain-related information at G1 to pass down the social chain.In studies where we have used calibration, we have found modulation of the autonomic response to be variable even during conditioning, which we assume is due to rapid nociceptor fatigue associated with painful stimuli, sometimes even at the upper temperatures used here (e.g., Greffrath et al., 2007).We therefore suspect that calibrating participants may weaken our effect.However, our focus is on relative within-subject changes between Tx and NT stimuli and we found no evidence to suggest that the 45°C temperature was experienced as particularly painful by participants.For ethical reasons, we employ a stringent thermal familiarisation procedure where all participants sample the full range of temperatures (up to 54°C) to take part in the study.All participants tolerated the stimuli well, even the upper temperatures, and none withdrew from the study.

5.
Please explain already at the start of the results, why "trial" was additionally included as a factor.
We have now restructured the manuscript so that the methods section appears before the results.We have added a sentence to the primary analysis section of the methods to clarify why trial was included in the model (see lines 393-397): Trial number was included in the analysis to assess extinction of the nocebo effect.

Line 95 psychological synchrony should be lower-case letters?
Due to reformatting, this text explaining the design of the study has been replaced with the Materials section.7. Out of curiosity and as a second manipulation check, I would also be interested whether the authors ran comparisons of their outcome measures with G1?I see comparisons between G2 and G3, but it would also be interesting how the nocebo effect differs from the experienced treatment effect in G1?
Apologies if we have misinterpreted this question, but we did not measure the nocebo effect in G1 Demonstrators.The reason being that employing a test phase for these participants would have broken the flow of the transmission chain.Instead, G1 participants were comparable to social models in previous studies, but without an awareness of the manipulation.However, it would be interesting to determine whether the magnitude of the conditioned effect among G1 differs from that of modelling.Unfortunately, this is beyond the scope of the present study.Our alternative interpretation of this question is whether the magnitude of the pain response at G1 predicted the nocebo effect at G2 which it did not (see 3.4.4.Interaction between Interpersonal Synchrony and the Pain Response expressed by the previous Demonstrator).We also depict the pain difference between Tx and NT stimuli in Figure 4a, which demonstrates that this difference was understandably much larger among G1 participants compared to G2 and G3.
8. Figure 3c: Is there a typo?In the title it says Nocebo (G2) but the Figure shows Nocebo (G3) and also lines for Nocebo (G2).
It is not a typo, Figure 3c (now 4c) depicts the interaction between the Nocebo effect of G2 and psychological synchrony on the nocebo effect of G3.We have amended the Figure caption to try and make this clearer.
Figure 4: 4a.represents the difference in pain ratings between Tx and NT trials for all Generations (error bars are +/-1SD, not SEM), please note that while the y-axis is labelled 'nocebo hyperalgesia', G1 represents a pain response to direct nociception; 4b.depicts the Generation by Psychological Synchrony interaction on nocebo hyperalgesia; 4c.depicts the interaction between G2 nocebo hyperalgesia (represented as slopes at +/-1SD and the mean) and Psychological Synchrony (G2/G3; xaxis) on G3 nocebo hyperalgesia (y-axis).

Did the authors check whether there was a difference between generations in the matching of Responder and Nonresponder dyads?
All participants who were G1 Demonstrators were Responders.We checked whether the frequency of Responders and Non-Responders changed across Generations 2 and 3 (see Lines 567-570): Of note, however, was that the proportion of Responders/Non-Responders did not differ significantly across G2 and G3 groups ( χ 2 (1,N=65)=0.001,p=.975, Cramer's V=.004) suggesting that there was limited change in receptivity to the manipulation at later stages of the chain (all G1 participants were Responders).
We did not explore the data beyond this.The primary reason being the large number of pre-registered analyses in the manuscript and the fact that transmission of nocebo hyperalgesia was found independent of Responder status at G2 and G3.A secondary reason is that our permutation analysis is not set up to specifically answer this question as it does not account for the trajectory of responses over time.For example, Non-Responders may have rapidly extinguished, but generated enough social information at the beginning of the block to elicit a response in the next participant.We think this would be a great question for future research but would likely require a different design.

Is this high rate of > 50% non-responders common regarding previous own and others' work? It might be helpful to put these numbers into perspective.
We are not aware of many studies that report response rates.However, previous research has suggested that around 36-39% of participants respond to placebo manipulations, while around 56% respond to nocebo manipulations.As such, the numbers in the present study appear comparable.We have added text to the discussion to highlight this point (please see lines 722-725): That similarity in experience may drive nocebo hyperalgesia is demonstrated by the prevalence of 'non-responders' across Generations.Non-responders are common in research concerning nocebo hyperalgesia and placebo analgesia, with some studies estimating as many as ~61 -64% of participants in placebo analgesia and 44% of participants in nocebo hyperalgesia failing to exhibit a significant response 20,71 .In the current study, approximately half of G2 participants (56%) failed to show a nocebo effect in their pain ratings, despite all witnessing a G1-Demonstrator experience 'treatment'-related pain.

Table 1: Please explain the abbreviations DET and LAM in the Table itself.
The table has been updated to say "Determinism (DET -recurrence in the EDA response)" and "Laminarity (LAM -stability in the EDA response)" 12. The fact that the authors preregistered the study only becomes apparent at the end, in the methods, I think this can already be highlighted much sooner Please also include the link to the preregistration directly in the paper.
We have now edited the manuscript so that the methods section appears before the results, meaning this information no longer appears at the end.We have added a link to the pre-registration and have also uploaded a pdf version to OSF.

Could the confusion of the cues have been due to color-blindness (Tritanomaly or Tritanopia)? Did the authors check for these issues/were the cues clearly separable?
Participants were screened to ensure that they had normal or corrected-to-normal vision (we realise that this information was missing from the original version of the manuscript and have now included it -lines 107-108).Some estimates regarding the prevalence rates of tritanopia are ~1/14,000 and ~1/1,000,000 for tritanomaly, meaning that the chances of one of our subjects being affected is low.As such, we don't believe that colour-blindness was a systematic confound in the present study.Robust conditioning effects have been demonstrated in previous studies using identical cues (e.g., Barnes et al., 2021), with clear differences in participant expectancy ratings between conditions, suggesting that the cues can be clearly distinguished by participants.

What role do the authors think that gender (and matching of gender between observer and demonstrator) played in their study and findings?
We pre-registered gender-match as a covariate based on previous research that suggests a potential role of participant gender in social modelling research.We did not observe any systematic effect of this covariate over the primary analyses and believe that other factors, such as interpersonal connection as studied here, may be stronger predictors.Other than 'gender-match', we did not look for differences in self-report gender across our analyses and are cautious of doing so.There are many different gender identities which can fluctuate over time.There are also multiple other features at the level of an individual that could generate stronger or weaker nocebo effects beyond gender which are difficult to control for in this type of research.
15.In the discussion, it might potentially be interesting to talk about bidirectional influences, so whether this chain can also reverse and influence future responses of the earlier chain members (considering the fact that synchrony is not unidirectional)?Just an idea :) This is an interesting idea, which we have added to the Discussion section (see Lines 756-761): Finally, our results were concerned with the unidirectional propagation of pain down the social transmission chain (i.e., moving from G1 to finish G3).Given that synchrony is bi-directional in nature, it may be interesting for future research to investigate whether any information is transmitted from the observer to the demonstrator-potentially influencing future experiences of the demonstrator as well.16.The authors also measured confidence in regard to the cues.Did I miss it are the results of these ratings reported somewhere?
No, you did not miss this information.There was such a large amount of data that we felt this variable added little to the overall conclusions, given that there was evidence of transmission even with the non-responders in the sample.We originally included the confidence measure as a manipulation check in the eventuality that evidence for the transmission of the nocebo effect was not found.In case the data is of interest to readers, we have now included it in the Supplementary Materials where we talk about differences between G2/G3 Responders and Non-Responders in more detail.
There was an effect of Responder on TENS Confidence (F(1, 62)=4.09,p=.022, ηp 2 =.12).Responders had higher confidence (M=5.13, SE=0.31, 95%CI [4.53, 5.75 This paper examined the effects of perceived pain on one's own pain experience, how this is passed down through generations, and modulated by synchrony (both physiological and behavioral) between the demonstrator and observer.I found the paper to be well-written, intriguing, and innovative.One of the most interesting findings was the evident transmission of nocebo pain up to the third generation.Below are several points that I believe could enhance this manuscript and connect it to more extensive work beyond the realm of nocebo effects.
Abstract -Upon my first reading, I must admit that I found the abstract rather puzzling, understanding it more clearly only after reading the entire paper.It begins with, "Witnessing another's pain can intensify pain in the observer.However, current research has concentrated on the observer's intrapersonal experience, overlooking social reciprocity."In discussing reciprocity, I expected the authors to examine the impact of the observer on the demonstrator (effects of G2 on G1) rather than how the effect disseminates to other members.The term "generations" in the abstract was also confusing, as I initially believed it referred to family generations or age variations.
Explaining the complexity of the design was difficult with such a limited wordcount.We have reordered the abstract and removed reference to social reciprocity, which we agree may be confusing in this instance.In reordering, we have also made clear that participants were randomised to 'Generation' to minimise confusion with respect to generations of family members.Based on the suggestion of R3, we have additionally highlighted that the G4-Observer was a confederate.

Introduction:
The section on nocebo is clear and informative.Nonetheless, presenting this solely in the context of nocebo effects neglects a vast realm of research on the evolutionary facets of social contagion and learning, including those specific to pain, stress, or fear, which could be responsible for these "side-effects".For instance, observing another's pain and subsequently experiencing or simulating it is considered fundamental for empathy and for learning to avoid potentially painful or stressful situations.In extreme scenarios (e.g., chronic stress or nocebo effects), this might have adverse impacts on the learner, yet it is generally an adaptive (and widely studied) mechanism.
Here's a recent paper on rodents that also reviews other relevant studies (including human ones):  , C., & Gazzola, V. (2023).Vicarious emotions of fear and pain in rodents.Affective Science, 1-10.
Discussion: Similar to the Introduction, while the paper predominantly focuses on nocebo effects, the impacts should be considered in the broader context of emotion contagion, especially concerning pain.
Many thanks for this suggestion and the recommended papers.The links between social contagion and the nocebo effect as studied in the current manuscript are clear.We have updated the introduction to include reference to this research (see Lines 24-36): Pain is a complex and near universal experience that involves psychological, cognitive, and social components [1][2][3] .Even though we often learn about pain through others, the influence of social factors remains understudied 4,5 .Understanding the impact that social interaction has on pain is important.
Research concerning both human and non-human animals has demonstrated that exposure to stress in others, including those experiencing pain, plays an important role when appraising and navigating the environment 6,7 .This type of social exposure allows us to understand and empathise with the experiences of others 8 but can also signal the presence of noxious stimuli or other environmental risk factors 9 .However, while often adaptive, our propensity to learn from others can have unwanted consequences.
The nocebo effect, a pervasive problem where negative expectancies amplify symptoms such as pain (termed nocebo hyperalgesia), concerns one such instance.While most research has focused on key mechanisms of nocebo hyperalgesia that are non-social, growing evidence indicates that social learning plays a significant role in generating maladaptive pain experiences 10,11 .
We now also introduce social contagion when appraising our results concerning interpersonal synchrony in the Discussion (see Lines 694-701): These results may be considered in light of the broader literature regarding social contagion 78 , the transfer of a behavioural or emotional response between conspecifics.Social contagion is generally considered to be adaptive, for example aiding interpersonal connection and avoiding threats in the environment 7 .However, it can also lead to the development of maladaptive responses such as chronic stress 6 .Interpersonal synchrony may be one factor that facilitates this social transmission 8,79,80 , but with respect to pain has only been demonstrated to confer empathetic and pro-social benefits in human participants [44][45][46] .For example, previous research has demonstrated that neural and physiological synchrony can attenuate the experience of pain 47,48,[50][51][52] .However, our results are the first to demonstrate that physiological synchrony is not purely beneficial and can play a maladaptive role in the subsequent experience of pain.
Given limitations on space, or preference is to focus on nocebo hyperalgesia for the rest of this section, given that this is what we set out specifically to test.

Methods:
Preregistration: I was unable to locate the preregistration on the OSF, making it impossible to compare the analyses.
Apologies, we thought that the pre-registration was uploaded.We have now included a link to the pre-registration in the paper and have also uploaded a pdf version to OSF.

How was the participant number determined?
Given the novelty of the procedure, we had no existing data regarding the expected effect size for the nocebo effect among G3 participants.However, we did have pilot data from a social modelling study in EEG using the same temperature stimuli.This provided an estimate for the effect at G2.As we thought that the nocebo effect may reduce across generations, we were conservative and rounded then doubled the estimated sample size.Our total sample size is outlined in our pre-registration: We found a large social-modelling effect size in a previous study (#52363): Cohen's dz = 1.67 (Cohen's f = 0.835) which would result in a total sample size of 16 in each generation for the interaction term (alpha=0.5;power=90%).As we expect a reduction in the size of the socially modelled nocebo effect by position in the chain, we rounded and doubled this estimate to 30 participants in each condition.As some participants may withdraw from the testing session, we will test until there are 30 participants at G3.
The term Psychological Synchrony might be somewhat confusing, as synchrony often suggests continuous temporal dynamics (as the authors themselves concede).A more accurate term would be more fitting.Perhaps terms like "Expectation of Pain Accuracy" or "Prediction of Pain" would be more appropriate.
We agree with this point regarding the temporal characteristics of the EDA and expectancy outcomes (with expectancy ratings only being continuous at the level of the trial).We employed the term 'synchrony' across the EDA and expectancy measures to broadly denote that we were taking about intercorrelations at the level of the dyad for both outcomes.We really like 'pain prediction' as alternative terminology.However, reflecting on this change to the manuscript in conjunction with our pre-registration, we realised that readers may erroneously believe that we made no specific hypotheses about observer/demonstrator expectancy ratings because we use 'synchrony' to refer to both outcomes (EDA and expectancy) throughout.For consistency between documents, our preference would be to retain the term 'psychological synchrony'.However, we have added the following text to make clear that we are not talking about continuous temporal dynamics (see lines 343-349): We note that this measure of 'Psychological Synchrony' differs from physiological measures of synchrony as the expectancy metric was only continuous at the level of the trial.Instead, in the present study, we use the terms physiological and psychological synchrony to broadly refer to intercorrelations between participants at the level of the individual dyads.As timeseries data at the trial level was not available (i.e., there was only one expectancy measure per trial), correlation coefficients were calculated for each dyad to represent the extent of synchronous responding.
We also make this clear again on lines 535-539: One difference between physiological and psychological measures of synchrony (as presented above) is that the physiological metrics are calculated across the entire time course of the experimental block and are therefore not tied in a phasic manner to the presentation of the thermal stimuli, whereas the Expectancy ratings are specific to the upcoming stimulus on each trial.

Reviewer #3 (Remarks to the Author):
This paper investigates multi-generational social transmission of nocebo hyperalgesia.In particular it investigates if nocebo hyperalgesia can be transmitted across generational chains of participants in the presence of sham treatments and mechanistically if psychological or physiological synchrony between demonstrator and observer contributes to such transmission.The results support both these effects.The study is well-designed and analyses are appropriate.
Interpretations of the results are in line with the data presented.I also appreciated the strong open science practices followed.In sum, this is a very good paper and I have only some minor comments, mostly suggestions for clarification.Thanks to the author for a fascinating read!
The methods were very clear both with regard to procedure and hypotheses.Unfortunately, I didn't find the description in section 2.1 as clear.I think information about what dyad pairs are the target of analysis could come earlier, together with hypothesis.I also first found (both here and in abstract) the information that there are 3 generations but a G4 participant somewhat confusing.Consider revisiting this section.
We have updated the abstract to make clear that the G4 Observer was a confederate early on.Because we have updated the order of the manuscript, the methods now appear before the results, with the design of the study presented in more depth.Section 2.1 is therefore now covered in section 2.5 'Design and Procedure'.Hopefully this new structure makes both design and the analysis plan clearer.
to the manuscript "Your pain increases my pain: Exploring the effect of interpersonal synchrony on the transmission of nocebo hyperalgesia" Note for all Reviewers:

Table 1 :
Please explain the abbreviations DET and LAM in the Table itself.