The influence of interpersonal synchrony and autism on impressions of dyadic interactions: a preregistered study

The influence of interpersonal synchrony and autism on impressions of dyadic interactions: a preregistered study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The influence of interpersonal synchrony and autism on impressions of dyadic interactions: a preregistered study Irene S. Plank, Ralf Tepest, Kai Vogeley, Christine M. Falter-Wagner This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5950403/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 12 Jun, 2025 Read the published version in Molecular Autism → Version 1 posted 12 You are reading this latest preprint version Abstract Background: Humans form almost instantaneous impressions of everyone they encounter. These impressions set the first tone for how they approach and interact with others. Research on impression formation unveiled that impressions formed by autistic and non-autistic people are often less favourable when rating an autistic person. This effect is partly explainable by differences in motion dynamics. Methods: In this preregistered study, we systematically assessed impressions formed by 27 autistic and 36 non-autistic comparison observers when watching videos showing silent, dyadic interactions between either two non-autistic or between an autistic and a non-autistic person. We used an eye tracker to capture their gaze patterns while observing these interactions. Of each dyadic interaction, a video vignette with high and a vignette with low interpersonal synchrony was extracted using Motion Energy Analysis so that we could investigate the effects of interpersonal synchrony and diagnosis, respectively. Results: Interactions were rated less favourably when the observed dyad included an autistic adult. Additionally, interactions showing low interpersonal synchrony were rated less favourably than interactions showing high interpersonal synchrony, regardless of dyad type. Both the effect of interpersonal synchrony and the effect of dyad type on the impressions were independent of the diagnostic status of the observer. Nonetheless, gaze patterns revealed differences between autistic and comparison observers, but were unrelated to interpersonal synchrony and dyad type Limitations: In this study, we investigated limited influences on impression formation, specifically interpersonal synchrony and autism. There are many more potentially interesting aspects of individuals that impact impression formation, such as facial expressiveness, gaze behaviour and linguistic content of conversations, which should be investigated systematically and in a controlled fashion in future research. Conclusions: Both the interaction partners in a dyad and the synchrony of their motion influence the impressions autistic and comparison observers form of the interaction, such that vignettes showing high interpersonal synchrony are perceived as more pleasant. Furthermore, interactions of dyads consisting of one autistic and one non-autistic person are perceived as less pleasant than those of two non-autistic people, which was the case for autistic and comparison observers likewise. impression formation interpersonal synchrony observed interactions autism spectrum disorder behavioural coordination dyadic interactions Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Autism spectrum disorder and interpersonal synchrony Several psychiatric diagnoses are associated with changes in behaviour affecting social interactions ( 1 ). Autism spectrum disorder (ASD) is characterised by differences in social interaction and communication, along with repetitive and restricted behaviours ( 2 ). While symptoms of this neurodevelopmental disorder emerge during childhood, they persist across the lifespan profoundly affecting autistic adults’ lives ( 3 ). A relevant aspect of behaviour that is affected in ASD is the coordination of behaviour with an interaction partner, i.e., interpersonal synchrony (IPS, 4–8). IPS is crucial for facilitating social interactions ( 9 , 10 ). Establishing social connections or rapport with others critically depends on experiencing smooth reciprocity in interactions with them ( 11 – 13 ). For instance, IPS of movement has been shown to be decreased in dyads consisting of two autistic interaction partners as well as in mixed dyads consisting of one autistic and one non-autistic interaction partner ( 5 ). This effect of decreased IPS patterns can be traced back to attenuations in adaptation of the two interaction partners, i.e., leading and following behaviours. These patterns have been used to train machine learning algorithms which detected autistic interactions based on IPS and facial expressions as well as based on speech patterns and IPS of speech ( 7 , 14 ). Despite reports of differences on the production of IPS, imaging studies revealed that processing of IPS appears to be associated with similar neural correlates in autistic and non-autistic adults when they observe an interaction ( 15 ) or when they are part of an interaction ( 16 ). Bierlich and colleagues ( 15 ) collected ratings of the naturalness of the observed interactions, which did not differ between autistic and non-autistic comparison observers, possibly indicating that perception and interpretation of IPS is spared in ASD. Impressions of autistic people Several impression formation studies have shown that autistic people receive less favourable ratings by others than non-autistic people, but underlying mechanisms are unclear. Several studies have asked participants to watch videos of others talking and to report their impressions, including on the observed persons likeability, awkwardness or trustworthiness. Observed autistic people have been consistently rated less favourably ( 17 – 22 ). Less favourable impressions have potentially vast implications for social ( 19 , 20 ) and professional participation of autistic adults ( 23 ). Based on reported differences in IPS associated with ASD, a recent study investigated the effect of IPS on impressions formed about autistic and non-autistic people by non-autistic observers ( 24 ). Specifically, silent video segments of dyadic interactions either between two non-autistic or including one autistic person were presented to non-autistic observers who were asked to rate their impressions of one of the two interaction partners. IPS was evaluated as either leading or following behaviour to assess the differential influence on impressions of just one of the two interaction partners. Results showed that leading results in more favourable impressions of non-autistic people only: impressions of autistic people were not influenced by IPS. This effect suggests that autistic people did not benefit from the positive effect of IPS even when they interacted with non-autistic people. Another possible explanation for less favourable ratings is readability: in a sample of non-autistic observers, Alkhaldi and colleagues ( 25 ) found a relationship between independently rated readability and impressions, such that people regardless of ASD diagnosis were rated more favourable by non-autistic observers the easier they were to read. Impressions formed by autistic observers Reports on differences between impressions formed by autistic and non-autistic participants showed mixed results, with some findings showing more favourable impressions being formed by autistic adults of virtual characters ( 26 ) and people based on brief videos ( 18 ). Other studies with virtual characters who were modelled in their behaviour after autistic and non-autistic people showed more positive ratings by non-autistic people ( 27 ). Focusing on the interaction between diagnostic status of the observer and the observed, DeBrabander and colleagues ( 18 ) asked autistic and non-autistic adults to rate brief videos of autistic and non-autistic adults. They used multiple ratings of character impressions, i.e., likeability, and socialisation intentions, i.e., wanting to hang out with this person. While both autistic and non-autistic observers rated autistic observed people as less favourable on half of the character impressions and three of the four socialisation intentions, there was less difference between the intentions to hang out with a person dependent on their diagnostic state in autistic observers. Additionally, while non-autistic observers reported that they were more likely to sit near non-autistic people, they were equally likely as autistic observers to sit near an autistic person. This data shows that less favourable impressions of autistic people are not limited to non-autistic observers but also extend to autistic observers. Impressions of dyads including an autistic person Since interactive behaviour influences the impression formed of the interaction partners, we expect interactive behaviour to also influence impressions of the interaction itself. In a study by Crompton and colleagues ( 28 ), non-autistic, autistic and mixed dyads completed social interactions and reported their feelings of rapport on five dimensions: ease, enjoyment, success, friendliness and awkwardness. These dimensions with awkwardness being reverse coded were then combined to a single rapport score and the interaction partners’ scores were averaged to compute a rapport score for the interaction. The authors found evidence for higher mean rapport reported in the non-autistic dyads, followed by the autistic dyads, with the mixed dyads reporting the lowest rapport values. Afterwards, they asked an independent sample of autistic and non-autistic participants to report their impression of the dyads’ rapport based on observing some of the conversations via video recordings, with rapport again being comprised of ratings on how easy, enjoyable, friendly, successful and awkward the observers considered the interaction. The authors found evidence for impressions of higher rapport for the non-autistic and the autistic dyads compared to the impressions of the mixed dyads, with comparable results for autistic and comparison observers. This pattern of results could suggest that while impressions formed of autistic people by observers are less favourable, this does not necessarily translate to less perceived rapport of observed interactions including autistic people. Crompton and colleagues ( 28 ) postulate this might be due to rapport impressions being based on stable interpersonal coordination in their study, which might be increased in homogeneous dyads due to more similar interpersonal styles. Purpose of this study While past research has shown that attenuated IPS is an important characteristic of ASD and that impressions can be influenced by IPS, the link between perceived IPS and autism is unclear. Therefore, the goal of this study was to investigate the influence of IPS in interactions of autistic and non-autistic people on impressions formed by a new sample of autistic and non-autistic comparison observers. We presented participants with video segments of interactions of non-autistic or mixed dyads portraying either high or low IPS which was extracted from the videos using Motion Energy Analysis ( 29 ). After each vignette, we asked participants to rate how pleasant they find the just observed interaction. Additionally, we collected eye-tracking data to extract dwell times and assess whether autistic and comparison observers focus on the same aspects of the observed interactions. We hypothesised that ratings would be more positive for interactions of non-autistic dyads than interactions of mixed dyads as well as more positive for interactions with high compared to low IPS. Furthermore, we expected differences between the ratings given by autistic and comparison observers. Specifically, we expected a decreased effect of dyad type on the ratings of autistic compared to comparison observers. Last, we hypothesised a decreased effect of IPS for mixed dyads compared to non-autistic dyads. Concerning the dwell times, we expected differences between the dwell times to areas of interests (AOIs) between non-autistic and mixed dyads, between interactions with high and low IPS as well as between autistic and comparison observers. Method This study was preregistered on the Open Science Framework: https://osf.io/36tuk . The experiment was presented using PsychToolBox 3 ( 30 ) in MATLAB R2023a ( 31 ). Data was preprocessed and analysed using MATLAB R2023a and R 4.2.2 in RStudio 2024.0.4 ( 32 ). All scripts as well as anonymised, aggregated data can be retrieved from GitHub: https://github.com/IreneSophia/PESI . Variance is reported in standard errors throughout the manuscript and credible intervals are reported in square brackets. Participants This study was approved by the Ethics committee of the Medical Faculty of the LMU Munich (Reference number: 23–0268). Participants were recruited through the internal participant database and local network. We aimed to analyse at least 27 participants per group to achieve a power of 90% to detect a medium group effect ( d = 0.45). The power analysis was conducted with PANGEA v0.2 ( 33 ) based on a mixed design with three population-level predictors (observers: autistic or non-autistic comparison group; dyad type: mixed or non-autistic; IPS: high or low) and two group-level predictors (participants and trials). Due to technical difficulties, eye-tracking data was only collected for 45 participants (19 autistic and 26 non-autistic). Thus, we recruited more participants than initially planned (total sample: 70 participants). Inclusion criteria were as follows: no current neurological diagnoses, between 18 and 45 years old, no intellectual disability (i.e., intelligence estimate above 70), normal or corrected-to-normal vision and written informed consent. Additionally, autistic participants had an existing ASD diagnosis, while comparison participants had no psychiatric diagnoses. We excluded data of seven participants from analysis, five autistic and two comparison participants, because only less than two thirds of the ratings were completed. Finally, we included 27 autistic and 36 comparison participants in the behavioural analysis (12 autistic women, 20 comparison women, no difference in gender distribution between diagnostic groups: log BF 10 = -0.695 based on a Bayesian contingency table). There was no difference in age or intelligence estimate between the autistic and the comparison group according to Bayesian t -tests (age: X̄ autistic = 28.66 ± 1.26 years, X̄ comparison = 27.70 ± 0.92 years, log BF 10 = -1.297; intelligence estimate: X̄ autistic = 112.31 ± 3.97, X̄ comparison = 114.19 ± 2.14, log BF 10 = -1.289). Procedure Participants completed the interaction observation task to assess the here presented hypotheses. Additionally, they completed two unrelated tasks, a task modelled after Allenmark and colleagues ( 34 ) and a visual acuity task ( 35 ), which will be presented elsewhere. The order of tasks was counterbalanced across participants. They completed a demographic questionnaire as well as the following standardised questionnaires and tests: the Beck Depression Inventory (BDI-II, 36), the Culture Fair Intelligence Test 20-R (CFT 20-R, 37), the d2 attention test ( 38 ), the questionnaire for Intolerance for Uncertainty (UI, 39), the Ishihara test for colour blindness ( 40 ), the Landolt C for vision acuity ( 41 ), the Ritvo Autism Diagnostic Scale – Revised (RADS-R, 42) as well as the State and Trait Anxiety Inventory (STAI, 43). Summary scores for both groups as well as group comparisons are presented in the supplementary materials. The interaction observation task was presented using a Lenovo Thinkpad P52 laptop with an external monitor (1920x1080 pixels, 533x300mm, 120Hz) and an external number pad. During the interaction observation task, eye movements of both eyes were captured with a LiveTrack Lightning eye tracker (Cambridge RS, sampling rate 500Hz) while participants placed their head in a headrest to ensure a stable viewing distance of 57cm. The eye tracker was calibrated before each task block using nine points and data was only collected if at least one eye had an accuracy below 0.5°. For the analysis of the fixations, we focused on the eye with the better accuracy during calibration for each participant. In total, testing took two to three hours. Stimuli and Interaction Observation Task In the interaction observation task, we used silent videos of dyadic interactions either between two non-autistic adults or between an autistic and a non-autistic adult (mixed dyad) as captured by Georgescu and colleagues ( 5 ). We decided against including a third dyad type, namely autistic dyads, to keep the interaction observation task at an acceptable length for observers. The videos were processed with a Gaussian filter to reduce the visual presentation to the outlines of the interaction partners, both anonymising the dyads and reducing information to the interaction partners’ movements (15, for more information on stimulus creation see 24). Each stimulus captured a 10-second segment of these interactions. The segments were chosen to create a balanced sample of segments portraying high and low IPS of both mixed and non-autistic dyads such that there were no differences in overall motion (see supplementary materials). Specifically, four segments of high and four segments of low IPS were chosen for each of the dyads to model effects of IPS independent of dyadic confounders. Therefore, the task design consisted of one between-subjects factor (autistic and comparison observers) and two within-subject factors, namely IPS and dyad type (see Fig. 1 ). IPS was computed based on Motion Energy Analysis ( 29 ) as applied previously ( 5 , 7 , 24 ) and calculated by detecting the peak of the output of the windowed-lagged cross-correlation of motion quantity as implemented in the rMEA package. In total, 64 segments were presented to each participant in two blocks. Participants were able to take a break between blocks. Each frame of the video segment was presented with an approximate visual angle of 26° by 18°, depending on the exact frame. Before each segment, a fixation cross was presented for 50ms at the centre of the segment presentation. After each video segment, participants were asked to rate how pleasant they imagined the interactions to be (“Wie angenehm stellen Sie sich diese Interaktion vor?”, from “not at all” to “very”; see Fig. 2 ). Participants had 4s to complete their rating, if they did not complete their choice the trial was excluded from the analysis. Ratings were converted into values from 0 to 100. To create frame-by-frame areas of interest (AOIs) dynamically tracking the heads and hands of the interaction partners, we used DeepLabCut™ which allows to estimate pose of user-defined body parts using deep learning ( 44 ). AOIs were circle-shaped and had a diameter of 210 pixels for the heads and 140 pixels for the hands, corresponding to visual angles of 5.85° and 3.91°, respectively. Body AOIs were the same for all segments of one dyad and chosen to capture all possible body positions (see Fig. 2 ). Preprocessing and analyses To improve model fit, we aggregated ratings, resulting in two values per participant for each dyad: the mean rating for segments of this dyad with high IPS and the mean rating for segments with low IPS. Fixations and saccades were detected using the algorithm developed by Nyström and Holmqvist ( 45 ). We classified for each fixation whether its starting point was located on a head, hand or body AOI to compute dwell times for each of the AOIs by summing fixation durations. In case of overlap, fixations were considered for all relevant AOIs, i.e., all fixations on the hand or hand were also counted as body fixations. Dwell times were aggregated by computing the median dwell time per dyad, condition and AOI, resulting in six values per participant for each dyad (each three AOI values for the segments with high and low IPS). To consider overall differences in fixation durations, we computed dwell times as a proportion of total duration of all fixations regardless of location. We used Bayesian linear mixed models as implemented in the brms package ( 46 ), assessing our hypotheses with the brms::hypothesis() function using α = 0.05 for directed and α = 0.025 for undirected hypotheses. We used two models to test our hypotheses, one focusing on relevant predictors for the impression ratings and one for the dwell times. Both models included the population-level predictors diagnostic status of the observer (autistic or comparison), dyad type (mixed or non-autistic) and IPS (high or low). The model assessing dwell times additionally included the population-level predictor AOI (hand, head, body). We included the dyad and participants on the group-level in both models. While we only included intercepts on the group-level for the model assessing ratings, we also included the slopes for IPS, dyad type and AOI as well as all interactions for the participants and the slopes for diagnostic status of the observer, IPS and AOI as well as all interactions for dyads on the group-level. We checked the reliability and computational faithfulness of our models using prior predictive checks, simulation-based calibration and posterior predictive checks as proposed by Schad and colleagues ( 47 ) before commencing inference testing (for details see the supplementary materials). In addition to our hypothesis-guided confirmatory testing, we computed two exploratory analyses. First, we used a Bayesian linear mixed model using a Poisson likelihood to assess possible influences of dyad type, IPS, diagnostic status of the observer and their interactions on the number of saccades produced. Second, we used a Bayesian multiple linear regression with a Gaussian likelihood to investigate whether gaze behaviour can predict mean impressions of how pleasant the interaction was. Specifically, we included the four gaze predictors: mean dwell times on heads, hands or bodies as well as mean number of saccades. All predictors were standardised with a z -score normalisation. For both models, we added random intercepts for the participant and the observed dyad on the group-level. Results Impression ratings The model investigating aggregated impression ratings revealed support for our hypotheses postulating decreased ratings for mixed dyads compared to non-autistic dyads ( estimate = -3.87 [-6.79, -1.01], posterior probability = 0.984) and increased ratings for segments with high IPS as opposed to segments with low IPS ( estimate = 0.83 [0.16, 1.49], posterior probability = 0.978). Specifically, the model predicts a mean difference of 7.742 [0.803, 14.907] between non-autistic and mixed dyads as well as a mean differences of 1.662 [0.039, 3.204] between high and low IPS (see Fig. 3 ). However, there is no support for our hypotheses regarding differences between the ratings of autistic and comparison observers ( estimate = 1.07 [-1.61, 3.78], posterior probability = 0.786) or the interaction of IPS and dyad type ( estimate = -0.06 [-0.71, 0.6], posterior probability = 0.555) or dyad type and diagnostic status of the observer ( estimate = 0.24 [-0.41, 0.89], posterior probability = 0.739). Dwell times The model focusing on dwell times revealed support for our expectation of differences between autistic and comparison observers dependent on the specific AOI ( estimate = -2.20% [-4.11, -0.31], posterior probability = 0.987). Further exploration shows that this is driven by decreased dwell times on the head region for autistic observers ( estimate = 4.97% [0.53, 9.50], posterior probability = 0.986). Specifically, this model predicts a difference of dwell times on the heads of the interaction partners of 9.945% [1.055, 18.995] between autistic and comparison observers (see Fig. 4 ). There is no support for our hypotheses regarding interactions between IPS and AOI ( estimate = 0.13% [-1.55, 1.85], posterior probability = 0.570) or dyad type and AOI ( estimate = -0.19% [-3.19, 2.87], posterior probability = 0.555). Exploratory analyses In general, observers produced on average 19.31 +- 0.7 saccades per 10-second segment. There were no credible differences between any of the task conditions or interactions. However, autistic observers produced fewer saccades than non-autistic observers ( estimate = 0.06 [0, 0.13], posterior probability = 0.98). Specifically, the model predicts that autistic observers produce on average 17.714 [15.936, 19.568] saccades, while comparison observers are predicted to produce 20.151 [18.327, 22.031] saccades on average. Predicting impression ratings with gaze behaviour revealed that more favourable impressions were associated with reduced dwell times on the hands (Scaled dwell times on hands: estimate = -2.27 [-3.83, -0.7]) but no other gaze behaviour (see supplementary materials). Discussion In this study, we investigated influences on impressions of observed interactions, specifically between either two non-autistic interaction partners or between one autistic and one non-autistic interaction partner. We found that the dyad type influenced the impressions formed by others, such that interactions of mixed dyads were rated less favourable, irrespective of the diagnostic status of the observer. Additionally, higher IPS elicited more favourable ratings than lower IPS independent of dyad type in both autistic and comparison observers. Thus, segments of mixed dyads showing high IPS were rated as less pleasant than non-autistic dyads showing comparably high IPS. While impression ratings were comparable between autistic and comparison observers, autistic observers spent less time dwelling on the head regions of the interaction partners. This difference in dwell time suggests that autistic observers used different aspects of the perceived interaction to infer a similar conclusion regarding their impression of how pleasant the interaction is, consistent with previous literature ( 48 ). Differences in impressions of non-autistic and mixed dyads mirror previous results on impressions of autistic and non-autistic people, indicating that autistic people themselves ( 18 , 20 ) as well as interactions of mixed dyads ( 28 ) evoke less favourable impressions. By reducing the information our observers received from interactants on motion dynamics, we were able to show that IPS is key to less favourable impressions of others. Specifically, we omitted audio to rule out observers being influenced by content or diverse speech patterns associated with ASD ( 14 , 49 , 50 ). Additionally, our segments were only “thin slices” of 10s to capture almost instantaneous impressions of the interactions ( 20 , 51 ). Therefore, we captured differences in impressions formed for mixed and non-autistic dyads based on the individual interaction partners’ movements as well as the coordination of these movements. Our results indicate that smooth, dynamic interactions with high IPS lead to more favourable impressions and could support Crompton and colleagues’ ( 28 ) suggestion that stable interpersonal coordination may facilitate higher rapport ratings. This effect is also in line with a study by Miles and colleagues ( 13 ) showing evidence that IPS of two walkers influenced rapport impression regardless of whether walking scenes were presented visually or auditorily. Importantly, previous literature has shown both mixed and autistic dyads producing decreased overall IPS ( 5 ). However, while overall IPS is an important characteristic of an interaction, social interactions show fluctuations of IPS while they unfold ( 52 ). Thus, we decided to use short segments portraying high and low IPS for each dyad to ensure that the observed effect of increased impression ratings could be clearly attributed to high or low IPS. Since non-autistic dyads seem to produce more IPS overall, they could benefit more from the positive effect it has on impressions formed by observers. Future studies could pair methods capturing the development of experiencing how pleasant an interaction is with capturing impressions by observers to assess how close experience is to observation. This study did not reveal any differences in impressions formed by autistic and comparison observers, neither in terms of overall nor in terms of interactive effects. These comparable impressions are in line with studies on observed interactions ( 28 ) and on impression formation of individual people ( 18 ). However, other studies suggest differences in impression formations between autistic and comparison observers ( 26 , 27 ) and intentions based on impressions ( 18 ). In the current study, IPS had a similar effect on impressions by autistic and comparison observers. This similarity of impressions between autistic and comparison observers is in line with recent evidence showing spared neural correlates of IPS processing based on observing segments of the same social interactions as used in this study ( 15 ). Bierlich and colleagues also did not find any differences in naturalness impressions between autistic and comparison observers. This pattern of results could suggest that autistic and comparison observers arrive at comparable interpretations when observing social interactions showing different levels of IPS. Less favourable impressions, both for an individual person and for interactions they are part of, can have negative implications for their everyday life by reducing the willingness of others to interact with them ( 19 , 20 ). These implications are especially important since impressions can have long lasting effects on relationship development ( 53 ). Suggestions on how to navigate and approach these implications exist both on the side of the observer and of the person who is the target of the impression. Movement interventions based on imitation and synchronisation could potentially to some extent give autistic adults the tools to increase IPS ( 54 ). Concerning the observer, diagnostic labels can lead to more positive impressions, possibly indicating the capacity to adjust expectations ( 21 , 55 ). Impressions of autistic individuals by comparison observers might improve if the comparison observers have more knowledge on ASD ( 21 ). In a recent study, autism awareness training did not lead to more favourable impressions of autistic people’s character after a short conversation between autistic and non-autistic participants. However, the autism awareness training increased socialisation intentions with the autistic people in non-autistic interaction partners; thus, mitigating the negative effect of less favourable impressions ( 56 ). This improvement could be due to reframing an interaction as characteristic for autism instead of characteristic of a less pleasant interaction. However, while information on autism is widely distributed on social media with more than a dozen billion views on TikTok alone, a lot of the distributed information is inaccurate or overgeneralised ( 57 ). Therefore, improving the quality of psychoeducation on autism should be a priority of policy-making. Despite comparable impressions, autistic and comparison observers produced slightly different gaze behaviour while watching the interactions. Dwell times for the head of interactants were reduced in autistic compared to comparison observers. Reduced dwell times or speed to fixation in response to social stimuli like heads, faces or eyes in ASD has been shown throughout the literature ( 58 , 59 ). Additionally, autistic observers produced fewer saccades compared to comparison observers. Although these results suggest differences in attention to social stimuli, neither dwell times to heads nor number of saccades predicted impression ratings, suggesting that these differences in gaze behaviour did not influence the formation of impressions. Limitations The exact impact of the effects of IPS and dyad type on impressions in real-life cannot be estimated on the basis of this study. Real-life impression formation depends on a wide variety of information, including facial expressiveness, gaze behaviour and conversational content. In this study, we explicitly focused on manipulating two possible influences separately: IPS and dyad type. Segments showing high IPS were estimated to be rated 1.662 points more favourable on our scale from 0 to 100, with 95% of the estimated values between 0.039 and 3.204 points. This effect is potentially negligible when observing interactions in real-life. However, this effect only captures IPS of movement, while in reality we observe multiple channels in which interaction partners can be in sync or not. Small effects on each channel might add up to significantly less favourable impressions. The effect of dyad type was estimated between 0.803 and 14.907 points with an average of 7.742 points difference between mixed dyads consisting of one autistic and one non-autistic person and dyads consisting of two non-autistic people. This effect is almost five times as large as the effect of IPS. Yet, it is still unclear what differences between these dyad types cause the influence on impressions that is independent of IPS. In this study, observers only saw 10s segments of people’s outlines, so they received no auditory information regarding the interaction and only limited information on appearances. Future studies should dissect the exact movement patterns to further carve out which behavioural differences cause less favourable impressions for these mixed dyads. Furthermore, it is paramount to extend the research to interactions of two autistic people to further carve out how autism interacts with impressions. Last, the sample analysed here consisted of highly intelligent adults; thus, generalisability is limited to a subset of adults with ASD. Conclusions This study illuminates predictors of impression formation of dyadic interactions, particularly those involving autistic adults. Results highlight the importance of motion dynamics in shaping these impressions, with smoother, more coordinated interactions as measured by IPS being perceived as more favourable, specifically more pleasant. In addition to this effect of IPS on impressions, interactions between autistic and non-autistic adults were generally rated as less pleasant than interactions between two non-autistic adults. While both autistic and comparison observers formed similar impressions, differences in their gaze patterns could suggest distinct strategies for processing social information. These findings underscore the importance of IPS in fostering positive social interactions. Future research should examinate other predictors of less favourable impressions of autistic people. Abbreviations AOI Areas of Interest ASD Autism spectrum disorder BDI Beck Depression Inventory CFT Culture Fair Intelligence Test COMP comparison group of non-autistic observers IPS interpersonal synchrony RADS-R Ritvo Autism Diagnostic Scale – Revised STAI State and Trait Anxiety Inventory UI Intolerance for Uncertainty Declarations Ethics approval and consent to participate This study was approved by the Ethics committee of the LMU Munich (Reference number: 23–0268) and conducted in concordance with the Declaration of Helsinki. All participants were informed of the study procedure, study aim, associated risks and benefits as well as data processing and data protection, before they signed a written consent form. Consent for publication We received consent for publication from the two interaction partners depicted in Fig. 2. Competing interests The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as competing interests. Funding This project was funded by the “Verein zur Förderung von Wissenschaft und Forschung an der Medizinischen Fakultät der LMU München e.V.” and by the “Stiftungen zugunsten der Medizinischen Fakultät – Cluster 2” of the LMU Klinikum. KV received funding from the Federal German Ministry of Education and Research (grant number 01GP2215). CFW was supported by German Research Council Funding (Deutsche Forschungsgemeinschaft, DFG, FA 876/3–1, FA 876/5–1). Author Contribution I.S.P., K.V. and C.F.W. conceptualised the research project. K.V. provided the stimuli used in the experiment, while I.S.P. and C.F.W. provided resources and acquired the funding. R.T. filtered and anonymised the videos. I.S.P. wrote the preregistration on which C.F.W. provided feedback. I.S.P. designed and programmed the experimental setup, preprocessed and analysed the data as well as wrote the first draft of the manuscript. All authors were involved in the revision of the manuscript and have approved the submitted version. Acknowledgement We would like to thank all the people who have assisted us in this project. First, A. L. Georgescu for sharing the stimulus material and L. S. Traiger for extracting motion quantity with Motion Energy Analysis. Second, A. M. Bierlich for providing constructive feedback during the conceptualisation and analysis of the study. Third, S. Coenen for organising and conducting the data collection. Last but not least, we want to thank S. Nothaft and Y. W. Foo for assisting during data collection as well as training DeepLabCutTM for the extraction of Areas of Interest for the eye tracking analysis. Data Availability The dataset supporting the conclusions of this article is available in the GitHub repository, https://doi.org/10.5281/zenodo.14793465 or https://github.com/IreneSophia/PESI. References American Psychiatric Association. The Diagnostic and Statistical Manual of Mental Disorders [Internet]. fifth ed., text rev. 2022 [cited 2023 Jul 28]. Available from: https://doi.org/10.1176/appi.books.9780890425787 World Health Organization. International Classification of Diseases, Eleventh Revision (ICD-11) [Internet]. 2019. Available from: https://icd.who.int/browse11 Thapar A, Cooper M, Rutter M. Neurodevelopmental disorders. Lancet Psychiatry. 2017;4(4):339–46. Bierlich AM, Scheel N, Koehler JC, Bloch C, Plank IS, Falter-Wagner C. Attenuated behavioral interpersonal synchrony in autistic adults is not explained by perception [Internet]. OSF; 2024 [cited 2024 Dec 14]. Available from: https://osf.io/j4gdx Georgescu AL, Koeroglu S, de Hamilton C, Vogeley AF, Falter-Wagner K, Tschacher CM. Reduced nonverbal interpersonal synchrony in autism spectrum disorder independent of partner diagnosis: a motion energy study. Mol Autism. 2020;11(1):1–14. Glass D, Yuill N. Social motor synchrony in autism spectrum conditions: A systematic review. Autism. 2024;28(7):1638–53. Koehler JC, Dong MS, Bierlich AM, Fischer S, Späth J, Plank IS, et al. Machine learning classification of autism spectrum disorder based on reciprocity in naturalistic social interactions. Transl Psychiatry. 2024;14(1):1–9. McNaughton KA, Redcay E. Interpersonal Synchrony in Autism. Curr Psychiatry Rep [Internet]. 2020;22(3). Available from: https://pubmed.ncbi.nlm.nih.gov/32025922/ Bowsher-Murray C, Gerson S, von dem Hagen E, Jones CRG. The Components of Interpersonal Synchrony in the Typical Population and in Autism: A Conceptual Analysis. Front Psychol [Internet]. 2022 [cited 2023 Jul 5];13. Available from: https://www.frontiersin.org/articles/ 10.3389/fpsyg.2022.897015 Hoehl S, Fairhurst M, Schirmer A. Interactional synchrony: Signals, mechanisms and benefits. Soc Cogn Affect Neurosci. 2021;16(1–2):5–18. Hove MJ, Risen JL. It’s All in the Timing: Interpersonal Synchrony Increases Affiliation. Soc Cogn. 2009;27(6):949–60. Marsh KL, Richardson MJ, Schmidt RC. Social Connection Through Joint Action and Interpersonal Coordination. Top Cogn Sci. 2009;1(2):320–39. Miles LK, Nind LK, Macrae CN. The rhythm of rapport: Interpersonal synchrony and social perception. J Exp Soc Psychol. 2009;45:585–9. Plank IS, Koehler JC, Nelson A, Koutsouleris N, Falter-Wagner C. Automated extraction of speech and turn-taking parameters in autism allows for diagnostic classification using a multivariable prediction model [Internet]. PsyArXiv; 2023 [cited 2023 Jul 24]. Available from: https://psyarxiv.com/upz57/ Bierlich AM, Scheel NT, Traiger LS, Keeser D, Tepest R, Georgescu AL, et al. Neural Mechanisms of Social Interaction Perception: Observing Interpersonal Synchrony Modulates Action Observation Network Activation and Is Spared in Autism. Hum Brain Mapp. 2024;45(15):e70052. Bierlich AM, Plank IS, Scheel NT, Keeser D, Falter-Wagner CM. Neural processing of social reciprocity in autism [Internet]. medRxiv; 2024 [cited 2024 Dec 14]. p. 2024.01.31.24302051. Available from: https://www.medrxiv.org/content/ 10.1101/2024.01.31.24302051v1 Alkhaldi RS, Sheppard E, Burdett E, Mitchell P. Do Neurotypical People Like or Dislike Autistic People? Autism Adulthood. 2021;3(3):275–9. DeBrabander KM, Morrison KE, Jones DR, Faso DJ, Chmielewski M, Sasson NJ. Do First Impressions of Autistic Adults Differ Between Autistic and Nonautistic Observers? Autism Adulthood. 2019;1(4):250–7. Morrison KE, DeBrabander KM, Jones DR, Faso DJ, Ackerman RA, Sasson NJ. Outcomes of real-world social interaction for autistic adults paired with autistic compared to typically developing partners. Autism. 2020;24(5):1067–80. Sasson NJ, Faso DJ, Nugent J, Lovell S, Kennedy DP, Grossman RB. Neurotypical Peers are Less Willing to Interact with Those with Autism based on Thin Slice Judgments. Sci Rep. 2017;7(October 2016):1–10. Sasson NJ, Morrison KE. First impressions of adults with autism improve with diagnostic disclosure and increased autism knowledge of peers. Autism. 2019;23(1):50–9. Cola ML, Plate S, Yankowitz L, Yankowitz L, Petrulla V, Bateman L, et al. Sex differences in the first impressions made by girls and boys with autism. Mol Autism. 2020;11(1):1–12. Whelpley CE, May CP. Seeing is Disliking: Evidence of Bias Against Individuals with Autism Spectrum Disorder in Traditional Job Interviews. J Autism Dev Disord. 2023;53(4):1363–74. Plank IS, Traiger LS, Nelson AM, Koehler JC, Lang SF, Tepest R, et al. The role of interpersonal synchrony in forming impressions of autistic and non-autistic adults. Sci Rep. 2023;13(1):15306. Alkhaldi RS, Sheppard E, Mitchell P. Is There a Link Between Autistic People Being Perceived Unfavorably and Having a Mind That Is Difficult to Read? J Autism Dev Disord. 2019;49(10):3973–82. Schwartz C, Dratsch T, Vogeley K, Bente G. Brief Report: Impression Formation in High-Functioning Autism: Role of Nonverbal Behavior and Stereotype Activating Information. J Autism Dev Disord. 2014;44(7):1759–65. Bloch C, Tepest R, Koeroglu S, Feikes K, Jording M, Vogeley K et al. Interacting with autistic virtual characters: intrapersonal synchrony of nonverbal behavior affects participants’ perception. Eur Arch Psychiatry Clin Neurosci [Internet]. 2024 Jan 25 [cited 2024 Apr 17]; Available from: https://doi.org/10.1007/s00406-023-01750-3 Crompton CJ, Sharp M, Axbey H, Fletcher-Watson S, Flynn EG, Ropar D. Neurotype-Matching, but Not Being Autistic, Influences Self and Observer Ratings of Interpersonal Rapport. Front Psychol. 2020;11(October):1–12. Ramseyer FT. Motion energy analysis (MEA): A primer on the assessment of motion from video. J Couns Psychol. 2020;67(4):536–49. Kleiner M, Brainard D, Pelli D. What’s new in Psychtoolbox-3? 2007;14. The Mathworks Inc. MATLAB [Computer software]. 2022. RStudio Team. RStudio: Integrated Development Environment for R [Internet]. Boston, MA: RStudio, PBC. 2020. Available from: http://www.rstudio.com/ Westfall J. PANGEA: Power ANalysis for General Anova designs. University of Texas at Austin; 2016. Allenmark F, Shi Z, Pistorius RL, Theisinger LA, Koutsouleris N, Falkai P, et al. Acquisition and Use of ‘Priors’ in Autism: Typical in Deciding Where to Look, Atypical in Deciding What Is There. J Autism Dev Disord. 2021;51(10):3744–58. Falter CM, Elliott MA, Bailey AJ. Enhanced visual temporal resolution in autism spectrum disorders. PLoS ONE. 2012;7(3):1–6. Kühner C, Bürger C, Keller F, Hautzinger M. Reliabilität und Validität des revidierten Beck-Depressionsinventars (BDI-II). Nervenarzt. 2007;78(6):651–6. Weiß R. CFT 20-R: Grundintelligenztest Skala 2-Revision. Hogrefe; 2006. Bates ME, Lemay EP. The d2 test of attention: Construct validity and extensions in scoring techniques. J Int Neuropsychol Soc. 2004;10(3):392–400. Gerlach AL, Andor T, Patzelt J. Die Bedeutung von Unsicherheitsintoleranz für die Generalisierte Angststörung Modellüberlegungen und Entwicklung einer deutschen Version der Unsicherheitsintoleranz-Skala. Z Für Klin Psychol Psychother. 2008;37(3). Clark JH. The Ishihara Test for Color Blindness. Am J Physiol Opt. 1924;5:269–76. Wesemann W, Schiefer U, Bach M. New DIN norms for determination of visual acuity. Ophthalmologe. 2010;107(9):821–6. Rausch J, Fangmeier T, Falter-Wagner CM, Ackermann H, Espelöer J, Hölzel LP et al. A novel screening instrument for the assessment of autism in German language: validation of the German version of the RAADS-R, the RADS-R. Eur Arch Psychiatry Clin Neurosci [Internet]. 2024 Oct 27 [cited 2024 Nov 19]; Available from: https://doi.org/10.1007/s00406-024-01894-w Laux L, Glanzmann P, Schaffner CD. Das state-trait-angstinventar [The state-trait anxiety inventory]. Göttingen: Hogrefe; 1981. Mathis A, Mamidanna P, Cury KM, Abe T, Murthy VN, Mathis MW, et al. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nat Neurosci. 2018;21(9):1281–9. Nyström M, Holmqvist K. An adaptive algorithm for fixation, saccade, and glissade detection in eyetracking data. Behav Res Methods. 2010;42(1):188–204. Bürkner PC. brms: An R package for Bayesian multilevel models using Stan. J Stat Softw. 2017;80(1). Schad DJ, Betancourt M, Vasishth S. Toward a principled Bayesian workflow in cognitive science [Internet]. arXiv; 2020 [cited 2023 Sep 26]. Available from: http://arxiv.org/abs/1904.12765 Bloch C, Viswanathan S, Tepest R, Jording M, Falter-Wagner CM, Vogeley K. Differentiated, rather than shared, strategies for time-coordinated action in social and non-social domains in autistic individuals. Cortex. 2023;166:207–32. Fusaroli R, Grossman R, Bilenberg N, Cantio C, Jepsen JRM, Weed E. Toward a cumulative science of vocal markers of autism: A cross-linguistic meta-analysis-based investigation of acoustic markers in American and Danish autistic children. Autism Res. 2022;15(4):653–64. Ochi K, Ono N, Owada K, Kojima M, Kuroda M, Sagayama S, et al. Quantification of speech and synchrony in the conversation of adults with autism spectrum disorder. PLoS ONE. 2019;14(12):1–22. Willis J, Todorov A. Making Up Your Mind After a 100-Ms Exposure to a Face. 2015;17(7):592–8. Mayo O, Gordon I. In and out of synchrony—Behavioral and physiological dynamics of dyadic interpersonal coordination. Psychophysiology. 2020;57(6):1–15. Human LJ, Sandstrom GM, Biesanz JC, Dunn EW. Accurate First Impressions Leave a Lasting Impression: The Long-Term Effects of Distinctive Self-Other Agreement on Relationship Development. Soc Psychol Personal Sci. 2013;4(4):395–402. Koehne S, Behrends A, Fairhurst MT, Dziobek I. Fostering Social Cognition through an Imitation- and Synchronization-Based Dance/Movement Intervention in Adults with Autism Spectrum Disorder: A Controlled Proof-of-Concept Study. Psychother Psychosom. 2015;85(1):27–35. Brosnan M, Mills E. The effect of diagnostic labels on the affective responses of college students towards peers with ‘Asperger’s Syndrome’ and ‘Autism Spectrum Disorder’. Autism. 2016;20(4):388–94. Jones DR, Morrison KE, DeBrabander KM, Ackerman RA, Pinkham AE, Sasson NJ. Greater Social Interest Between Autistic and Non-autistic Conversation Partners Following Autism Acceptance Training for Non-autistic People. Front Psychol [Internet]. 2021 [cited 2023 Jul 5];12. Available from: https://www.frontiersin.org/articles/ 10.3389/fpsyg.2021.739147 Aragon-Guevara D, Castle G, Sheridan E, Vivanti G. The Reach and Accuracy of Information on Autism on TikTok. J Autism Dev Disord [Internet]. 2023 Aug 6 [cited 2024 Dec 14]; Available from: https://doi.org/10.1007/s10803-023-06084-6 Chita-Tegmark M. Social attention in ASD: A review and meta-analysis of eye-tracking studies. Res Dev Disabil. 2016;48:79–93. Setien-Ramos I, Lugo-Marín J, Gisbert-Gustemps L, Díez-Villoria E, Magán-Maganto M, Canal-Bedia R, et al. Eye-tracking studies in adults with autism spectrum disorder: A systematic review and meta-analysis. J Autism Dev Disord. 2023;53(6):2430–43. Additional Declarations No competing interests reported. Supplementary Files analysisPESImerged.pdf Cite Share Download PDF Status: Published Journal Publication published 12 Jun, 2025 Read the published version in Molecular Autism → Version 1 posted Editorial decision: Revision requested 07 Apr, 2025 Reviews received at journal 29 Mar, 2025 Reviews received at journal 11 Mar, 2025 Reviews received at journal 16 Feb, 2025 Reviewers agreed at journal 13 Feb, 2025 Reviewers agreed at journal 12 Feb, 2025 Reviewers agreed at journal 11 Feb, 2025 Reviewers agreed at journal 11 Feb, 2025 Reviewers invited by journal 10 Feb, 2025 Editor assigned by journal 10 Feb, 2025 Submission checks completed at journal 04 Feb, 2025 First submitted to journal 03 Feb, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5950403","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":427443030,"identity":"c885da18-3fac-495e-a3a3-e7b9ae770268","order_by":0,"name":"Irene S. Plank","email":"data:image/png;base64,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","orcid":"","institution":"LMU University Hospital, LMU Munich","correspondingAuthor":true,"prefix":"","firstName":"Irene","middleName":"S.","lastName":"Plank","suffix":""},{"id":427443031,"identity":"620260b3-3f47-45e1-8f16-7e4c95e57270","order_by":1,"name":"Ralf Tepest","email":"","orcid":"","institution":"University of Cologne","correspondingAuthor":false,"prefix":"","firstName":"Ralf","middleName":"","lastName":"Tepest","suffix":""},{"id":427443032,"identity":"ca00fe9e-b6a9-4a22-afab-8c385d26adf3","order_by":2,"name":"Kai Vogeley","email":"","orcid":"","institution":"University of Cologne","correspondingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"Vogeley","suffix":""},{"id":427443033,"identity":"ade166d0-b528-4f37-ad59-9707fd46569b","order_by":3,"name":"Christine M. Falter-Wagner","email":"","orcid":"","institution":"LMU University Hospital, LMU Munich","correspondingAuthor":false,"prefix":"","firstName":"Christine","middleName":"M.","lastName":"Falter-Wagner","suffix":""}],"badges":[],"createdAt":"2025-02-03 10:53:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5950403/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5950403/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13229-025-00668-y","type":"published","date":"2025-06-12T15:57:54+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":78656909,"identity":"dbf86b02-f766-47f8-b0b2-72846838f21c","added_by":"auto","created_at":"2025-03-17 09:37:42","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":17383,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eSchematics of the mixed 2x2x2 design. Autistic and non-autistic comparison (COMP) observers watched segments of four types: mixed dyad producing high IPS, mixed dyad producing low IPS, non-autistic dyad producing high IPS and non-autistic dyad producing low IPS.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure1design.png","url":"https://assets-eu.researchsquare.com/files/rs-5950403/v1/96e20b471220dac7495c1a3a.png"},{"id":78656323,"identity":"3b492224-63ed-497d-801e-3aab7655cc12","added_by":"auto","created_at":"2025-03-17 09:29:42","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":41148,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eA still frame from a video segment, overlayed with the body (pink), hands (green) and head (blue) areas of interest (AOIs), followed by the impression rating. This segment was created with lab members to maintain the privacy of the interaction partners. It was processed with the same filter as the original stimuli.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure2taskoverview.png","url":"https://assets-eu.researchsquare.com/files/rs-5950403/v1/546782fff655d4cc224b5dec.png"},{"id":78656328,"identity":"7c2c8557-b140-467a-a8e6-e6011795a0f0","added_by":"auto","created_at":"2025-03-17 09:29:42","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":33130,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eRaincloud plot visualising the mean impression ratings for each participant for each dyad type and IPS. The graph shows reduced ratings for mixed dyads as well as segments with low IPS, independent of the two groups of observers.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure3ratings.png","url":"https://assets-eu.researchsquare.com/files/rs-5950403/v1/6089429e05c6d9b9e96b2ce3.png"},{"id":78656326,"identity":"345d5a87-dda3-489f-8a5c-aae50dd552ef","added_by":"auto","created_at":"2025-03-17 09:29:42","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":67529,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eRaincloud plots visualising the mean dwell times of each participant. There is a clear pattern of differences in dwell times between autistic and non-autistic comparison observers, regardless of the dyad type and IPS. Specifically, autistic observers seem to dwell less on the head regions of the observed interaction partners.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure4dwelltimes.png","url":"https://assets-eu.researchsquare.com/files/rs-5950403/v1/953ac4902fa47bdbea927002.png"},{"id":84726540,"identity":"f774bc40-4c22-4e9a-be63-7ccf7b10d621","added_by":"auto","created_at":"2025-06-16 16:06:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":862555,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5950403/v1/3332a04c-6689-4489-b03b-76728d516b24.pdf"},{"id":78656344,"identity":"ce21e17d-7c73-4731-9fff-1e20e45a5da0","added_by":"auto","created_at":"2025-03-17 09:29:43","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":15097228,"visible":true,"origin":"","legend":"","description":"","filename":"analysisPESImerged.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5950403/v1/59e19d824c73efe26c9a50cf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"The influence of interpersonal synchrony and autism on impressions of dyadic interactions: a preregistered study","fulltext":[{"header":"Background","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eAutism spectrum disorder and interpersonal synchrony\u003c/h2\u003e \u003cp\u003eSeveral psychiatric diagnoses are associated with changes in behaviour affecting social interactions (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Autism spectrum disorder (ASD) is characterised by differences in social interaction and communication, along with repetitive and restricted behaviours (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). While symptoms of this neurodevelopmental disorder emerge during childhood, they persist across the lifespan profoundly affecting autistic adults\u0026rsquo; lives (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). A relevant aspect of behaviour that is affected in ASD is the coordination of behaviour with an interaction partner, i.e., interpersonal synchrony (IPS, 4\u0026ndash;8). IPS is crucial for facilitating social interactions (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Establishing social connections or rapport with others critically depends on experiencing smooth reciprocity in interactions with them (\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). For instance, IPS of movement has been shown to be decreased in dyads consisting of two autistic interaction partners as well as in mixed dyads consisting of one autistic and one non-autistic interaction partner (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). This effect of decreased IPS patterns can be traced back to attenuations in adaptation of the two interaction partners, i.e., leading and following behaviours. These patterns have been used to train machine learning algorithms which detected autistic interactions based on IPS and facial expressions as well as based on speech patterns and IPS of speech (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Despite reports of differences on the production of IPS, imaging studies revealed that processing of IPS appears to be associated with similar neural correlates in autistic and non-autistic adults when they observe an interaction (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) or when they are part of an interaction (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Bierlich and colleagues (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) collected ratings of the naturalness of the observed interactions, which did not differ between autistic and non-autistic comparison observers, possibly indicating that perception and interpretation of IPS is spared in ASD.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eImpressions of autistic people\u003c/h2\u003e \u003cp\u003eSeveral impression formation studies have shown that autistic people receive less favourable ratings by others than non-autistic people, but underlying mechanisms are unclear. Several studies have asked participants to watch videos of others talking and to report their impressions, including on the observed persons likeability, awkwardness or trustworthiness. Observed autistic people have been consistently rated less favourably (\u003cspan additionalcitationids=\"CR18 CR19 CR20 CR21\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). Less favourable impressions have potentially vast implications for social (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e) and professional participation of autistic adults (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). Based on reported differences in IPS associated with ASD, a recent study investigated the effect of IPS on impressions formed about autistic and non-autistic people by non-autistic observers (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). Specifically, silent video segments of dyadic interactions either between two non-autistic or including one autistic person were presented to non-autistic observers who were asked to rate their impressions of one of the two interaction partners. IPS was evaluated as either leading or following behaviour to assess the differential influence on impressions of just one of the two interaction partners. Results showed that leading results in more favourable impressions of non-autistic people only: impressions of autistic people were not influenced by IPS. This effect suggests that autistic people did not benefit from the positive effect of IPS even when they interacted with non-autistic people. Another possible explanation for less favourable ratings is readability: in a sample of non-autistic observers, Alkhaldi and colleagues (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e) found a relationship between independently rated readability and impressions, such that people regardless of ASD diagnosis were rated more favourable by non-autistic observers the easier they were to read.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eImpressions formed by autistic observers\u003c/h3\u003e\n\u003cp\u003eReports on differences between impressions formed by autistic and non-autistic participants showed mixed results, with some findings showing more favourable impressions being formed by autistic adults of virtual characters (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e) and people based on brief videos (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Other studies with virtual characters who were modelled in their behaviour after autistic and non-autistic people showed more positive ratings by non-autistic people (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). Focusing on the interaction between diagnostic status of the observer and the observed, DeBrabander and colleagues (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e) asked autistic and non-autistic adults to rate brief videos of autistic and non-autistic adults. They used multiple ratings of character impressions, i.e., likeability, and socialisation intentions, i.e., wanting to hang out with this person. While both autistic and non-autistic observers rated autistic observed people as less favourable on half of the character impressions and three of the four socialisation intentions, there was less difference between the intentions to hang out with a person dependent on their diagnostic state in autistic observers. Additionally, while non-autistic observers reported that they were more likely to sit near non-autistic people, they were equally likely as autistic observers to sit near an autistic person. This data shows that less favourable impressions of autistic people are not limited to non-autistic observers but also extend to autistic observers.\u003c/p\u003e\n\u003ch3\u003eImpressions of dyads including an autistic person\u003c/h3\u003e\n\u003cp\u003eSince interactive behaviour influences the impression formed of the interaction partners, we expect interactive behaviour to also influence impressions of the interaction itself. In a study by Crompton and colleagues (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e), non-autistic, autistic and mixed dyads completed social interactions and reported their feelings of rapport on five dimensions: ease, enjoyment, success, friendliness and awkwardness. These dimensions with awkwardness being reverse coded were then combined to a single rapport score and the interaction partners\u0026rsquo; scores were averaged to compute a rapport score for the interaction. The authors found evidence for higher mean rapport reported in the non-autistic dyads, followed by the autistic dyads, with the mixed dyads reporting the lowest rapport values. Afterwards, they asked an independent sample of autistic and non-autistic participants to report their impression of the dyads\u0026rsquo; rapport based on observing some of the conversations via video recordings, with rapport again being comprised of ratings on how easy, enjoyable, friendly, successful and awkward the observers considered the interaction. The authors found evidence for impressions of higher rapport for the non-autistic and the autistic dyads compared to the impressions of the mixed dyads, with comparable results for autistic and comparison observers. This pattern of results could suggest that while impressions formed of autistic people by observers are less favourable, this does not necessarily translate to less perceived rapport of observed interactions including autistic people. Crompton and colleagues (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e) postulate this might be due to rapport impressions being based on stable interpersonal coordination in their study, which might be increased in homogeneous dyads due to more similar interpersonal styles.\u003c/p\u003e\n\u003ch3\u003ePurpose of this study\u003c/h3\u003e\n\u003cp\u003eWhile past research has shown that attenuated IPS is an important characteristic of ASD and that impressions can be influenced by IPS, the link between perceived IPS and autism is unclear. Therefore, the goal of this study was to investigate the influence of IPS in interactions of autistic and non-autistic people on impressions formed by a new sample of autistic and non-autistic comparison observers. We presented participants with video segments of interactions of non-autistic or mixed dyads portraying either high or low IPS which was extracted from the videos using Motion Energy Analysis (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). After each vignette, we asked participants to rate how pleasant they find the just observed interaction. Additionally, we collected eye-tracking data to extract dwell times and assess whether autistic and comparison observers focus on the same aspects of the observed interactions. We hypothesised that ratings would be more positive for interactions of non-autistic dyads than interactions of mixed dyads as well as more positive for interactions with high compared to low IPS. Furthermore, we expected differences between the ratings given by autistic and comparison observers. Specifically, we expected a decreased effect of dyad type on the ratings of autistic compared to comparison observers. Last, we hypothesised a decreased effect of IPS for mixed dyads compared to non-autistic dyads. Concerning the dwell times, we expected differences between the dwell times to areas of interests (AOIs) between non-autistic and mixed dyads, between interactions with high and low IPS as well as between autistic and comparison observers.\u003c/p\u003e"},{"header":"Method","content":"\u003cp\u003eThis study was preregistered on the Open Science Framework: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osf.io/36tuk\u003c/span\u003e\u003cspan address=\"https://osf.io/36tuk\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The experiment was presented using PsychToolBox 3 (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e) in MATLAB R2023a (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Data was preprocessed and analysed using MATLAB R2023a and R 4.2.2 in RStudio 2024.0.4 (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). All scripts as well as anonymised, aggregated data can be retrieved from GitHub: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/IreneSophia/PESI\u003c/span\u003e\u003cspan address=\"https://github.com/IreneSophia/PESI\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Variance is reported in standard errors throughout the manuscript and credible intervals are reported in square brackets.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003e This study was approved by the Ethics committee of the Medical Faculty of the LMU Munich (Reference number: 23\u0026ndash;0268). Participants were recruited through the internal participant database and local network. We aimed to analyse at least 27 participants per group to achieve a power of 90% to detect a medium group effect (\u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.45). The power analysis was conducted with PANGEA v0.2 (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e) based on a mixed design with three population-level predictors (observers: autistic or non-autistic comparison group; dyad type: mixed or non-autistic; IPS: high or low) and two group-level predictors (participants and trials). Due to technical difficulties, eye-tracking data was only collected for 45 participants (19 autistic and 26 non-autistic). Thus, we recruited more participants than initially planned (total sample: 70 participants). Inclusion criteria were as follows: no current neurological diagnoses, between 18 and 45 years old, no intellectual disability (i.e., intelligence estimate above 70), normal or corrected-to-normal vision and written informed consent. Additionally, autistic participants had an existing ASD diagnosis, while comparison participants had no psychiatric diagnoses. We excluded data of seven participants from analysis, five autistic and two comparison participants, because only less than two thirds of the ratings were completed. Finally, we included 27 autistic and 36 comparison participants in the behavioural analysis (12 autistic women, 20 comparison women, no difference in gender distribution between diagnostic groups: log\u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e = -0.695 based on a Bayesian contingency table). There was no difference in age or intelligence estimate between the autistic and the comparison group according to Bayesian \u003cem\u003et\u003c/em\u003e-tests (age: X̄\u003csub\u003eautistic\u003c/sub\u003e = 28.66\u0026thinsp;\u0026plusmn;\u0026thinsp;1.26 years, X̄\u003csub\u003ecomparison\u003c/sub\u003e = 27.70\u0026thinsp;\u0026plusmn;\u0026thinsp;0.92 years, log\u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e = -1.297; intelligence estimate: X̄\u003csub\u003eautistic\u003c/sub\u003e = 112.31\u0026thinsp;\u0026plusmn;\u0026thinsp;3.97, X̄\u003csub\u003ecomparison\u003c/sub\u003e = 114.19\u0026thinsp;\u0026plusmn;\u0026thinsp;2.14, log\u003cem\u003eBF\u003c/em\u003e\u003csub\u003e\u003cem\u003e10\u003c/em\u003e\u003c/sub\u003e = -1.289).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eProcedure\u003c/h3\u003e\n\u003cp\u003eParticipants completed the interaction observation task to assess the here presented hypotheses. Additionally, they completed two unrelated tasks, a task modelled after Allenmark and colleagues (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e) and a visual acuity task (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e), which will be presented elsewhere. The order of tasks was counterbalanced across participants. They completed a demographic questionnaire as well as the following standardised questionnaires and tests: the Beck Depression Inventory (BDI-II, 36), the Culture Fair Intelligence Test 20-R (CFT 20-R, 37), the d2 attention test (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e), the questionnaire for Intolerance for Uncertainty (UI, 39), the Ishihara test for colour blindness (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e), the Landolt C for vision acuity (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e), the Ritvo Autism Diagnostic Scale \u0026ndash; Revised (RADS-R, 42) as well as the State and Trait Anxiety Inventory (STAI, 43). Summary scores for both groups as well as group comparisons are presented in the supplementary materials. The interaction observation task was presented using a Lenovo Thinkpad P52 laptop with an external monitor (1920x1080 pixels, 533x300mm, 120Hz) and an external number pad. During the interaction observation task, eye movements of both eyes were captured with a LiveTrack Lightning eye tracker (Cambridge RS, sampling rate 500Hz) while participants placed their head in a headrest to ensure a stable viewing distance of 57cm. The eye tracker was calibrated before each task block using nine points and data was only collected if at least one eye had an accuracy below 0.5\u0026deg;. For the analysis of the fixations, we focused on the eye with the better accuracy during calibration for each participant. In total, testing took two to three hours.\u003c/p\u003e\n\u003ch3\u003eStimuli and Interaction Observation Task\u003c/h3\u003e\n\u003cp\u003eIn the interaction observation task, we used silent videos of dyadic interactions either between two non-autistic adults or between an autistic and a non-autistic adult (mixed dyad) as captured by Georgescu and colleagues (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). We decided against including a third dyad type, namely autistic dyads, to keep the interaction observation task at an acceptable length for observers. The videos were processed with a Gaussian filter to reduce the visual presentation to the outlines of the interaction partners, both anonymising the dyads and reducing information to the interaction partners\u0026rsquo; movements (15, for more information on stimulus creation see 24). Each stimulus captured a 10-second segment of these interactions. The segments were chosen to create a balanced sample of segments portraying high and low IPS of both mixed and non-autistic dyads such that there were no differences in overall motion (see supplementary materials). Specifically, four segments of high and four segments of low IPS were chosen for each of the dyads to model effects of IPS independent of dyadic confounders. Therefore, the task design consisted of one between-subjects factor (autistic and comparison observers) and two within-subject factors, namely IPS and dyad type (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIPS was computed based on Motion Energy Analysis (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e) as applied previously (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e) and calculated by detecting the peak of the output of the windowed-lagged cross-correlation of motion quantity as implemented in the rMEA package. In total, 64 segments were presented to each participant in two blocks. Participants were able to take a break between blocks. Each frame of the video segment was presented with an approximate visual angle of 26\u0026deg; by 18\u0026deg;, depending on the exact frame. Before each segment, a fixation cross was presented for 50ms at the centre of the segment presentation. After each video segment, participants were asked to rate how pleasant they imagined the interactions to be (\u0026ldquo;Wie angenehm stellen Sie sich diese Interaktion vor?\u0026rdquo;, from \u0026ldquo;not at all\u0026rdquo; to \u0026ldquo;very\u0026rdquo;; see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Participants had 4s to complete their rating, if they did not complete their choice the trial was excluded from the analysis. Ratings were converted into values from 0 to 100.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo create frame-by-frame areas of interest (AOIs) dynamically tracking the heads and hands of the interaction partners, we used DeepLabCut\u0026trade; which allows to estimate pose of user-defined body parts using deep learning (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e). AOIs were circle-shaped and had a diameter of 210 pixels for the heads and 140 pixels for the hands, corresponding to visual angles of 5.85\u0026deg; and 3.91\u0026deg;, respectively. Body AOIs were the same for all segments of one dyad and chosen to capture all possible body positions (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePreprocessing and analyses\u003c/h2\u003e \u003cp\u003e To improve model fit, we aggregated ratings, resulting in two values per participant for each dyad: the mean rating for segments of this dyad with high IPS and the mean rating for segments with low IPS. Fixations and saccades were detected using the algorithm developed by Nystr\u0026ouml;m and Holmqvist (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e). We classified for each fixation whether its starting point was located on a head, hand or body AOI to compute dwell times for each of the AOIs by summing fixation durations. In case of overlap, fixations were considered for all relevant AOIs, i.e., all fixations on the hand or hand were also counted as body fixations. Dwell times were aggregated by computing the median dwell time per dyad, condition and AOI, resulting in six values per participant for each dyad (each three AOI values for the segments with high and low IPS). To consider overall differences in fixation durations, we computed dwell times as a proportion of total duration of all fixations regardless of location.\u003c/p\u003e \u003cp\u003eWe used Bayesian linear mixed models as implemented in the brms package (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e), assessing our hypotheses with the brms::hypothesis() function using α\u0026thinsp;=\u0026thinsp;0.05 for directed and α\u0026thinsp;=\u0026thinsp;0.025 for undirected hypotheses. We used two models to test our hypotheses, one focusing on relevant predictors for the impression ratings and one for the dwell times. Both models included the population-level predictors diagnostic status of the observer (autistic or comparison), dyad type (mixed or non-autistic) and IPS (high or low). The model assessing dwell times additionally included the population-level predictor AOI (hand, head, body). We included the dyad and participants on the group-level in both models. While we only included intercepts on the group-level for the model assessing ratings, we also included the slopes for IPS, dyad type and AOI as well as all interactions for the participants and the slopes for diagnostic status of the observer, IPS and AOI as well as all interactions for dyads on the group-level. We checked the reliability and computational faithfulness of our models using prior predictive checks, simulation-based calibration and posterior predictive checks as proposed by Schad and colleagues (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e) before commencing inference testing (for details see the supplementary materials).\u003c/p\u003e \u003cp\u003eIn addition to our hypothesis-guided confirmatory testing, we computed two exploratory analyses. First, we used a Bayesian linear mixed model using a Poisson likelihood to assess possible influences of dyad type, IPS, diagnostic status of the observer and their interactions on the number of saccades produced. Second, we used a Bayesian multiple linear regression with a Gaussian likelihood to investigate whether gaze behaviour can predict mean impressions of how pleasant the interaction was. Specifically, we included the four gaze predictors: mean dwell times on heads, hands or bodies as well as mean number of saccades. All predictors were standardised with a \u003cem\u003ez\u003c/em\u003e-score normalisation. For both models, we added random intercepts for the participant and the observed dyad on the group-level.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eImpression ratings\u003c/h2\u003e \u003cp\u003eThe model investigating aggregated impression ratings revealed support for our hypotheses postulating decreased ratings for mixed dyads compared to non-autistic dyads (\u003cem\u003eestimate\u003c/em\u003e = -3.87 [-6.79, -1.01], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.984) and increased ratings for segments with high IPS as opposed to segments with low IPS (\u003cem\u003eestimate\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.83 [0.16, 1.49], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.978). Specifically, the model predicts a mean difference of 7.742 [0.803, 14.907] between non-autistic and mixed dyads as well as a mean differences of 1.662 [0.039, 3.204] between high and low IPS (see Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). However, there is no support for our hypotheses regarding differences between the ratings of autistic and comparison observers (\u003cem\u003eestimate\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.07 [-1.61, 3.78], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.786) or the interaction of IPS and dyad type (\u003cem\u003eestimate\u003c/em\u003e = -0.06 [-0.71, 0.6], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.555) or dyad type and diagnostic status of the observer (\u003cem\u003eestimate\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.24 [-0.41, 0.89], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.739).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eDwell times\u003c/h2\u003e \u003cp\u003eThe model focusing on dwell times revealed support for our expectation of differences between autistic and comparison observers dependent on the specific AOI (\u003cem\u003eestimate\u003c/em\u003e = -2.20% [-4.11, -0.31], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.987). Further exploration shows that this is driven by decreased dwell times on the head region for autistic observers (\u003cem\u003eestimate\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4.97% [0.53, 9.50], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.986). Specifically, this model predicts a difference of dwell times on the heads of the interaction partners of 9.945% [1.055, 18.995] between autistic and comparison observers (see Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). There is no support for our hypotheses regarding interactions between IPS and AOI (\u003cem\u003eestimate\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.13% [-1.55, 1.85], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.570) or dyad type and AOI (\u003cem\u003eestimate\u003c/em\u003e = -0.19% [-3.19, 2.87], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.555).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eExploratory analyses\u003c/h2\u003e \u003cp\u003eIn general, observers produced on average 19.31 +- 0.7 saccades per 10-second segment. There were no credible differences between any of the task conditions or interactions. However, autistic observers produced fewer saccades than non-autistic observers (\u003cem\u003eestimate\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.06 [0, 0.13], \u003cem\u003eposterior probability\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.98). Specifically, the model predicts that autistic observers produce on average 17.714 [15.936, 19.568] saccades, while comparison observers are predicted to produce 20.151 [18.327, 22.031] saccades on average. Predicting impression ratings with gaze behaviour revealed that more favourable impressions were associated with reduced dwell times on the hands (Scaled dwell times on hands: \u003cem\u003eestimate\u003c/em\u003e = -2.27 [-3.83, -0.7]) but no other gaze behaviour (see supplementary materials).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we investigated influences on impressions of observed interactions, specifically between either two non-autistic interaction partners or between one autistic and one non-autistic interaction partner. We found that the dyad type influenced the impressions formed by others, such that interactions of mixed dyads were rated less favourable, irrespective of the diagnostic status of the observer. Additionally, higher IPS elicited more favourable ratings than lower IPS independent of dyad type in both autistic and comparison observers. Thus, segments of mixed dyads showing high IPS were rated as less pleasant than non-autistic dyads showing comparably high IPS. While impression ratings were comparable between autistic and comparison observers, autistic observers spent less time dwelling on the head regions of the interaction partners. This difference in dwell time suggests that autistic observers used different aspects of the perceived interaction to infer a similar conclusion regarding their impression of how pleasant the interaction is, consistent with previous literature (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDifferences in impressions of non-autistic and mixed dyads mirror previous results on impressions of autistic and non-autistic people, indicating that autistic people themselves (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e) as well as interactions of mixed dyads (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e) evoke less favourable impressions. By reducing the information our observers received from interactants on motion dynamics, we were able to show that IPS is key to less favourable impressions of others. Specifically, we omitted audio to rule out observers being influenced by content or diverse speech patterns associated with ASD (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e). Additionally, our segments were only “thin slices” of 10s to capture almost instantaneous impressions of the interactions (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e). Therefore, we captured differences in impressions formed for mixed and non-autistic dyads based on the individual interaction partners’ movements as well as the coordination of these movements.\u003c/p\u003e \u003cp\u003eOur results indicate that smooth, dynamic interactions with high IPS lead to more favourable impressions and could support Crompton and colleagues’ (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e) suggestion that stable interpersonal coordination may facilitate higher rapport ratings. This effect is also in line with a study by Miles and colleagues (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e) showing evidence that IPS of two walkers influenced rapport impression regardless of whether walking scenes were presented visually or auditorily. Importantly, previous literature has shown both mixed and autistic dyads producing decreased overall IPS (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). However, while overall IPS is an important characteristic of an interaction, social interactions show fluctuations of IPS while they unfold (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). Thus, we decided to use short segments portraying high and low IPS for each dyad to ensure that the observed effect of increased impression ratings could be clearly attributed to high or low IPS. Since non-autistic dyads seem to produce more IPS overall, they could benefit more from the positive effect it has on impressions formed by observers. Future studies could pair methods capturing the development of experiencing how pleasant an interaction is with capturing impressions by observers to assess how close experience is to observation.\u003c/p\u003e \u003cp\u003eThis study did not reveal any differences in impressions formed by autistic and comparison observers, neither in terms of overall nor in terms of interactive effects. These comparable impressions are in line with studies on observed interactions (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e) and on impression formation of individual people (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). However, other studies suggest differences in impression formations between autistic and comparison observers (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e) and intentions based on impressions (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). In the current study, IPS had a similar effect on impressions by autistic and comparison observers. This similarity of impressions between autistic and comparison observers is in line with recent evidence showing spared neural correlates of IPS processing based on observing segments of the same social interactions as used in this study (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Bierlich and colleagues also did not find any differences in naturalness impressions between autistic and comparison observers. This pattern of results could suggest that autistic and comparison observers arrive at comparable interpretations when observing social interactions showing different levels of IPS.\u003c/p\u003e \u003cp\u003eLess favourable impressions, both for an individual person and for interactions they are part of, can have negative implications for their everyday life by reducing the willingness of others to interact with them (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). These implications are especially important since impressions can have long lasting effects on relationship development (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). Suggestions on how to navigate and approach these implications exist both on the side of the observer and of the person who is the target of the impression. Movement interventions based on imitation and synchronisation could potentially to some extent give autistic adults the tools to increase IPS (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e). Concerning the observer, diagnostic labels can lead to more positive impressions, possibly indicating the capacity to adjust expectations (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). Impressions of autistic individuals by comparison observers might improve if the comparison observers have more knowledge on ASD (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). In a recent study, autism awareness training did not lead to more favourable impressions of autistic people’s character after a short conversation between autistic and non-autistic participants. However, the autism awareness training increased socialisation intentions with the autistic people in non-autistic interaction partners; thus, mitigating the negative effect of less favourable impressions (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e). This improvement could be due to reframing an interaction as characteristic for autism instead of characteristic of a less pleasant interaction. However, while information on autism is widely distributed on social media with more than a dozen billion views on TikTok alone, a lot of the distributed information is inaccurate or overgeneralised (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e). Therefore, improving the quality of psychoeducation on autism should be a priority of policy-making.\u003c/p\u003e \u003cp\u003eDespite comparable impressions, autistic and comparison observers produced slightly different gaze behaviour while watching the interactions. Dwell times for the head of interactants were reduced in autistic compared to comparison observers. Reduced dwell times or speed to fixation in response to social stimuli like heads, faces or eyes in ASD has been shown throughout the literature (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e). Additionally, autistic observers produced fewer saccades compared to comparison observers. Although these results suggest differences in attention to social stimuli, neither dwell times to heads nor number of saccades predicted impression ratings, suggesting that these differences in gaze behaviour did not influence the formation of impressions.\u003c/p\u003e "},{"header":"Limitations","content":"\u003cp\u003eThe exact impact of the effects of IPS and dyad type on impressions in real-life cannot be estimated on the basis of this study. Real-life impression formation depends on a wide variety of information, including facial expressiveness, gaze behaviour and conversational content. In this study, we explicitly focused on manipulating two possible influences separately: IPS and dyad type. Segments showing high IPS were estimated to be rated 1.662 points more favourable on our scale from 0 to 100, with 95% of the estimated values between 0.039 and 3.204 points. This effect is potentially negligible when observing interactions in real-life. However, this effect only captures IPS of movement, while in reality we observe multiple channels in which interaction partners can be in sync or not. Small effects on each channel might add up to significantly less favourable impressions. The effect of dyad type was estimated between 0.803 and 14.907 points with an average of 7.742 points difference between mixed dyads consisting of one autistic and one non-autistic person and dyads consisting of two non-autistic people. This effect is almost five times as large as the effect of IPS. Yet, it is still unclear what differences between these dyad types cause the influence on impressions that is independent of IPS. In this study, observers only saw 10s segments of people’s outlines, so they received no auditory information regarding the interaction and only limited information on appearances. Future studies should dissect the exact movement patterns to further carve out which behavioural differences cause less favourable impressions for these mixed dyads. Furthermore, it is paramount to extend the research to interactions of two autistic people to further carve out how autism interacts with impressions. Last, the sample analysed here consisted of highly intelligent adults; thus, generalisability is limited to a subset of adults with ASD.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study illuminates predictors of impression formation of dyadic interactions, particularly those involving autistic adults. Results highlight the importance of motion dynamics in shaping these impressions, with smoother, more coordinated interactions as measured by IPS being perceived as more favourable, specifically more pleasant. In addition to this effect of IPS on impressions, interactions between autistic and non-autistic adults were generally rated as less pleasant than interactions between two non-autistic adults. While both autistic and comparison observers formed similar impressions, differences in their gaze patterns could suggest distinct strategies for processing social information. These findings underscore the importance of IPS in fostering positive social interactions. Future research should examinate other predictors of less favourable impressions of autistic people.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAOI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAreas of Interest\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eASD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAutism spectrum disorder\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBDI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eBeck Depression Inventory\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCFT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCulture Fair Intelligence Test\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCOMP\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ecomparison group of non-autistic observers\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIPS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003einterpersonal synchrony\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRADS-R\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRitvo Autism Diagnostic Scale \u0026ndash; Revised\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSTAI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eState and Trait Anxiety Inventory\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eUI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eIntolerance for Uncertainty\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":" \u003cp\u003e \u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e \u003cp\u003e This study was approved by the Ethics committee of the LMU Munich (Reference number: 23\u0026ndash;0268) and conducted in concordance with the Declaration of Helsinki. All participants were informed of the study procedure, study aim, associated risks and benefits as well as data processing and data protection, before they signed a written consent form.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eWe received consent for publication from the two interaction partners depicted in Fig.\u0026nbsp;2.\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis project was funded by the \u0026ldquo;Verein zur F\u0026ouml;rderung von Wissenschaft und Forschung an der Medizinischen Fakult\u0026auml;t der LMU M\u0026uuml;nchen e.V.\u0026rdquo; and by the \u0026ldquo;Stiftungen zugunsten der Medizinischen Fakult\u0026auml;t \u0026ndash; Cluster 2\u0026rdquo; of the LMU Klinikum. KV received funding from the Federal German Ministry of Education and Research (grant number 01GP2215). CFW was supported by German Research Council Funding (Deutsche Forschungsgemeinschaft, DFG, FA 876/3\u0026ndash;1, FA 876/5\u0026ndash;1).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eI.S.P., K.V. and C.F.W. conceptualised the research project. K.V. provided the stimuli used in the experiment, while I.S.P. and C.F.W. provided resources and acquired the funding. R.T. filtered and anonymised the videos. I.S.P. wrote the preregistration on which C.F.W. provided feedback. I.S.P. designed and programmed the experimental setup, preprocessed and analysed the data as well as wrote the first draft of the manuscript. All authors were involved in the revision of the manuscript and have approved the submitted version.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe would like to thank all the people who have assisted us in this project. First, A. L. Georgescu for sharing the stimulus material and L. S. Traiger for extracting motion quantity with Motion Energy Analysis. Second, A. M. Bierlich for providing constructive feedback during the conceptualisation and analysis of the study. Third, S. Coenen for organising and conducting the data collection. Last but not least, we want to thank S. Nothaft and Y. W. Foo for assisting during data collection as well as training DeepLabCutTM for the extraction of Areas of Interest for the eye tracking analysis.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe dataset supporting the conclusions of this article is available in the GitHub repository, https://doi.org/10.5281/zenodo.14793465 or https://github.com/IreneSophia/PESI.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAmerican Psychiatric Association. The Diagnostic and Statistical Manual of Mental Disorders [Internet]. fifth ed., text rev. 2022 [cited 2023 Jul 28]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1176/appi.books.9780890425787\u003c/span\u003e\u003cspan address=\"10.1176/appi.books.9780890425787\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organization. International Classification of Diseases, Eleventh Revision (ICD-11) [Internet]. 2019. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://icd.who.int/browse11\u003c/span\u003e\u003cspan address=\"https://icd.who.int/browse11\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThapar A, Cooper M, Rutter M. Neurodevelopmental disorders. Lancet Psychiatry. 2017;4(4):339\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBierlich AM, Scheel N, Koehler JC, Bloch C, Plank IS, Falter-Wagner C. Attenuated behavioral interpersonal synchrony in autistic adults is not explained by perception [Internet]. OSF; 2024 [cited 2024 Dec 14]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osf.io/j4gdx\u003c/span\u003e\u003cspan address=\"https://osf.io/j4gdx\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeorgescu AL, Koeroglu S, de Hamilton C, Vogeley AF, Falter-Wagner K, Tschacher CM. Reduced nonverbal interpersonal synchrony in autism spectrum disorder independent of partner diagnosis: a motion energy study. Mol Autism. 2020;11(1):1\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlass D, Yuill N. Social motor synchrony in autism spectrum conditions: A systematic review. Autism. 2024;28(7):1638\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoehler JC, Dong MS, Bierlich AM, Fischer S, Sp\u0026auml;th J, Plank IS, et al. Machine learning classification of autism spectrum disorder based on reciprocity in naturalistic social interactions. Transl Psychiatry. 2024;14(1):1\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcNaughton KA, Redcay E. Interpersonal Synchrony in Autism. Curr Psychiatry Rep [Internet]. 2020;22(3). Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubmed.ncbi.nlm.nih.gov/32025922/\u003c/span\u003e\u003cspan address=\"https://pubmed.ncbi.nlm.nih.gov/32025922/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBowsher-Murray C, Gerson S, von dem Hagen E, Jones CRG. The Components of Interpersonal Synchrony in the Typical Population and in Autism: A Conceptual Analysis. Front Psychol [Internet]. 2022 [cited 2023 Jul 5];13. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.frontiersin.org/articles/\u003c/span\u003e\u003cspan address=\"https://www.frontiersin.org/articles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpsyg.2022.897015\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2022.897015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoehl S, Fairhurst M, Schirmer A. Interactional synchrony: Signals, mechanisms and benefits. Soc Cogn Affect Neurosci. 2021;16(1\u0026ndash;2):5\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHove MJ, Risen JL. It\u0026rsquo;s All in the Timing: Interpersonal Synchrony Increases Affiliation. Soc Cogn. 2009;27(6):949\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarsh KL, Richardson MJ, Schmidt RC. Social Connection Through Joint Action and Interpersonal Coordination. Top Cogn Sci. 2009;1(2):320\u0026ndash;39.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiles LK, Nind LK, Macrae CN. The rhythm of rapport: Interpersonal synchrony and social perception. J Exp Soc Psychol. 2009;45:585\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePlank IS, Koehler JC, Nelson A, Koutsouleris N, Falter-Wagner C. Automated extraction of speech and turn-taking parameters in autism allows for diagnostic classification using a multivariable prediction model [Internet]. PsyArXiv; 2023 [cited 2023 Jul 24]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://psyarxiv.com/upz57/\u003c/span\u003e\u003cspan address=\"https://psyarxiv.com/upz57/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBierlich AM, Scheel NT, Traiger LS, Keeser D, Tepest R, Georgescu AL, et al. Neural Mechanisms of Social Interaction Perception: Observing Interpersonal Synchrony Modulates Action Observation Network Activation and Is Spared in Autism. Hum Brain Mapp. 2024;45(15):e70052.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBierlich AM, Plank IS, Scheel NT, Keeser D, Falter-Wagner CM. Neural processing of social reciprocity in autism [Internet]. medRxiv; 2024 [cited 2024 Dec 14]. p. 2024.01.31.24302051. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.medrxiv.org/content/\u003c/span\u003e\u003cspan address=\"https://www.medrxiv.org/content/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/2024.01.31.24302051v1\u003c/span\u003e\u003cspan address=\"10.1101/2024.01.31.24302051v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlkhaldi RS, Sheppard E, Burdett E, Mitchell P. Do Neurotypical People Like or Dislike Autistic People? Autism Adulthood. 2021;3(3):275\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeBrabander KM, Morrison KE, Jones DR, Faso DJ, Chmielewski M, Sasson NJ. Do First Impressions of Autistic Adults Differ Between Autistic and Nonautistic Observers? Autism Adulthood. 2019;1(4):250\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorrison KE, DeBrabander KM, Jones DR, Faso DJ, Ackerman RA, Sasson NJ. Outcomes of real-world social interaction for autistic adults paired with autistic compared to typically developing partners. Autism. 2020;24(5):1067\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSasson NJ, Faso DJ, Nugent J, Lovell S, Kennedy DP, Grossman RB. Neurotypical Peers are Less Willing to Interact with Those with Autism based on Thin Slice Judgments. Sci Rep. 2017;7(October 2016):1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSasson NJ, Morrison KE. First impressions of adults with autism improve with diagnostic disclosure and increased autism knowledge of peers. Autism. 2019;23(1):50\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCola ML, Plate S, Yankowitz L, Yankowitz L, Petrulla V, Bateman L, et al. Sex differences in the first impressions made by girls and boys with autism. Mol Autism. 2020;11(1):1\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhelpley CE, May CP. Seeing is Disliking: Evidence of Bias Against Individuals with Autism Spectrum Disorder in Traditional Job Interviews. J Autism Dev Disord. 2023;53(4):1363\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePlank IS, Traiger LS, Nelson AM, Koehler JC, Lang SF, Tepest R, et al. The role of interpersonal synchrony in forming impressions of autistic and non-autistic adults. Sci Rep. 2023;13(1):15306.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlkhaldi RS, Sheppard E, Mitchell P. Is There a Link Between Autistic People Being Perceived Unfavorably and Having a Mind That Is Difficult to Read? J Autism Dev Disord. 2019;49(10):3973\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwartz C, Dratsch T, Vogeley K, Bente G. Brief Report: Impression Formation in High-Functioning Autism: Role of Nonverbal Behavior and Stereotype Activating Information. J Autism Dev Disord. 2014;44(7):1759\u0026ndash;65.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBloch C, Tepest R, Koeroglu S, Feikes K, Jording M, Vogeley K et al. Interacting with autistic virtual characters: intrapersonal synchrony of nonverbal behavior affects participants\u0026rsquo; perception. Eur Arch Psychiatry Clin Neurosci [Internet]. 2024 Jan 25 [cited 2024 Apr 17]; Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00406-023-01750-3\u003c/span\u003e\u003cspan address=\"10.1007/s00406-023-01750-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrompton CJ, Sharp M, Axbey H, Fletcher-Watson S, Flynn EG, Ropar D. Neurotype-Matching, but Not Being Autistic, Influences Self and Observer Ratings of Interpersonal Rapport. Front Psychol. 2020;11(October):1\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamseyer FT. Motion energy analysis (MEA): A primer on the assessment of motion from video. J Couns Psychol. 2020;67(4):536\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKleiner M, Brainard D, Pelli D. What\u0026rsquo;s new in Psychtoolbox-3? 2007;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe Mathworks Inc. MATLAB [Computer software]. 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRStudio Team. RStudio: Integrated Development Environment for R [Internet]. Boston, MA: RStudio, PBC. 2020. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.rstudio.com/\u003c/span\u003e\u003cspan address=\"http://www.rstudio.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWestfall J. PANGEA: Power ANalysis for General Anova designs. University of Texas at Austin; 2016.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAllenmark F, Shi Z, Pistorius RL, Theisinger LA, Koutsouleris N, Falkai P, et al. Acquisition and Use of \u0026lsquo;Priors\u0026rsquo; in Autism: Typical in Deciding Where to Look, Atypical in Deciding What Is There. J Autism Dev Disord. 2021;51(10):3744\u0026ndash;58.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFalter CM, Elliott MA, Bailey AJ. Enhanced visual temporal resolution in autism spectrum disorders. PLoS ONE. 2012;7(3):1\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eK\u0026uuml;hner C, B\u0026uuml;rger C, Keller F, Hautzinger M. Reliabilit\u0026auml;t und Validit\u0026auml;t des revidierten Beck-Depressionsinventars (BDI-II). Nervenarzt. 2007;78(6):651\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei\u0026szlig; R. CFT 20-R: Grundintelligenztest Skala 2-Revision. Hogrefe; 2006.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBates ME, Lemay EP. The d2 test of attention: Construct validity and extensions in scoring techniques. J Int Neuropsychol Soc. 2004;10(3):392\u0026ndash;400.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGerlach AL, Andor T, Patzelt J. Die Bedeutung von Unsicherheitsintoleranz f\u0026uuml;r die Generalisierte Angstst\u0026ouml;rung Modell\u0026uuml;berlegungen und Entwicklung einer deutschen Version der Unsicherheitsintoleranz-Skala. Z F\u0026uuml;r Klin Psychol Psychother. 2008;37(3).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark JH. The Ishihara Test for Color Blindness. Am J Physiol Opt. 1924;5:269\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWesemann W, Schiefer U, Bach M. New DIN norms for determination of visual acuity. Ophthalmologe. 2010;107(9):821\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRausch J, Fangmeier T, Falter-Wagner CM, Ackermann H, Espel\u0026ouml;er J, H\u0026ouml;lzel LP et al. A novel screening instrument for the assessment of autism in German language: validation of the German version of the RAADS-R, the RADS-R. Eur Arch Psychiatry Clin Neurosci [Internet]. 2024 Oct 27 [cited 2024 Nov 19]; Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00406-024-01894-w\u003c/span\u003e\u003cspan address=\"10.1007/s00406-024-01894-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaux L, Glanzmann P, Schaffner CD. Das state-trait-angstinventar [The state-trait anxiety inventory]. G\u0026ouml;ttingen: Hogrefe; 1981.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMathis A, Mamidanna P, Cury KM, Abe T, Murthy VN, Mathis MW, et al. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nat Neurosci. 2018;21(9):1281\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNystr\u0026ouml;m M, Holmqvist K. An adaptive algorithm for fixation, saccade, and glissade detection in eyetracking data. Behav Res Methods. 2010;42(1):188\u0026ndash;204.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026uuml;rkner PC. brms: An R package for Bayesian multilevel models using Stan. J Stat Softw. 2017;80(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchad DJ, Betancourt M, Vasishth S. Toward a principled Bayesian workflow in cognitive science [Internet]. arXiv; 2020 [cited 2023 Sep 26]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://arxiv.org/abs/1904.12765\u003c/span\u003e\u003cspan address=\"http://arxiv.org/abs/1904.12765\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBloch C, Viswanathan S, Tepest R, Jording M, Falter-Wagner CM, Vogeley K. Differentiated, rather than shared, strategies for time-coordinated action in social and non-social domains in autistic individuals. Cortex. 2023;166:207\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFusaroli R, Grossman R, Bilenberg N, Cantio C, Jepsen JRM, Weed E. Toward a cumulative science of vocal markers of autism: A cross-linguistic meta-analysis-based investigation of acoustic markers in American and Danish autistic children. Autism Res. 2022;15(4):653\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOchi K, Ono N, Owada K, Kojima M, Kuroda M, Sagayama S, et al. Quantification of speech and synchrony in the conversation of adults with autism spectrum disorder. PLoS ONE. 2019;14(12):1\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWillis J, Todorov A. Making Up Your Mind After a 100-Ms Exposure to a Face. 2015;17(7):592\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMayo O, Gordon I. In and out of synchrony\u0026mdash;Behavioral and physiological dynamics of dyadic interpersonal coordination. Psychophysiology. 2020;57(6):1\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuman LJ, Sandstrom GM, Biesanz JC, Dunn EW. Accurate First Impressions Leave a Lasting Impression: The Long-Term Effects of Distinctive Self-Other Agreement on Relationship Development. Soc Psychol Personal Sci. 2013;4(4):395\u0026ndash;402.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoehne S, Behrends A, Fairhurst MT, Dziobek I. Fostering Social Cognition through an Imitation- and Synchronization-Based Dance/Movement Intervention in Adults with Autism Spectrum Disorder: A Controlled Proof-of-Concept Study. Psychother Psychosom. 2015;85(1):27\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrosnan M, Mills E. The effect of diagnostic labels on the affective responses of college students towards peers with \u0026lsquo;Asperger\u0026rsquo;s Syndrome\u0026rsquo; and \u0026lsquo;Autism Spectrum Disorder\u0026rsquo;. Autism. 2016;20(4):388\u0026ndash;94.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones DR, Morrison KE, DeBrabander KM, Ackerman RA, Pinkham AE, Sasson NJ. Greater Social Interest Between Autistic and Non-autistic Conversation Partners Following Autism Acceptance Training for Non-autistic People. Front Psychol [Internet]. 2021 [cited 2023 Jul 5];12. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.frontiersin.org/articles/\u003c/span\u003e\u003cspan address=\"https://www.frontiersin.org/articles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpsyg.2021.739147\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2021.739147\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAragon-Guevara D, Castle G, Sheridan E, Vivanti G. The Reach and Accuracy of Information on Autism on TikTok. J Autism Dev Disord [Internet]. 2023 Aug 6 [cited 2024 Dec 14]; Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10803-023-06084-6\u003c/span\u003e\u003cspan address=\"10.1007/s10803-023-06084-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChita-Tegmark M. Social attention in ASD: A review and meta-analysis of eye-tracking studies. Res Dev Disabil. 2016;48:79\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSetien-Ramos I, Lugo-Mar\u0026iacute;n J, Gisbert-Gustemps L, D\u0026iacute;ez-Villoria E, Mag\u0026aacute;n-Maganto M, Canal-Bedia R, et al. Eye-tracking studies in adults with autism spectrum disorder: A systematic review and meta-analysis. J Autism Dev Disord. 2023;53(6):2430\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"molecular-autism","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mola","sideBox":"Learn more about [Molecular Autism](http://molecularautism.biomedcentral.com/)","snPcode":"13229","submissionUrl":"https://submission.nature.com/new-submission/13229/3","title":"Molecular Autism","twitterHandle":"@MolecularAutism","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"impression formation, interpersonal synchrony, observed interactions, autism spectrum disorder, behavioural coordination, dyadic interactions","lastPublishedDoi":"10.21203/rs.3.rs-5950403/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5950403/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground:\u003c/h2\u003e \u003cp\u003eHumans form almost instantaneous impressions of everyone they encounter. These impressions set the first tone for how they approach and interact with others. Research on impression formation unveiled that impressions formed by autistic and non-autistic people are often less favourable when rating an autistic person. This effect is partly explainable by differences in motion dynamics.\u003c/p\u003e\u003ch2\u003eMethods:\u003c/h2\u003e \u003cp\u003eIn this preregistered study, we systematically assessed impressions formed by 27 autistic and 36 non-autistic comparison observers when watching videos showing silent, dyadic interactions between either two non-autistic or between an autistic and a non-autistic person. We used an eye tracker to capture their gaze patterns while observing these interactions. Of each dyadic interaction, a video vignette with high and a vignette with low interpersonal synchrony was extracted using Motion Energy Analysis so that we could investigate the effects of interpersonal synchrony and diagnosis, respectively.\u003c/p\u003e\u003ch2\u003eResults:\u003c/h2\u003e \u003cp\u003eInteractions were rated less favourably when the observed dyad included an autistic adult. Additionally, interactions showing low interpersonal synchrony were rated less favourably than interactions showing high interpersonal synchrony, regardless of dyad type. Both the effect of interpersonal synchrony and the effect of dyad type on the impressions were independent of the diagnostic status of the observer. Nonetheless, gaze patterns revealed differences between autistic and comparison observers, but were unrelated to interpersonal synchrony and dyad type\u003c/p\u003e\u003ch2\u003eLimitations:\u003c/h2\u003e \u003cp\u003eIn this study, we investigated limited influences on impression formation, specifically interpersonal synchrony and autism. There are many more potentially interesting aspects of individuals that impact impression formation, such as facial expressiveness, gaze behaviour and linguistic content of conversations, which should be investigated systematically and in a controlled fashion in future research.\u003c/p\u003e\u003ch2\u003eConclusions:\u003c/h2\u003e \u003cp\u003eBoth the interaction partners in a dyad and the synchrony of their motion influence the impressions autistic and comparison observers form of the interaction, such that vignettes showing high interpersonal synchrony are perceived as more pleasant. Furthermore, interactions of dyads consisting of one autistic and one non-autistic person are perceived as less pleasant than those of two non-autistic people, which was the case for autistic and comparison observers likewise.\u003c/p\u003e","manuscriptTitle":"The influence of interpersonal synchrony and autism on impressions of dyadic interactions: a preregistered study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-17 09:29:38","doi":"10.21203/rs.3.rs-5950403/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-04-08T02:55:02+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-03-29T21:03:28+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-03-11T19:32:33+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-17T04:46:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"66391711139194372146395702707430826109","date":"2025-02-13T14:43:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"185927257685273424274188176061214196685","date":"2025-02-13T03:50:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"128574351570079113016516924578804134310","date":"2025-02-11T15:50:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"301791347508991349933003255982477871629","date":"2025-02-11T08:35:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-02-11T03:29:45+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-02-10T20:07:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-02-04T05:40:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"Molecular Autism","date":"2025-02-03T10:40:43+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"molecular-autism","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mola","sideBox":"Learn more about [Molecular Autism](http://molecularautism.biomedcentral.com/)","snPcode":"13229","submissionUrl":"https://submission.nature.com/new-submission/13229/3","title":"Molecular Autism","twitterHandle":"@MolecularAutism","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"96c29cab-f0d1-4248-a100-4829f66c642d","owner":[],"postedDate":"March 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-06-16T16:02:25+00:00","versionOfRecord":{"articleIdentity":"rs-5950403","link":"https://doi.org/10.1186/s13229-025-00668-y","journal":{"identity":"molecular-autism","isVorOnly":false,"title":"Molecular Autism"},"publishedOn":"2025-06-12 15:57:54","publishedOnDateReadable":"June 12th, 2025"},"versionCreatedAt":"2025-03-17 09:29:38","video":"","vorDoi":"10.1186/s13229-025-00668-y","vorDoiUrl":"https://doi.org/10.1186/s13229-025-00668-y","workflowStages":[]},"version":"v1","identity":"rs-5950403","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5950403","identity":"rs-5950403","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}