Vicarious experience modulates implicit and explicit evaluation of virtual agents delivering pain and touch over an embodied avatar

Vicarious experience modulates implicit and explicit evaluation of virtual agents delivering pain and touch over an embodied avatar | 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 Vicarious experience modulates implicit and explicit evaluation of virtual agents delivering pain and touch over an embodied avatar Matteo P. Lisi, Martina Fusaro, Francesca March, Salvatore Maria Aglioti This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8765445/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Witnessing pain and touch on one’s embodied avatar in immersive virtual environments can elicit vicarious (un)pleasant sensations. What remains unclear is whether observing pain and touch on one’s virtual body can modulate attitudes towards the touching agent. In this study, participants underwent a full-body illusion in Immersive Virtual Reality and observed different virtual agents delivering either a painful (stab) or a pleasant (caress) virtual stimulus on their embodied avatar's right hand. We assessed the pre-post changes induced by virtual somatosensory stimuli on (1) explicit judgements of the virtual agent's trustworthiness and attractiveness; (2) implicit evaluation of the virtual agent's facial attractiveness (via a mouse-tracking procedure), and (3) the physiological arousal and gaze during the regulation of the virtual agent-human interpersonal distance. We found that trustworthiness increased and comfort distance decreased with the perceived touch pleasantness. Analysis of hand movements revealed decreased attraction toward the stabbing agents. Consistent with this pattern, stabbing agents elicited stronger cardiac deceleration– an index of freezing–during the subsequent approach. Gaze increased toward the eye region for both stabbers and caressers, suggesting heightened monitoring of the agent’s intentions. Together, these findings indicate that vicarious somatosensory experience can influence the social encoding of virtual agents at both explicit and implicit levels, as inferred from behavioural, physiological, and gaze signatures. These results contribute to understanding the psychological impact of embodied interactions in immersive digital spaces. Impression updating Virtual agents Virtual pain and touch Psychophysiology Eye-tracking Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Immersive Virtual Reality (IVR) enables users to experience a compelling illusion of being present in a virtual environment and to perceive a virtual body as their own (Aglioti et al., 2025 ; Maselli & Slater, 2013 ; Slater et al., 2010 ). This embodied experience allows researchers to examine how observing tactile or threatening stimuli on one’s virtual body shapes cognition, emotion, and social behaviour. Previous research has shown that witnessing a painful event, such as a stab, delivered to a virtual body evokes unpleasant feelings and elevates physiological arousal (González-Franco et al., 2014 ; Lisi et al., 2024 ; Nicolardi et al., 2025 ). Conversely, positive somatosensory stimuli, such as virtual caresses applied to embodied avatars, induce pleasant sensations (Fusaro et al., 2021 ; Mello, Fusaro, et al., 2022 ; Verga, Di Marco, et al., 2025 ), with effects varying with body site and identity of the agent delivering touch (Fusaro et al., 2021 ; Lisi et al., 2025 ; Mello, Fusaro, et al., 2022 ). What remains unclear is whether vicarious somatosensation shapes impressions concerning the virtual agent who touches or harms us. Interindividual physical contact is a core channel for social affiliation, bonding, and well-being (Fusaro et al., 2022 ; Suvilehto et al., 2023 ). The experience of touch reflects an interplay of bottom-up, mechanical factors and top-down, cognitive appraisals. Work on affective touch highlighted the role of C-tactile (CT) afferents in encoding gentle stroking, while recognizing that affective touch is not reducible to CT input (Case et al., 2023 ; Schirmer et al., 2023 ) and remains strongly context-dependent (Ellingsen et al., 2016 ; Mello et al., 2024 ; Sailer et al., 2024 ) and influenced by individual differences (Mello et al., 2025 ; Verga, Lisi, et al., 2025 ). Top-down social factors further modulate touch perception (Saarinen et al., 2021 ). Perceived appropriateness increases linearly with relationship strength, and touch from women is generally perceived as more acceptable (Suvilehto et al., 2015 , 2019 ). These priors shape not only explicit responses but also somatosensory cortex activity (Gazzola et al., 2012 ; Suvilehto et al., 2021 ). In a similar vein, attractiveness cues influence perceived pleasantness of touch. For example, attractive faces observed while receiving stroking were associated with increased touch pleasantness, particularly for the slow touch condition (Novembre et al., 2021 ). This was also associated with increased heart rate variability, possibly signalling a positive reaction at the autonomic level. By contrast, evidence that touch changes attitudes or behavior is mixed. Research on the Midas touch phenomenon suggests that even minimal interpersonal contact can increase compliance and positive evaluations of the toucher (Crusco & Wetzel, 1984 ; Seger et al., 2014 ), and that similar effects can extend to mediated and virtual interactions ( virtual Midas touch ; (Haans et al., 2014 ; Harjunen et al., 2018 ; Świdrak et al., 2020 ). However, more recent work indicates that these beneficial effects are far from universal and critically depend on contextual and relational factors, such as perceived appropriateness, naturalness, and credibility of the touching agent, with several studies reporting null or even negative effects of virtual social touch (Ma et al., 2025 ; Saini et al., 2024 ; Świdrak et al., 2021 ). Harmful actions directed at one’s body provide highly diagnostic cues about an agent’s intentions, whereas affiliative touch is inherently more ambiguous and context-sensitive. This asymmetry suggests that pain and pleasant touch may differentially shape trust, avoidance, and approach tendencies toward the agent, particularly when such interactions are experienced as happening to one’s own embodied self. Understanding the reactivity to a virtual toucher is crucial in immersive digital spaces, where human-virtual agent interactions are increasingly common in contexts such as healthcare (Fusaro et al., 2025 ; Pandhare et al., 2025 ), sales (Mull et al., 2015 ; Zhu et al., 2024 ), education (Makransky et al., 2019 ), and gaming (Hensel et al., 2020 ), and where first impressions drive engagement (Lee et al., 2018 ; Ter Stal et al., 2020 ; Yang et al., 2025 ). Virtual agents are often perceived and judged using the same heuristics applied to humans: people form rapid first impressions and evaluate agents along aesthetic (attractiveness) and moral (trustworthiness) dimensions, much as they do with human partners (Cafaro et al., 2012 ; Cihodaru-Ștefanache & Podina, 2025 ; Oosterhof & Todorov, 2008 ). People also tend to regulate their nonverbal behavior accordingly, including interpersonal distance (IPD) - the preferred physical distance from others- which serves as a sensitive nonverbal index of approach-avoidance (Bailenson et al., 2003 ; Nolte et al., 2025 ). In both real and virtual contexts, IPD tracks perceived social comfort (Candini et al., 2021 ; Fusaro et al., 2023 ; Iachini et al., 2016 ; Lisi, Fusaro, et al., 2021 ). Moreover, perceived violations of preferred IPD responses are usually associated with increased arousal (Candini et al., 2021 ; Kroczek et al., 2020 ; Perry et al., 2013 ; Placidi et al., 2025 ; Vabba, Placidi, et al., 2025 ; Wieser et al., 2010 ) and gaze aversion (Argyle & Dean, 1965 ; Bailenson et al., 2001 ). Virtual agents can thus be seen as a specific class of artificial social partners that can adopt a human-like appearance and use both verbal and nonverbal signals for communication. However, it is still unclear whether and how first impressions and nonverbal behaviours such as IPD are updated based on bodily interactions with them. The present study. We test whether vicarious experience of pain vs. pleasant touch delivered over one’s embodied avatar updates the social encoding and subsequent nonverbal interactions with the touching agent. Participants experienced a full-body illusion in IVR and then observed different agents deliver either a painful (stab) or pleasant (caress) stimulus to their avatar’s hand. We assessed pre- to post-changes in explicit judgments (trustworthiness, attractiveness), implicit evaluation of facial attractiveness (mouse-tracking), nonverbal and physiological (skin conductance responses, heart rate, and gaze) responses during IPD regulation with each agent. This design allows us to ask whether the type of virtual somatosensation shifts both explicit appraisals and implicit approach–avoidance toward the sender, advancing our understanding of embodied interactions with virtual agents. We expected that trustworthiness and attractiveness would increase for the caressers and decrease for the stabbers, and we anticipated that these effects might be moderated by subjective touch (un)pleasantness. Physiologically, we expected reduced autonomic arousal (lower skin conductance responses and heart rate) during the subsequent approach of the caressers, and an increased arousal for the stabbers. Concerning gaze, we predicted a threat-monitoring shift following stabbing interactions, such that visual attention would be reallocated away from the agent’s eyes toward action-relevant regions (e.g., the moving hand/effector) whereas pleasant touch would be associated with relatively greater eye-region engagement during subsequent interactions. Materials and methods Participants 40 neurotypical, right-handed, cisgender, heterosexual participants took part in the study (20 women, mean age = 24.38, sd = 2.05, age range = 20–30; for all demographics see Table S1 ). An a priori power analysis conducted in MorePower (Campbell & Thompson, 2012 ) indicated that N = 40 was required to detect an effect of Cohen’s f = 0.46 with 80% power at α = 0.05. Anxiety was measured by means of the State-Trait Anxiety Inventory (STAI; Spielberger et al., 1983 ), Italian version (Pedrabissi & Santinello, 1989 ). Prior VR experience was assessed with a single item (“How much experience have you had with Immersive Virtual Reality so far?”) rated on a 0-100 VAS scale. All participants gave written consent to take part in this study. The protocol was approved by the Santa Lucia Foundation Committee (protocol CE/2022_012) and complied with the Declaration of Helsinki (2013). Experimental stimuli and setup The virtual scenario was designed using Blender 2.9 ( https://www.blender.org/ ) and implemented in Unity v2020.3.11f ( https://unity.com/ ). Two female and two male avatars were selected among a pool extracted from the Rocketbox Library (Gonzalez-Franco et al., 2020 ), based on their perceived age, gender, and ethnicity (Tab.S3). These attributes were rated by an independent sample of 59 Prolific participants (Palan & Schitter, 2018 ). The scenario was presented by means of HTC Vive Pro Eye ( https://www.vive.com/sea/product/vive-pro-eye/overview/ ). In order to realize naturalistic movements, we used the Xsense motion capture suit ( https://www.xsens.com/ ) to record the kinematics of an actor either gently caressing or stabbing the right hand of another actor with the right hand (see Fig. S1 in the Supplementary Information). A VIVE controller was used as a prop to reproduce the knife stabbing. The actor was instructed to perform the caress at a speed of about 3 cm/s. The actors’ kinematics were transferred to the virtual agents by means of Motion Builder 2022 ( https://www.autodesk.com/ ) and rendered in Unity. Through recorded kinematics, we were able to keep the toucher’s movements constant and control for other emotional interference that can be conveyed in traditional experimental settings by confederates through other nonverbal cues. Galvanic skin response (GSR) and electrocardiogram (ECG) were recorded as indices of physiological reactivity. Signals were amplified using an ADInstruments PowerLab 8/35 system equipped with an ML116 GSR Amplifier (75 Hz AC excitation; low constant voltage of 22 mVrms) and dedicated GSR sensors comprising two bipolar finger electrodes. GSR electrodes were attached to the distal phalanges of the right index and middle fingers, and the signal was sampled at 1 kHz. For ECG acquisition, two pre-gelled electrodes (DORMO) were placed on the back of each hand, with the reference electrode positioned on the left ankle. ECG signals were sampled at 1 kHz and low-pass filtered at 30 Hz. All recordings were acquired using LabChart 8 (ADInstruments, Inc.). Participants’ gaze was recorded using the HTC Vive Pro Eye built-in Tobii eye-tracker (120 Hz; nominal accuracy of 0.5°-1.1°). A 5-point calibration (SteamVR) was performed before each task (Keshava et al., 2020 ; Schuetz & Fiehler, 2022 ). In the IPD task, the agents’ eye region was defined in Unity using a collider-based region of interest (ROI). Gaze samples were mapped onto scene objects via ray casting; samples intersecting the eye-ROI collider were counted as looks to the eyes (Clay et al., 2019 ). Invalid samples were removed, and linear interpolation was applied. Trials with > 20% invalid gaze were excluded (11 trials, 0.8% of the dataset). For the Touch task, pupil diameter (mm) was downsampled to 30 Hz. Samples marked as invalid by the eye-tracker and samples outside the 1.5–9 mm range were removed (Kret & Sjak-Shie, 2019 ). Missing data were linearly interpolated. Trials were excluded if more than 20% of the samples were invalid (12 trials, 1.5% of the dataset). Pupil diameter was then averaged across the two eyes. To obtain a baseline-corrected measure, we subtracted the mean pupil diameter during the first second of the trial (agent appearance) from the mean pupil diameter during the remainder of the trial. General Procedure Participants first filled out the questionnaires (STAI, VR Experience), then performed a baseline mouse-tracking task. They subsequently donned the VR headset and performed the baseline interpersonal distance (IPD) task, which included virtual agents’ evaluations. Next, they completed the virtual touch task. Immediately afterward, they repeated the IPD task and the virtual agents’ evaluations to assess pre–post changes. Finally, participants completed the mouse-tracking task again. Touch Task Participants observed a gender-matched virtual body seated on a chair, with the right arm resting on a desk, and were instructed to focus on their right virtual hand. At trial onset, a virtual agent appeared standing to the participant’s right, and participants were instructed to look at the agent (to facilitate the agent’s recognition). After 1 s, the agent initiated a movement (caress or stab), and participants were instructed to attend to the agent’s hand. The action lasted ~ 5 s, and the agent then remained still for 1s before disappearing (see Fig. 1 ). After a 500 ms delay, a visual analogue scale (VAS) appeared. On each trial, participants provided ratings of the agent’s attractiveness and trustworthiness, as well as the (un)pleasantness (0 = extremely unpleasant; 50 = neutral; 100 = extremely pleasant) and vicariousness of the stimulation (“It was as if I was actually feeling the stimulus on my body”). At the end of the task, body ownership illusion was assessed using selected items from the Avatar Embodiment Questionnaire (Peck & Gonzalez-Franco, 2021 ; see Tab.S2 in Supplementary Material). Skin conductance responses (SCR) were extracted as the average phasic driver within the 7-s trial window using Ledalab (Benedek & Kaernbach, 2010 ). Heart rate (HR) in bpm and pupil size in mm were extracted as the mean over the same 7 s window and baseline-corrected using the 2 s immediately preceding trial onset. Interpersonal Distance Task In each trial, the virtual agent appeared 2.25 m in front of the participant. After 2 s, the agent walked toward the participant at a constant speed until reaching a distance of 0.25 m, and then stood at that distance for the remainder of the trial (see Fig. 2 ). The agent maintained direct gaze toward the participant (i.e., the VR camera) throughout the trial. Each trial lasted 8 s and was followed by a 20 s inter-trial interval. During the final repetition, after the inter-trial interval, participants rated the agent’s attractiveness and trustworthiness. Participants were instructed to press a button on the controller when they first felt the agent’s approach become uncomfortable (comfort point). The agent continued its approach regardless of the comfort-point response, allowing physiological responses to be analyzed over a uniform time-window, including the period after participants’ preferred distance was exceeded. Preferred distance was computed as the Euclidean distance (cm) between the agent’s nose and the participant’s nose (with a marker placed on the VR Camera in Unity) at the moment of button press. Skin conductance responses (SCR) were extracted as the average phasic driver within the 8-s trial window using Ledalab (Benedek & Kaernbach, 2010 ). Heart rate (HR) in bpm was extracted over the same 8 s window and baseline-corrected using the 2 s immediately preceding trial onset. Eye-tracking data for the IPD task were analysed using the eyetrackingR package (Dink & Ferguson, 2015 ). The dependent measure was the proportion of looks to the agent’s eye-ROI (eyes vs elsewhere) within each trial. To accommodate boundary values (0 and 1), we used empirical logit transformation ( elog ; Barr, 2008 ). We conducted a time-window analysis with TimeWindow as a categorical factor with three levels: Far (0–2 seconds), Approach (2–5.43 seconds), and Near (5.43-8 seconds). Mouse-Tracking Task Participants completed an alternate forced-choice task adapted from Faust and colleagues (Faust et al., 2019 ) using the MouseTracker software (Freeman & Ambady, 2010 ). For each trial, participants were presented with three two-digit numbers: the first one (“basis”) was located at the bottom of the screen, and the other two (“targets”) were located at the top left and right corners of the screen (see Fig. 3 ). Participants were instructed to click on the target that was numerically closer to the basis. Next to each target, one face was presented. Participants were asked to ignore the faces and complete the task correctly, while also being as fast as possible. The maximum difference in value between the basis and the targets was five units. For instance, if the basis was 22, each of the two targets was not larger than 27. This is to ensure the difficulty of the task across all trials was consistent. Participants were asked to execute the task accurately and fast and to ignore the faces. Data were analysed using the mousetrap package in R (Wulff et al., 2025 ). The preprocessing included time and length normalization. Trajectories were remapped on the right side. We excluded trials with errors (i.e., the choice of number was incorrect), and trials that were outside the range defined by the mean response time ± 2 standard deviations (Wulff et al., 2025 ). As outcomes, we considered the signed Maximum Absolute Deviation (MAD) from the ideal path connecting the start and the end of the trajectory; the Area Under the Curve, i.e., the amount of area between the ideal and the observed trajectory, and the response time (Stillman et al., 2018 ). Results Analytical strategy Analyses were conducted in RStudio within a Bayesian framework (Kruschke & Liddell, 2018 ) using the brms (Bürkner, 2017 ), bayestestR (Makowski, Ben-Shachar, & Lüdecke, 2019 ), and emmeans (Lenth & Piaskowski, 2025 ) packages. For each outcome collected during the IPD task (attractiveness and trustworthiness ratings, comfort distance point, heart rate deceleration, skin conductance response, and fixations), we ran two complementary analyses. Firstly, we fit a linear mixed-effects model with the interaction between Session (Pre, Post) and Stimulation (Stab, Caress) as a fixed effect, and by-Stimulation slopes and by-participant intercepts as random effects. Secondly, we computed an update score Δ defined as Post-Pre (positive values= increase), and modelled the effects of Stimulation and Touch Pleasantness, with by-participant intercepts as a random effect. All dependent variables were scaled and centred. We used weakly informative priors on the coefficients (Normal, M = 0 and SD = 1). For the mouse-tracking task, we fit a linear mixed-effect model with the interaction between Session and Condition (Congruent, Incongruent) as fixed effects, and the intercept over the participants as random effects. For the physiological (SCR, HR) and gaze indices, STAI-state and STAI-trait were entered as covariates to control for possible effects from state and trait anxiety. For touch (un)Pleasantness and Vicariousness, we included VR Experience as a predictor to control for possible effects of prior VR exposure. All models’ syntax is reported in the Supplementary Information (Table S4). For detecting an effect, we established a probability of direction > 95% (Makowski, Ben-Shachar, Chen, et al., 2019); we also report 89% highest posterior density intervals, mean, and standard deviation of the posterior distributions. For all coefficients, we verified that the Effective Sample Size (ESS) was higher than 1000 and that Rhat, indicating convergence of the chains, was < 1.05. Touch Task Body ownership The analysis on the body ownership ratings showed that scores were higher for the ownership items compared to the control ones (0.95 ± 0.13, HDI [0.72, 1.18], pd = 100%). Touch (un)Pleasantness Results showed a significantly greater Pleasantness for the caresses compared to the stabs (1.13 ± 0.18, [0.85, 1.41], pd = 100.00%). We also observed an effect of reported VR experience, so that the higher the VR experience, the lower Pleasantness for the caresses (-0.28 ± 0.13, [-0.49, -0.07], pd = 98.38%). Touch Vicariousness Vicariousness was higher for caresses compared to stabs (0.40 ± 0.10, [0.24, 0.55], pd = 99.99%). VR Experience decreased vicariousness for the caresses (-0.29 ± 0.14, [-0.51, -0.06], pd = 97.82%) but not for the stabs (-0.12 ± 0.15, [-0.36, 0.12], pd = 78.16%). Physiological reactivity to touch Skin conductance responses were higher for stabs compared to caresses (0.017 ± 0.008, [-0.004, 0.03], pd = 99.03%). Heart rate responses did not show a difference between the two conditions (-0.03 ± 0.04, [-0.01, 0.03], pd = 78.10%). Pupil size analysis showed that stabs increased pupil size compared to caresses (0.58 ± 0.08, [0.46, 0.71], pd = 100%; see Fig.S2), and the magnitude of pupil size increase was inversely predicted by (un)Pleasantness, so that the lower the perceived pleasantness, the stronger the pupil dilation (-0.08 ± 0.04, [0.15, -0.01], pd = 97.49%). Pre-post changes Attractiveness Neither the stabbing (-0.18 ± 0.14 [-0.40, 0.04], pd = 90.44%) nor the caressing agents (0.09 ± 0.14 [-0.14, 0.30], pd = 74.44%) showed a credible change of attractiveness from pre to post. In Δ analysis, Touch (un)Pleasantness did not modulate attractiveness change for caressing (0.13 ± 0.12 [-0.06, 0.32], pd = 85.50%) nor stabbing (0.20 ± 0.17 [-0.06, 0.49], pd = 88.64%) agents. These results suggest that Attractiveness judgments are not affected by the type of somatosensory interaction or by its perceived (un)Pleasantness. Trustworthiness We observed a significant decrease in Trustworthiness for the stabbing avatar from pre to post treatment (-0.57 ± 0.13 [-0.78, -0.37], pd = 100%), while no difference was found for the caressing agents (0.17 ± 0.13, [-0.03, 0.38], pd = 89.98%). Analysis on the Δ values showed that higher Touch (un)Pleasantness (higher ratings) predicted a larger increase in trustworthiness for the caressing agents (0.25 ± 0.11, [0.07, 0.43], pd = 98.60%) but not for the stabbing ones (0.17 ± 0.18, [-0.11, 0.46], pd = 83.12%). These results suggest that the virtual harm reduced trust irrespective of individual appraisal, while caress raised trust only when experienced as pleasant (see Fig. 4 ). Mouse-tracking task We observed a decrease in Maximum Absolute Deviation for the congruent trials (those in which the caressing agents were placed next to the correct response; -0.13 ± 0.07 [-0.24, -0.03], pd = 97.88%). This decrease was not found for the incongruent condition (trials in which the stabbing agents were placed next to the correct response; -0.005 ± 0.07 [-0.11, 0.11], pd = 52.40%). Similarly, Area Under the Curve was reduced for the congruent trials from pre to post Touch Task (-0.11 ± 0.06 [-0.20, -0.02], pd = 97.28%), while no difference for incongruent trials was found (-0.03 ± 0.06 [-0.13, 0.05], pd = 71.38%). Analysis of the response times showed a general decrease for both the congruent (-0.26 ± 0.04 [-0.33, -0.19], pd = 100%) and incongruent (-0.18 ± 0.04 [-0.25, -0.11], pd = 100%) conditions from pre to post Touch Task; however, only in post Touch Task session, the congruent trials showed shorter decision times than the incongruent ones (Pre: -0.06 ± 0.08 [-0.18, 0.06], pd = 79.45%; Post: -0.14 ± 0.07 [-0.25, -0.02], pd = 97.02%), suggesting a reduced conflict for trials in which the caressers were associated with the right response target compared to trials in which the stabbers were associated with the right response target (see Fig. 5 ). The emergence of a post-task congruency advantage suggests relatively greater facilitation when caressing agents were paired with the correct response compared with when stabbing agents were paired with the correct response—consistent with facilitation by caress-associated stimuli and comparatively greater conflict in stabbing-associated trials. Interpersonal distance For stabbing agents, the comfort-distance increased (i.e., more space between the participant and the avatar) from Pre to Post (0.27 ± 0.04 [0.21, 0.33], pd = 100%), also for the caressing agents to a smaller extent (0.07 ± 0.03 [0.01, 0.12], pd = 97.09%). Analysis on the Δ values showed that higher Touch (un)Pleasantness (higher ratings) predicted a larger decrease in interpersonal distance for the caressing agents (-0.26 ± 0.10, [-0.47, -0.07], pd = 99.62%) but not for the stabbing ones (-0.23 ± 0.19, [-0.63, 0.13], pd = 89.46%). These results suggest that the virtual harm increased preferred interpersonal distance irrespective of individual appraisal, while caresses decreased interpersonal distance only when experienced as pleasant (see Fig. 6 ). Physiological and gaze correlates of interpersonal distance For both stabbing (-0.14 ± 0.03 [-0.18, -0.09], pd = 100%) and caressing avatars (-0.14 ± 0.03 [-0.18, -0.08], pd = 100%), we observed a decrease in SCR from pre to post treatment. No other credible differences were found for SCR analysis (see Supplementary Information). We found an increase of HR deceleration for stabbers from pre to post touch task (-0.11 ± 0.06 [-0.21, -0.02], pd = 96.71%). No difference for the caressing agents was found (0.03 ± 0.06 [-0.06, 0.13], pd = 69.88%). Analysis on the Δ values revealed that touch (un)Pleasantness predicted a decreased heart rate deceleration in subsequent approach for caressing (0.23 ± 0.12 [0.04, 0.42], pd = 97.44%) but not stabbing agents (-0.08 ± 0.17 [-0.35, 0.20], pd = 67.53%). We observed an increase in gaze for both the stabbing (0.09 ± 0.05 [0.003, 0.18], pd = 95.89%) and the caressing (0.12 ± 0.06 [0.03, 0.21], pd = 98.31%) agents in the approach phase from pre to post. Moreover, we observed an increase in gaze for the stabbing agents in the appearance phase (at far distances) from pre to post (0.26 ± 0.06 [0.16, 0.35], pd = 98.31%). No other differences were found. Overall, these effects are consistent with the ambiguous function of eye gaze and suggest that stabbing may have increased visual attention toward the stabbing agent at far distances, potentially reflecting face identification and/or assessment of the agent’s intentions. Discussion The extent to which social touch and pain shape subsequent interactions with virtual agents remains largely unknown. Here, we investigated how sensations evoked by virtual pain versus pleasant touch delivered on one’s own virtual body modulate evaluations of and nonverbal behaviour toward the touching agents. Participants experienced a full-body illusion, and observed agents deliver either a stab or a caress to their virtual right hand. We assessed changes in explicit impressions (trustworthiness, attractiveness), implicit evaluation via mouse-tracking, preferred interpersonal distance and physiological responses during the virtual agent’s approach. Stabbing agents were evaluated as less trustworthy, preferred at a greater interpersonal distance, and associated with stronger heart rate deceleration during their approach. A more complex picture emerged for caressing agents: here, it was not the caress per se, but its subjectively experienced (un)pleasantness that predicted higher trustworthiness ratings, reduced preferred interpersonal distance, and reduced heart rate deceleration during the virtual agent’s approach. In this sense, our findings refine the notion of a virtual Midas touch by showing that touch-related effects on social evaluation are not automatic, but depend on whether the somatosensory interaction is experienced as affiliative or aversive when enacted on an embodied virtual body. Our results on impression updating are consistent with the view that judgments of physical attractiveness are relatively resistant to change, whereas moral judgments (such as trustworthiness) are more plastic and experience-dependent (Glennon & Zeki, 2022 ). Previous work shows that trustworthiness can be rapidly updated when new information about an agent’s behaviour is encountered (Chang et al., 2010 ; Filkowski et al., 2016 ). Neuroimaging studies show that a distributed brain network underpins belief updating about social agents and integration of new social information into value representations (Mende-Siedlecki, 2018 ), supporting the idea that moral impressions are continuously revised as behaviourally diagnostic evidence accumulates. In our study, painful stimulations delivered by the stabbing agents affected subsequent evaluations independently of individual differences in reported (un)pleasantness, likely reflecting the fact that stabbing was perceived as uniformly unpleasant across participants. By contrast, the impact of caresses depended on their subjective pleasantness, consistent with the inherently more ambiguous and context-sensitive nature of affiliative touch. This asymmetry fits with the high salience of threat cues and the well-established negativity bias in impression formation, according to which negative information tends to be prioritized when forming or revising interpersonal impressions, especially when it signals potential harm (Baumeister et al., 2001 ; Skowronski & Carlston, 1989 ). In this sense, pain inflicted by an agent can serve as highly diagnostic behavioural information about that agent’s intentions, while touch is subjected to a more ambiguous interpretation. Our mouse-tracking results indicate that exposure to the touch task altered motor conflict in the numerical task. In congruent trials—where the correct response was mapped to the caressing agents and the competing option to the stabbing agents—we observed reduced maximum absolute deviation and area under the curve, suggesting more direct trajectories and less competition from the alternative option. Response times decreased from pre- to post-session for both congruent and incongruent trials, consistent with learning, but were also shorter for congruent than incongruent trials at post, indicating that the mapping in which caressers signalled the correct choice became relatively easier to resolve. While this mapping does not allow us to disentangle the separate contribution of approach tendencies toward caressers and avoidance tendencies toward stabbers, these effects suggest that virtual touch and pain might act as strong diagnostic behavioural information leading to update not only explicit, but also implicit evaluations, which are typically harder to change (Cone & Ferguson, 2015 ; Mann et al., 2019 ). Importantly, these effects emerged in a task that did not require any explicit evaluation of the agents, suggesting that vicarious touch and pain influenced social processing at an implicit level. In the interpersonal distance task, participants increased the preferred distance towards the stabbing agents. For the caressing agents, higher subjective touch pleasantness predicted a reduction in preferred distance, suggesting that pleasant social touch promotes approach when experienced as pleasant. Previous research indicates that perceived threat enlarges personal space (Abra et al., 2024 ; Cartaud, Quesque, et al., 2020; Holt et al., 2022 ; Lisi, Scattolin, et al., 2021 ). This pattern suggests that virtual agents who inflict pain are treated as threatening social partners, whereas affiliative agents gain proximity only when their tactile behaviour is interpreted as appropriate and benign. At the physiological level, these results were not paralleled by an increase of SCR, as both stabbing and caressing agents were associated with reduced SCR during approach compared to baseline, suggesting habituation of the initial orienting response. Previous studies have reported that angry virtual faces elicit higher SCR and larger preferred distances than neutral faces (Cartaud, Ott, et al., 2020 ), and that fearful expressions increase SCR, particularly at close distances relative to positive expressions (Ellena et al., 2020 ). In our case, the absence of SCR differences between conditions may reflect rapid habituation to repeated approaches by the same avatars or a more general arousal response to the approaching virtual body itself. Heart rate, by contrast, showed a deceleration during the subsequent stabbing agents’ approach, consistent with a freezing-like response to threat (Roelofs et al., 2010 ). Heart rate deceleration accompanied by relative bodily immobility has been repeatedly observed in healthy individuals exposed to physical threat and is interpreted as fear bradycardia—a defensive state that enhances vigilance while preparing the organism for rapid action (Azevedo et al., 2005 ; Roelofs et al., 2010 ). At the neurobiological level, such freezing responses are thought to involve projections from the amygdala to the periaqueductal gray (Applegate et al., 1983 ; Hermans et al., 2013 ). Recent VR work similarly reports lower heart rate when participants face avatars displaying aggressive postures compared to neutral ones (Mello, Dupont, et al., 2022 ) or when they face threatening and immoral characters such as criminals (Placidi et al., 2025 ). By contrast, when participants can actively escape or stop the approaching threat, heart rate acceleration is more likely, reflecting action preparation and active coping rather than passive freezing (Lu et al., 2023 ). It is worth noting that our task did not allow participants to avoid the approaching agents, which may have favoured bradycardia over acceleratory cardiac responses. Thus, heart rate deceleration in the present task is more consistent with a defensive, freezing-like response than with nonspecific autonomic arousal. Gaze to the eye region increased during the approach phase for both caressing and stabbing agents. For stabbing agents, this increase was also found in the early, far-distance phase. According to equilibrium theory, interpersonal distance and mutual gaze are jointly regulated indices of interpersonal intimacy that are adjusted to maintain an optimal level of comfort (Argyle & Dean, 1965 ; Bailenson et al., 2001 ). On this basis, we expected greater intimacy—reflected in more eye contact—towards caressing agents only. However, our data only partially supported this prediction, as increased eye contact was not restricted to caressing agents. Increased eye contact with stabbing agents could reflect functions other than intimacy signalling, such as face identification (Royer et al., 2018 ) or assessment of the agent’s intentions (Hietanen, 2018 ). More broadly, these results are consistent with the notion that direct gaze is intrinsically ambiguous (Kleinke, 1986 ): sustained eye contact can signal attraction (Hoffmann et al., 2024 ), but in other contexts it can signal dominance or threat (Ellsworth & Langer, 1976 ; Terburg et al., 2011 ). During the agents’ approach, rather than directly coding valence, gaze may convey the intensity and motivational relevance of the interaction. Across several outcome measures—trustworthiness, heart rate deceleration, and interpersonal distance—the positive effects of caressing agents in our study were not driven by the caress per se, but by how pleasant the touch was subjectively experienced. This fits with work showing that touch pleasantness is highly variable across individuals and contexts, rather than a fixed property of a given tactile stimulus (Saarinen et al., 2021 ). One important source of individual variability lies in internal states: longing for touch has been associated with increased pleasantness ratings of social touch, suggesting that internal “need” states bias hedonic appraisal (Abra et al., 2024 ; Meijer et al., 2022 ). More broadly, the experiential outcome of touch depends on whether the interaction matches the recipient’s expectations and current social needs (Sailer & Leknes, 2022 ). Importantly, we did not directly assess such individual differences (e.g., longing for touch, current social needs, or attitudes towards technology-mediated touch), which represents a limitation of the present study. Moreover, while previous studies (Fusaro et al., 2021 ; Mello, Fusaro, et al., 2022 ; Verga, Di Marco, et al., 2025 ) and anedoctal reports (Freeman et al., 2022 ; Wiederhold, 2022 ) suggest that merely vision of touch on one’s embodied avatar can trigger derivative tactile sensations, it is worth noting that (un)pleasantness effect might by boosted by a congruent haptic feedback (Maunsbach et al., 2023 ; Seinfeld et al., 2022 ; Smekal et al., 2025 ; Sun et al., 2024 ). Future studies should better investigate the interplay between individual longing for touch, attitudes towards technology-mediated touch, virtual agents’ characteristics, and haptic feedback quality in affecting touch reactions and subsequent interactions. Taken together, our findings show that how virtual agents touching or harming the embodied avatar of a human participant affects how they are subsequently perceived and approached, at both explicit and implicit levels. This suggests that, in immersive social spaces, agents’ tactile behaviours can function as diagnostic social signals that calibrate users’ feelings of safety, intimacy, and engagement. More broadly, these results provide an empirical basis for investigating how threatening and affiliative somatosensory cues contribute to more complex social processes, including moral decision-making in dynamic exchanges with virtual agents (Vabba, Lisi, et al., 2025 ). Declarations Funding M.P.L.: BEFORERC 2024, awarded by La Sapienza University of Rome. S.M.A.: Brain Magnet Program awarded by the Italian Institute of Technology. Author Contribution M.P.L., M.F., and S.M.A. conceived and designed research; M.P.L. and F.M. performed experiments; M.P.L., M.F., and F.M. analyzed data; M.P.L., M.F., and F.M. prepared the virtual reality set-up; M.P.L. drafted the manuscript; M.P.L., M.F., F.M., and S.M.A. edited, revised, and approved the final version of the manuscript. Data Availability Data and scripts used for the analyses are available at the following link: https://osf.io/8j7us/overview?view\_only=b4624112f87144f7880111d02d3762c2 References Abra Y, Mirams L, Fairhurst MT (2024) The space between us: The effect of perceived threat on discomfort distance and perceived pleasantness of interpersonal vicarious touch. Heliyon 10(16):e36487. https://doi.org/10.1016/j.heliyon.2024.e36487 Aglioti SM, Frisanco A, Lisi MP (2025) The body in the brain, in the age of the metaverse. Proceedings of the Paris Institute for Advanced Study . https://doi.org/10.5281/ZENODO.17127881 Applegate C, Kapp B, Underwood M, Mcnall C (1983) Autonomic and somatomotor effects of amygdala central N. stimulation in awake rabbits☆. Physiol Behav 31(3):353–360. https://doi.org/10.1016/0031-9384(83)90201-9 Argyle M, Dean J (1965) Eye-Contact, Distance and Affiliation. Sociometry 28(3):289. https://doi.org/10.2307/2786027 Azevedo TM, Volchan E, Imbiriba LA, Rodrigues EC, Oliveira JM, Oliveira LF, Lutterbach LG, Vargas CD (2005) A freezing-like posture to pictures of mutilation. Psychophysiology 42(3):255–260. https://doi.org/10.1111/j.1469-8986.2005.00287.x Bailenson JN, Blascovich J, Beall AC, Loomis JM (2001) Equilibrium Theory Revisited: Mutual Gaze and Personal Space in Virtual Environments. Presence: Teleoperators Virtual Environ 10(6):583–598. https://doi.org/10.1162/105474601753272844 Bailenson JN, Blascovich J, Beall AC, Loomis JM (2003) Interpersonal Distance in Immersive Virtual Environments. Pers Soc Psychol Bull 29(7):819–833. https://doi.org/10.1177/0146167203029007002 Barr DJ (2008) Analyzing ‘visual world’ eyetracking data using multilevel logistic regression. J Mem Lang 59(4):457–474. https://doi.org/10.1016/j.jml.2007.09.002 Baumeister RF, Bratslavsky E, Finkenauer C, Vohs KD (2001) Bad is Stronger than Good. Rev Gen Psychol 5(4):323–370. https://doi.org/10.1037/1089-2680.5.4.323 Benedek M, Kaernbach C (2010) A continuous measure of phasic electrodermal activity. J Neurosci Methods 190(1):80–91. https://doi.org/10.1016/j.jneumeth.2010.04.028 Bürkner P-C (2017) brms: An R Package for Bayesian Multilevel Models Using Stan . J Stat Softw 80(1). https://doi.org/10.18637/jss.v080.i01 Cafaro A, Vilhjálmsson HH, Bickmore T, Heylen D, Jóhannsdóttir KR, Valgarðsson GS (2012) First Impressions: Users’ Judgments of Virtual Agents’ Personality and Interpersonal Attitude in First Encounters. In Y. Nakano, M. Neff, A. Paiva, & M. Walker (A c. Di), Intelligent Virtual Agents (Vol. 7502, pp. 67–80). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-33197-8_7 Campbell JID, Thompson VA (2012) MorePower 6.0 for ANOVA with relational confidence intervals and Bayesian analysis. Behav Res Methods 44(4):1255–1265. https://doi.org/10.3758/s13428-012-0186-0 Candini M, Battaglia S, Benassi M, Di Pellegrino G, Frassinetti F (2021) The physiological correlates of interpersonal space. Sci Rep 11(1):2611. https://doi.org/10.1038/s41598-021-82223-2 Cartaud A, Ott L, Iachini T, Honoré J, Coello Y (2020) The influence of facial expression at perceptual threshold on electrodermal activity and social comfort distance. Psychophysiology 57(9):e13600. https://doi.org/10.1111/psyp.13600 Cartaud A, Quesque F, Coello Y (2020) Wearing a face mask against Covid-19 results in a reduction of social distancing. PLoS ONE 15(12):e0243023. https://doi.org/10.1371/journal.pone.0243023 Case LK, Madian N, McCall MV, Bradson ML, Liljencrantz J, Goldstein B, Alasha VJ, Zimmerman MS (2023) Aβ-CT Affective Touch: Touch Pleasantness Ratings for Gentle Stroking and Deep Pressure Exhibit Dependence on A-Fibers. Eneuro , 10 (5), ENEURO.0504-22.2023. https://doi.org/10.1523/ENEURO.0504-22.2023 Chang LJ, Doll BB, Van ’T, Wout M, Frank MJ, Sanfey AG (2010) Seeing is believing: Trustworthiness as a dynamic belief. Cogn Psychol 61(2):87–105. https://doi.org/10.1016/j.cogpsych.2010.03.001 Cihodaru-Ștefanache Ștefania, Podina IR (2025) The uncanny valley effect in embodied conversational agents: A critical systematic review of attractiveness, anthropomorphism, and uncanniness. Front Psychol 16:1625984. https://doi.org/10.3389/fpsyg.2025.1625984 Clay V, König P, König SU (2019) Eye tracking in virtual reality. J Eye Mov Res 12(1). https://doi.org/10.16910/jemr.12.1.3 Cone J, Ferguson MJ (2015) He did what? The role of diagnosticity in revising implicit evaluations. J Personal Soc Psychol 108(1):37–57. https://doi.org/10.1037/pspa0000014 Crusco AH, Wetzel CG (1984) The Midas Touch: The Effects of Interpersonal Touch on Restaurant Tipping. Pers Soc Psychol Bull 10(4):512–517. https://doi.org/10.1177/0146167284104003 Dink JW, Ferguson B (2015) eyetrackingR: An R Library for Eye-tracking Data Analysis. Available from http://www.eyetrackingr.com Ellena G, Battaglia S, Làdavas E (2020) The spatial effect of fearful faces in the autonomic response. Exp Brain Res 238(9):2009–2018. https://doi.org/10.1007/s00221-020-05829-4 Ellingsen D-M, Leknes S, Løseth G, Wessberg J, Olausson H (2016) The Neurobiology Shaping Affective Touch: Expectation, Motivation, and Meaning in the Multisensory Context. Frontiers in Psychology , 6 . https://doi.org/10.3389/fpsyg.2015.01986 Ellsworth PC, Langer EJ (1976) Staring and approach: An interpretation of the stare as a nonspecific activator. J Personal Soc Psychol 33(1):117–122. https://doi.org/10.1037/0022-3514.33.1.117 Faust NT, Chatterjee A, Christopoulos GI (2019) Beauty in the eyes and the hand of the beholder: Eye and hand movements’ differential responses to facial attractiveness. J Exp Soc Psychol 85:103884. https://doi.org/10.1016/j.jesp.2019.103884 Filkowski MM, Anderson IW, Haas BW (2016) Trying to trust: Brain activity during interpersonal social attitude change. Cogn Affect Behav Neurosci 16(2):325–338. https://doi.org/10.3758/s13415-015-0393-0 Freeman G, Zamanifard S, Maloney D, Acena D (2022) Disturbing the Peace: Experiencing and Mitigating Emerging Harassment in Social Virtual Reality. Proceedings of the ACM on Human-Computer Interaction , 6 (CSCW1), 1–30. https://doi.org/10.1145/3512932 Freeman JB, Ambady N (2010) MouseTracker: Software for studying real-time mental processing using a computer mouse-tracking method. Behav Res Methods 42(1):226–241. https://doi.org/10.3758/BRM.42.1.226 Fusaro M, Bufacchi RJ, Nicolardi V, Provenzano L (2022) The analgesic power of pleasant touch in individuals with chronic pain: Recent findings and new insights. Front Integr Nuerosci 16:956510. https://doi.org/10.3389/fnint.2022.956510 Fusaro M, Fanti V, Chakrabarti B (2023) Greater interpersonal distance in adults with autism. Autism Res 16(10):2002–2007. https://doi.org/10.1002/aur.3013 Fusaro M, Lisi MP, Era V, Porciello G, Candidi M, Aglioti SM, Boukarras S (2025) The Transformative Power of Virtual Reality: Redefining Interactions in Virtual Platforms, Education, Healthcare, and Workplaces. Topoi . https://doi.org/10.1007/s11245-025-10216-1 Fusaro M, Lisi MP, Tieri G, Aglioti SM (2021) Heterosexual, gay, and lesbian people’s reactivity to virtual caresses on their embodied avatars’ taboo zones. Scientific Reports , 11 (1), Articolo 1. https://doi.org/10.1038/s41598-021-81168-w Gazzola V, Spezio ML, Etzel JA, Castelli F, Adolphs R, Keysers C (2012) Primary somatosensory cortex discriminates affective significance in social touch. Proceedings of the National Academy of Sciences , 109 (25). https://doi.org/10.1073/pnas.1113211109 Glennon MJ, Zeki S (2022) The power of external influences to modify judgments of facial and moral beauty. PsyCh J 11(5):707–714. https://doi.org/10.1002/pchj.492 Gonzalez-Franco M, Ofek E, Pan Y, Antley A, Steed A, Spanlang B, Maselli A, Banakou D, Pelechano N, Orts-Escolano S, Orvalho V, Trutoiu L, Wojcik M, Sanchez-Vives MV, Bailenson J, Slater M, Lanier J (2020) The Rocketbox Library and the Utility of Freely Available Rigged Avatars. Front Virtual Real 1:561558. https://doi.org/10.3389/frvir.2020.561558 González-Franco M, Peck TC, Rodríguez-Fornells A, Slater M (2014) A threat to a virtual hand elicits motor cortex activation. Experimental Brain Research , 232 (3), Articolo 3. https://doi.org/10.1007/s00221-013-3800-1 Haans A, De Bruijn R, IJsselsteijn WA (2014) A Virtual Midas Touch? Touch, Compliance, and Confederate Bias in Mediated Communication. J Nonverbal Behav 38(3):301–311. https://doi.org/10.1007/s10919-014-0184-2 Harjunen VJ, Spapé M, Ahmed I, Jacucci G, Ravaja N (2018) Persuaded by the machine: The effect of virtual nonverbal cues and individual differences on compliance in economic bargaining. Comput Hum Behav 87:384–394. https://doi.org/10.1016/j.chb.2018.06.012 Hensel LB, Zhan J, Bjornsdottir RT, Garrod OGB, Schyns PG, Jack RE (2020) Psychologically Valid Social Face Features for Virtual Agents. Proceedings of the 20th ACM International Conference on Intelligent Virtual Agents , 1–3. https://doi.org/10.1145/3383652.3423899 Hermans EJ, Henckens MJAG, Roelofs K, Fernández G (2013) Fear bradycardia and activation of the human periaqueductal grey. NeuroImage 66:278–287. https://doi.org/10.1016/j.neuroimage.2012.10.063 Hietanen JK (2018) Affective Eye Contact: An Integrative Review. Front Psychol 9:1587. https://doi.org/10.3389/fpsyg.2018.01587 Hoffmann A, Schiestl S, Sinske P, Gondan M, Sachse P, Maran T (2024) Sharing and Receiving Eye-Contact Predicts Mate Choice After a 5-Minute Conversation: Evidence from a Speed-Dating Study. Arch Sex Behav 53(3):959–968. https://doi.org/10.1007/s10508-023-02806-0 Holt DJ, Zapetis SL, Babadi B, Zimmerman J, Tootell RBH (2022) Personal space increases during the COVID-19 pandemic in response to real and virtual humans. Front Psychol 13:952998. https://doi.org/10.3389/fpsyg.2022.952998 Iachini T, Coello Y, Frassinetti F, Senese VP, Galante F, Ruggiero G (2016) Peripersonal and interpersonal space in virtual and real environments: Effects of gender and age. J Environ Psychol 45:154–164. https://doi.org/10.1016/j.jenvp.2016.01.004 Keshava A, Aumeistere A, Izdebski K, Konig P (2020) Decoding Task From Oculomotor Behavior In Virtual Reality. ACM Symposium on Eye Tracking Research and Applications , 1–5. https://doi.org/10.1145/3379156.3391338 Kleinke CL (1986) Gaze and eye contact: A research review. Psychol Bull 100(1):78–100. https://doi.org/10.1037/0033-2909.100.1.78 Kret ME, Sjak-Shie EE (2019) Preprocessing pupil size data: Guidelines and code. Behav Res Methods 51(3):1336–1342. https://doi.org/10.3758/s13428-018-1075-y Kroczek LOH, Pfaller M, Lange B, Müller M, Mühlberger A (2020) Interpersonal Distance During Real-Time Social Interaction: Insights From Subjective Experience, Behavior, and Physiology. Front Psychiatry 11:561. https://doi.org/10.3389/fpsyt.2020.00561 Kruschke JK, Liddell TM (2018) The Bayesian New Statistics: Hypothesis testing, estimation, meta-analysis, and power analysis from a Bayesian perspective. Psychon Bull Rev 25(1):178–206. https://doi.org/10.3758/s13423-016-1221-4 Lee Y-H, Xiao M, Wells RH (2018) The Effects of Avatars’ Age on Older Adults’ Self-Disclosure and Trust. Cyberpsychology Behav Social Netw 21(3):173–178. https://doi.org/10.1089/cyber.2017.0451 Lenth R, Piaskowski J (2025) Estimated Marginal Means, aka Least-Squares Means . https://rvlenth.github.io/emmeans/ Lisi MP, Fusaro M, Aglioti SM (2024) Visual perspective and body ownership modulate vicarious pain and touch: A systematic review. Psychon Bull Rev. https://doi.org/10.3758/s13423-024-02477-5 Lisi MP, Fusaro M, Tieri G, Aglioti SM (2021) Humans adjust virtual comfort-distance towards an artificial agent depending on their sexual orientation and implicit prejudice against gay men. Comput Hum Behav 125:106948. https://doi.org/10.1016/j.chb.2021.106948 Lisi MP, Giometti A, Fusaro M, Aglioti SM (2025) Surrogate bias: Behavioral and physiological effects of seeing ingroup and outgroup virtual agents touching one’s embodied avatar Lisi MP, Scattolin M, Fusaro M, Aglioti SM (2021) A Bayesian approach to reveal the key role of mask wearing in modulating projected interpersonal distance during the first COVID-19 outbreak. PLoS ONE 16(8):e0255598. https://doi.org/10.1371/journal.pone.0255598 Lu J, Kemmerer SK, Riecke L, De Gelder B (2023) Early threat perception is independent of later cognitive and behavioral control. A virtual reality-EEG-ECG study. Cereb Cortex 33(13):8748–8758. https://doi.org/10.1093/cercor/bhad156 Ma HH, Lee M, Kim E, Pang S (2025) The Effects of Virtual Social Touch on Interpersonal Communication Outcomes in Virtual Environments: A Meta-Analysis. Int J Human–Computer Interact 1–17. https://doi.org/10.1080/10447318.2025.2526590 Makowski D, Ben-Shachar MS, Chen SHA, Lüdecke D (2019) Indices of Effect Existence and Significance in the Bayesian Framework. Front Psychol 10:2767. https://doi.org/10.3389/fpsyg.2019.02767 Makowski D, Ben-Shachar MS, Lüdecke D (2019) bayestestR: Describing Effects and their Uncertainty, Existence and Significance within the Bayesian Framework. J Open Source Softw 4(40):1541. https://doi.org/10.21105/joss.01541 Makransky G, Wismer P, Mayer RE (2019) A gender matching effect in learning with pedagogical agents in an immersive virtual reality science simulation. J Comput Assist Learn 35(3):349–358. https://doi.org/10.1111/jcal.12335 Mann TC, Cone J, Heggeseth B, Ferguson MJ (2019) Updating implicit impressions: New evidence on intentionality and the affect misattribution procedure. J Personal Soc Psychol 116(3):349–374. https://doi.org/10.1037/pspa0000146 Maselli A, Slater M (2013) The building blocks of the full body ownership illusion. Frontiers in Human Neuroscience , 7 . https://doi.org/10.3389/fnhum.2013.00083 Maunsbach M, Hornbæk K, Seifi H (2023) Mediated Social Touching: Haptic Feedback Affects Social Experience of Touch Initiators. 2023 IEEE World Haptics Conference (WHC) , 93–100. https://doi.org/10.1109/WHC56415.2023.10224506 Meijer LL, Hasenack B, Kamps JCC, Mahon A, Titone G, Dijkerman HC, Keizer A (2022) Affective touch perception and longing for touch during the COVID-19 pandemic. Sci Rep 12(1):3887. https://doi.org/10.1038/s41598-022-07213-4 Mello M, Dupont L, Engelen T, Acciarino A, De Borst AW, De Gelder B (2022) The influence of body expression, group affiliation and threat proximity on interactions in virtual reality. Curr Res Behav Sci 3:100075. https://doi.org/10.1016/j.crbeha.2022.100075 Mello M, Fusaro M, Aglioti SM, Minio-Paluello I (2024) Exploring social touch in autistic and non-autistic adults via a self-report body-painting task: The role of sex, social context and body area. Autism 28(8):1985–2001. https://doi.org/10.1177/13623613231218314 Mello M, Fusaro M, Aglioti SM, Minio-Paluello I (2025) Centrality of Touch Avoidance in Social Touch Experiences in Autism. J Autism Dev Disord. https://doi.org/10.1007/s10803-025-07053-x Mello M, Fusaro M, Tieri G, Aglioti SM (2022) Wearing same- and opposite-sex virtual bodies and seeing them caressed in intimate areas. Quarterly Journal of Experimental Psychology , 75 (3), Articolo 3. https://doi.org/10.1177/17470218211031557 Mende-Siedlecki P (2018) Changing our minds: The neural bases of dynamic impression updating. Curr Opin Psychol 24:72–76. https://doi.org/10.1016/j.copsyc.2018.08.007 Mull I, Wyss J, Moon E, Lee S-E (2015) An exploratory study of using 3D avatars as online salespeople: The effect of avatar type on credibility, homophily, attractiveness and intention to interact. J Fashion Mark Manage 19(2):154–168. https://doi.org/10.1108/JFMM-05-2014-0033 Nicolardi V, Lisi MP, Mello M, Fusaro M, Tieri G, Aglioti SM (2025) Taking an embodied avatar’s perspective modulates the temporal dynamics of vicarious pain and pleasure: A Virtual Reality and EEG study. Soc Cognit Affect Neurosci nsaf035. https://doi.org/10.1093/scan/nsaf035 Nolte D, Hjoj R, Sánchez Pacheco T, Huang A, König P (2025) Investigating proxemics behaviors towards individuals, pairs, and groups in virtual reality. Virtual Reality 29(2):58. https://doi.org/10.1007/s10055-025-01127-y Novembre G, Etzi R, Morrison I (2021) Hedonic Responses to Touch are Modulated by the Perceived Attractiveness of the Caresser. Neuroscience 464:79–89. https://doi.org/10.1016/j.neuroscience.2020.10.007 Oosterhof NN, Todorov A (2008) The functional basis of face evaluation. Proceedings of the National Academy of Sciences , 105 (32), 11087–11092. https://doi.org/10.1073/pnas.0805664105 Palan S, Schitter C (2018) Prolific.ac—A subject pool for online experiments. J Behav Experimental Finance 17:22–27. https://doi.org/10.1016/j.jbef.2017.12.004 Pandhare A, Skulikj D, Lee J, Mikhiel P, Banakou D (2025) vER: Virtual Human Companionship for Patients in Preoperative Care. Proceedings of the 2025 ACM International Conference on Interactive Media Experiences , 468–472. https://doi.org/10.1145/3706370.3731639 Peck TC, Gonzalez-Franco M (2021) Avatar Embodiment. A Standardized Questionnaire. Front Virtual Real 1:575943. https://doi.org/10.3389/frvir.2020.575943 Pedrabissi L, Santinello M (1989) STAI state-trait anxiety inventory forma Y Manuale. Organizzazioni Speciali Perry A, Rubinsten O, Peled L, Shamay-Tsoory SG (2013) Don’t stand so close to me: A behavioral and ERP study of preferred interpersonal distance. NeuroImage 83:761–769. https://doi.org/10.1016/j.neuroimage.2013.07.042 Placidi V, Lisi MP, Vabba A, Frisanco A, Aglioti SM (2025) Virtual embodiment of civic heroes preserves conflict about anti-mafia intentions and enhances physiological discomfort in response to mafia-related crime . PsyArXiv. https://doi.org/10.31234/osf.io/va2k5_v1 Roelofs K, Hagenaars MA, Stins J (2010) Facing Freeze: Social Threat Induces Bodily Freeze in Humans. Psychol Sci 21(11):1575–1581. https://doi.org/10.1177/0956797610384746 Royer J, Blais C, Charbonneau I, Déry K, Tardif J, Duchaine B, Gosselin F, Fiset D (2018) Greater reliance on the eye region predicts better face recognition ability. Cognition 181:12–20. https://doi.org/10.1016/j.cognition.2018.08.004 Saarinen A, Harjunen V, Jasinskaja-Lahti I, Jääskeläinen IP, Ravaja N (2021) Social touch experience in different contexts: A review. Neurosci Biobehavioral Reviews 131:360–372. https://doi.org/10.1016/j.neubiorev.2021.09.027 Sailer U, Friedrich Y, Asgari F, Hassenzahl M, Croy I (2024) Determinants for positive and negative experiences of interpersonal touch: Context matters. Cogn Emot 38(4):565–586. https://doi.org/10.1080/02699931.2024.2311800 Sailer U, Leknes S (2022) Meaning makes touch affective. Curr Opin Behav Sci 44:101099. https://doi.org/10.1016/j.cobeha.2021.101099 Saini G, Zain M, Noronha M, Bonin RP, Lomanowska AM (2024) Virtually simulated interpersonal touch negatively affects perceived closeness and social affiliation to an avatar partner. Sci Rep 14(1):1347. https://doi.org/10.1038/s41598-024-51773-6 Schirmer A, Croy I, Ackerley R (2023) What are C-tactile afferents and how do they relate to affective touch? Neurosci Biobehavioral Reviews 151:105236. https://doi.org/10.1016/j.neubiorev.2023.105236 Schuetz I, Fiehler K (2022) Eye tracking in virtual reality: Vive pro eye spatial accuracy, precision, and calibration reliability. J Eye Mov Res 15(3). https://doi.org/10.16910/jemr.15.3.3 Seger CR, Smith ER, Percy EJ, Conrey FR (2014) Reach Out and Reduce Prejudice: The Impact of Interpersonal Touch on Intergroup Liking. Basic Appl Soc Psychol 36(1):51–58. https://doi.org/10.1080/01973533.2013.856786 Seinfeld S, Schmidt I, Müller J (2022) Evoking realistic affective touch experiences in virtual reality (arXiv:2202.13389). arXiv. http://arxiv.org/abs/2202.13389 Skowronski JJ, Carlston DE (1989) Negativity and extremity biases in impression formation: A review of explanations. Psychol Bull 105(1):131–142. https://doi.org/10.1037/0033-2909.105.1.131 Slater M, Spanlang B, Sanchez-Vives MV, Blanke O (2010) First Person Experience of Body Transfer in Virtual Reality. PLoS ONE 5(5):e10564. https://doi.org/10.1371/journal.pone.0010564 Smekal V, Hecquard J, Kühne S, Occidental N, Lécuyer A, Macé M, De Gelder B (2025) Influence of Haptic Feedback on Perception of Threat and Peripersonal Space in Social VR. IEEE Trans Vis Comput Graph 31(5):2986–2994. https://doi.org/10.1109/TVCG.2025.3549884 Spielberger CD, Gorsuch R, Lushene R, Vagg PR, Jacobs GA (1983) Manual for the state-trait anxiety inventory. Consulting Psychologists Stillman PE, Shen X, Ferguson MJ (2018) How Mouse-tracking Can Advance Social Cognitive Theory. Trends Cogn Sci 22(6):531–543. https://doi.org/10.1016/j.tics.2018.03.012 Sun W, Banakou D, Świdrak J, Valori I, Slater M, Fairhurst MT (2024) Multisensory experiences of affective touch in virtual reality enhance engagement, body ownership, pleasantness, and arousal modulation. Virtual Reality 28(4):162. https://doi.org/10.1007/s10055-024-01056-2 Suvilehto JT, Cekaite A, Morrison I (2023) The why, who and how of social touch. Nat Reviews Psychol 2(10):606–621. https://doi.org/10.1038/s44159-023-00217-5 Suvilehto JT, Glerean E, Dunbar RIM, Hari R, Nummenmaa L (2015) Topography of social touching depends on emotional bonds between humans. Proceedings of the National Academy of Sciences , 112 (45), 13811–13816. https://doi.org/10.1073/pnas.1519231112 Suvilehto JT, Nummenmaa L, Harada T, Dunbar RIM, Hari R, Turner R, Sadato N, Kitada R (2019) Cross-cultural similarity in relationship-specific social touching. Proceedings of the Royal Society B: Biological Sciences , 286 (1901), 20190467. https://doi.org/10.1098/rspb.2019.0467 Suvilehto JT, Renvall V, Nummenmaa L (2021) Relationship-specific Encoding of Social Touch in Somatosensory and Insular Cortices. Neuroscience 464:105–116. https://doi.org/10.1016/j.neuroscience.2020.09.015 Świdrak J, Pochwatko G, Insabato A (2021) Does an agent’s touch always matter? Study on virtual Midas touch, masculinity, social status, and compliance in Polish men. J Multimodal User Interfaces 15(2):163–174. https://doi.org/10.1007/s12193-020-00351-x Świdrak J, Pochwatko G, Navarro X, Osęka L, Doliński D (2020) The joint influence of social status and personal attitudes in a contact and open versus a noncontact and homophobic culture on the virtual Midas touch. Virtual Reality 24(4):619–633. https://doi.org/10.1007/s10055-019-00423-8 Ter Stal S, Tabak M, Akker OD, Beinema H, T., Hermens H (2020) Who Do You Prefer? The Effect of Age, Gender and Role on Users’ First Impressions of Embodied Conversational Agents in eHealth. Int J Human–Computer Interact 36(9):881–892. https://doi.org/10.1080/10447318.2019.1699744 Terburg D, Hooiveld N, Aarts H, Kenemans JL, Van Honk J (2011) Eye Tracking Unconscious Face-to-Face Confrontations: Dominance Motives Prolong Gaze to Masked Angry Faces. Psychol Sci 22(3):314–319. https://doi.org/10.1177/0956797611398492 Vabba A, Lisi MP, Placidi V, Provenzano L, Zahid MN, Aglioti SM (2025) An immersive virtual reality (IVR) and neuroscience-based approach to study and change attitudes and behaviour towards mafia-type organized crime . PsyArXiv. https://doi.org/10.31234/osf.io/sbnxg_v1 Vabba A, Placidi V, Lisi MP, Aglioti SM (2025) Where the Proteus effect stops – embodying fictional mafia criminals leads to paradoxical effects on implicit attitudes towards organized crime . PsyArXiv. https://doi.org/10.31234/osf.io/hcyt8_v1 Verga C, Di Marco S, Giove F, Aglioti SM, Lisi MP (2025) Temporal dynamics of full-body ownership and vicarious touch sensations as inferred from embodying a virtual avatar. Exp Brain Res 243(4):103. https://doi.org/10.1007/s00221-025-07055-2 Verga C, Lisi MP, Aglioti SM, Fusaro M (2025) Translating, adapting and validating the Touch Experiences and Attitudes Questionnaire (TEAQ) for the Italian population . https://doi.org/10.31234/osf.io/4scwh_v2 Wiederhold BK (2022) Sexual Harassment in the Metaverse. Cyberpsychology, Behavior, and Social Networking , 25 (8), Articolo 8. https://doi.org/10.1089/cyber.2022.29253.editorial Wieser MJ, Pauli P, Grosseibl M, Molzow I, Mühlberger A (2010) Virtual Social Interactions in Social Anxiety—The Impact of Sex, Gaze, and Interpersonal Distance. Cyberpsychology, Behavior, and Social Networking , 13 (5), 547–554. https://doi.org/10.1089/cyber.2009.0432 Wulff DU, Kieslich PJ, Henninger F, Haslbeck JMB, Schulte-Mecklenbeck M (2025) Movement tracking of psychological processes: A tutorial using mousetrap. Behav Res Methods 57(11):307. https://doi.org/10.3758/s13428-025-02695-2 Yang F-C, Acevedo P, Guo S, Choi M, Mousas C (2025) Embodied Conversational Agents in Extended Reality: A Systematic Review. IEEE Access 13:79805–79824. https://doi.org/10.1109/ACCESS.2025.3566698 Zhu S, Hu W, Li W, Dong Y (2024) Virtual Agents in Immersive Virtual Reality Environments: Impact of Humanoid Avatars and Output Modalities on Shopping Experience. Int J Human–Computer Interact 40(19):5771–5793. https://doi.org/10.1080/10447318.2023.2241293 Additional Declarations No competing interests reported. Supplementary Files SITouchingAgentsEvaluation2.2.26.docx Cite Share Download PDF Status: Posted Version 1 posted 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-8765445","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":587740342,"identity":"9507da75-1954-4b65-a2a6-a5630dcf115b","order_by":0,"name":"Matteo P. Lisi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIiWNgGAWjYBADGTYGxsYHDAwSYN4BBgN8ipkZG4AkD1BLswFUCyNQCz49UC1AzCYBEzrAgMca3fbzxx98qGDg4eM/3FbNu8Min1/s8IMDDAV/cGoxO5PM2DjjDNBhEoltt3nPSFjOnJ1mgNdhZgeSGZt520BaGIFa2iQMDG4nENBy/jFj81+QFv6DbcUQLekf8Gu5AbSFEaSFIbGNGaIlh4AtNx4bzuw5IwHyS7PkXKAWydk5BQcSDIzxOCzxwYcfFTZy8v3HH35421ZnwC+dvvnDhz9yOLVAgQQaP4GQhlEwCkbBKBgFeAEAreZNm71yJS8AAAAASUVORK5CYII=","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":true,"prefix":"","firstName":"Matteo","middleName":"P.","lastName":"Lisi","suffix":""},{"id":587740343,"identity":"69539455-69f6-426f-a189-ffc05cbf6071","order_by":1,"name":"Martina Fusaro","email":"","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Martina","middleName":"","lastName":"Fusaro","suffix":""},{"id":587740344,"identity":"dee21eaa-4773-4208-99a5-f76b67ec8820","order_by":2,"name":"Francesca March","email":"","orcid":"","institution":"University of Camerino","correspondingAuthor":false,"prefix":"","firstName":"Francesca","middleName":"","lastName":"March","suffix":""},{"id":587740345,"identity":"66a5d39b-34e5-4b0f-9b23-547d422fbc02","order_by":3,"name":"Salvatore Maria Aglioti","email":"","orcid":"","institution":"Sapienza University of Rome","correspondingAuthor":false,"prefix":"","firstName":"Salvatore","middleName":"Maria","lastName":"Aglioti","suffix":""}],"badges":[],"createdAt":"2026-02-02 13:23:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8765445/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8765445/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102250603,"identity":"1e516fb6-951d-4a7c-b61a-211b2cf247fc","added_by":"auto","created_at":"2026-02-09 19:24:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":986932,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTouch task.\u003c/strong\u003e Participants observed a virtual body from a 1PP sitting on a chair with the right hand placed on a desk. At the beginning of the trial, a virtual agent appeared standing on the right side of the participant’s virtual body and either stabbed (A, B) or caressed (C, D) the participant’s virtual right hand.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/31313798cb94644503577b01.png"},{"id":102250602,"identity":"bdac9f6c-cfd2-4b7d-9072-687add5048d6","added_by":"auto","created_at":"2026-02-09 19:24:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":856276,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eInterpersonal distance task\u003c/strong\u003e. Participants observed the virtual agent appearing at 2.25 in front of them. After 2 seconds, the virtual agent walked toward the participant and stopped at 0.25 m and kept standing close until the end of the trial. Participants had to press the trigger button on the right controller when they first felt the virtual agent’s approach as uncomfortable. SCR, HR, and gaze were recorded throughout the trial.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/eeb5195251cb9ba4fcd88d55.png"},{"id":102250606,"identity":"b93c7a82-3c48-4427-94aa-06e917376c9d","added_by":"auto","created_at":"2026-02-09 19:24:42","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":157337,"visible":true,"origin":"","legend":"\u003cp\u003eMouse-tracking task. Participants were instructed to click the number that was closer to the basis number. In this trial, participants should click on the number closer to 24, which is 22.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/5a82bd67ec8ed4e32f06d7d6.png"},{"id":102250622,"identity":"9483d954-030b-4e50-a2a9-93b1fb34b605","added_by":"auto","created_at":"2026-02-09 19:24:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":178066,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between trustworthiness updating and of touch (un)Pleasantness. Higher pleasantness for the caress predicted a positive increase in trustworthiness towards the caressing agents.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/a1a120ab21e276b9e9329c55.png"},{"id":102250607,"identity":"61c463e0-2aa9-4547-8251-cebb9051820e","added_by":"auto","created_at":"2026-02-09 19:24:42","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":268453,"visible":true,"origin":"","legend":"\u003cp\u003eMaximum Absolute Deviation, an index of motor conflict in the mouse-tracking task, decreased after the Touch task in the Congruent condition, in which caressing agents were paired with the correct response option.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/2782b258e485c74edb09cfe2.png"},{"id":102250605,"identity":"64ac5df4-3f19-41fd-b85a-bd5e6bed911d","added_by":"auto","created_at":"2026-02-09 19:24:42","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":191582,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/07ffcdcc37964a20a5110761.png"},{"id":102745206,"identity":"78e798b2-f91d-4289-8bfe-dc56e26e4a3a","added_by":"auto","created_at":"2026-02-16 08:43:40","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":328928,"visible":true,"origin":"","legend":"\u003cp\u003eThe plot shows the proportion of gaze directed to the virtual agent’s eye region. Eye-region gaze increased during the approach phase for both the caressing and stabbing agents, and for the stabbing agent during the standing-far phase.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/ec49f86fafb9e115afebd7eb.png"},{"id":102750164,"identity":"7559e333-17c9-4ccd-9ea0-b4a82af6bf5b","added_by":"auto","created_at":"2026-02-16 09:17:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3694703,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/f297636b-b9ae-4ffd-8762-9bd0ce4154d8.pdf"},{"id":102250608,"identity":"1fcefc03-fc58-4e6a-85e0-104d56bb9d12","added_by":"auto","created_at":"2026-02-09 19:24:42","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":533535,"visible":true,"origin":"","legend":"","description":"","filename":"SITouchingAgentsEvaluation2.2.26.docx","url":"https://assets-eu.researchsquare.com/files/rs-8765445/v1/0fe82c8dbb5913cc0310fa1c.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Vicarious experience modulates implicit and explicit evaluation of virtual agents delivering pain and touch over an embodied avatar","fulltext":[{"header":"Introduction","content":"\u003cp\u003eImmersive Virtual Reality (IVR) enables users to experience a compelling illusion of being present in a virtual environment and to perceive a virtual body as their own (Aglioti et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Maselli \u0026amp; Slater, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Slater et al., \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). This embodied experience allows researchers to examine how observing tactile or threatening stimuli on one\u0026rsquo;s virtual body shapes cognition, emotion, and social behaviour. Previous research has shown that witnessing a painful event, such as a stab, delivered to a virtual body evokes unpleasant feelings and elevates physiological arousal (Gonz\u0026aacute;lez-Franco et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Lisi et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Nicolardi et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Conversely, positive somatosensory stimuli, such as virtual caresses applied to embodied avatars, induce pleasant sensations (Fusaro et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Mello, Fusaro, et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Verga, Di Marco, et al., \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), with effects varying with body site and identity of the agent delivering touch (Fusaro et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Lisi et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Mello, Fusaro, et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). What remains unclear is whether vicarious somatosensation shapes impressions concerning the virtual agent who touches or harms us.\u003c/p\u003e \u003cp\u003eInterindividual physical contact is a core channel for social affiliation, bonding, and well-being (Fusaro et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Suvilehto et al., \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The experience of touch reflects an interplay of bottom-up, mechanical factors and top-down, cognitive appraisals. Work on affective touch highlighted the role of C-tactile (CT) afferents in encoding gentle stroking, while recognizing that affective touch is not reducible to CT input (Case et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Schirmer et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and remains strongly context-dependent (Ellingsen et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Mello et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Sailer et al., \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) and influenced by individual differences (Mello et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Verga, Lisi, et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTop-down social factors further modulate touch perception (Saarinen et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Perceived appropriateness increases linearly with relationship strength, and touch from women is generally perceived as more acceptable (Suvilehto et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2015\u003c/span\u003e, \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). These priors shape not only explicit responses but also somatosensory cortex activity (Gazzola et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Suvilehto et al., \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In a similar vein, attractiveness cues influence perceived pleasantness of touch. For example, attractive faces observed while receiving stroking were associated with increased touch pleasantness, particularly for the slow touch condition (Novembre et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This was also associated with increased heart rate variability, possibly signalling a positive reaction at the autonomic level.\u003c/p\u003e \u003cp\u003eBy contrast, evidence that touch changes attitudes or behavior is mixed. Research on the \u003cem\u003eMidas touch\u003c/em\u003e phenomenon suggests that even minimal interpersonal contact can increase compliance and positive evaluations of the toucher (Crusco \u0026amp; Wetzel, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e1984\u003c/span\u003e; Seger et al., \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), and that similar effects can extend to mediated and virtual interactions (\u003cem\u003evirtual Midas touch\u003c/em\u003e; (Haans et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Harjunen et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Świdrak et al., \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, more recent work indicates that these beneficial effects are far from universal and critically depend on contextual and relational factors, such as perceived appropriateness, naturalness, and credibility of the touching agent, with several studies reporting null or even negative effects of virtual social touch (Ma et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Saini et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Świdrak et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHarmful actions directed at one\u0026rsquo;s body provide highly diagnostic cues about an agent\u0026rsquo;s intentions, whereas affiliative touch is inherently more ambiguous and context-sensitive. This asymmetry suggests that pain and pleasant touch may differentially shape trust, avoidance, and approach tendencies toward the agent, particularly when such interactions are experienced as happening to one\u0026rsquo;s own embodied self. Understanding the reactivity to a virtual toucher is crucial in immersive digital spaces, where human-virtual agent interactions are increasingly common in contexts such as healthcare (Fusaro et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Pandhare et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), sales (Mull et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Zhu et al., \u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), education (Makransky et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and gaming (Hensel et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and where first impressions drive engagement (Lee et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Ter Stal et al., \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Yang et al., \u003cspan citationid=\"CR114\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Virtual agents are often perceived and judged using the same heuristics applied to humans: people form rapid first impressions and evaluate agents along aesthetic (attractiveness) and moral (trustworthiness) dimensions, much as they do with human partners (Cafaro et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Cihodaru-Ștefanache \u0026amp; Podina, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Oosterhof \u0026amp; Todorov, \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). People also tend to regulate their nonverbal behavior accordingly, including interpersonal distance (IPD) - the preferred physical distance from others- which serves as a sensitive nonverbal index of approach-avoidance (Bailenson et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Nolte et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In both real and virtual contexts, IPD tracks perceived social comfort (Candini et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Fusaro et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Iachini et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Lisi, Fusaro, et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Moreover, perceived violations of preferred IPD responses are usually associated with increased arousal (Candini et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Kroczek et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Perry et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Placidi et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Vabba, Placidi, et al., \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Wieser et al., \u003cspan citationid=\"CR112\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) and gaze aversion (Argyle \u0026amp; Dean, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1965\u003c/span\u003e; Bailenson et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). Virtual agents can thus be seen as a specific class of artificial social partners that can adopt a human-like appearance and use both verbal and nonverbal signals for communication. However, it is still unclear whether and how first impressions and nonverbal behaviours such as IPD are updated based on bodily interactions with them.\u003c/p\u003e \u003cp\u003e\u003cb\u003eThe present study.\u003c/b\u003e We test whether vicarious experience of pain vs. pleasant touch delivered over one\u0026rsquo;s embodied avatar updates the social encoding and subsequent nonverbal interactions with the touching agent. Participants experienced a full-body illusion in IVR and then observed different agents deliver either a painful (stab) or pleasant (caress) stimulus to their avatar\u0026rsquo;s hand. We assessed pre- to post-changes in explicit judgments (trustworthiness, attractiveness), implicit evaluation of facial attractiveness (mouse-tracking), nonverbal and physiological (skin conductance responses, heart rate, and gaze) responses during IPD regulation with each agent. This design allows us to ask whether the type of virtual somatosensation shifts both explicit appraisals and implicit approach\u0026ndash;avoidance toward the sender, advancing our understanding of embodied interactions with virtual agents. We expected that trustworthiness and attractiveness would increase for the caressers and decrease for the stabbers, and we anticipated that these effects might be moderated by subjective touch (un)pleasantness. Physiologically, we expected reduced autonomic arousal (lower skin conductance responses and heart rate) during the subsequent approach of the caressers, and an increased arousal for the stabbers. Concerning gaze, we predicted a threat-monitoring shift following stabbing interactions, such that visual attention would be reallocated away from the agent\u0026rsquo;s eyes toward action-relevant regions (e.g., the moving hand/effector) whereas pleasant touch would be associated with relatively greater eye-region engagement during subsequent interactions.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003e40 neurotypical, right-handed, cisgender, heterosexual participants took part in the study (20 women, mean age\u0026thinsp;=\u0026thinsp;24.38, sd\u0026thinsp;=\u0026thinsp;2.05, age range\u0026thinsp;=\u0026thinsp;20\u0026ndash;30; for all demographics see Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). An a priori power analysis conducted in MorePower (Campbell \u0026amp; Thompson, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) indicated that N\u0026thinsp;=\u0026thinsp;40 was required to detect an effect of Cohen\u0026rsquo;s \u003cem\u003ef\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.46 with 80% power at α\u0026thinsp;=\u0026thinsp;0.05. Anxiety was measured by means of the State-Trait Anxiety Inventory (STAI; Spielberger et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e1983\u003c/span\u003e), Italian version (Pedrabissi \u0026amp; Santinello, \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e1989\u003c/span\u003e). Prior VR experience was assessed with a single item (\u0026ldquo;How much experience have you had with Immersive Virtual Reality so far?\u0026rdquo;) rated on a 0-100 VAS scale. All participants gave written consent to take part in this study. The protocol was approved by the Santa Lucia Foundation Committee (protocol CE/2022_012) and complied with the Declaration of Helsinki (2013).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eExperimental stimuli and setup\u003c/h3\u003e\n\u003cp\u003eThe virtual scenario was designed using Blender 2.9 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.blender.org/\u003c/span\u003e\u003cspan address=\"https://www.blender.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e)\u003c/span\u003e and implemented in Unity v2020.3.11f (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://unity.com/\u003c/span\u003e\u003cspan address=\"https://unity.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e).\u003c/span\u003e Two female and two male avatars were selected among a pool extracted from the Rocketbox Library (Gonzalez-Franco et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), based on their perceived age, gender, and ethnicity (Tab.S3). These attributes were rated by an independent sample of 59 Prolific participants (Palan \u0026amp; Schitter, \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The scenario was presented by means of HTC Vive Pro Eye (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.vive.com/sea/product/vive-pro-eye/overview/\u003c/span\u003e\u003cspan address=\"https://www.vive.com/sea/product/vive-pro-eye/overview/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e).\u003c/span\u003e In order to realize naturalistic movements, we used the Xsense motion capture suit (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.xsens.com/\u003c/span\u003e\u003cspan address=\"https://www.xsens.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e)\u003c/span\u003e to record the kinematics of an actor either gently caressing or stabbing the right hand of another actor with the right hand (see Fig.\u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e in the Supplementary Information). A VIVE controller was used as a prop to reproduce the knife stabbing. The actor was instructed to perform the caress at a speed of about 3 cm/s. The actors\u0026rsquo; kinematics were transferred to the virtual agents by means of Motion Builder 2022 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.autodesk.com/\u003c/span\u003e\u003cspan address=\"https://www.autodesk.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e)\u003c/span\u003e and rendered in Unity. Through recorded kinematics, we were able to keep the toucher\u0026rsquo;s movements constant and control for other emotional interference that can be conveyed in traditional experimental settings by confederates through other nonverbal cues.\u003c/p\u003e \u003cp\u003eGalvanic skin response (GSR) and electrocardiogram (ECG) were recorded as indices of physiological reactivity. Signals were amplified using an ADInstruments PowerLab 8/35 system equipped with an ML116 GSR Amplifier (75 Hz AC excitation; low constant voltage of 22 mVrms) and dedicated GSR sensors comprising two bipolar finger electrodes. GSR electrodes were attached to the distal phalanges of the right index and middle fingers, and the signal was sampled at 1 kHz. For ECG acquisition, two pre-gelled electrodes (DORMO) were placed on the back of each hand, with the reference electrode positioned on the left ankle. ECG signals were sampled at 1 kHz and low-pass filtered at 30 Hz. All recordings were acquired using LabChart 8 (ADInstruments, Inc.). Participants\u0026rsquo; gaze was recorded using the HTC Vive Pro Eye built-in Tobii eye-tracker (120 Hz; nominal accuracy of 0.5\u0026deg;-1.1\u0026deg;). A 5-point calibration (SteamVR) was performed before each task (Keshava et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Schuetz \u0026amp; Fiehler, \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In the IPD task, the agents\u0026rsquo; eye region was defined in Unity using a collider-based region of interest (ROI). Gaze samples were mapped onto scene objects via ray casting; samples intersecting the eye-ROI collider were counted as looks to the eyes (Clay et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Invalid samples were removed, and linear interpolation was applied. Trials with \u0026gt;\u0026thinsp;20% invalid gaze were excluded (11 trials, 0.8% of the dataset). For the Touch task, pupil diameter (mm) was downsampled to 30 Hz. Samples marked as invalid by the eye-tracker and samples outside the 1.5\u0026ndash;9 mm range were removed (Kret \u0026amp; Sjak-Shie, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Missing data were linearly interpolated. Trials were excluded if more than 20% of the samples were invalid (12 trials, 1.5% of the dataset). Pupil diameter was then averaged across the two eyes. To obtain a baseline-corrected measure, we subtracted the mean pupil diameter during the first second of the trial (agent appearance) from the mean pupil diameter during the remainder of the trial.\u003c/p\u003e\n\u003ch3\u003eGeneral Procedure\u003c/h3\u003e\n\u003cp\u003eParticipants first filled out the questionnaires (STAI, VR Experience), then performed a baseline mouse-tracking task. They subsequently donned the VR headset and performed the baseline interpersonal distance (IPD) task, which included virtual agents\u0026rsquo; evaluations. Next, they completed the virtual touch task. Immediately afterward, they repeated the IPD task and the virtual agents\u0026rsquo; evaluations to assess pre\u0026ndash;post changes. Finally, participants completed the mouse-tracking task again.\u003c/p\u003e\n\u003ch3\u003eTouch Task\u003c/h3\u003e\n\u003cp\u003eParticipants observed a gender-matched virtual body seated on a chair, with the right arm resting on a desk, and were instructed to focus on their right virtual hand. At trial onset, a virtual agent appeared standing to the participant\u0026rsquo;s right, and participants were instructed to look at the agent (to facilitate the agent\u0026rsquo;s recognition).\u003c/p\u003e \u003cp\u003e After 1 s, the agent initiated a movement (caress or stab), and participants were instructed to attend to the agent\u0026rsquo;s hand. The action lasted\u0026thinsp;~\u0026thinsp;5 s, and the agent then remained still for 1s before disappearing (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). After a 500 ms delay, a visual analogue scale (VAS) appeared. On each trial, participants provided ratings of the agent\u0026rsquo;s attractiveness and trustworthiness, as well as the (un)pleasantness (0\u0026thinsp;=\u0026thinsp;extremely unpleasant; 50\u0026thinsp;=\u0026thinsp;neutral; 100\u0026thinsp;=\u0026thinsp;extremely pleasant) and vicariousness of the stimulation (\u0026ldquo;It was as if I was actually feeling the stimulus on my body\u0026rdquo;). At the end of the task, body ownership illusion was assessed using selected items from the Avatar Embodiment Questionnaire (Peck \u0026amp; Gonzalez-Franco, \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; see Tab.S2 in Supplementary Material). Skin conductance responses (SCR) were extracted as the average phasic driver within the 7-s trial window using Ledalab (Benedek \u0026amp; Kaernbach, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Heart rate (HR) in bpm and pupil size in mm were extracted as the mean over the same 7 s window and baseline-corrected using the 2 s immediately preceding trial onset.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eInterpersonal Distance Task\u003c/h3\u003e\n\u003cp\u003eIn each trial, the virtual agent appeared 2.25 m in front of the participant. After 2 s, the agent walked toward the participant at a constant speed until reaching a distance of 0.25 m, and then stood at that distance for the remainder of the trial (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The agent maintained direct gaze toward the participant (i.e., the VR camera) throughout the trial. Each trial lasted 8 s and was followed by a 20 s inter-trial interval. During the final repetition, after the inter-trial interval, participants rated the agent\u0026rsquo;s attractiveness and trustworthiness. Participants were instructed to press a button on the controller when they first felt the agent\u0026rsquo;s approach become uncomfortable (comfort point). The agent continued its approach regardless of the comfort-point response, allowing physiological responses to be analyzed over a uniform time-window, including the period after participants\u0026rsquo; preferred distance was exceeded. Preferred distance was computed as the Euclidean distance (cm) between the agent\u0026rsquo;s nose and the participant\u0026rsquo;s nose (with a marker placed on the VR Camera in Unity) at the moment of button press.\u003c/p\u003e \u003cp\u003eSkin conductance responses (SCR) were extracted as the average phasic driver within the 8-s trial window using Ledalab (Benedek \u0026amp; Kaernbach, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Heart rate (HR) in bpm was extracted over the same 8 s window and baseline-corrected using the 2 s immediately preceding trial onset. Eye-tracking data for the IPD task were analysed using the \u003cem\u003eeyetrackingR\u003c/em\u003e package (Dink \u0026amp; Ferguson, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The dependent measure was the proportion of looks to the agent\u0026rsquo;s eye-ROI (eyes vs elsewhere) within each trial. To accommodate boundary values (0 and 1), we used empirical logit transformation (\u003cem\u003eelog\u003c/em\u003e; Barr, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). We conducted a time-window analysis with TimeWindow as a categorical factor with three levels: Far (0\u0026ndash;2 seconds), Approach (2\u0026ndash;5.43 seconds), and Near (5.43-8 seconds).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eMouse-Tracking Task\u003c/h2\u003e \u003cp\u003eParticipants completed an alternate forced-choice task adapted from Faust and colleagues (Faust et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) using the MouseTracker software (Freeman \u0026amp; Ambady, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). For each trial, participants were presented with three two-digit numbers: the first one (\u0026ldquo;basis\u0026rdquo;) was located at the bottom of the screen, and the other two (\u0026ldquo;targets\u0026rdquo;) were located at the top left and right corners of the screen (see Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Participants were instructed to click on the target that was numerically closer to the basis. Next to each target, one face was presented. Participants were asked to ignore the faces and complete the task correctly, while also being as fast as possible. The maximum difference in value between the basis and the targets was five units. For instance, if the basis was 22, each of the two targets was not larger than 27. This is to ensure the difficulty of the task across all trials was consistent. Participants were asked to execute the task accurately and fast and to ignore the faces.\u003c/p\u003e \u003cp\u003eData were analysed using the \u003cem\u003emousetrap\u003c/em\u003e package in R (Wulff et al., \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The preprocessing included time and length normalization. Trajectories were remapped on the right side. We excluded trials with errors (i.e., the choice of number was incorrect), and trials that were outside the range defined by the mean response time\u0026thinsp;\u0026plusmn;\u0026thinsp;2 standard deviations (Wulff et al., \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). As outcomes, we considered the signed Maximum Absolute Deviation (MAD) from the ideal path connecting the start and the end of the trajectory; the Area Under the Curve, i.e., the amount of area between the ideal and the observed trajectory, and the response time (Stillman et al., \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eAnalytical strategy\u003c/h2\u003e \u003cp\u003eAnalyses were conducted in RStudio within a Bayesian framework (Kruschke \u0026amp; Liddell, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) using the brms (B\u0026uuml;rkner, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), bayestestR (Makowski, Ben-Shachar, \u0026amp; L\u0026uuml;decke, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and emmeans (Lenth \u0026amp; Piaskowski, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) packages. For each outcome collected during the IPD task (attractiveness and trustworthiness ratings, comfort distance point, heart rate deceleration, skin conductance response, and fixations), we ran two complementary analyses. Firstly, we fit a linear mixed-effects model with the interaction between Session (Pre, Post) and Stimulation (Stab, Caress) as a fixed effect, and by-Stimulation slopes and by-participant intercepts as random effects. Secondly, we computed an update score Δ defined as Post-Pre (positive values= increase), and modelled the effects of Stimulation and Touch Pleasantness, with by-participant intercepts as a random effect. All dependent variables were scaled and centred. We used weakly informative priors on the coefficients (Normal, M\u0026thinsp;=\u0026thinsp;0 and SD\u0026thinsp;=\u0026thinsp;1). For the mouse-tracking task, we fit a linear mixed-effect model with the interaction between Session and Condition (Congruent, Incongruent) as fixed effects, and the intercept over the participants as random effects. For the physiological (SCR, HR) and gaze indices, STAI-state and STAI-trait were entered as covariates to control for possible effects from state and trait anxiety. For touch (un)Pleasantness and Vicariousness, we included VR Experience as a predictor to control for possible effects of prior VR exposure. All models\u0026rsquo; syntax is reported in the Supplementary Information (Table S4). For detecting an effect, we established a probability of direction\u0026thinsp;\u0026gt;\u0026thinsp;95% (Makowski, Ben-Shachar, Chen, et al., 2019); we also report 89% highest posterior density intervals, mean, and standard deviation of the posterior distributions. For all coefficients, we verified that the Effective Sample Size (ESS) was higher than 1000 and that Rhat, indicating convergence of the chains, was \u0026lt;\u0026thinsp;1.05.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eTouch Task\u003c/h2\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003eBody ownership\u003c/h2\u003e \u003cp\u003eThe analysis on the body ownership ratings showed that scores were higher for the ownership items compared to the control ones (0.95\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13, HDI [0.72, 1.18], pd\u0026thinsp;=\u0026thinsp;100%).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eTouch (un)Pleasantness\u003c/h2\u003e \u003cp\u003eResults showed a significantly greater Pleasantness for the caresses compared to the stabs (1.13\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18, [0.85, 1.41], pd\u0026thinsp;=\u0026thinsp;100.00%). We also observed an effect of reported VR experience, so that the higher the VR experience, the lower Pleasantness for the caresses (-0.28\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13, [-0.49, -0.07], pd\u0026thinsp;=\u0026thinsp;98.38%).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eTouch Vicariousness\u003c/h2\u003e \u003cp\u003eVicariousness was higher for caresses compared to stabs (0.40\u0026thinsp;\u0026plusmn;\u0026thinsp;0.10, [0.24, 0.55], pd\u0026thinsp;=\u0026thinsp;99.99%). VR Experience decreased vicariousness for the caresses (-0.29\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14, [-0.51, -0.06], pd\u0026thinsp;=\u0026thinsp;97.82%) but not for the stabs (-0.12\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15, [-0.36, 0.12], pd\u0026thinsp;=\u0026thinsp;78.16%).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003ePhysiological reactivity to touch\u003c/h2\u003e \u003cp\u003eSkin conductance responses were higher for stabs compared to caresses (0.017\u0026thinsp;\u0026plusmn;\u0026thinsp;0.008, [-0.004, 0.03], pd\u0026thinsp;=\u0026thinsp;99.03%). Heart rate responses did not show a difference between the two conditions (-0.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04, [-0.01, 0.03], pd\u0026thinsp;=\u0026thinsp;78.10%). Pupil size analysis showed that stabs increased pupil size compared to caresses (0.58\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08, [0.46, 0.71], pd\u0026thinsp;=\u0026thinsp;100%; see Fig.S2), and the magnitude of pupil size increase was inversely predicted by (un)Pleasantness, so that the lower the perceived pleasantness, the stronger the pupil dilation (-0.08\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04, [0.15, -0.01], pd\u0026thinsp;=\u0026thinsp;97.49%).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003ePre-post changes\u003c/h2\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003eAttractiveness\u003c/h2\u003e \u003cp\u003eNeither the stabbing (-0.18\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14 [-0.40, 0.04], pd\u0026thinsp;=\u0026thinsp;90.44%) nor the caressing agents (0.09\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14 [-0.14, 0.30], pd\u0026thinsp;=\u0026thinsp;74.44%) showed a credible change of attractiveness from pre to post. In Δ analysis, Touch (un)Pleasantness did not modulate attractiveness change for caressing (0.13\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12 [-0.06, 0.32], pd\u0026thinsp;=\u0026thinsp;85.50%) nor stabbing (0.20\u0026thinsp;\u0026plusmn;\u0026thinsp;0.17 [-0.06, 0.49], pd\u0026thinsp;=\u0026thinsp;88.64%) agents. These results suggest that Attractiveness judgments are not affected by the type of somatosensory interaction or by its perceived (un)Pleasantness.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eTrustworthiness\u003c/h2\u003e \u003cp\u003eWe observed a significant decrease in Trustworthiness for the stabbing avatar from pre to post treatment (-0.57\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13 [-0.78, -0.37], pd\u0026thinsp;=\u0026thinsp;100%), while no difference was found for the caressing agents (0.17\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13, [-0.03, 0.38], pd\u0026thinsp;=\u0026thinsp;89.98%). Analysis on the Δ values showed that higher Touch (un)Pleasantness (higher ratings) predicted a larger increase in trustworthiness for the caressing agents (0.25\u0026thinsp;\u0026plusmn;\u0026thinsp;0.11, [0.07, 0.43], pd\u0026thinsp;=\u0026thinsp;98.60%) but not for the stabbing ones (0.17\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18, [-0.11, 0.46], pd\u0026thinsp;=\u0026thinsp;83.12%). These results suggest that the virtual harm reduced trust irrespective of individual appraisal, while caress raised trust only when experienced as pleasant (see Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eMouse-tracking task\u003c/h2\u003e \u003cp\u003eWe observed a decrease in Maximum Absolute Deviation for the congruent trials (those in which the caressing agents were placed next to the correct response; -0.13\u0026thinsp;\u0026plusmn;\u0026thinsp;0.07 [-0.24, -0.03], pd\u0026thinsp;=\u0026thinsp;97.88%). This decrease was not found for the incongruent condition (trials in which the stabbing agents were placed next to the correct response; -0.005\u0026thinsp;\u0026plusmn;\u0026thinsp;0.07 [-0.11, 0.11], pd\u0026thinsp;=\u0026thinsp;52.40%). Similarly, Area Under the Curve was reduced for the congruent trials from pre to post Touch Task (-0.11\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06 [-0.20, -0.02], pd\u0026thinsp;=\u0026thinsp;97.28%), while no difference for incongruent trials was found (-0.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06 [-0.13, 0.05], pd\u0026thinsp;=\u0026thinsp;71.38%). Analysis of the response times showed a general decrease for both the congruent (-0.26\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04 [-0.33, -0.19], pd\u0026thinsp;=\u0026thinsp;100%) and incongruent (-0.18\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04 [-0.25, -0.11], pd\u0026thinsp;=\u0026thinsp;100%) conditions from pre to post Touch Task; however, only in post Touch Task session, the congruent trials showed shorter decision times than the incongruent ones (Pre: -0.06\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08 [-0.18, 0.06], pd\u0026thinsp;=\u0026thinsp;79.45%; Post: -0.14\u0026thinsp;\u0026plusmn;\u0026thinsp;0.07 [-0.25, -0.02], pd\u0026thinsp;=\u0026thinsp;97.02%), suggesting a reduced conflict for trials in which the caressers were associated with the right response target compared to trials in which the stabbers were associated with the right response target (see Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The emergence of a post-task congruency advantage suggests relatively greater facilitation when caressing agents were paired with the correct response compared with when stabbing agents were paired with the correct response\u0026mdash;consistent with facilitation by caress-associated stimuli and comparatively greater conflict in stabbing-associated trials.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eInterpersonal distance\u003c/h2\u003e \u003cp\u003eFor stabbing agents, the comfort-distance increased (i.e., more space between the participant and the avatar) from Pre to Post (0.27\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04 [0.21, 0.33], pd\u0026thinsp;=\u0026thinsp;100%), also for the caressing agents to a smaller extent (0.07\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03 [0.01, 0.12], pd\u0026thinsp;=\u0026thinsp;97.09%). Analysis on the Δ values showed that higher Touch (un)Pleasantness (higher ratings) predicted a larger decrease in interpersonal distance for the caressing agents (-0.26\u0026thinsp;\u0026plusmn;\u0026thinsp;0.10, [-0.47, -0.07], pd\u0026thinsp;=\u0026thinsp;99.62%) but not for the stabbing ones (-0.23\u0026thinsp;\u0026plusmn;\u0026thinsp;0.19, [-0.63, 0.13], pd\u0026thinsp;=\u0026thinsp;89.46%). These results suggest that the virtual harm increased preferred interpersonal distance irrespective of individual appraisal, while caresses decreased interpersonal distance only when experienced as pleasant (see Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003ePhysiological and gaze correlates of interpersonal distance\u003c/h2\u003e \u003cp\u003eFor both stabbing (-0.14\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03 [-0.18, -0.09], pd\u0026thinsp;=\u0026thinsp;100%) and caressing avatars (-0.14\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03 [-0.18, -0.08], pd\u0026thinsp;=\u0026thinsp;100%), we observed a decrease in SCR from pre to post treatment. No other credible differences were found for SCR analysis (see Supplementary Information).\u003c/p\u003e \u003cp\u003eWe found an increase of HR deceleration for stabbers from pre to post touch task (-0.11\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06 [-0.21, -0.02], pd\u0026thinsp;=\u0026thinsp;96.71%). No difference for the caressing agents was found (0.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06 [-0.06, 0.13], pd\u0026thinsp;=\u0026thinsp;69.88%). Analysis on the Δ values revealed that touch (un)Pleasantness predicted a decreased heart rate deceleration in subsequent approach for caressing (0.23\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12 [0.04, 0.42], pd\u0026thinsp;=\u0026thinsp;97.44%) but not stabbing agents (-0.08\u0026thinsp;\u0026plusmn;\u0026thinsp;0.17 [-0.35, 0.20], pd\u0026thinsp;=\u0026thinsp;67.53%).\u003c/p\u003e \u003cp\u003eWe observed an increase in gaze for both the stabbing (0.09\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05 [0.003, 0.18], pd\u0026thinsp;=\u0026thinsp;95.89%) and the caressing (0.12\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06 [0.03, 0.21], pd\u0026thinsp;=\u0026thinsp;98.31%) agents in the approach phase from pre to post. Moreover, we observed an increase in gaze for the stabbing agents in the appearance phase (at far distances) from pre to post (0.26\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06 [0.16, 0.35], pd\u0026thinsp;=\u0026thinsp;98.31%). No other differences were found. Overall, these effects are consistent with the ambiguous function of eye gaze and suggest that stabbing may have increased visual attention toward the stabbing agent at far distances, potentially reflecting face identification and/or assessment of the agent\u0026rsquo;s intentions.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe extent to which social touch and pain shape subsequent interactions with virtual agents remains largely unknown. Here, we investigated how sensations evoked by virtual pain versus pleasant touch delivered on one\u0026rsquo;s own virtual body modulate evaluations of and nonverbal behaviour toward the touching agents. Participants experienced a full-body illusion, and observed agents deliver either a stab or a caress to their virtual right hand. We assessed changes in explicit impressions (trustworthiness, attractiveness), implicit evaluation via mouse-tracking, preferred interpersonal distance and physiological responses during the virtual agent\u0026rsquo;s approach. Stabbing agents were evaluated as less trustworthy, preferred at a greater interpersonal distance, and associated with stronger heart rate deceleration during their approach. A more complex picture emerged for caressing agents: here, it was not the caress per se, but its subjectively experienced (un)pleasantness that predicted higher trustworthiness ratings, reduced preferred interpersonal distance, and reduced heart rate deceleration during the virtual agent\u0026rsquo;s approach.\u003c/p\u003e \u003cp\u003eIn this sense, our findings refine the notion of a virtual Midas touch by showing that touch-related effects on social evaluation are not automatic, but depend on whether the somatosensory interaction is experienced as affiliative or aversive when enacted on an embodied virtual body. Our results on impression updating are consistent with the view that judgments of physical attractiveness are relatively resistant to change, whereas moral judgments (such as trustworthiness) are more plastic and experience-dependent (Glennon \u0026amp; Zeki, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Previous work shows that trustworthiness can be rapidly updated when new information about an agent\u0026rsquo;s behaviour is encountered (Chang et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Filkowski et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Neuroimaging studies show that a distributed brain network underpins belief updating about social agents and integration of new social information into value representations (Mende-Siedlecki, \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), supporting the idea that moral impressions are continuously revised as behaviourally diagnostic evidence accumulates. In our study, painful stimulations delivered by the stabbing agents affected subsequent evaluations independently of individual differences in reported (un)pleasantness, likely reflecting the fact that stabbing was perceived as uniformly unpleasant across participants. By contrast, the impact of caresses depended on their subjective pleasantness, consistent with the inherently more ambiguous and context-sensitive nature of affiliative touch. This asymmetry fits with the high salience of threat cues and the well-established negativity bias in impression formation, according to which negative information tends to be prioritized when forming or revising interpersonal impressions, especially when it signals potential harm (Baumeister et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Skowronski \u0026amp; Carlston, \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e1989\u003c/span\u003e). In this sense, pain inflicted by an agent can serve as highly diagnostic behavioural information about that agent\u0026rsquo;s intentions, while touch is subjected to a more ambiguous interpretation.\u003c/p\u003e \u003cp\u003eOur mouse-tracking results indicate that exposure to the touch task altered motor conflict in the numerical task. In congruent trials\u0026mdash;where the correct response was mapped to the caressing agents and the competing option to the stabbing agents\u0026mdash;we observed reduced maximum absolute deviation and area under the curve, suggesting more direct trajectories and less competition from the alternative option. Response times decreased from pre- to post-session for both congruent and incongruent trials, consistent with learning, but were also shorter for congruent than incongruent trials at post, indicating that the mapping in which caressers signalled the correct choice became relatively easier to resolve. While this mapping does not allow us to disentangle the separate contribution of approach tendencies toward caressers and avoidance tendencies toward stabbers, these effects suggest that virtual touch and pain might act as strong diagnostic behavioural information leading to update not only explicit, but also implicit evaluations, which are typically harder to change (Cone \u0026amp; Ferguson, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Mann et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Importantly, these effects emerged in a task that did not require any explicit evaluation of the agents, suggesting that vicarious touch and pain influenced social processing at an implicit level.\u003c/p\u003e \u003cp\u003e In the interpersonal distance task, participants increased the preferred distance towards the stabbing agents. For the caressing agents, higher subjective touch pleasantness predicted a reduction in preferred distance, suggesting that pleasant social touch promotes approach when experienced as pleasant. Previous research indicates that perceived threat enlarges personal space (Abra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Cartaud, Quesque, et al., 2020; Holt et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Lisi, Scattolin, et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This pattern suggests that virtual agents who inflict pain are treated as threatening social partners, whereas affiliative agents gain proximity only when their tactile behaviour is interpreted as appropriate and benign. At the physiological level, these results were not paralleled by an increase of SCR, as both stabbing and caressing agents were associated with reduced SCR during approach compared to baseline, suggesting habituation of the initial orienting response. Previous studies have reported that angry virtual faces elicit higher SCR and larger preferred distances than neutral faces (Cartaud, Ott, et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and that fearful expressions increase SCR, particularly at close distances relative to positive expressions (Ellena et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In our case, the absence of SCR differences between conditions may reflect rapid habituation to repeated approaches by the same avatars or a more general arousal response to the approaching virtual body itself.\u003c/p\u003e \u003cp\u003eHeart rate, by contrast, showed a deceleration during the subsequent stabbing agents\u0026rsquo; approach, consistent with a freezing-like response to threat (Roelofs et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Heart rate deceleration accompanied by relative bodily immobility has been repeatedly observed in healthy individuals exposed to physical threat and is interpreted as fear bradycardia\u0026mdash;a defensive state that enhances vigilance while preparing the organism for rapid action (Azevedo et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Roelofs et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). At the neurobiological level, such freezing responses are thought to involve projections from the amygdala to the periaqueductal gray (Applegate et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e1983\u003c/span\u003e; Hermans et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Recent VR work similarly reports lower heart rate when participants face avatars displaying aggressive postures compared to neutral ones (Mello, Dupont, et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) or when they face threatening and immoral characters such as criminals (Placidi et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). By contrast, when participants can actively escape or stop the approaching threat, heart rate acceleration is more likely, reflecting action preparation and active coping rather than passive freezing (Lu et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). It is worth noting that our task did not allow participants to avoid the approaching agents, which may have favoured bradycardia over acceleratory cardiac responses. Thus, heart rate deceleration in the present task is more consistent with a defensive, freezing-like response than with nonspecific autonomic arousal.\u003c/p\u003e \u003cp\u003e Gaze to the eye region increased during the approach phase for both caressing and stabbing agents. For stabbing agents, this increase was also found in the early, far-distance phase. According to equilibrium theory, interpersonal distance and mutual gaze are jointly regulated indices of interpersonal intimacy that are adjusted to maintain an optimal level of comfort (Argyle \u0026amp; Dean, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1965\u003c/span\u003e; Bailenson et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). On this basis, we expected greater intimacy\u0026mdash;reflected in more eye contact\u0026mdash;towards caressing agents only. However, our data only partially supported this prediction, as increased eye contact was not restricted to caressing agents. Increased eye contact with stabbing agents could reflect functions other than intimacy signalling, such as face identification (Royer et al., \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) or assessment of the agent\u0026rsquo;s intentions (Hietanen, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). More broadly, these results are consistent with the notion that direct gaze is intrinsically ambiguous (Kleinke, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e1986\u003c/span\u003e): sustained eye contact can signal attraction (Hoffmann et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), but in other contexts it can signal dominance or threat (Ellsworth \u0026amp; Langer, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1976\u003c/span\u003e; Terburg et al., \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). During the agents\u0026rsquo; approach, rather than directly coding valence, gaze may convey the intensity and motivational relevance of the interaction.\u003c/p\u003e \u003cp\u003e Across several outcome measures\u0026mdash;trustworthiness, heart rate deceleration, and interpersonal distance\u0026mdash;the positive effects of caressing agents in our study were not driven by the caress per se, but by how pleasant the touch was subjectively experienced. This fits with work showing that touch pleasantness is highly variable across individuals and contexts, rather than a fixed property of a given tactile stimulus (Saarinen et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). One important source of individual variability lies in internal states: longing for touch has been associated with increased pleasantness ratings of social touch, suggesting that internal \u0026ldquo;need\u0026rdquo; states bias hedonic appraisal (Abra et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Meijer et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). More broadly, the experiential outcome of touch depends on whether the interaction matches the recipient\u0026rsquo;s expectations and current social needs (Sailer \u0026amp; Leknes, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Importantly, we did not directly assess such individual differences (e.g., longing for touch, current social needs, or attitudes towards technology-mediated touch), which represents a limitation of the present study. Moreover, while previous studies (Fusaro et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Mello, Fusaro, et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Verga, Di Marco, et al., \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) and anedoctal reports (Freeman et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Wiederhold, \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) suggest that merely vision of touch on one\u0026rsquo;s embodied avatar can trigger derivative tactile sensations, it is worth noting that (un)pleasantness effect might by boosted by a congruent haptic feedback (Maunsbach et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Seinfeld et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Smekal et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Future studies should better investigate the interplay between individual longing for touch, attitudes towards technology-mediated touch, virtual agents\u0026rsquo; characteristics, and haptic feedback quality in affecting touch reactions and subsequent interactions.\u003c/p\u003e \u003cp\u003eTaken together, our findings show that how virtual agents touching or harming the embodied avatar of a human participant affects how they are subsequently perceived and approached, at both explicit and implicit levels. This suggests that, in immersive social spaces, agents\u0026rsquo; tactile behaviours can function as diagnostic social signals that calibrate users\u0026rsquo; feelings of safety, intimacy, and engagement. More broadly, these results provide an empirical basis for investigating how threatening and affiliative somatosensory cues contribute to more complex social processes, including moral decision-making in dynamic exchanges with virtual agents (Vabba, Lisi, et al., \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e "},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eM.P.L.: BEFORERC 2024, awarded by La Sapienza University of Rome.\u003c/p\u003e \u003cp\u003eS.M.A.: Brain Magnet Program awarded by the Italian Institute of Technology.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eM.P.L., M.F., and S.M.A. conceived and designed research; M.P.L. and F.M. performed experiments; M.P.L., M.F., and F.M. analyzed data; M.P.L., M.F., and F.M. prepared the virtual reality set-up; M.P.L. drafted the manuscript; M.P.L., M.F., F.M., and S.M.A. edited, revised, and approved the final version of the manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData and scripts used for the analyses are available at the following link: https://osf.io/8j7us/overview?view\\_only=b4624112f87144f7880111d02d3762c2\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbra Y, Mirams L, Fairhurst MT (2024) The space between us: The effect of perceived threat on discomfort distance and perceived pleasantness of interpersonal vicarious touch. Heliyon 10(16):e36487. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.heliyon.2024.e36487\u003c/span\u003e\u003cspan address=\"10.1016/j.heliyon.2024.e36487\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAglioti SM, Frisanco A, Lisi MP (2025) The body in the brain, in the age of the metaverse. \u003cem\u003eProceedings of the Paris Institute for Advanced Study\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/ZENODO.17127881\u003c/span\u003e\u003cspan address=\"10.5281/ZENODO.17127881\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eApplegate C, Kapp B, Underwood M, Mcnall C (1983) Autonomic and somatomotor effects of amygdala central N. stimulation in awake rabbits☆. Physiol Behav 31(3):353\u0026ndash;360. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0031-9384(83)90201-9\u003c/span\u003e\u003cspan address=\"10.1016/0031-9384(83)90201-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArgyle M, Dean J (1965) Eye-Contact, Distance and Affiliation. Sociometry 28(3):289. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2786027\u003c/span\u003e\u003cspan address=\"10.2307/2786027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAzevedo TM, Volchan E, Imbiriba LA, Rodrigues EC, Oliveira JM, Oliveira LF, Lutterbach LG, Vargas CD (2005) A freezing-like posture to pictures of mutilation. Psychophysiology 42(3):255\u0026ndash;260. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1469-8986.2005.00287.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1469-8986.2005.00287.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBailenson JN, Blascovich J, Beall AC, Loomis JM (2001) Equilibrium Theory Revisited: Mutual Gaze and Personal Space in Virtual Environments. Presence: Teleoperators Virtual Environ 10(6):583\u0026ndash;598. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1162/105474601753272844\u003c/span\u003e\u003cspan address=\"10.1162/105474601753272844\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBailenson JN, Blascovich J, Beall AC, Loomis JM (2003) Interpersonal Distance in Immersive Virtual Environments. Pers Soc Psychol Bull 29(7):819\u0026ndash;833. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0146167203029007002\u003c/span\u003e\u003cspan address=\"10.1177/0146167203029007002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarr DJ (2008) Analyzing \u0026lsquo;visual world\u0026rsquo; eyetracking data using multilevel logistic regression. J Mem Lang 59(4):457\u0026ndash;474. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jml.2007.09.002\u003c/span\u003e\u003cspan address=\"10.1016/j.jml.2007.09.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaumeister RF, Bratslavsky E, Finkenauer C, Vohs KD (2001) Bad is Stronger than Good. Rev Gen Psychol 5(4):323\u0026ndash;370. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/1089-2680.5.4.323\u003c/span\u003e\u003cspan address=\"10.1037/1089-2680.5.4.323\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenedek M, Kaernbach C (2010) A continuous measure of phasic electrodermal activity. J Neurosci Methods 190(1):80\u0026ndash;91. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jneumeth.2010.04.028\u003c/span\u003e\u003cspan address=\"10.1016/j.jneumeth.2010.04.028\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026uuml;rkner P-C (2017) brms: An \u003cem\u003eR\u003c/em\u003e Package for Bayesian Multilevel Models Using \u003cem\u003eStan\u003c/em\u003e. J Stat Softw 80(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18637/jss.v080.i01\u003c/span\u003e\u003cspan address=\"10.18637/jss.v080.i01\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCafaro A, Vilhj\u0026aacute;lmsson HH, Bickmore T, Heylen D, J\u0026oacute;hannsd\u0026oacute;ttir KR, Valgar\u0026eth;sson GS (2012) First Impressions: Users\u0026rsquo; Judgments of Virtual Agents\u0026rsquo; Personality and Interpersonal Attitude in First Encounters. In Y. Nakano, M. Neff, A. Paiva, \u0026amp; M. Walker (A c. Di), \u003cem\u003eIntelligent Virtual Agents\u003c/em\u003e (Vol. 7502, pp. 67\u0026ndash;80). Springer Berlin Heidelberg. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-642-33197-8_7\u003c/span\u003e\u003cspan address=\"10.1007/978-3-642-33197-8_7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCampbell JID, Thompson VA (2012) MorePower 6.0 for ANOVA with relational confidence intervals and Bayesian analysis. Behav Res Methods 44(4):1255\u0026ndash;1265. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13428-012-0186-0\u003c/span\u003e\u003cspan address=\"10.3758/s13428-012-0186-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCandini M, Battaglia S, Benassi M, Di Pellegrino G, Frassinetti F (2021) The physiological correlates of interpersonal space. Sci Rep 11(1):2611. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-82223-2\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-82223-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCartaud A, Ott L, Iachini T, Honor\u0026eacute; J, Coello Y (2020) The influence of facial expression at perceptual threshold on electrodermal activity and social comfort distance. Psychophysiology 57(9):e13600. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/psyp.13600\u003c/span\u003e\u003cspan address=\"10.1111/psyp.13600\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCartaud A, Quesque F, Coello Y (2020) Wearing a face mask against Covid-19 results in a reduction of social distancing. PLoS ONE 15(12):e0243023. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0243023\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0243023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCase LK, Madian N, McCall MV, Bradson ML, Liljencrantz J, Goldstein B, Alasha VJ, Zimmerman MS (2023) Aβ-CT Affective Touch: Touch Pleasantness Ratings for Gentle Stroking and Deep Pressure Exhibit Dependence on A-Fibers. \u003cem\u003eEneuro\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(5), ENEURO.0504-22.2023. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1523/ENEURO.0504-22.2023\u003c/span\u003e\u003cspan address=\"10.1523/ENEURO.0504-22.2023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang LJ, Doll BB, Van \u0026rsquo;T, Wout M, Frank MJ, Sanfey AG (2010) Seeing is believing: Trustworthiness as a dynamic belief. Cogn Psychol 61(2):87\u0026ndash;105. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cogpsych.2010.03.001\u003c/span\u003e\u003cspan address=\"10.1016/j.cogpsych.2010.03.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCihodaru-Ștefanache Ștefania, Podina IR (2025) The uncanny valley effect in embodied conversational agents: A critical systematic review of attractiveness, anthropomorphism, and uncanniness. Front Psychol 16:1625984. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2025.1625984\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2025.1625984\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClay V, K\u0026ouml;nig P, K\u0026ouml;nig SU (2019) Eye tracking in virtual reality. J Eye Mov Res 12(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.16910/jemr.12.1.3\u003c/span\u003e\u003cspan address=\"10.16910/jemr.12.1.3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCone J, Ferguson MJ (2015) He did what? The role of diagnosticity in revising implicit evaluations. J Personal Soc Psychol 108(1):37\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/pspa0000014\u003c/span\u003e\u003cspan address=\"10.1037/pspa0000014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrusco AH, Wetzel CG (1984) The Midas Touch: The Effects of Interpersonal Touch on Restaurant Tipping. Pers Soc Psychol Bull 10(4):512\u0026ndash;517. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0146167284104003\u003c/span\u003e\u003cspan address=\"10.1177/0146167284104003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDink JW, Ferguson B (2015) \u003cem\u003eeyetrackingR: An R Library for Eye-tracking Data Analysis. Available from\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.eyetrackingr.com\u003c/span\u003e\u003cspan address=\"http://www.eyetrackingr.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEllena G, Battaglia S, L\u0026agrave;davas E (2020) The spatial effect of fearful faces in the autonomic response. Exp Brain Res 238(9):2009\u0026ndash;2018. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00221-020-05829-4\u003c/span\u003e\u003cspan address=\"10.1007/s00221-020-05829-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEllingsen D-M, Leknes S, L\u0026oslash;seth G, Wessberg J, Olausson H (2016) The Neurobiology Shaping Affective Touch: Expectation, Motivation, and Meaning in the Multisensory Context. \u003cem\u003eFrontiers in Psychology\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2015.01986\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2015.01986\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEllsworth PC, Langer EJ (1976) Staring and approach: An interpretation of the stare as a nonspecific activator. J Personal Soc Psychol 33(1):117\u0026ndash;122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/0022-3514.33.1.117\u003c/span\u003e\u003cspan address=\"10.1037/0022-3514.33.1.117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaust NT, Chatterjee A, Christopoulos GI (2019) Beauty in the eyes and the hand of the beholder: Eye and hand movements\u0026rsquo; differential responses to facial attractiveness. J Exp Soc Psychol 85:103884. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jesp.2019.103884\u003c/span\u003e\u003cspan address=\"10.1016/j.jesp.2019.103884\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFilkowski MM, Anderson IW, Haas BW (2016) Trying to trust: Brain activity during interpersonal social attitude change. Cogn Affect Behav Neurosci 16(2):325\u0026ndash;338. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13415-015-0393-0\u003c/span\u003e\u003cspan address=\"10.3758/s13415-015-0393-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFreeman G, Zamanifard S, Maloney D, Acena D (2022) Disturbing the Peace: Experiencing and Mitigating Emerging Harassment in Social Virtual Reality. \u003cem\u003eProceedings of the ACM on Human-Computer Interaction\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(CSCW1), 1\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/3512932\u003c/span\u003e\u003cspan address=\"10.1145/3512932\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFreeman JB, Ambady N (2010) MouseTracker: Software for studying real-time mental processing using a computer mouse-tracking method. Behav Res Methods 42(1):226\u0026ndash;241. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/BRM.42.1.226\u003c/span\u003e\u003cspan address=\"10.3758/BRM.42.1.226\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFusaro M, Bufacchi RJ, Nicolardi V, Provenzano L (2022) The analgesic power of pleasant touch in individuals with chronic pain: Recent findings and new insights. Front Integr Nuerosci 16:956510. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fnint.2022.956510\u003c/span\u003e\u003cspan address=\"10.3389/fnint.2022.956510\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFusaro M, Fanti V, Chakrabarti B (2023) Greater interpersonal distance in adults with autism. Autism Res 16(10):2002\u0026ndash;2007. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/aur.3013\u003c/span\u003e\u003cspan address=\"10.1002/aur.3013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFusaro M, Lisi MP, Era V, Porciello G, Candidi M, Aglioti SM, Boukarras S (2025) The Transformative Power of Virtual Reality: Redefining Interactions in Virtual Platforms, Education, Healthcare, and Workplaces. \u003cem\u003eTopoi\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11245-025-10216-1\u003c/span\u003e\u003cspan address=\"10.1007/s11245-025-10216-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFusaro M, Lisi MP, Tieri G, Aglioti SM (2021) Heterosexual, gay, and lesbian people\u0026rsquo;s reactivity to virtual caresses on their embodied avatars\u0026rsquo; taboo zones. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), Articolo 1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-81168-w\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-81168-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGazzola V, Spezio ML, Etzel JA, Castelli F, Adolphs R, Keysers C (2012) Primary somatosensory cortex discriminates affective significance in social touch. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e109\u003c/em\u003e(25). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.1113211109\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1113211109\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlennon MJ, Zeki S (2022) The power of external influences to modify judgments of facial and moral beauty. PsyCh J 11(5):707\u0026ndash;714. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/pchj.492\u003c/span\u003e\u003cspan address=\"10.1002/pchj.492\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonzalez-Franco M, Ofek E, Pan Y, Antley A, Steed A, Spanlang B, Maselli A, Banakou D, Pelechano N, Orts-Escolano S, Orvalho V, Trutoiu L, Wojcik M, Sanchez-Vives MV, Bailenson J, Slater M, Lanier J (2020) The Rocketbox Library and the Utility of Freely Available Rigged Avatars. Front Virtual Real 1:561558. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/frvir.2020.561558\u003c/span\u003e\u003cspan address=\"10.3389/frvir.2020.561558\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonz\u0026aacute;lez-Franco M, Peck TC, Rodr\u0026iacute;guez-Fornells A, Slater M (2014) A threat to a virtual hand elicits motor cortex activation. \u003cem\u003eExperimental Brain Research\u003c/em\u003e, \u003cem\u003e232\u003c/em\u003e(3), Articolo 3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00221-013-3800-1\u003c/span\u003e\u003cspan address=\"10.1007/s00221-013-3800-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaans A, De Bruijn R, IJsselsteijn WA (2014) A Virtual Midas Touch? Touch, Compliance, and Confederate Bias in Mediated Communication. J Nonverbal Behav 38(3):301\u0026ndash;311. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10919-014-0184-2\u003c/span\u003e\u003cspan address=\"10.1007/s10919-014-0184-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarjunen VJ, Spap\u0026eacute; M, Ahmed I, Jacucci G, Ravaja N (2018) Persuaded by the machine: The effect of virtual nonverbal cues and individual differences on compliance in economic bargaining. Comput Hum Behav 87:384\u0026ndash;394. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.chb.2018.06.012\u003c/span\u003e\u003cspan address=\"10.1016/j.chb.2018.06.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHensel LB, Zhan J, Bjornsdottir RT, Garrod OGB, Schyns PG, Jack RE (2020) Psychologically Valid Social Face Features for Virtual Agents. \u003cem\u003eProceedings of the 20th ACM International Conference on Intelligent Virtual Agents\u003c/em\u003e, 1\u0026ndash;3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/3383652.3423899\u003c/span\u003e\u003cspan address=\"10.1145/3383652.3423899\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHermans EJ, Henckens MJAG, Roelofs K, Fern\u0026aacute;ndez G (2013) Fear bradycardia and activation of the human periaqueductal grey. NeuroImage 66:278\u0026ndash;287. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroimage.2012.10.063\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2012.10.063\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHietanen JK (2018) Affective Eye Contact: An Integrative Review. Front Psychol 9:1587. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2018.01587\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2018.01587\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoffmann A, Schiestl S, Sinske P, Gondan M, Sachse P, Maran T (2024) Sharing and Receiving Eye-Contact Predicts Mate Choice After a 5-Minute Conversation: Evidence from a Speed-Dating Study. Arch Sex Behav 53(3):959\u0026ndash;968. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10508-023-02806-0\u003c/span\u003e\u003cspan address=\"10.1007/s10508-023-02806-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolt DJ, Zapetis SL, Babadi B, Zimmerman J, Tootell RBH (2022) Personal space increases during the COVID-19 pandemic in response to real and virtual humans. Front Psychol 13:952998. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2022.952998\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2022.952998\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIachini T, Coello Y, Frassinetti F, Senese VP, Galante F, Ruggiero G (2016) Peripersonal and interpersonal space in virtual and real environments: Effects of gender and age. J Environ Psychol 45:154\u0026ndash;164. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvp.2016.01.004\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvp.2016.01.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeshava A, Aumeistere A, Izdebski K, Konig P (2020) Decoding Task From Oculomotor Behavior In Virtual Reality. \u003cem\u003eACM Symposium on Eye Tracking Research and Applications\u003c/em\u003e, 1\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/3379156.3391338\u003c/span\u003e\u003cspan address=\"10.1145/3379156.3391338\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKleinke CL (1986) Gaze and eye contact: A research review. Psychol Bull 100(1):78\u0026ndash;100. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/0033-2909.100.1.78\u003c/span\u003e\u003cspan address=\"10.1037/0033-2909.100.1.78\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKret ME, Sjak-Shie EE (2019) Preprocessing pupil size data: Guidelines and code. Behav Res Methods 51(3):1336\u0026ndash;1342. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13428-018-1075-y\u003c/span\u003e\u003cspan address=\"10.3758/s13428-018-1075-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKroczek LOH, Pfaller M, Lange B, M\u0026uuml;ller M, M\u0026uuml;hlberger A (2020) Interpersonal Distance During Real-Time Social Interaction: Insights From Subjective Experience, Behavior, and Physiology. Front Psychiatry 11:561. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyt.2020.00561\u003c/span\u003e\u003cspan address=\"10.3389/fpsyt.2020.00561\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKruschke JK, Liddell TM (2018) The Bayesian New Statistics: Hypothesis testing, estimation, meta-analysis, and power analysis from a Bayesian perspective. Psychon Bull Rev 25(1):178\u0026ndash;206. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13423-016-1221-4\u003c/span\u003e\u003cspan address=\"10.3758/s13423-016-1221-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee Y-H, Xiao M, Wells RH (2018) The Effects of Avatars\u0026rsquo; Age on Older Adults\u0026rsquo; Self-Disclosure and Trust. Cyberpsychology Behav Social Netw 21(3):173\u0026ndash;178. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/cyber.2017.0451\u003c/span\u003e\u003cspan address=\"10.1089/cyber.2017.0451\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLenth R, Piaskowski J (2025) \u003cem\u003eEstimated Marginal Means, aka Least-Squares Means\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://rvlenth.github.io/emmeans/\u003c/span\u003e\u003cspan address=\"https://rvlenth.github.io/emmeans/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLisi MP, Fusaro M, Aglioti SM (2024) Visual perspective and body ownership modulate vicarious pain and touch: A systematic review. Psychon Bull Rev. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13423-024-02477-5\u003c/span\u003e\u003cspan address=\"10.3758/s13423-024-02477-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLisi MP, Fusaro M, Tieri G, Aglioti SM (2021) Humans adjust virtual comfort-distance towards an artificial agent depending on their sexual orientation and implicit prejudice against gay men. Comput Hum Behav 125:106948. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.chb.2021.106948\u003c/span\u003e\u003cspan address=\"10.1016/j.chb.2021.106948\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLisi MP, Giometti A, Fusaro M, Aglioti SM (2025) \u003cem\u003eSurrogate bias: Behavioral and physiological effects of seeing ingroup and outgroup virtual agents touching one\u0026rsquo;s embodied avatar\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLisi MP, Scattolin M, Fusaro M, Aglioti SM (2021) A Bayesian approach to reveal the key role of mask wearing in modulating projected interpersonal distance during the first COVID-19 outbreak. PLoS ONE 16(8):e0255598. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0255598\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0255598\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu J, Kemmerer SK, Riecke L, De Gelder B (2023) Early threat perception is independent of later cognitive and behavioral control. A virtual reality-EEG-ECG study. Cereb Cortex 33(13):8748\u0026ndash;8758. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/cercor/bhad156\u003c/span\u003e\u003cspan address=\"10.1093/cercor/bhad156\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa HH, Lee M, Kim E, Pang S (2025) The Effects of Virtual Social Touch on Interpersonal Communication Outcomes in Virtual Environments: A Meta-Analysis. Int J Human\u0026ndash;Computer Interact 1\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/10447318.2025.2526590\u003c/span\u003e\u003cspan address=\"10.1080/10447318.2025.2526590\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMakowski D, Ben-Shachar MS, Chen SHA, L\u0026uuml;decke D (2019) Indices of Effect Existence and Significance in the Bayesian Framework. Front Psychol 10:2767. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2019.02767\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2019.02767\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMakowski D, Ben-Shachar MS, L\u0026uuml;decke D (2019) bayestestR: Describing Effects and their Uncertainty, Existence and Significance within the Bayesian Framework. J Open Source Softw 4(40):1541. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21105/joss.01541\u003c/span\u003e\u003cspan address=\"10.21105/joss.01541\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMakransky G, Wismer P, Mayer RE (2019) A gender matching effect in learning with pedagogical agents in an immersive virtual reality science simulation. J Comput Assist Learn 35(3):349\u0026ndash;358. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/jcal.12335\u003c/span\u003e\u003cspan address=\"10.1111/jcal.12335\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMann TC, Cone J, Heggeseth B, Ferguson MJ (2019) Updating implicit impressions: New evidence on intentionality and the affect misattribution procedure. J Personal Soc Psychol 116(3):349\u0026ndash;374. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/pspa0000146\u003c/span\u003e\u003cspan address=\"10.1037/pspa0000146\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaselli A, Slater M (2013) The building blocks of the full body ownership illusion. \u003cem\u003eFrontiers in Human Neuroscience\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fnhum.2013.00083\u003c/span\u003e\u003cspan address=\"10.3389/fnhum.2013.00083\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaunsbach M, Hornb\u0026aelig;k K, Seifi H (2023) Mediated Social Touching: Haptic Feedback Affects Social Experience of Touch Initiators. \u003cem\u003e2023 IEEE World Haptics Conference (WHC)\u003c/em\u003e, 93\u0026ndash;100. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/WHC56415.2023.10224506\u003c/span\u003e\u003cspan address=\"10.1109/WHC56415.2023.10224506\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeijer LL, Hasenack B, Kamps JCC, Mahon A, Titone G, Dijkerman HC, Keizer A (2022) Affective touch perception and longing for touch during the COVID-19 pandemic. Sci Rep 12(1):3887. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-022-07213-4\u003c/span\u003e\u003cspan address=\"10.1038/s41598-022-07213-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMello M, Dupont L, Engelen T, Acciarino A, De Borst AW, De Gelder B (2022) The influence of body expression, group affiliation and threat proximity on interactions in virtual reality. Curr Res Behav Sci 3:100075. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.crbeha.2022.100075\u003c/span\u003e\u003cspan address=\"10.1016/j.crbeha.2022.100075\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMello M, Fusaro M, Aglioti SM, Minio-Paluello I (2024) Exploring social touch in autistic and non-autistic adults via a self-report body-painting task: The role of sex, social context and body area. Autism 28(8):1985\u0026ndash;2001. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/13623613231218314\u003c/span\u003e\u003cspan address=\"10.1177/13623613231218314\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMello M, Fusaro M, Aglioti SM, Minio-Paluello I (2025) Centrality of Touch Avoidance in Social Touch Experiences in Autism. J Autism Dev Disord. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10803-025-07053-x\u003c/span\u003e\u003cspan address=\"10.1007/s10803-025-07053-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMello M, Fusaro M, Tieri G, Aglioti SM (2022) Wearing same- and opposite-sex virtual bodies and seeing them caressed in intimate areas. \u003cem\u003eQuarterly Journal of Experimental Psychology\u003c/em\u003e, \u003cem\u003e75\u003c/em\u003e(3), Articolo 3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/17470218211031557\u003c/span\u003e\u003cspan address=\"10.1177/17470218211031557\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMende-Siedlecki P (2018) Changing our minds: The neural bases of dynamic impression updating. Curr Opin Psychol 24:72\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.copsyc.2018.08.007\u003c/span\u003e\u003cspan address=\"10.1016/j.copsyc.2018.08.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMull I, Wyss J, Moon E, Lee S-E (2015) An exploratory study of using 3D avatars as online salespeople: The effect of avatar type on credibility, homophily, attractiveness and intention to interact. J Fashion Mark Manage 19(2):154\u0026ndash;168. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1108/JFMM-05-2014-0033\u003c/span\u003e\u003cspan address=\"10.1108/JFMM-05-2014-0033\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNicolardi V, Lisi MP, Mello M, Fusaro M, Tieri G, Aglioti SM (2025) Taking an embodied avatar\u0026rsquo;s perspective modulates the temporal dynamics of vicarious pain and pleasure: A Virtual Reality and EEG study. Soc Cognit Affect Neurosci nsaf035. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/scan/nsaf035\u003c/span\u003e\u003cspan address=\"10.1093/scan/nsaf035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNolte D, Hjoj R, S\u0026aacute;nchez Pacheco T, Huang A, K\u0026ouml;nig P (2025) Investigating proxemics behaviors towards individuals, pairs, and groups in virtual reality. Virtual Reality 29(2):58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10055-025-01127-y\u003c/span\u003e\u003cspan address=\"10.1007/s10055-025-01127-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNovembre G, Etzi R, Morrison I (2021) Hedonic Responses to Touch are Modulated by the Perceived Attractiveness of the Caresser. Neuroscience 464:79\u0026ndash;89. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroscience.2020.10.007\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroscience.2020.10.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOosterhof NN, Todorov A (2008) The functional basis of face evaluation. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e105\u003c/em\u003e(32), 11087\u0026ndash;11092. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.0805664105\u003c/span\u003e\u003cspan address=\"10.1073/pnas.0805664105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePalan S, Schitter C (2018) Prolific.ac\u0026mdash;A subject pool for online experiments. J Behav Experimental Finance 17:22\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jbef.2017.12.004\u003c/span\u003e\u003cspan address=\"10.1016/j.jbef.2017.12.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePandhare A, Skulikj D, Lee J, Mikhiel P, Banakou D (2025) vER: Virtual Human Companionship for Patients in Preoperative Care. \u003cem\u003eProceedings of the 2025 ACM International Conference on Interactive Media Experiences\u003c/em\u003e, 468\u0026ndash;472. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/3706370.3731639\u003c/span\u003e\u003cspan address=\"10.1145/3706370.3731639\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeck TC, Gonzalez-Franco M (2021) Avatar Embodiment. A Standardized Questionnaire. Front Virtual Real 1:575943. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/frvir.2020.575943\u003c/span\u003e\u003cspan address=\"10.3389/frvir.2020.575943\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePedrabissi L, Santinello M (1989) STAI state-trait anxiety inventory forma Y Manuale. Organizzazioni Speciali\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerry A, Rubinsten O, Peled L, Shamay-Tsoory SG (2013) Don\u0026rsquo;t stand so close to me: A behavioral and ERP study of preferred interpersonal distance. NeuroImage 83:761\u0026ndash;769. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroimage.2013.07.042\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2013.07.042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePlacidi V, Lisi MP, Vabba A, Frisanco A, Aglioti SM (2025) \u003cem\u003eVirtual embodiment of civic heroes preserves conflict about anti-mafia intentions and enhances physiological discomfort in response to mafia-related crime\u003c/em\u003e. PsyArXiv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.31234/osf.io/va2k5_v1\u003c/span\u003e\u003cspan address=\"10.31234/osf.io/va2k5_v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoelofs K, Hagenaars MA, Stins J (2010) Facing Freeze: Social Threat Induces Bodily Freeze in Humans. Psychol Sci 21(11):1575\u0026ndash;1581. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0956797610384746\u003c/span\u003e\u003cspan address=\"10.1177/0956797610384746\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoyer J, Blais C, Charbonneau I, D\u0026eacute;ry K, Tardif J, Duchaine B, Gosselin F, Fiset D (2018) Greater reliance on the eye region predicts better face recognition ability. Cognition 181:12\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cognition.2018.08.004\u003c/span\u003e\u003cspan address=\"10.1016/j.cognition.2018.08.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaarinen A, Harjunen V, Jasinskaja-Lahti I, J\u0026auml;\u0026auml;skel\u0026auml;inen IP, Ravaja N (2021) Social touch experience in different contexts: A review. Neurosci Biobehavioral Reviews 131:360\u0026ndash;372. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neubiorev.2021.09.027\u003c/span\u003e\u003cspan address=\"10.1016/j.neubiorev.2021.09.027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSailer U, Friedrich Y, Asgari F, Hassenzahl M, Croy I (2024) Determinants for positive and negative experiences of interpersonal touch: Context matters. Cogn Emot 38(4):565\u0026ndash;586. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/02699931.2024.2311800\u003c/span\u003e\u003cspan address=\"10.1080/02699931.2024.2311800\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSailer U, Leknes S (2022) Meaning makes touch affective. Curr Opin Behav Sci 44:101099. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cobeha.2021.101099\u003c/span\u003e\u003cspan address=\"10.1016/j.cobeha.2021.101099\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaini G, Zain M, Noronha M, Bonin RP, Lomanowska AM (2024) Virtually simulated interpersonal touch negatively affects perceived closeness and social affiliation to an avatar partner. Sci Rep 14(1):1347. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-024-51773-6\u003c/span\u003e\u003cspan address=\"10.1038/s41598-024-51773-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchirmer A, Croy I, Ackerley R (2023) What are C-tactile afferents and how do they relate to affective touch? Neurosci Biobehavioral Reviews 151:105236. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neubiorev.2023.105236\u003c/span\u003e\u003cspan address=\"10.1016/j.neubiorev.2023.105236\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchuetz I, Fiehler K (2022) Eye tracking in virtual reality: Vive pro eye spatial accuracy, precision, and calibration reliability. J Eye Mov Res 15(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.16910/jemr.15.3.3\u003c/span\u003e\u003cspan address=\"10.16910/jemr.15.3.3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeger CR, Smith ER, Percy EJ, Conrey FR (2014) Reach Out and Reduce Prejudice: The Impact of Interpersonal Touch on Intergroup Liking. Basic Appl Soc Psychol 36(1):51\u0026ndash;58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/01973533.2013.856786\u003c/span\u003e\u003cspan address=\"10.1080/01973533.2013.856786\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeinfeld S, Schmidt I, M\u0026uuml;ller J (2022) \u003cem\u003eEvoking realistic affective touch experiences in virtual reality\u003c/em\u003e (arXiv:2202.13389). arXiv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://arxiv.org/abs/2202.13389\u003c/span\u003e\u003cspan address=\"http://arxiv.org/abs/2202.13389\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSkowronski JJ, Carlston DE (1989) Negativity and extremity biases in impression formation: A review of explanations. Psychol Bull 105(1):131\u0026ndash;142. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/0033-2909.105.1.131\u003c/span\u003e\u003cspan address=\"10.1037/0033-2909.105.1.131\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSlater M, Spanlang B, Sanchez-Vives MV, Blanke O (2010) First Person Experience of Body Transfer in Virtual Reality. PLoS ONE 5(5):e10564. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0010564\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0010564\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmekal V, Hecquard J, K\u0026uuml;hne S, Occidental N, L\u0026eacute;cuyer A, Mac\u0026eacute; M, De Gelder B (2025) Influence of Haptic Feedback on Perception of Threat and Peripersonal Space in Social VR. IEEE Trans Vis Comput Graph 31(5):2986\u0026ndash;2994. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/TVCG.2025.3549884\u003c/span\u003e\u003cspan address=\"10.1109/TVCG.2025.3549884\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpielberger CD, Gorsuch R, Lushene R, Vagg PR, Jacobs GA (1983) Manual for the state-trait anxiety inventory. Consulting Psychologists\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStillman PE, Shen X, Ferguson MJ (2018) How Mouse-tracking Can Advance Social Cognitive Theory. Trends Cogn Sci 22(6):531\u0026ndash;543. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tics.2018.03.012\u003c/span\u003e\u003cspan address=\"10.1016/j.tics.2018.03.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun W, Banakou D, Świdrak J, Valori I, Slater M, Fairhurst MT (2024) Multisensory experiences of affective touch in virtual reality enhance engagement, body ownership, pleasantness, and arousal modulation. Virtual Reality 28(4):162. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10055-024-01056-2\u003c/span\u003e\u003cspan address=\"10.1007/s10055-024-01056-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuvilehto JT, Cekaite A, Morrison I (2023) The why, who and how of social touch. Nat Reviews Psychol 2(10):606\u0026ndash;621. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s44159-023-00217-5\u003c/span\u003e\u003cspan address=\"10.1038/s44159-023-00217-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuvilehto JT, Glerean E, Dunbar RIM, Hari R, Nummenmaa L (2015) Topography of social touching depends on emotional bonds between humans. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e112\u003c/em\u003e(45), 13811\u0026ndash;13816. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.1519231112\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1519231112\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuvilehto JT, Nummenmaa L, Harada T, Dunbar RIM, Hari R, Turner R, Sadato N, Kitada R (2019) Cross-cultural similarity in relationship-specific social touching. \u003cem\u003eProceedings of the Royal Society B: Biological Sciences\u003c/em\u003e, \u003cem\u003e286\u003c/em\u003e(1901), 20190467. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rspb.2019.0467\u003c/span\u003e\u003cspan address=\"10.1098/rspb.2019.0467\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuvilehto JT, Renvall V, Nummenmaa L (2021) Relationship-specific Encoding of Social Touch in Somatosensory and Insular Cortices. Neuroscience 464:105\u0026ndash;116. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroscience.2020.09.015\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroscience.2020.09.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eŚwidrak J, Pochwatko G, Insabato A (2021) Does an agent\u0026rsquo;s touch always matter? Study on virtual Midas touch, masculinity, social status, and compliance in Polish men. J Multimodal User Interfaces 15(2):163\u0026ndash;174. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12193-020-00351-x\u003c/span\u003e\u003cspan address=\"10.1007/s12193-020-00351-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eŚwidrak J, Pochwatko G, Navarro X, Osęka L, Doliński D (2020) The joint influence of social status and personal attitudes in a contact and open versus a noncontact and homophobic culture on the virtual Midas touch. Virtual Reality 24(4):619\u0026ndash;633. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10055-019-00423-8\u003c/span\u003e\u003cspan address=\"10.1007/s10055-019-00423-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTer Stal S, Tabak M, Akker OD, Beinema H, T., Hermens H (2020) Who Do You Prefer? The Effect of Age, Gender and Role on Users\u0026rsquo; First Impressions of Embodied Conversational Agents in eHealth. Int J Human\u0026ndash;Computer Interact 36(9):881\u0026ndash;892. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/10447318.2019.1699744\u003c/span\u003e\u003cspan address=\"10.1080/10447318.2019.1699744\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTerburg D, Hooiveld N, Aarts H, Kenemans JL, Van Honk J (2011) Eye Tracking Unconscious Face-to-Face Confrontations: Dominance Motives Prolong Gaze to Masked Angry Faces. Psychol Sci 22(3):314\u0026ndash;319. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0956797611398492\u003c/span\u003e\u003cspan address=\"10.1177/0956797611398492\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVabba A, Lisi MP, Placidi V, Provenzano L, Zahid MN, Aglioti SM (2025) \u003cem\u003eAn immersive virtual reality (IVR) and neuroscience-based approach to study and change attitudes and behaviour towards mafia-type organized crime\u003c/em\u003e. PsyArXiv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.31234/osf.io/sbnxg_v1\u003c/span\u003e\u003cspan address=\"10.31234/osf.io/sbnxg_v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVabba A, Placidi V, Lisi MP, Aglioti SM (2025) \u003cem\u003eWhere the Proteus effect stops \u0026ndash; embodying fictional mafia criminals leads to paradoxical effects on implicit attitudes towards organized crime\u003c/em\u003e. PsyArXiv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.31234/osf.io/hcyt8_v1\u003c/span\u003e\u003cspan address=\"10.31234/osf.io/hcyt8_v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerga C, Di Marco S, Giove F, Aglioti SM, Lisi MP (2025) Temporal dynamics of full-body ownership and vicarious touch sensations as inferred from embodying a virtual avatar. Exp Brain Res 243(4):103. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00221-025-07055-2\u003c/span\u003e\u003cspan address=\"10.1007/s00221-025-07055-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerga C, Lisi MP, Aglioti SM, Fusaro M (2025) \u003cem\u003eTranslating, adapting and validating the Touch Experiences and Attitudes Questionnaire (TEAQ) for the Italian population\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.31234/osf.io/4scwh_v2\u003c/span\u003e\u003cspan address=\"10.31234/osf.io/4scwh_v2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWiederhold BK (2022) Sexual Harassment in the Metaverse. \u003cem\u003eCyberpsychology, Behavior, and Social Networking\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(8), Articolo 8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/cyber.2022.29253.editorial\u003c/span\u003e\u003cspan address=\"10.1089/cyber.2022.29253.editorial\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWieser MJ, Pauli P, Grosseibl M, Molzow I, M\u0026uuml;hlberger A (2010) Virtual Social Interactions in Social Anxiety\u0026mdash;The Impact of Sex, Gaze, and Interpersonal Distance. \u003cem\u003eCyberpsychology, Behavior, and Social Networking\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(5), 547\u0026ndash;554. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/cyber.2009.0432\u003c/span\u003e\u003cspan address=\"10.1089/cyber.2009.0432\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWulff DU, Kieslich PJ, Henninger F, Haslbeck JMB, Schulte-Mecklenbeck M (2025) Movement tracking of psychological processes: A tutorial using mousetrap. Behav Res Methods 57(11):307. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13428-025-02695-2\u003c/span\u003e\u003cspan address=\"10.3758/s13428-025-02695-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang F-C, Acevedo P, Guo S, Choi M, Mousas C (2025) Embodied Conversational Agents in Extended Reality: A Systematic Review. IEEE Access 13:79805\u0026ndash;79824. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/ACCESS.2025.3566698\u003c/span\u003e\u003cspan address=\"10.1109/ACCESS.2025.3566698\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhu S, Hu W, Li W, Dong Y (2024) Virtual Agents in Immersive Virtual Reality Environments: Impact of Humanoid Avatars and Output Modalities on Shopping Experience. Int J Human\u0026ndash;Computer Interact 40(19):5771\u0026ndash;5793. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/10447318.2023.2241293\u003c/span\u003e\u003cspan address=\"10.1080/10447318.2023.2241293\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":true,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Impression updating, Virtual agents, Virtual pain and touch, Psychophysiology, Eye-tracking","lastPublishedDoi":"10.21203/rs.3.rs-8765445/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8765445/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWitnessing pain and touch on one\u0026rsquo;s embodied avatar in immersive virtual environments can elicit vicarious (un)pleasant sensations. What remains unclear is whether observing pain and touch on one\u0026rsquo;s virtual body can modulate attitudes towards the touching agent. In this study, participants underwent a full-body illusion in Immersive Virtual Reality and observed different virtual agents delivering either a painful (stab) or a pleasant (caress) virtual stimulus on their embodied avatar's right hand. We assessed the pre-post changes induced by virtual somatosensory stimuli on (1) explicit judgements of the virtual agent's trustworthiness and attractiveness; (2) implicit evaluation of the virtual agent's facial attractiveness (via a mouse-tracking procedure), and (3) the physiological arousal and gaze during the regulation of the virtual agent-human interpersonal distance. We found that trustworthiness increased and comfort distance decreased with the perceived touch pleasantness. Analysis of hand movements revealed decreased attraction toward the stabbing agents. Consistent with this pattern, stabbing agents elicited stronger cardiac deceleration\u0026ndash; an index of freezing\u0026ndash;during the subsequent approach. Gaze increased toward the eye region for both stabbers and caressers, suggesting heightened monitoring of the agent\u0026rsquo;s intentions. Together, these findings indicate that vicarious somatosensory experience can influence the social encoding of virtual agents at both explicit and implicit levels, as inferred from behavioural, physiological, and gaze signatures. These results contribute to understanding the psychological impact of embodied interactions in immersive digital spaces.\u003c/p\u003e","manuscriptTitle":"Vicarious experience modulates implicit and explicit evaluation of virtual agents delivering pain and touch over an embodied avatar","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-09 19:24:32","doi":"10.21203/rs.3.rs-8765445/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f43a4548-0af7-4371-8ff6-901805612172","owner":[],"postedDate":"February 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-24T15:10:03+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-09 19:24:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8765445","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8765445","identity":"rs-8765445","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}