A Scoping Review of Nonverbal Synchrony Between Individuals With Social Anxiety and Others

Methods

In preparing this protocol, the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) (Tricco et al., 2018) statement was used (Supplement File 1). This scoping review also adopted the framework by the Joanna Briggs Institute (JBI) (Peters et al., 2017), which involves following a five-stage approach: Stage 1: Identify the research question; Stage 2: Identify relevant studies; Stage 3: Select studies for review; Stage 4: Chart the data; and Stage 5: Collate, summarize and report the results. The protocol for this study was registered at osf.io (https://osf.io/vqh3d).

Stage 1: Identify the Research Questions

We were interested in the nonverbal synchrony of individuals with social anxiety; therefore, our research questions were as follows:

1. What are the characteristics of nonverbal synchrony in individuals with SAD/social anxiety?

Stage 2: Identify Relevant Studies

We used the population, concept, and context (PCC) framework by the JBI for this scoping review (Peters et al., 2017) to define the inclusion criteria as follows:

Stage 3: Select Studies

The original literature search was conducted on December 19, 2023. We selected studies following the PRISMA flow diagram. We used the PCC framework, and the process was carried out independently by two researchers (AY and AM). The two authors compared their lists, and any differences in opinion were resolved by discussion and, where this failed, through arbitration by a third researcher (NN).

Stage 4: Chart the Data

Data extraction was carried out by one researcher (AY) using standard data extraction forms including country, journal, diagnosis, sample, age, methods, study type, control, measured outcomes, and conclusion. Another researcher (AM) confirmed the data extraction. If necessary, we contacted the authors of these studies for clarifications.

Stage 5: Collate, Summarize, and Report the Results

We presented the results of the search in a PRISMA flow diagram. We organized the extracted data described above as a qualitative synthesis.

Study Characteristics

Publication years ranged from 2014 to 2024. The review identified six experimental studies, five individual counseling studies (four studies combining CBT and PDT, and one PDT-only study), and one diagnostic interview study. Six studies targeted patients diagnosed with social anxiety disorder, whereas six recruited individuals with high social anxiety. The studies were conducted in Germany (n = 5), Israel (n = 4), Israel and the United States (n = 1), UK (n = 1) and France (n = 1). Recruitment sources included university campuses (n = 5), the Social Anxiety Psychotherapy Research Network Project (SOPHO-NET) (n = 2), a combination of SOPHO-NET and German psychiatric departments (n = 3), French psychiatric departments (n = 1), and online advertisements (n = 1). Diagnostic measures for social anxiety disorder included the ADIS (n = 3), the SCID plus an LSAS score ≥30 (n = 4), the SPIN (n = 2), the SIAS (n = 1), the DSM-5 (n = 1), and the SCID-IV (n = 1). All participants were adults.

Measurements of Nonverbal Synchrony

Across the included studies, nonverbal synchrony in individuals with high social anxiety was assessed via body movement, vocal, and physiological indicators (heart rate, skin-conductance response). The measurement contexts fell into three categories: clinical settings (during therapy or diagnostic interviews), conversations with opposite-sex strangers (light or intimacy-generating conversation, and light conversations with same-sex or opposite-sex strangers), and collaborative tasks (pendulum movement, drumming). Six studies explored synchrony in clinical contexts (Altmann et al., 2020; Reinecke et al., 2022; Schoenherr et al., 2019, 2021a, 2021b; Shatz, Oren-Yagoda, & Aderka, 2024), four in conversational contexts (Asher et al., 2020, 2021; Günak et al., 2020; Shatz, Oren-Yagoda, & Aderka, 2024), and two in collaborative-task contexts (Gordon et al., 2023; Varlet et al., 2014). Nonverbal synchrony was most quantified via video-based body movement analysis; nine studies employed motion energy analysis (MEA). One study recorded hand movements with a hand-held pendulum monitored by a motion-capture system, and another used the NEUROGES®-ELAN hand-movement coding system. Vocal synchrony was examined in one study, and heart-rate synchrony (photoplethysmography) in another, and cardiac interbeat interval (IBI) synchrony and skin conductance level (SCL) synchrony (via electrocardiography and electrodermal activity, respectively), together with behavioral drumming coordination, were assessed in one further study (Gordon et al., 2023).

Measures of Symptoms

Self-report instruments included measures of social-anxiety symptoms (LSAS, SR-LSAS, Mini-SPIN), interpersonal problems (Inventory of Interpersonal Problems-32; Horowitz et al., 1988), therapeutic alliance (Helping Alliance Questionnaire; Alexander & Luborsky, 1986), emotional state (Positive and Negative Affect Schedule; Watson et al., 1988), depressive symptoms (Beck Depression Inventory; Beck et al., 1961), attachment anxiety and avoidance (Experiences in Close Relationships; Brennan et al., 1998), social thoughts and beliefs (Social Thoughts and Beliefs Scale; Turner et al., 2003), Brief Fear of Negative Evaluation Scale revised (BFNE-II; Carleton et al., 2007), and Quality of Interaction Scale (QI; Berry & Hansen, 1996). Although many studies assessed symptoms, depression, anxiety, and alliance, none measured process factors such as self-focused attention, negative self-evaluation.

Nonverbal Synchrony in Social Anxiety Is Context-dependent and Role-Sensitive

The reviewed studies suggest that disrupted nonverbal synchrony in individuals with SAD/social anxiety is a context- and role-contingent difficulty when evaluative concern intensifies or when the individual is required to assume a high social load role (e.g., leader). Across conversational paradigms, several findings suggest that synchrony may be preserved or even compensated for in relatively low-demand interactions (e.g., small talk). In contrast, it becomes more disrupted when the interaction requires intimate self-disclosure and thus heightens the fear of negative evaluation. Asher et al. (2020, 2021) indicate that synchrony is more likely to deteriorate in contexts involving closer interpersonal engagement, which is consistent with the notion that perceived social threat disrupts synchrony. Shatz, Asher, and Aderka (2024) also report on context-dependent changes that are compatible with a compensatory account. Specifically, some individuals with SAD/social anxiety may attempt to maintain synchrony in low threat settings, but show reduced synchrony when intimacy and evaluative demands increase.

Interactional asymmetry in collaborative tasks also affects synchrony. In particular, Varlet et al. (2014) suggest that synchrony disruptions are most apparent when individuals with SAD/social anxiety must take the lead, implying that performance evaluation and responsibility for coordination can destabilize synchrony even when the task is explicitly cooperative. This role dependence implies that the responsibility for coordination can destabilize synchrony even in explicitly cooperative tasks (Varlet et al., 2014). In clinical assessment contexts, Shatz, Oren-Yagoda, and Aderka (2024) describe reduced “pacing” during diagnostic interviews, which is consistent with the idea that highly evaluative, asymmetric interactions can hinder interpersonal synchronization.

Clinical interaction studies suggest that evaluative pressure is modulated by relationship-specific safety and support. For instance, Altmann et al. (2020) and Schoenherr et al. (2019, 2021a) indicate that higher synchrony early in treatment is associated with more favorable relational/process outcomes (e.g., stronger therapeutic alliance, fewer interpersonal difficulties), implying that a supportive therapeutic relationship may buffer the negative effects of evaluation that are more typical in interactions with strangers. To integrate these cross-context comparisons, we propose an “individual–relationship–context” framework in which synchrony difficulties depend on the interplay of (i) heightened evaluative concern (e.g., intimacy, diagnostic scrutiny), (ii) socially demanding roles (e.g., leadership), and (iii) relational conditions that either mitigate or amplify threats. These findings suggest that reduced synchrony is not a uniform marker but varies systematically by context and role, highlighting the need to consider specific contexts in future mechanism-focused research and clinical assessment.

Measurement Approaches Based on Bodily Movement, Vocal, and Physiological Indicators

Across the included studies, measurement approaches showed both convergence and heterogeneity. Most included studies relied on video-based body movement synchrony, with nine of the twelve studies utilizing Motion Energy Analysis (MEA) to quantify frame-by-frame pixel changes as an index of whole-body movement (Altmann, 2013). This preference suggests that the field favors an accessible, non-invasive approach that can be applied across various contexts, including conversations, diagnostic interviews, and psychotherapy sessions. The remaining studies employed alternative methods, such as structured behavioral ratings, vocal synchrony indices, or physiological coupling measures.

Quantitative findings also indicate that measurement choice can shape conclusions. For instance, Günak et al. (2020) found that observer-rated joint action showed an indirect effect on likability (b = −0.55 to −1.89), whereas MEA-based synchrony was not associated with the corresponding outcome (p = 0.789). This divergence suggests that automated movement covariation and human-rated coordination may capture different aspects of social interaction. Physiological measures also showed modality-specific patterns. Gordon et al. (2023) noted that group-level social anxiety was negatively associated with heart-rate timing synchrony indexed by interbeat intervals (p = 0.001), but positively associated with skin conductance level synchrony (p = 0.036). These opposing patterns imply that different physiological channels may reflect distinct processes, such as coordination and stress contagion.

Several studies used directional or time-lagged indices that go beyond symmetric correlation-based metrics. For example, in diagnostic interviews, pacing was significantly reduced in participants with SAD (p = 0.004, partial η ² = 0.095), whereas leading was not significantly different (Shatz, Oren-Yagoda, & Aderka, 2024). Such indices may be particularly informative in asymmetric, evaluative interactions because they distinguish responsiveness from initiative. Finally, despite the recent emergence of inter-brain synchrony (InBS) research in social neuroscience, no included study assessed InBS in SAD or social anxiety. Overall, the operationalization of synchrony varies across modalities and contexts. These measurement choices, including modality and directionality, may shape conclusions and should be reported more transparently in future research.

Social Anxiety Indicators and Clinical Outcomes Associated With Nonverbal Synchrony

The included studies suggest that nonverbal synchrony relates to both symptom severity and clinically relevant relational or process outcomes. In non-clinical and experimental interactions, higher social anxiety severity was consistently associated with greater difficulty in achieving synchrony (Asher et al., 2020; Reinecke et al., 2022; Shatz, Asher, & Aderka, 2024). This pattern indicates that synchrony difficulties scale with symptom severity. In psychotherapy contexts, several studies suggest prognostic value. Schoenherr et al. (2019) reported that a 1% increase in early-session synchrony was associated with a 5% reduction in early dropout risk (hazard ratio = 0.95, p < 0.05). Altmann et al. (2020) also reported that body-movement synchrony predicted improvements in therapeutic alliance and interpersonal problems (IIP) (β = 0.20–0.30). These findings support the view that body synchrony can function as a process-relevant marker in treatment.

However, these findings also indicate that higher synchrony is not universally beneficial. Altmann et al. (2020) reported that greater patient-led synchrony at Session 8 was associated with worse end-of-treatment IIP and depressive symptoms (p < 0.05). Vocal synchrony results point in a similar direction. Schoenherr, Strauss, Stangier, and Altmann (2021) found that higher vocal synchrony was positively associated with attachment anxiety and avoidance at treatment end (β = 0.25–0.32). Taken together, these results suggest that the meaning of synchrony depends on who drives it and in which modality it is expressed. Therapist-led synchrony may facilitate a secure therapeutic environment, whereas patient-led or excessive synchrony might reflect maladaptive emotion regulation or over-accommodation.

In addition to social anxiety severity, depressive symptoms in some studies were found to covary with synchrony. This raises the possibility that synchrony reflects broader interpersonal and affective functioning, rather than SAD specific mechanisms alone. Importantly, no included study directly tested whether core cognitive mechanisms in SAD, such as self-focused attention, negative self-evaluation, or fear of negative evaluation, explain synchrony disruptions. Future work should integrate multimodal and directional synchrony analyses with designs that can test mechanisms and track changes during the course of treatment.

Limitations and Future Directions

Several limitations should be noted. First, generalizability is constrained by our search being limited to English- and Japanese-language literature and by the included studies being concentrated in Western and Middle Eastern contexts. The direct comparison of findings is also hindered by the equivalent treatment of clinical and laboratory settings. Conceptually, the review is limited by the absence of studies examining cognitive mediators such as metacognition, and whether the observed patterns are specific to SAD remains unclear given the lack of clinical control groups (e.g., other anxiety or depressive disorders). Finally, no included studies assessed inter-brain synchrony (InBS). Future research employing hyperscanning techniques (e.g., EEG or fNIRS) should test whether individuals with social anxiety exhibit altered synchrony in neural circuits related to social cognition or threat processing (Saul et al., 2022). Identifying such neural patterns, alongside cognitive and behavioral measures, will be essential to refine pathophysiological models of SAD and to explore whether interventions designed to enhance multimodal synchrony can directly improve treatment outcomes.