Introduction

Hearing loss is much more than “not to understand what has been said”. Chronic hearing impairment has a tremendous input on everyday life, mainly due to multiple problems in communication, which subsequently have a significant impact on the interaction between the hearing-impaired subject and the close partner (Govender et al., 2014; Scarinci et al., 2009). In general, communication is an active, bidirectional process of a mutual exchange with the goal of understanding the interlocutor’s mental states and behavior. It follows certain implicit rules which are often violated in the case of a hearing disorder (Hétu, 1996). Hearing loss triggers a set of disturbances, such as misunderstandings, frequent interruptions in the flow of conversation (Ekberg et al., 2016; Lind et al., 2004, 2006), limited turnovers (Jepsen & Liebst, 2021), and therefore cause a decrease in spontaneity and depth in communication which is important for the quality and intimacy of any relationship (Govender et al., 2014; Hallam et al., 2008).

Rhythmic entrainment, mutual focus and attunement with a mutually reinforcing feedback loop between inputs in a harmonious interaction process are the key elements of interactional rituals as the micro-interactional glue that holds social life together (Johannessen, 2023) and lead to high emotional energy, shared mood and social integration (Collins, 2004). In spoiled communication caused by auditory engulfment, the impaired rhythmic entrainment in face-to-face interaction results in a decrease in emotional energy and alientation from interaction, as certain types of behavior of individuals with hearing impairment (e.g. irregular flow of speech by the PHI, disturbed turn-taking, requests of repetition) are judged to be inappropriate to the implicit communication rules to those who are not impaired.

Given the emotional and social function of communication, these communicative difficulties cause a high emotional burden of stigmatization and have far-reaching psychosocial consequences, such as fear, shame and depression (Carlsson et al., 2015; Contrera et al., 2017; Danermark, 1998; Lawrence et al., 2020) with a negative impact on the general quality of life of the hearing-impaired person (Ciorba et al., 2012; Gopinath et al., 2012; Hogan et al., 2009) and that of the close partner, called third-party disability (Scarinci et al., 2011, 2012).

In a recent pilot study (Völter et al., 2022b), we explored the phenomenon of nonverbal synchrony (Tschacher & Ramseyer, 2017), which is generally present in human interaction (Nowak et al., 2020; Ramseyer & Tschacher, 2006), and which has been shown to be suitable for assessment in dyads with people suffering from hearing impairment (PHI) and their significant others (SO). The present study is an extension of our previous work in terms of experimentally assessing the effect of an inner ear prosthetic device called cochlear implant (CI) on social facets and nonverbal behavior pre- and postoperation.

Nonverbal Communication and Hearing Impairment

Information about emotions is strongly conveyed by nonverbal communication using signals such as gestures, facial expressions, eye contact, posture, proximity-distance behavior and body tension. A comparison of the ability to recognize emotions from facial expressions and posture between deaf and hearing people showed that people who were deaf in childhood are superior to hearing people in recognizing emotions from facial expressions or posture (Ferrari et al., 2019). Acquired or progressive deafness, in contrast, leads to lower emotional sensitivity to visual stimuli (Ambert-Dahan et al., 2017). Our group has recently focused on a dyadic aspect of nonverbal communication – nonverbal synchrony – which is closely related to co-adaptation at a micro-social level, and which has been shown to influence shared experiences, or something commonly known as interpersonal chemistry (Reis et al., 2022). Many of these bidirectional nonverbal processes in nonverbal synchrony lead to a connection between interacting people (Lakin et al., 2003), and most forms of synchrony usually occur outside conscious awareness and cannot be directly controlled (Ramseyer & Tschacher, 2006).

Imitation is a specific form of the multitude of adaptation processes which can easily be observed in some forms of learning and social, emotional and cognitive development in infancy (Bell, 2020; Friedman, 2020) and the more general forms of synchrony documented in various domains remain relevant throughout adulthood (Nowak et al., 2020). Recent meta-analyses concluded that various forms of synchronized phenomena may reflect satisfaction and social connectedness in relationships (Cacioppo et al., 2014; Coutinho et al., 2019; Vicaria & Dickens, 2016), increase positive affect (Nyman-Salonen et al., 2021; Tschacher et al., 2014) and provide information about the patient-therapist relationship (Altmann et al., 2020; Cohen et al., 2021; Ramseyer & Tschacher, 2011) (for a review see Zilcha-Mano, 2024) as well as the success of psychotherapeutic treatments (e.g. Altmann et al., 2020; Ramseyer & Tschacher, 2011). In subjects with hearing loss, a significant other is directly affected by the impaired communication which may then lead to e.g. the experience of discomfort and negative emotions,  because relevant aspects of social bonding and positive affect are associated with interpersonal synchrony (Mogan et al., 2017). In other words: nonverbal synchrony promised to be a suitable variable for the assessment of social aspects in dyads with one or two partners suffering from hearing impairment (Völter et al., 2022).

Assessment of Relationship Quality

Traditionally, the relationship between conversation partners or between a person with hearing loss and a relative has been assessed through interviews or questionnaires (Anderson & Noble, 2005; Barker et al., 2017; Hallam et al., 2008; Knussen et al., 2004), such as the CADI Subscale from the Carers’ Perceived Problem Checklist (Nolan & Grant, 1989; Nolan et al., 1990) or the Relationship Assessment Scale (Hendrick, 1988). More specifically, self-report assessments such as the Inclusion of the Other in the Self Scale/IOS (Aron et al., 1992), the Relationship Closeness Inventory (Berscheid et al., 1989) or the Personal Acquaintance Measure (Starzyk et al., 2006) can be used to quantify the closeness between persons. Since interaction and emotions are often not correctly acknowledged by the subject, more objective and automated methods might be a good option (Dunbar et al., 2022) such as Motion Energy Analysis (MEA), which uses computer-vision algorithms to recognize movement patterns that allow quantifying nonverbal synchrony between two interacting people (Ramseyer, 2020).

Cochlear Implantation

In the past years, cochlear implantation has become the method of choice for severe to profound hearing loss in cases where hearing aids are not sufficient (Boisvert et al., 2020; Dazert et al., 2020). But, an intense postoperative auditory rehabilitation is mandatory to adapt the brain to the new incoming signal transmitted by the cochlear implant (Anderson & Kraus, 2013; Bronus et al., 2011; Diller, 2009). This is a complex multidimensional process which usually takes place in specialized rehabilitation centers over a period of several months or years, and which goes beyond treating sensory loss such as including audioprocessor fittings, speech therapy by an experienced speech and language pathologist as well as counseling (Boothroyd, 2007; Grenness et al., 2014a, b; Lorens et al., 2023).

In general, a significant improvement in speech perception, but also in health-related quality of life and cognition (Völter et al., 2023) as well as a decrease in depression or anxiety in CI users have been reported in various studies (Bergman et al., 2020; Choi et al., 2016; Lenarz et al., 2012; Olze et al., 2011; Völter et al., 2020). However, current therapeutic strategies primarily focus on the hearing-impaired person alone and the audiological performance in clinical settings (Boisvert et al., 2020; Carlsson et al., 2015; Dazert et al., 2020; Moberly et al., 2020). But, the hearing-impaired is embedded in a person`s entourage influenced by supporting or hampering external contextual factors and does not act in a “vacuum” as described in the biopsychosocial framework of the ICF model endorsed by the WHO (2001) and later adapted to hearing loss by Hickson and Scarinci (Hickson & Scarinci, 2007).

So far, the microcosmos of the hearing-impaired person (PHI) and the close partner (significant others: SO) has received little attention in auditory rehabilitation after cochlear implantation (Ekberg et al., 2015; Hétu et al., 1993; Lorens et al., 2023; Scarinci et al., 2021) and a micro-sociological perspective on the interaction between PHI and the SO has not been extensively taken, although Danermark already stressed in 1998 the micro-sociological point of view to hearing impairment, and suggested that broken face-to-face interaction dynamics cause negative socioemotional outcomes for the individual (Danermark, 1998). Recently, Jepsen and Liebst (2021) developed a micro-sociological framework based on a qualitative analysis of eight interviews with adults who had received CIs in childhood, and suggested impaired verbal interactions as a source of emotional energy drain, subsequent disintegration and estrangement in the social bond (Jepsen & Liebst, 2021).

Therefore, the aim of the present study was to analyze nonverbal synchrony – defined as the dynamic coordination of head- and body-movement – in people with hearing loss and their significant others (1) before and (2) after auditory rehabilitation via cochlear implantation in a natural setting of conversation.

Method and Material

Nonverbal synchrony, hearing status and psychosocial factors of 30 adults with postlingual hearing impairment (PHI; m = 18, f = 12) and their significant others (SO; m = 7, f = 23) (Table 1) were assessed in mean 2.33 days (SD = 12.1) after initial fitting of the speech processor (T1) and 6 months (SD = 1.23) later (T2).

Table 1 Demographics of the PHI
Full size table

Subjects

15 dyads involved spouses, 14 dyads parents and child, and one dyad a very close friend. PHI were on average 59.43 years (SD = 14.15) and SO 50.10 years (SD = 15.54). Mean age difference was 9.33 years (SD = 20.54). Degree of hearing impairment was classified according to WHO (2021). Eleven PHI had complete deafness (> 95 dB), 6 PHI a profound (80 to < 95dB), 6 PHI a severe (65 to < 80dB), 5 PHI a moderately severe (50 to < 65 dB) and 2 PHI a moderate (35 to < 35 dB) hearing loss. Duration of hearing loss was in mean 32.80 years (SD = 20.12); 19 PHI suffered from hearing loss for more than 20 years.

Setting and Videorecording

Interactions of the dyads were recorded by a HD camera (Portable Video Lab) using VideoSyncProStudio (Mangold®). Dyads sat next to each other in a V-angle in front of a static background. At T1, dyads were asked to plan an imaginary party together and at T2, they had to arrange a vacation. After receiving basic instructions, dyads were left alone for the 10- minute duration of the task.

Assessment of Synchrony

To analyze nonverbal synchrony, 10- minute lasting videos of the interactions were submitted to an automatized objective quantification of movement using Motion Energy Analysis (MEA 4.10 (Ramseyer, 2020)) (Fig. 1). Previous studies done by our group have shown that signals obtained in the head and the upper body region correlated differentially with outcomes of a psychological intervention at the micro- versus macrolevel. In one of our studies, the head region was associated with global outcome after therapy, while the upper body was associated with session-level micro-outcome (Ramseyer & Tschacher, 2014). We chose two regions of interest (ROIs), namely the head (= head) and the combined regions of head and upper body (= body).

Fig. 1
figure 1

A: Still frames (1–3) from original video with PHI sitting on the left and SO on the right side. ROIs indicated by shaded areas in Panel A1*. Lower panels (A1*-3*) with bright pixels indicating areas where movements have taken place. B: Quantified movements of PHI (blue line) and of SO (green line); Y-axis: extent of movement captured by motion energy analysis, X-axis: time in minutes (Color figure online)

Full size image

Frame-differences were first generated by MEA 4.10 (Ramseyer, 2020), and then time-series of these frame-differences were analyzed in R (R version 4.0.3) using the package rMEA (Kleinbub & Ramseyer, 2021). In line with previous studies using MEA (Lin et al., 2023; Nelson et al., 2014; Tschacher et al., 2014; Völter et al., 2022b), cross-correlations were calculated in segments of 30 s (winSec = 30) with a maximum lag of ± 5 s (lagSec = 5) and without overlap (incSec = 30). General strength of synchrony for each dyad (general synchrony) as well as type of synchrony were assessed. Movements occurring at the exact same time (lag-0) were labelled synchronic movements, while instances with either PHI or SOs leading were labelled PHI leading and SO leading, respectively. Furthermore, strength of real synchrony was compared to pseudosynchrony (Moulder et al., 2018; Ramseyer & Tschacher, 2010) by generating N = 1000 pseudodyads using a between-subject shuffling design drawn from all available time-series (procedure shuffle in rMEA; size = 1000).

Based on our previous use of the methodology, we decided to retain formerly used parameters for the quantification of synchrony, and to apply the same specifications used in other studies. However, it should be noted that apart from our windowed lagged cross-correlation there are a number of alternative approaches for the quantification of dyadic synchrony such as the cross-wavelet analysis (e.g. Fujiwara & Daibo, 2018), which employs cross-wavelet coherence to evaluate the convergence of rhythm in dyads or the cross-reccurrence analysis (e.g. Abney et al., 2015) which also been successfully applied for the quantification of this dyadic aspect.

Questionnaires

To measure the burden hearing impairment of PHI poses on the SO (called third-party disability), the German version (Völter et al., 2022) of the SOS-Hear Questionnaire (Scarinci et al., 2009) with lower scores in case of less burden was filled out by the SO.

The PHI filled out the Nijmegen Cochlear Implant Questionnaire (Hinderink et al., 2000) to identify the hearing-related quality of life (HRQoL) with lower scores indicating less HRQoL and the Communication Profile for the hearing impaired (CPHI; including the subscales of maladaptive behaviors, verbal strategies, nonverbal strategies, attitudes of others and behaviors of others) by Demorest and Erdman (1987) to analyze communication problems and strategies. The questionnaire was translated into German for this study. Less communication problems lead to higher scores.

The Inclusion of Other in the Self Scale (IOS-scale) (Aron et al., 1992) is a graphic-based scale and was used to assess the perceived closeness to the other member of the dyad (high closeness = 7, low closeness = 1).

Audiometric Assessment

Speech perception of the CI-implanted ear was examined using the German Freiburger Monosyllabic Speech Test at 65 dB and 80 dB before implantation and at T2. For T1 the preoperative hearing with a hearing aid on the side where the CI was later implanted has been used.

Statistical Analysis

Statistical analysis was conducted using Jamovi (Version 2.3.26) software. First, descriptive statistics for nonverbal synchrony and questionnaires were explored with correlational models that were subsequently expanded to mixed models assessing the temporal change pre- and post-operation. Impact of gender on nonverbal synchrony was determined using one-way analysis of variance. Given the novel nature of this patient population and the small dataset, no correction for multiple testing was applied and the significance level was set to p < .05. Effect sizes were reported for all tested associations. A previous study with a highly similar protocol but a one-time assessment only (Völter et al., 2022) provided an effect size of Cohen’s d = 0.42 for nonverbal synchrony. In terms of sensitivity, our sample size of N = 30 dyads required an effect size of 0.53 assuming an alpha error probablity of 0.05 and power of (1-beta) = 0.8.

Results

Hearing After Cochlear Implantation

Preoperative speech understanding with hearing aid on the later CI-supplied side significantly increased in the German Freiburger Monosyllabic Speech Test at 65 dB from 11.3 to 50.8% (t(29) = 9.91; p < .001; d = 1.81) and at 80 dB from 21.0 to 61.5% (t(29) = 9.24; p < .001; d = 1.69), thus there was a mean increase in hearing of 39.5% (SD = 21.8%) at 65 dB and of 40.5% (SD = 24.0%) at 80 dB (Table 2). These improvements are in the expected range after this kind of surgery.

Table 2 Means (M), Standard Deviation (SD), Effect Sizes (Cohen’s d) of Outcome Assessments at T1 and T2. Correlations (Person’s r) of Synchrony with Pre-Operation Scores (T1; n = 23–30) and Post-Operation Scores (T2; n = 23–30). Associations: Upper rows = body synchrony (head & body); lower row = head synchrony (head only) (ROIs)
Full size table

Synchrony vs. Pseudosynchrony; Synchrony Pre-Surgery (T1) vs. Post-Surgery (T2)

Nonverbal synchrony in the 30 dyads differed at all times of assessment significantly from pseudosynchrony with an overall effect size of Cohen’s dhead = 0.70; dbody = 0.52, and dhead_T1 = 0.94; dbody_T1 = 0.72 at T1, and dhead_T2 = 0.43; dbody_T2 = 0.46 at T2. The strength of synchrony thus decreased with a medium sized effect after surgery (see also Fig. 2). In terms of statistical change based on p-values, only head-synchrony for SO leading from T1 to T2 could be classified as a significant change [t(29) = 2.14; p = .041; d = 0.39]. The other comparisons were all above trend level (p > .10) with low to small effect sizes (d = 0.17 to 0.32) (Table 3).

Fig. 2
figure 2

Real synchrony (head and body combined) at T1, T2 and pseudosynchrony. The bold blue line represents the group’s average cross-correlation at T1, the bold green line at T2, and the bold grey line the average cross-correlation of pseudo dyads. Thin lines in the background depict dyad-level cross-correlations. X-axis: time-delay in lags of ± 5 s, Y-axis: absolute values of the cross-correlations (Color figure online)

Full size image
Table 3 Nonverbal synchrony at T1 and T2 (zSync; bootstrap-standardized values)
Full size table

For illustrative purposes, heatmaps with cross-correlations of nonverbal synchrony at T1 and T2 are provided in Fig. 3. This exemplary dyad had an aggregated cross-correlation score of zSync (general synchrony) = 2.12 at T1 (PHI leading = 2.50; SO leading = 0.98) and zSync (general synchrony) = -0.66 at T2 (PHI leading = -0.48; SO leading = -0.61).

Fig. 3
figure 3

Heatmaps of nonverbal synchrony at T1 and T2. Example of one dyad. X-axis = time in segments of 30 s, Y-axis = time-delay in lags; 0 = synchronic movement, 0 to 5 = PHI leading, 0 to -5 = SO leading. Warmer colors indicate higher correlations. PHI = People with hearing impairment, SO = Significant other (Color figure online)

Full size image

Self-Report Pre-Surgery (T1) vs. Post-Surgery (T2)

Hearing-related quality of life significantly improved from T1 to T2 as measured by the Nijmegen Questionnaire of the PHI (54.74 vs. 65.75; p = .004; d = 0.58), the Communication Profile (3.58 vs. 3.74; p = .016; d = 0.54) and the burden of the SO assessed in the SOS-Hear questionnaire (1.59 vs. 1.13; p = .005; d = 0.56). Social aspects thus showed an improvement with a medium effect-size (Table 2). The IOS did not change beyond small effect-sizes from T1 to T2 (p > .05; d = 0.15 to 0.16).

Associations Between Synchrony and Self-Report

General synchronization; General synchronization at T1 negatively correlated with the duration of hearing impairment (r = -.412, p = .024): higher synchrony was present in PHI with a shorter duration of hearing impairment (further bivariate correlations are provided in Table 2). Furthermore, there were several significant associations between higher synchronization and self-report at T1, e.g. in dyads where the quality of life (Nijmegen Questionnaire) reported by the PHI was low (r = -.420; p = .021), and in dyads where the perceived closeness (IOS) reported by the PHI was low (r = -.394; p = .031). In other words: unfavorable self-report pre-operation was associated with higher amounts of nonverbal synchrony. At T2, general synchronization was less associated (all p > .05) with self-report (see Table 2 for details).

PHI leading, SO leading and synchronic movements; At T1 leading by PHI was negatively related to the Nijmegen Questionnaire of PHI (r = -.372; p = .043), i.e. dyads with PHI reporting lower hearing-related quality of life were chracterized by more leading from the PHI. Leading by the SO negatively correlated with the rating of closeness (IOS) reported by the PHI (r = -.421; p = .021) and in the closeness reported by the SOs themselves (r = -.429; p = .018). Dyads with closer relationships were thus characterized by less leading of the SO. In addition, synchronic movements were significantly higher in PHI with less hearing-related quality of life in the Nijmegen Questionnaire (r = -.433; p = .017). At T2, no associations were found with relevant strength, (see Table 2 for differences regarding ROIs, facets of synchrony, and times of assessment).

Demographic and hearing aspects and synchrony; At T1 speech perception at 80 dB negatively correlated with synchronic movements (r = -.365; p = .048). There were no relevant associations between nonverbal synchrony (general synchronization, PHI leading, SO leading, synchronic movements) and age of PHI and SO, the age difference in the dyad and the amount of hearing loss (p > .05). Further, gender of PHI or SO did not impact on nonverbal synchrony (p > .05). In contrast, at T2 no association of demographic or hearing aspects and synchrony could be found.

Associations between changes in synchrony, hearing and self-report; Changes in the hearing-related questionnaires (Nijmegen Questionnaire, Communication Profile, SOS-Hear) and the speech perception were not related to a change in synchrony (p > .05). A change in the perceived closeness by the SO in the IOS questionnaire negatively correlated with PHI leading (r=-.421; p = .020). The larger the difference in closeness from T1 to T2 was, the smaller the difference-score from T1 to T2 in PHI leading. A smaller difference score (including also negative scores) can represent only small changes but also a decrease of the leading by the PHI.

Combined associations, mixed models; Apart from the correlational associations between facets of synchrony and self-report, complexer models integrating self-report, synchrony and hearing-change were explored by means of mixed models. For the selection of the models, we relied on the associations reported above. Here we report models including questionnaires that documented changes from pre-to-post cochlea implantation. From T1 to T2, a decrease of synchrony was associated with an increase of social well-being reported in the Nijmegen Questionnaire [F(1,57.8) = 7.26; p = .009], and – at a trend-level – less problems in the SOS-Hear [F(1,57.6) = 3.35; p = .072]. No association was found in the communication profile [F(1,28.0) = 0.00; p = .950]. Figure 4 depicts the association between Nijmegen Questionnaire and general synchrony.

Fig. 4
figure 4

Mixed model for the association between Nijmegen Questionnaire (Nijmegen PHI) and general synchronization. Assessments at T1: grey circles; at T2: black triangles

Full size image

Discussion

To the best of our knowledge, the present study is the first to examine nonverbal interaction quantified as nonverbal synchrony in dyads of hearing-impaired people and their close relatives before and after auditory rehabilitation via cochlear implantation.

Hearing-impaired people showed nonverbal synchrony at levels clearly different from synchrony that would be expected by chance (pseudosynchrony). The amount of nonverbal synchrony in dyads consisting of a PHI and their SO was lower compared to synchrony found in the same task performed by students, where effect sizes ranged in the medium to high range (Cohen’s d = 1.11 (Tschacher et al., 2014), d = 0.80 (Nelson et al., 2016) and d = 0.89 (Georgescu et al., 2020)). The findings using the same task described here differ in one important aspect: dyads in the above mentioned experimental conditions were not acquainted, while the level of knowing each other in the present sample may be considered the opposite. Dyads had been together for an average of 32 years. Level of acquaintance has been shown to affect nonverbal synchrony in two recent studies based on MEA. Counterintuitively, the level of synchrony was unrelated or negatively associated with the quality of the relationship or conversation between friends (Fujiwara et al., 2020; Lin et al., 2023). Apart from this difference in relationship history which may affect the amount of synchrony, it has to be noted that people with hearing loss generally have a stronger withdrawal from social activities (Saporta et al., 2022), given that the limited auditory perception usually hampers turn-taking in conversation. These obstacles to normal communication lead to difficulties in coordinating movement, which makes interactional bonds more difficult and promoting withdrawal and isolation (Jepsen & Liebst, 2021).

An interesting fact lies in the finding that a higher degree of synchrony correlated with a shorter duration of hearing loss. This could be due to the fact, that at the onset of hearing loss, the dyad faces the task of mustering more energy to overcome the difficult communicative situation. The higher degree of nonverbal synchrony in our sample thus fits to several studies that documented increased synchrony in challenging listening situations such as in background noise (Boker et al., 2002; Miles et al., 2023). Situations with obstacles for “normal” verbal communication thus appear to foster the recruitment of other means – such as nonverbal communication – that could facilitate interaction and social connectedness (Hadley & Ward, 2021).

A further or additional aspect for higher synchrony could stem from the finding of higher synchrony being associated with situations where people are in a position or situation that requires them to add additional effort for a “smooth” interaction. A number of recent studies in very different settings points in this direction, for example the observation that therapists and their patients display higher synchrony after a rupture in the therapeutic relationship (Deres-Cohen et al., 2021), or the finding that white doctors treating black/african-american cancer patients also showed higher levels of synchrony in comparison to white doctors treating white patients (Hamel et al., 2022). Furthermore, a variation in terms of cognitive demands during an interaction was tested in an experimental task setting, where subjects were asked to tell a lie, which leads to an increase of imitation compared to subjects who were telling the truth (van der Zee et al., 2021).

The situation of hearing impairment could thus be viewed as both acoustically demanding (requiring more attention and cognitive effort), as well as emotionally challenging because of the frustration associated with impaired communication. The dyads in our sample who reported less closeness, could thus be examples of distorted communication which leads to more synchrony, but also exerts a toll on relationship quality. The fact that associations between closeness and synchrony were higher at T1 than at T2 suggests that after CI-implantation synchrony diminishes, but more importantly its association with closeness is less evident. In other studies on nonverbal synchrony, a general trend for a decrease across time was reported for therapy-process (Ramseyer & Tschacher, 2011) and in sessions of coaching (Erdös & Ramseyer, 2021).

Taken together, we think that apart from the impaired communicative environment caused by hearing impairment, an additional social aspect was emphasized in our study, which is embodied in dyads, and which is less easily changed by the surgical procedure. This social aspect which is inherent in the interactive situation created by our assessment could thus more or less reflect the social communication dynamics that take more time to change, despite the fact that the acoustic barrier may have been alleviated by cochlear implantation. This means that listening effort in hearing-impaired is not only determined by hampered speech understanding, but also strongly influenced by the high effort needed to create social bonding in this condition (Hughes et al., 2018).

We think that this social bonding provides an important additional cue for future rehabilitative efforts. Comparable efforts in the direction of including significant others have been called for in the domain of chronic pain (Martire et al., 2019). Only by sufficiently assessing and supporting such social aspects of hearing impairment, patients will be enabled to fully profit from surgery in the long term follow-up. Providing social assisstance – including the siginificant others – should thus be an important addition to the normal (common) aftercare following cochlear implantation.

Apart from these general observations, we found a number of relevant associations between leading of the SO and unfavourable self-reports: both PHI as well as SO reported low personal closeness at T1 when leading of SO was high. The socially relevant question of who is following whom in terms of movement dynamics was thus an indicator of less connection in dyads. After successful amelioration of hearing after cochlear implantation, it may be assumed that PHI possess more resources apart from trying to hear the significant other, which could therefore motivate the SO not to take the lead, as Lazzarotto et al. (2019) described. An SO’s helpful attitude towards a PHI is thus embodied in her or his refrainment from excessive leading at the level of nonverbal interaction dynamics. Being able to enter a communicative situation with a PHI at eye-level (i.e. with an equally distributed responsibility) manifests itself at the quantifiable level of not-leading in a nonverbal manner and it goes along with a relationship that is described as being close. In other words: the question of who is leading whom may be quite relevant for relationship quality, and being able to better hear (in case of successful CI implantation) affects this association. The differentiation of leading versus following has also been reported as a relevant factor in psychotherapy dyads (Altmann et al., 2020), and it could be included in future evaluations of a dyad’s dynamics.

The present data underlines the fact that auditory rehabilitation should not only focus on the cure or compensation of the sensory limitation, but also consider a better understanding and a potential alleviation of the emotional burden from stigmatization and impaired identity PHI are faced with. The acceptance of the hearing impairment by the PHI is a condition of a successful rehabilitation and the involvement of SO in rehabilitation settings might make SO more concerned about the possible difficulties caused by hearing loss (Hofsöe et al., 2018; Yorgason et al., 2007; Scarinci et al., 2021). Regardless of the communicative strategy adopted by hearing-impaired people, negative emotional consequences might remain, and we would like to stress that they are often not sufficiently acknowledged in auditory rehabilitation. PHI and their SOs have to recognize the critical role of emotional aspects of a hearing loss like shame and embarrassment and they have to learn to successfully cope with these emotions (Danermark, 1998). Negative emotions have to be noted and should be turned into positive emotions by psychotherapeutical sessions to gain positive emotional energy and reward feelings that might foster interaction in dyads. Furthermore, SOs should be sensitive to recognize what kind of support the PHI may need.

Our micro-sociological perspective offers a bottom-up approach to analyze the interactive problems hearing-impaired dyads encounter in daily life. Whereas self-reports often give only an imprecise information about interpersonal relation, observational data by movement analysis before cochlear implantation might objectively uncover imbalances in a dyad’s relationship. If future research confirmed our finding that increased leading by SO could be a sign of embodied relationship strain, strategies to counter such a tendency could be developed and taught in the postoperative rehabilitation on the level of psychoeducation and of practical nonverbal behavior, such as a perceptual training for a better recognition of emotions and for nonverbal communicative signals (Schweinberger & Eiff, 2022). Given the empirically based assessment in an ecologically valid setting, such a pre-assessment could guide further provision of psychological support (Timmer et al., 2023).

While our conclusions open up a wide range of potential applications, we note a number of limitations in the present study. Besides the small sample size of 30 dyads which limits generalizability, we would like to emphasize the exploratory character of the present study. The multi-faceted aspect of our synchrony-facets makes a fully conclusive analysis of post-hoc power difficult, but by averaging all facets (head, body, T1, T2) of synchrony vs. pseudosynchrony we achieved an average of d = 0.61, which leads to a power of (1-beta) 0.89 which we achieved in our sample. Furthermore, the short follow-up period – suitable for improvements in auditory and cognitive abilities – may not be a good choice for longer-term changes in psychosocial domains and even more for changes in patterns of nonverbal behavior. A further evaluation with a larger sample and specifically a longer follow-up assessment would be needed in order to answer these relevant questions.

Conclusion

The present study showed that nonverbal synchrony correlated with the quality of self-reported relationship, particularly with the perceived closeness in existing long-term relationships. High levels of synchrony might reflect the effort to improve interpersonal interaction and could also signal situations that require high cognitive effort, such as the strain put on both interaction partners when communication is hindered by hearing impairment. In our sample, higher synchrony was associated with less favorable outcomes and with interpersonal strain. Auditory rehabilitation can lower the emotional burden hearing loss poses on dyads, However, despite auditory restoration psychological support might be helpful to balance the longstanding desintegrated interactions caused by spoiled communication. The straightforward measure of nonverbal synchrony allows a valid insight into social interaction processes and might prescribe the inclusion of further psychological therapeutical strategies in the postoperative rehabilitation regimens after implantation.