Introduction

This study examined interactional synchrony in conversations about emotional episodes using a between-participants pseudosynchrony experimental paradigm. Japanese university students were the participants. First, we confirmed the occurrence of interactional synchrony in conversations about both positive and negative episodes. Next, we compared interactional synchrony in conversations about positive episodes with negative episodes. From a cultural perspective, we examined the validity of the measurement of perceived synchrony in conversations about emotional episodes using a between-participants pseudosynchrony experimental paradigm in Japan. Previous studies had examined perceived synchrony only in Western cultures. If we replicated this paradigm in Japan, then the findings would strengthen the cross-cultural validity of this paradigm for measuring the movement characteristics of synchrony.

Issues about the Methodology for Studying Interactional Synchrony

Interactional synchrony refers to the coordination of movement between individuals in both timing and form during interpersonal communication. These phenomena are observed in newborn infants (Condon & Sander, 1974), and it is reported that these phenomena are related to language acquisition (e.g., Wylie, 1985) and rapport (e.g., Tickle-Degnen & Rosenthal, 1987). Therefore, many researchers have been interested in investigating the nature of these phenomena and have introduced theories explaining these phenomena. Because of this broad range of theoretical applicability, interactional synchrony has been measured in many different ways (Bernieri & Rosenthal, 1991). These methodologies can be divided into two types: behavior coding and ratings.

Most previous studies used a behavior coding method. In this approach, researchers must rigidly define units of analysis and identify them. Condon coded individual body parts until the precise point of a change in movement (i.e., initiation, termination, change in speed, or change in direction) can be identified (Condon, 1970; Condon & Ogston, 1966). This approach has two limitations. One limitation is that researchers must reduce each complex interpersonal communication into a simple index to track a few components within a given interpersonal interaction (Bernieri, Davis, Rosenthal, & Knee, 1994). For example, Matarazzo, Saslow, Wiens, Weitman, and Allen (1964) paid attention to head movements while Tronick, Als, and Brazelton (1977) examined facial expressions. Their results about each communication channel (i.e., head movement or facial expression) were interesting, but ignoring the other communication channels (e.g., gaze, speech, body orientation, and so on) was problematic. Another limitation is that researchers could not establish the existence of genuine non-random interactional synchrony that occurs beyond a chance level, since the statistical baseline is not considered (Cappella, 1981). For those reasons, an alternative way to measure interactional synchrony had been sought (Bernieri & Rosenthal, 1991).

An alternative to the behavior coding method is the gestalt approach: the rating method (Bernieri & Rosenthal, 1991). Based on the conceptualization of social affordance (Baron & Boudreau, 1987), it is hypothesized that observers could directly perceive the degree of interactional synchrony. This rating method is called “the pseudosynchrony experimental paradigm.” The validity of this method has been confirmed repeatedly (Bernieri, 1988; Bernieri et al., 1994; Bernieri, Resnick, & Rosenthal, 1988). The procedure is as follows. After judges observed dyadic interaction clips with the sound turned off, they rated their gestalt impressions of interactional synchrony as a whole. Judges used a rating form based on three aspects of interactional synchrony: (1) simultaneous movement, (2) tempo similarity, and (3) coordination and smoothness. An important aspect of this paradigm is the generation of a perceived synchrony control stimulus as a means to represent the existence of genuine non-random interactional synchrony that occurs beyond chance level. Two different kinds of dyadic interaction clips were constructed as stimuli, either from genuine or pseudo-interactions. Then, the extent of perceived synchrony was compared. If the former were higher than the latter, researchers would conclude that interactional synchrony had occurred.

In the pseudosynchrony experimental paradigm, the extent of perceived synchrony of genuine interactions is compared with that of pseudo-interactions. Genuine interaction clips were the originally recorded interactions without experimental manipulation. In contrast, the pseudo-interaction clips were artificially synthesized interactions. They were constructed by isolating the video images of each interactant and then pairing them with the video images of other interactants recorded in other interactions. There are two types: internal pseudo-interaction and external pseudo-interaction.Footnote 1

In internal pseudo-interaction, individual behaviors from the same interaction were paired, but these interaction clips were constructed by pairing the recorded behavior of one interactant in one time period with the behavior of the partner in another time period (Bernieri et al., 1994). For example, if a certain interaction continued for 3 min, a segment of an interactant in the first minute was paired with the partner she was interacting with in the last minute. It was thought that the sequential phases of conversation change as time goes by, and this was manipulated in the internal pseudo-interaction. But because the partners had actually interacted, the atmosphere between the interactants that was formed in face-to-face communication (e.g., rapport; Tickle-Degnen & Rosenthal, 1987) was not manipulated in this type of pseudo-interaction (Bernieri, 1988; Bernieri et al., 1988, 1994).

The other type was external pseudo-interaction. In external pseudo-interaction, individual behaviors from different interactions were paired, so the two partners had not actually interacted with each other. Those interaction clips were constructed by pairing the recorded behaviors from the same time period in different interactions (Bernieri et al., 1994). Because the partners had not actually interacted, the atmosphere between interactants that was formed in face-to-face communication was manipulated. Because those interactants were paired from the same time period, the sequential phases of conversation were not manipulated.

Conversation about Emotional Episodes

We focused on conversations about emotional episodes in this study. Communication behaviors change dynamically according to the emotional tones of conversations (Planalp, 1999). When people are happy, proud, angry, fearful, sad, shamed, or embarrassed, they tend to display their emotions through a variety of communication channels (e.g., facial expression, body movement, gesture, gaze, nodding, paralanguage, and verbal behavior). Also, inter or intra-personal communication channels function reciprocally and complementarily (e.g., Argyle & Dean, 1965). Therefore, it might be difficult to understand interactional synchrony in a conversation about emotional episodes by only focusing on a particular communication channel. For example, using the behavior coding method and examining posture, Charney (1966) reported that interactional synchrony occurred in conversations about positive emotional episodes but not about negative emotional episodes. Therefore, to understand interactional synchrony in conversations about emotional episodes, it is important to track the dynamics of interpersonal communication as a whole using the pseudosynchrony experimental paradigm.

Kimura, Yogo, and Daibo (2004) studied interactional synchrony in conversations about emotional episodes using a within-participants pseudosynchrony experimental paradigm, the same method introduced by Bernieri et al. (1988). The experimenter presented genuine interaction clips and two types of pseudo-interaction clips (internal and external pseudo-interactions) of conversations about both positive and negative episodes. Positive episodes were the happiest events in recent days, and negative episodes were the saddest events in recent days. Then the judges were asked to rate the degree of perceived synchrony. The results showed that perceived synchrony was significantly higher for genuine interactions than for external pseudo-interactions for both types of emotional episodes. However, perceived synchrony was not significantly higher for genuine interactions than for internal pseudo-interactions involving either of the types of emotional episodes. In Bernieri et al. (1988), which studied perceived synchrony using the pseudosynchrony experimental paradigm, the judges rated the degree of perceived synchrony highest for genuine interactions, followed by internal, and external pseudo-interactions.

Do those results indicate that the judges could not perceive the differences between genuine interactions and internal pseudo-interactions in Kimura et al. (2004)? There were two possible answers for this question. One possibility was that judges actually could not perceive these differences. In Kimura et al. (2004), perceived synchrony was significantly higher for genuine interactions than for external pseudo-interactions, but was not significantly higher for genuine interactions than for internal pseudo-interactions for both emotional episodes. It was thought that while the atmosphere between interactants, which had been formed in face-to-face communication, would have influenced perceived synchrony strongly, the sequential phases of conversation would not have influenced it as much. That was consistent with Bernieri et al. (1988) who found that while perceived synchrony from internal pseudo-interactions showed smaller differences than genuine interactions, external pseudo-interactions showed larger differences than genuine interactions. However, if judges actually could not perceive the differences between genuine interactions and internal pseudo-interactions in Kimura et al. (2004), the validity of the measurement using this paradigm would be open to question.

The other possibility was that the within-participants design, in which the same judges rated all the types of the interactions, caused a contrast effect that influenced the perceived synchrony. Although Bernieri et al. (1988) also used a within-participants design, they had the raters come back on different days so they would never see the same targets in both pseudo and genuine interaction conditions on the same day. In contrast, Kimura et al. (2004) conducted the experiment all in the same day. That might make those tasks more difficult for the judges to differentiate the genuine or pseudo-interactions. This difference might be reflected in the results of Kimura et al. (2004). As previously mentioned, Bernieri et al. (1988) stated that external pseudo-interactions were discriminative as stimuli. In a within-participants design, when external pseudo-interactions were compared with internal pseudo-interactions, contrast effects might make features of internal pseudo-interactions indistinctive. As a result, there might not be differences of perceived synchrony between genuine interactions and internal pseudo-interactions. So in this study, we examined the validity of these two possibilities by using a between-participants design in which different sets of judges rated each of the interaction types. Considering the consistent findings about the validity of this method (Bernieri, 1988; Bernieri et al., 1988, 1994), we speculated that the first possibility that judges could not perceive the differences between genuine interactions and internal pseudo-interactions would not be supported while the second possibility that a within-participants design caused contrast effects would be supported.

In this study, we compared perceived synchrony in conversations about positive episodes with conversations about negative episodes. According to Buck, Losow, Murphy, and Costanzo (1992), positive expressions were facilitated while negative expressions were inhibited in interpersonal communication. Japanese people especially inhibit negative expressions in public (e.g., Matsumoto & Ekman, 1989). In addition, Bernieri, Gillis, Davis, and Grahe (1996) reported that judgments of rapport were driven primarily by expressivity (expressivity halo effect). Kimura, Yogo, and Daibo (2005) reported that this expressivity halo effect was also observed in Japan, a high context culture (Takai & Ota, 1994). Taking into account this effect, it would be predicted that the perceived synchrony would be rated higher in conversations about positive episodes whose expressivity was facilitated than in conversations about negative episodes whose expressivity was inhibited.

Perceived Synchrony and Culture

Previous studies about perceived synchrony have been conducted in the U.S. (Bernieri, 1988; Bernieri et al., 1988, 1994). The results supported the validity of measurement using this paradigm. Moreover, in the study about judgments of rapport that were related to interactional synchrony (e.g., Tickle-Degnen & Rosenthal, 1987), the tendency for Americans to judge rapport was similar to that for Greeks (Bernieri & Gillis, 1995). Accordingly, it was thought that people in Western cultures have similar tendencies in the perception of interpersonal communication.

However, extending the cultural perspective to the comparison of Western cultures with Eastern cultures suggested cultural differences in cognitive styles (e.g., Markus & Kitayama, 1991; Nisbett, 2003). For example, according to Nisbett (2003), while people in Western cultures perceived the environment as a combination of elements, people in Eastern cultures perceived the environment as a whole. Considering these cultural differences of cognitive styles in relation to the measurement of interactional synchrony using the pseudosynchrony experimental paradigm, it remains to be seen whether the results in Japan would be the same as in the U.S. There could be the possibility that cultural differences in cognitive styles could influence perceived synchrony. The differences between Kimura et al. (2004) and previous studies (Bernieri, 1988; Bernieri et al., 1988, 1994) might be a result of cultural factors. If this is the case, then the use of the paradigm in Japan might not be valid. Therefore, from a cultural perspective, we were interested in exploring the validity of the pseudosynchrony experimental paradigm in Japan.

Hypotheses in this Study

In this study, we used the between-participants pseudosynchrony experimental paradigm and tested the following two hypotheses.

Hypothesis 1: Interactional synchrony will occur in conversations about negative episodes as well as in conversations about positive episodes.

Hypothesis 2: Perceived synchrony will be higher in conversations about positive episodes than in conversations about negative episodes.

Method

Participants

Eighty-one Japanese university students (21 male and 60 female) participated in the experiment as judges. This is a completely different sample from Kimura et al. (2004). All participants were unfamiliar with the interactants in the stimuli. Because of the difference in the sample size between males and females, and to compare with the results in Kimura et al. (2004), only the data from female university students were analyzed in the subsequent analyses in this study.Footnote 2

Materials

Conversation about Emotional Episodes

We prepared interaction video clips as stimuli. The interactants in the video clips were the 42 Japanese female university students (i.e., 21 dyads) from Kimura et al. (2004)’s study of perceived synchrony. The procedures were as follows. Female interactants were combined into dyads. They had not previously been acquainted with each other. The dyads were randomly assigned to one of two conditions (positive or negative emotional episodes condition). When the dyads arrived at the lab, they were ushered into a small experimental cubicle and seated at a table. They were asked to recall memories about the happiest (or saddest) events in recent days and describe them on sheet of paper. After that, they were asked to move to the next room, sit on chairs, and talk for 3 min about the positive (or negative) emotional episodes that they had written down. A video camera recorded their entire bodies during the conversation. During the interaction, they were given the role as a narrator or listener and were told not to change roles, but we did not put any other restrictions on them.

Pilot Study for Stimuli Selection

Kimura et al. (2004) conducted a pilot study for stimuli selection. Nine naïve undergraduate and graduate students (5 male and 4 female) participated in the study. All stimuli were the conversations of the 21 dyads described above. Every stimulus was presented on a screen without sound, and participants were asked to rate three aspects of the interaction: (1) simultaneous movement, (2) tempo similarity, and (3) coordination and smoothness (based on Bernieri et al., 1988; Bernieri & Rosenthal, 1991). Each rating was made on a 9-point Likert scale.Footnote 3 The Cronbach alpha reliability coefficient for the 3 items was .93. We constructed our rating scales to represent these three aspects of perceived synchrony (positive episodes M = 17.05, SD = 2.22; negative episodes M = 15.21, SD = 2.38; t(15) = 1.64, ns). Based on those scores, we selected three dyads, the highest, the middle, and the lowest, from each emotional episode as stimuli for the following experiment.

Stimuli Tape Editing

Thirty-second video segments of the narrator and listener were extracted from the 3 min in each interaction. We used a split-screen special effects generator for the construction of video clips that paired the recorded image of the narrator in one time-period with the image of the listener recorded in any other time-period. This allowed us to combine individual sequences to form artificially re-created on-screen dyadic interactions that never took place. In order to control for editing effects, we adapted the same procedure to create genuine dyadic interaction video clips. Three types of dyadic interaction clips (genuine interaction, internal pseudo-interaction, and external pseudo-interaction) were constructed from these individual behavior segments. The total number of stimuli was 18: 2 (emotional tone of conversation: positive and negative episodes) × 3 (interaction type: genuine interaction, internal pseudo-interaction, and external pseudo-interaction) × 3 (score of perceived synchrony: high, middle, and low).

Genuine interaction. The first type of interaction clip re-created the originally recorded interaction. That is, the two interactants on screen had actually interacted, and the sequential phases in the conversation were maintained.

Internal pseudo-interaction. The second type of interaction clips were edited into ‘‘altered time frame’’ clips as described in Bernieri et al. (1988). These pseudo-interaction clips were constructed by pairing the recorded behavior of the narrator in one time period with the behavior of the listener in another time period. For example, a segment of the narrator in the last 30 s was paired with the listener’s first 30 s of interaction. Similarly, a segment of the narrator in the first 30 s was paired with the segment of the listener’s last 30 s of the interaction. Thus, the time frame was altered but the dyads had actually interacted.

External pseudo-interaction. The third type of interaction clips were edited into ‘‘switched partner’’ clips as described in Bernieri et al. (1988). In the external pseudo-interaction, individual behaviors from different interactions were paired, which means that the people appearing on the video monitor were actually interacting with a different partner off screen. For this type of clip, the time frame was held constant, and the behavioral sequences were recorded from the same 30 s of each interaction.

Perceived Synchrony Criterion Measurement

The participant judges used a rating form based on the three aspects of interactional synchrony according to Bernieri et al. (1988; Bernieri & Rosenthal, 1991). A cover sheet explained what each rating was designed to measure, and the 60 judges were told that the rating definitions could be interpreted “loosely and liberally.” Each was rated on a 9-point Likert scale. The descriptions provided to the judges were as follows.

Simultaneous movement. This reflects the quantity or degree of movement that appears to begin or end at the same moment. If a narrator begins to turn her head at the precise moment that a listener raises her face, it is an instance of simultaneous movement.

Tempo similarity. This reflects the degree of the similarity of interactants’ tempos of behavior. Assume that all people have built-in tempos or speeds at which their behavior is set (much like the tempo an orchestra follows at a concert).

Coordination and smoothness. This reflects the extent to which the behaviors of these interactants fit together evenly. You are asked how smooth the interactants’ flow of behavior is.

Judgment Procedure

The 60 female participants were randomly assigned into three groups: (a) genuine interaction condition, (b) internal pseudo-interaction condition, and (c) external pseudo-interaction condition. Each condition consisted of 20 female participants. In each condition, six stimuli video clips were presented without sound. Those stimuli consisted of three conversation clips about positive emotional episodes and three conversation clips about negative emotional episodes. Every time a stimulus was presented, participants rated the three aspects of the interactions as described above: (1) simultaneous movement, (2) tempo similarity, and (3) coordination and smoothness.

Results

The rating scales represent the three aspects of perceived synchrony: simultaneous movement, tempo similarity, and coordination and smoothness. The mean ratings for the positive and negative emotional episode conditions appear in Table 1.Footnote 4 The ratings on the three scales were highly inter-correlated. The Cronbach alpha reliability coefficient for the entire scale was .89 for conversations about positive emotional episodes and .90 for conversations about negative emotional episodes. We decided to average the three variables to form a composite variable of perceived global synchrony. The resulting data analyses are based on this composite measure.

Table 1 Mean ratings for three aspects of perceived synchrony (N = 60)
Full size table

Interactional Synchrony in Conversations about Emotional Episodes

To test Hypothesis 1, we conducted analyses of the scores for interactional synchrony in the conversations about positive and negative episodes (see Table 2). To examine perceived synchrony in the conversations about positive emotional episodes, we conducted a one-way ANOVA on the scores of perceived synchrony in the conversations about positive emotional episodes. The main effect of type of interaction was significant (p(2, 57) = 5.52, p < .01). Then, we compared the three types of interactions using Tukey’s method. The results show that the scores of perceived synchrony were significantly higher for the genuine interactions (M = 5.42, SD = 1.08) than for the internal (M = 4.77, SD = 1.02) and external pseudo-interactions (M = 4.39, SD = .95).Footnote 5 These results suggest that interactional synchrony was perceived in the conversations about positive emotional episodes.

Table 2 Mean scores of perceived synchrony for positive and negative episodes (N  = 20 per each condition)
Full size table

Next, we conducted a one-way ANOVA on the scores of perceived synchrony in the conversations about negative emotional episodes. The main effect of type of interaction was significant (F(2, 57) = 18.16, p < .0001). Then, we compared the three types of interactions using Tukey’s method. The results show that the genuine interactions (M = 4.59, SD = .81) yielded the highest scores of synchrony; internal pseudo-interaction (M = 3.75, SD = 1.05) was second highest; and external pseudo-interaction (M = 2.84, SD = .97) was the lowest. These differences were statistically significant in all three of the pairwise comparisons among conditions. These results suggest that interactional synchrony was perceived in the conversations about negative emotional episodes.

These results support Hypothesis 1. In Kimura et al. (2004), perceived synchrony was not significantly higher for genuine interactions than for internal pseudo-interactions for both emotional episodes. Our present study suggests that the results in Kimura et al. (2004) might have occurred because the within-participants experimental design led to contrast effects, instead of being attributed to the inability of the judges to perceive the differences between genuine interactions and internal pseudo-interactions (i.e., the sequential phases of conversation).

To examine in more detail the interactional synchrony in the conversations about emotional episodes, we compared perceived synchrony in the conversations about positive episodes with negative episodes (the test of Hypothesis 2). We conducted a 3 × 2 ANOVA (type of interaction × emotional tone of episode) using the three types of interaction clips as a between-participants factor and the emotional tone of the conversation (positive and negative episodes) as a within-participants factor.Footnote 6 The results showed no interaction between those two factors (F(2, 57) = 2.25, ns). But the main effect for type of interaction was significant (F(2, 57) = 15.97, p < .001). Then, we compared the three types using Tukey’s method. The results showed that genuine interactions were rated highest in the scores of synchrony, internal pseudo-interactions were the second, and external pseudo-interactions were the lowest. These differences were statistically significant. The main effect of the emotional tone of conversation was also significant (F(1, 57) = 59.93, p < .001). The perceived synchrony was higher for the positive episode conditions than for the negative episode conditions. These results support Hypothesis 2.

Discussion

When we compared perceived synchrony in the conversations about positive episodes with negative episodes, perceived synchrony was rated as higher in the positive episode conditions than in negative episode conditions. There could be two interpretations. One interpretation was that those differences were due to actual synchrony. According to previous studies (e.g., Buck et al., 1992; Matsumoto & Ekman, 1989), expressivity was higher for positive emotions than for negative emotions. So, positive and negative emotions might produce different degrees of synchrony. The other interpretation would be associated with cognitive bias, the expressivity halo effect (Bernieri et al., 1996; Kimura et al., 2005). Expressive behavior was more active in conversations about positive episodes than in conversations about negative episodes. In Kimura et al. (2005), participants considered the degree of rapport to be higher for positive episodes than in negative episodes. Tickle-Degnen and Rosenthal (1987) reported a positive relationship between rapport and synchrony. Therefore, the degree of perceived synchrony might be rated as higher for positive episodes in which expressivity was facilitated than for negative episodes in which expressivity was inhibited. However, it is difficult to determine which interpretation is acceptable from this study.

If the expressivity halo effect influences perceived synchrony, this could be a limitation of the pseudosynchrony experimental paradigm because measurement in this paradigm depends on judges’ perceptions. If this paradigm is influenced by a cognitive bias like the expressivity halo effect, there could be a possibility that interactional synchrony measured by the pseudosynchrony experimental paradigm would not match that measured by a behavior coding method.

Despite this possibility, studying interactional synchrony as measured by the pseudosynchrony experimental paradigm might have a special benefit. Perceived synchrony as rated by the judges reflected interpersonal communication of the type that is observed in everyday life. What people perceive in daily life is not synchrony measured by a coding method using objective criteria, but synchrony perceived under various cognitive limitations (e.g., cognitive bias). Interactional synchrony measured by the pseudosynchrony experimental paradigm is of ecological significance (e.g., Gillis, Bernieri, & Wooten, 1995; Grahe & Bernieri, 2002). In addition, people usually judge rapport from interpersonal communication as observers and these judgments are closely related to social adaptation (Bernieri & Gillis, 1995; Berinieri et al., 1996; Hall & Bernieri, 2001; Kimura et al., 2005; Tickle-Degnen & Rosenthal, 1987). Perceived synchrony measured by the pseudosynchrony experimental paradigm from the perspective of observers is associated with how people adapt in society, and studying it has a special ecological value.

Perceived Synchrony from a Cultural Perspective

In this study, we examined the validity of the measurement of perceived synchrony using the pseudosynchrony experimental paradigm in Japan. Consistent with previous studies in Western cultures (Bernieri, 1988; Bernieri et al., 1988, 1994), perceived synchrony was higher in genuine interactions than in pseudo-interactions. Most studies of perceived synchrony have been conducted in the United States. Moreover, according to Bernieri and Gillis (1995), Americans and Greeks had similar characteristics of judgments of rapport, which was reported to be related to synchrony (e.g., Tickle-Degnen & Rosenthal, 1987). These results are consistent but did not go beyond Western cultures. Cross-cultural studies suggest that there are cultural differences in cognitive styles. Takai and Ota (1994) reported that people in a high-context culture like Japan were more sensitive to the social environment than people in a low-context culture like the United States. Also, it is known that cognitive styles in Western cultures are different from those in Eastern cultures (Markus & Kitayama, 1991; Nisbett, 2003). Taking into account these things, there would have been a possibility that the results from the measurement of perceived synchrony using the pseudosynchrony experimental paradigm in Japan might not have been consistent with previous studies in Western culture.

We applied the pseudosynchrony experimental paradigm to Japanese university students. The results are consistent with previous studies in Western cultures. This supports the validity of measuring perceived synchrony using the pseudosynchrony experimental paradigm. In spite of differences in cognitive styles between Western and Eastern cultures, these results support the universality of perceived synchrony. Interactional synchrony was also observed in early infancy (Bernieri et al., 1988). All this suggests that perception of interactional synchrony is inherent in human beings as a fundamental biological mechanism.

Future Directions

This study used a rating method and confirmed that interactional synchrony was perceived in conversations about negative episodes as well as positive episodes. In addition, the degree of perceived synchrony was higher for positive episodes than for negative episodes. These results are consistent with Charny (1966), who used a coding method, in terms of the direction of effect but did not match in terms of the magnitude. These results illustrate the similarities and differences between using a coding method and a rating method for measuring synchrony. Some researchers studying interpersonal communication have used both methods. In research on the judgment of rapport, several studies examined how people used objective cues, as measured by the coding method, or subjective cues, as measured by the rating method, when they perceived interpersonal communication (e.g., Gillis et al., 1995; Grahe & Bernieri, 2002). In future research, interactional synchrony should be studied with both methods at one time to further understanding of interpersonal communication about emotional episodes.