3.1 Static cues

Physical attractiveness is one of the most commonly studied static cues. According to the what-is-beautiful-is-good hypothesis, people perceive good looking people as possessing an array of positive internal characteristics, such as intelligence, sociability, and trustworthiness (Dion, Berscheid and Walster 1972). Meta-analyses suggest that this beauty bias is stronger for judgments about social attributes, such as confidence and extraversion, than cognitive attributes, such as intelligence and mental health (Eagly et al. 1991). There is also an actual (rather than merely perceived) relationship between social skills and physical attractiveness, with attractive individuals appearing more self-confident than less attractive individuals (Feingold 1992), presumably because people treat attractive people better, leading them to feel more confident. Even in virtual environments, people assigned to be represented by attractive avatars engage in more confident behavior, such as standing closer to other avatars, than do people assigned to less attractive avatars (Yee and Bailenson 2007), which suggests that people believe attractive individuals act differently than unattractive individuals. However, being physically attractive does not lead to uniformly positive perceptions. The what-is-beautiful-is-conceited hypothesis specifies that people who are especially attractive are also perceived as snobby, materialistic, vain, and self-involved (Dermer and Thiel 1975).

Given that people associate physical attractiveness with a host of positive internal characteristics, it is not surprising that physical attractiveness often acts as a screening device narrowing the field of potential dating partners. Dating service studies have shown that physical appearance is one of the best predictors of whether people click on someone’s videotaped profile (Woll 1986) or desire further interaction after speed-dating (Kurzban and Weeden 2005). On Facebook, people are especially likely to want to pursue friendships with members of the opposite sex who post attractive pictures of themselves (Wang et al. 2010).

The most physically attractive people, however, do not always get the most dates. The matching hypothesis suggests that individuals tend to pair up with people who are similar to themselves in terms of level of physical attractiveness (Berscheid and Walster 1974). So people who consider themselves moderately attractive will look for partners who are also moderately attractive rather than looking for the most attractive person available. This is because people worry about being rejected by individuals who are more attractive than themselves. People also generalize similarity in physical attractiveness to similarity in other areas, leading them to think they will be more compatible with someone who possesses a level of physical attractiveness comparable to their own. The matching hypothesis applies to friendships as well as romantic relationships (Feingold 1988), differentiates between people who are really dating versus those who are posing as a couple (Murstein 1972), helps determine who has a good time on a date (Berscheid and Walster 1974), and predicts who will have a happy relationship (Zajonc et al. 1987).

In addition to general physical attractiveness, specific nonverbal behaviors such as voice and body type influence initial impressions. The what-sounds-beautiful-is-good hypothesis specifies that people with attractive voices are rated more favorably on characteristics such as agreeability, friendliness, and trustworthiness (Zuckerman and Driver 1989). People also believe themselves to be more similar to individuals with attractive versus unattractive voices (Miyake and Zuckerman 1993). In terms of body type, within the United States, mesomorphs, whose weight and height are in balance, are perceived somewhat more positively than ectomorphs, who are thin in relation to their height, and much more positively than endomorphs, who are heavy in relation to their height (Portnoy 1993). During speed-dating sessions, those with a healthy body mass index are more likely to be regarded favorably and considered for future dates that are those with an unhealthy body mass index (Kurzban and Weeden 2005).

3.2 Dynamic cues

Dynamic cues can modify or add to the impressions that people make based on appearance. Albada, Knapp, and Theune (2002) advanced interaction appearance theory to explain how interaction can modify people’s first impressions about attractiveness. This theory rests on four principles: (1) people tend to believe that an ideal partner should be physically attractive and a good communicator; (2) people communicate with a wide range of individuals, including some people whom they do not consider to be optimally physically attractive; (3) when people start to develop a satisfying relationship with a person who is a good communicator but not optimally physically attractive, their beliefs about what constitutes an ideal partner are challenged; and (4) to resolve this inconsistency, they re-evaluate their partner as being more physically attractive. Albana et al. (2002) showed that positive interaction modified original perceptions of physical attractiveness as predicted by the theory. Negative interaction had a stronger modifying effect than positive interaction, such that participants downgraded a person’s physical appearance following a negative interaction more than they upgraded a person’s physical appearance following a positive interaction. Thus, physical appearance alone is not enough to shape initial impressions, especially if a person’s communication does not match expectations based on such stereotypes as the what-is-goodis-beautiful hypothesis.

Other research has examined specific dynamic cues that are related to receiver judgments of personality. Much of the research in this area has utilized Brunswik’s (1956) lens model as a theoretical framework. This model focuses on encoding (or ecological validity) that displays an element of a sender’s personality. Decoding (or cue utilization) occurs when receivers make inferences about a sender’s personality based upon the sender’s behavior. Both encoding and decoding are influenced by the context in which communication occurs. Research in this area has shown that low levels of smiling are decoded as a sign that someone is a serious person, speaking with a hand over one’s mouth while smiling is decoded as a sign of shyness, and speaking in a tense voice and turning one’s head away while speaking are decoded as signs of being uptight (Ferrari and Swinkels 1996). In initial interactions, these first impressions often affect whether a person wants to get to know someone better.

Certain nonverbal cues also make a person appear more agreeable and likable. Of these, smiling is the most important. Compared to people with neutral facial expressions, people who are smiling are perceived as more sincere, sociable, and competent, as well as less independent and masculine (Reis et al. 2012). On Facebook, people who are smiling and engaging in social interaction with others in their photos tend to be rated as more expressive and outgoing than those who are not (Weisbuch, Ivcevic and Ambady 2009). On email, people perceive others as more likable when they include a smiley emoticon in their messages (Byron and Baldridge 2007).

These types of associations help explain a phenomenon called thin slice impressions (Ambady, Krabbenhoft and Hogan 2006). Thin slice impressions occur when people base initial judgments about a person on a short sample (less than five minutes long) of dynamic behavior, such as facial expressions, posture, voice, and gesture (Ambady et al. 2006). For example, a person who smiles and expresses vocal warmth is likely judged as friendly. These thin slice impressions are often related to enduring and accurate judgments about people. For example, Ambady and Rosenthal (1993) found that thin slices of teacher behavior sampled at the beginning of the semester predicted student and principal evaluations at the end of the semester. Thin slices of managers’ (DeGroot and Motowidlo 1999) and sales-people’s (Ambady et al. 2006) voices not only predict initial impressions, but are also related to job performance and supervisor evaluations.

Some scholars have cautioned that although thin slice impressions are often more accurate than chance, they cannot always be trusted (Ames et al. 2010). People vary in their confidence levels when judging people based on nonverbal cues. Those who are confident that their first impressions are right are not necessarily more accurate in their perceptions; however, there is evidence that a lack of confidence in the accuracy of first impressions based on nonverbal cues is indicative of inaccuracy (Ames et al. 2010). At least three factors may influence the accuracy of first impressions based on thin slices of behavior (Human et al. 2012). First, some senders are more transparent and easier to judge. Second, some people are better decoders of nonverbal information. Third, people who are motivated to access someone’s personality are often more accurate, especially in terms of noting the distinctive characteristics of an individual. (See Gifford, 2012, for other conditions that make it difficult to judge personality accurately from nonverbal behavior.)

4.1 Positive involvement behaviors

The nonverbal cues that communicate liking and intimacy have variously been referred to as immediacy or positive involvement behaviors. In line with Mehrabian’s (1981) original conceptualization, Andersen (1985) defined nonverbal immediacy as a set of behaviors that indicate physical and psychological closeness, approachability, sensory stimulation, interest, and interpersonal warmth. Prager and Roberts (2004) used the term positive involvement to describe behaviors that commonly reflect both positive affect and involvement, such as gaze, smiling, forward leans, affectionate touch, and affirming head nods. Some scholars prefer the term positive involvement because they view immediacy and affect as orthogonal dimensions that underlie broad concepts such as intimacy and dominance (Dillard, Solomon, and Palmer 1999). Under this view, affective cues such as smiling, scowling, vocal warmth, and sarcastic tone determine whether immediacy cues (such as forward leans, an animated voice, and eye contact) are interpreted as reflecting liking and intimacy, or something else, like dominance or aggression. Despite the use of different terms, there is consensus regarding the specific nonverbal cues that reflect positive involvement, liking, and intimacy. These include: close proxemic distancing, touch, direct body orientation, open posture, increased or sustained gaze that is evaluated as friendly rather than intimidating, smiling, and vocal qualities that communicate warmth, expressiveness, and relaxation (Andersen 1985; Coker and Burgoon 1987; Patterson 1983).

Research on positive involvement or immediacy is also consistent in demonstrating the importance of reciprocity. Indeed, several theories of nonverbal intimacy (see Burgoon, Stern and Dillman 1995; Chapter 10, Burgoon, Dunbar and White) rest on the following related principles: (1) Patterns of reciprocal positive involvement cues produce the most intimate interactions and occur in the closest and most satisfying relationships, and (2) Receivers are most likely to reciprocate or converge to the positive involvement behaviors of senders who they like or consider to be rewarding. In support of these principles, married partners engage in similar levels of public touch toward one another (Guerrero and Andersen 1994), and people who like one another echo each other’s body positions, adopt mirror-image postures, and use similar gestures (Maxwell et al. 1985).

Positive involvement behavior also changes as relationships move from casual to close. People are especially affectionate when moving their romantic relationships toward more commitment. Once a couple is fully committed, there is often a drop in immediacy. For example, a study that involved observing people waiting in lines at movie theaters and zoos showed that couples in serious dating relationships touched about twice as much as casually dating or married couples (Guerrero and Andersen 1991). A similar curvilinear relationship emerged in a study looking at level of intimacy and private touch (Emmers and Dindia 1995). These findings support a principle of nonverbal escalation, in that nonverbal immediacy cues tend to increase as a romantic relationship escalates toward commitment and sexual involvement, but then decrease and level off after the relationship is fully committed.

4.2 Flirtatious behavior

Positive involvement behavior communicates liking, and is therefore used to develop, intensify, and maintain a variety of relationships. Flirtatious behavior, on the other hand, expresses sexual and romantic interest, and is therefore inappropriate in most platonic relationships. Although studied most in the context of courtship, flirting also helps maintain established relationships (Henningsen 2004), including marriages (Frisby 2009). In general, flirtatious behavior is more indirect and ambiguous during the initial stages of an interaction (in part to save face if one’s advances are rejected), and then becomes bolder, more intense, and increasingly synchronized as interaction unfolds (Cunningham and Barbee 2008; Givens 2005).

Ambiguous flirtatious behavior includes coy smiles and eye contact that is characterized as darting, fleeting, or room-encompassing rather than focused on the (potential) partner (Coker and Burgoon 1987; Moore 2002). After two people have secured one another’s attention and determined that interest might be mutual, flirtatious behaviors subtly communicate approachability and openness. Specific cues include maintaining an open-stance, directing one’s full body or head and shoulders toward the partner, pointing one’s feet toward the partner, sitting or standing next to the partner, or leaning to decrease the distance between oneself and the partner (Givens 2005; Grammer, Kruck and Magnusson 1998; Moore 2010; Muehlenhard et al. 1986). Men may be more likely to secure and keep a woman’s attention if they have open body positions with legs or arms extended (Renninger and Wade 2004).

Eventually, flirtatious behavior may become bolder and less ambiguous. Koeppel et al. (1993) found that behavior moved from being interpreted as friendly, flirtatious, and eventually, seductive, based on increasingly higher levels of smiling, eye contact, and touch. Bold behaviors include extended mutual gaze; wider, larger smiles; and various forms of touch (Givens 2010; Moore 2002). Touch to relatively non-vulnerable areas of the body, such as the forearm, wrist, lower back and shoulders are perceived as less flirtatious than those to more vulnerable areas of the body such as the legs, neck, and face (Muehlenhard et al., 1986).

Flirtation also includes unique behaviors that go beyond positive involvement. Preening behaviors, such as holding one’s head and shoulders high, standing upright, sucking in one’s stomach, pushing out one’s chest (Grammer et al. 1998), smoothing or adjusting one’s hair or clothing (Givens 2005), and licking or puckering one’s lips (Moore 2002), help people maximize their attractiveness. Other behaviors fairly unique to flirting include laughing while tossing one’s head back, tossing or playing with one’s hair, and tilting one’s head while looking up or down at someone (Givens 2005; Moore 2002, 2010). Men’s voices tend to switch from being high-pitched and animated during the beginning phases of flirtation, to being more low-pitched and monotone as the interaction progresses (Anolli and Ciceri 2002).

Finally, flirtatious interaction is often characterized by increasingly synchronized behavior. Cunningham and Barbee (2008) noted that dating rituals such as dancing may help potential partners determine if they are compatible by providing a testing ground for synchronization. When partners flirt, they tend to match one another’s posture, body positioning, vocal behavior, and laughter (e.g., Grammer et al. 1998; Muehlenhard et al. 1986). Thus, just as reciprocity is important for developing and maintaining relationship via positive involvement cues, so is synchronization important for creating intimacy via flirtatious behavior.

5.1 The bright side of nonverbal dominance

Relational partners have many verbal and nonverbal behaviors at their disposal when they want to influence one another. Nonverbal cues reflecting both dominance and social skill are especially likely to be successful in gaining compliance, solving problems, and eliciting long-term change. These behaviors can be grouped into three broad categories based on whether they primarily reflect (a) poise and self-assurance, (b) panache or dynamism, or (c) skill in interaction management (Burgoon and Dunbar 2000). Behaviors reflecting poise and self-assurance include: asymmetrical leg and arm positions; sideways leaning; open arm and body positions; kinesic animation; a low amount of swiveling, adaptors, and random movement; fluent speech; facial pleasantness and smiling; eye contact; a moderately fast and loud voice; and relatively high levels talk time (Guerrero and Floyd 2006). People who have panache communicate in a “dramatic, memorable, and attention-grabbing communication style that is immediate, expressive, and energetic” and includes nonverbal behaviors such as close distancing, gaze and direct body orientation, forward lean, vocal and kinesic expressiveness, and faster, louder speech (Guerrero and Floyd 2006: 156). Finally, nonverbal behaviors that facilitate smooth interaction management and conversational control include speaking fluently, engaging in smooth turn-switching, talking for an extended time, and using eye contact when speaking, all of which can be perceived as socially skilled behaviors that contribute to better problem solving in relationships (Spitzberg and Hecht 1984).

Another bright side effect of nonverbal dominance is that, in some cases, appearing dominant leads to attraction. Sadalla, Kenrick, and Vershure (1987) demonstrated that women rated men using dominant nonverbal behavior (i.e., sitting close to a person behind a desk while leaning backward slightly in a relaxed fashion and engaging in frequent gesturing) as more attractive than men using submissive behavior (i.e., looking down, nodding, and sitting farther away). In a study by Ahmetoglu and Swami (2012), low dominance was represented by a closed body position, moderate dominance by an open body position, and high dominance by an open body position and gesturing. Women rated men in the high dominance condition as most attractive. Maner, DeWall and Galliot (2008) found that participants looked longer at male faces that were rated as dominant.

Research has qualified these findings by showing that dominance may only be related to attractiveness and dating desirability if a man is also rated as agreeable (Jensen-Campbell, Graziano, and West 1995) and competent (Touhey 2011). Thus, nonverbal behaviors that are perceived as a sign of both dominance and social skill are the most effective in fostering positive impressions. Dominant behaviors may also be especially likely to be perceived positively when they reflect prestige rather than aggression and are used in acceptable contexts, such as a game or sports contest (Snyder, Kirkpatrick, and Barrett 2008).

5.2 The dark side of nonverbal dominance

Rather than using dominant behaviors that reflect social skill, people sometimes use intimidation or control to get what they want. Intimidating nonverbal behaviors include: a hard stare, angry face, loud voice, extended silence, standing over someone, and encroaching upon someone’s personal space (Dovidio and Ellyson, 1985; Montepare and Dobish 2003). Behaviors perceived as threatening or controlling typically have negative effects on relationships. Ostrov and Collins (2007) found that partners who displayed intrusive touch and resource control (by not letting each other touch or sort the cards during a game), tended to report being more aggressive and having more arguments during a collaborative problem solving task. Raising one’s voice was also associated with intrusive touch for females, and resource control for males.

Dominance cues that reflect intimidation and control are sometimes a manifestation of power imbalances within relationships. For example, the chilling effect occurs when the less powerful person in a relationship is silent and withholds complaints because he or she worries that speaking up will produce negative relational consequences, such aggression or breakup (Roloff and Cloven 1990). Several conditions make people more susceptible to the chilling effect, including: being more dependent on, more invested in, and more committed to the relationship than one’s partner; and having less power than one’s partner (Roloff and Solomon 2002; Knobloch, and Fitzpatrick 2004).

Intimate terrorism is an enduring and strategic pattern of behavior that involves using threats and violence to control one’s partner (Johnson and Ferraro, 2000), leading to a lopsided power imbalance in a relationship (Olson, 2002). In around 87% of cases, men are the perpetrators of intimate partner violence and women are the victims (Graham-Kevan and Archer 2003). Johnson (2006) noted that intimate terrorism is related to the chilling effect, in that the individual who is being controlled usually becomes increasingly fearful of speaking out. Intimate terrorism typically increases in severity and frequency over time, and the perpetrators of intimate terrorism usually become more possessive and violent when their partners resist or rebel against their controlling behavior (Graham-Kevan and Archer 2003; Johnson and Ferraro, 2000). Violent tactile behaviors, as well as nonverbal forms of intimidation, are obvious instruments for inflicting intimate terrorism. Yet research has yet to uncover specific nonverbal behaviors that perpetrators use to control their partners.

Nonverbal behaviors are also implicated in the cycle of intimate terrorism in another way – perpetrators of intimate terrorism not only engage in violent behaviors, they also use especially affectionate, apologetic, and ingratiating behaviors to compensate for their hurtful behavior (Shackelford et al. 2005). These positive behaviors, many of which are undoubtedly nonverbal, such as increased affection, are also controlling because they are designed to keep the partner in the relationship despite its tumultuous nature.

5.3 Relative power

As research on the chilling effect and intimate terrorism shows, the power balance in a relationship is negotiated dyadically. Dyadic power theory (Dunbar 2004) helps explain how people negotiate this balance by focusing on relative power, which is the extent to which one partner has power in comparison to the other. According to the theory, people who have high relative power do not need to display dominant behavior because they already have considerable control in their relationships. This theory does not consider cases where power is rooted in intimate terrorism; instead power imbalances stem from one partner caring more, being more invested, having more resources, or depending more on the relationship. Dyadic power theory also predicts that people with low relative power are unlikely to exhibit much dominant behavior, either because of the chilling effect or because they do not believe their attempts at dominance will succeed (Dunbar and Burgoon 2005). Finally, dyadic power theory suggests that partners who have similar levels of power are most likely to engage in high levels of dominant behavior because they need to assert themselves, negotiate terms, and compete for resources. Thus, the theory predicts that there is an inverted u-shaped curvilinear relationship between relative power and nonverbal displays of dominance.

Research has shown that nonverbal behavior does indeed differ based on relative power, although not always in the directions originally predicted. In line with dyadic power theory, Dunbar and Burgoon (2005) found that people who perceived themselves as having slightly more or less power than their partners displayed more nonverbal dominance than did people who perceived themselves as having much more or less relative power. Contrary to the theory, however, people who perceived themselves as low in relative power interrupted more and used more illustrator gestures, perhaps as a way to try to gain the power they lack. Other contrary findings surfaced in a study manipulating the degree to which members of an unacquainted dyad had control of a task-related activity (Dunbar and Abra 2010). Participants in the low power position showed the least dominance, whereas those in the equal- and high-power positions did not differ significantly from one another. Research is needed to determine the exact nature of the association between relative power and displays of dominance.

6.1 Affectionate emotions: Joy and love

Joy and love are affectionate emotions that help sustain healthy relationships. Joy is expressed through many nonverbal cues, especially smiling. Scholars have distinguished between genuine smiles that reflect real happiness and fake smiles that are posed (Ekman, Davidson, and Friesen 1990). In line with the behavioral ecology approach, genuine smiles are also associated with cooperation and prosocial behavior (Brown, Palameta, and Moore 2003; Lakin, Chartrand, and Arkin 2002). In one study, people excluded from a group had more positive reactions to those who exhibited genuine versus fake smiles (Bernstein et al. 2010). Genuine smiling does more than signal that a person is happy, it also signals that a person is cooperative and agreeable, which can pave the way for positive social interaction.

Joy is also associated with vocal characteristics, body movement, and touch. Vocal characteristics include moderately loud volume, moderately high pitch, vocal animation, rapid and varied tempo, and laughter (see Burgoon et al. 2010 for a review). Joy is also associated with bouncy and bubbly behavior as well as bright, glowing faces (Shaver et al. 1987). When participants judged the emotions of computer-generated mannequins posed in various positions, mannequins with raised arms and a backward head tilt were rated as happy (Coulson 2004). Hertenstein et al. (2006, 2009) investigated how people communicate joy, along with other emotions, via touch. In the 2006 study, encoders from behind a curtain used swinging, shaking, and lifting hands as ways to communicate joy tactilely. In the 2009 study, when encoders and decoders were not separated by a curtain, encoders used squeezing, patting, lifting, shaking and hugging to communicate joy. Joy was decoded from these cues at a rate greater than chance.

Expressions of love occur less frequently than expressions of joy. Saying “I love you” is the most common and direct way of expressing love, but people also communicate love nonverbally. Marston and Hecht (1999: 286) noted that, “lovers report roughly equal numbers of verbal and nonverbal behaviors that communicate love in their relationships.” Nonverbal ways of communicating love include spending time together, doing special things for someone, showing affection, using positive forms of touch (such as hugging or kissing), giving gifts, sitting close, exchanging rings, and having sex (Lemieux 1996; Marston and Hecht 1999). Mutual gaze can also temporarily increase feelings of love (Kellerman, Lewis, and Laird 1989). Vocal qualities associated with love include softer voices and vocalizations such as “oooh” and “aaah” (Hatfield et al. 1995). Finally, love is associated with interlocking fingers, stroking and rubbing someone’s hand (Hertenstein et al. 2006), and moderately intense hugging (Hertenstein et al. 2009), with decoders able to interpret touch as reflecting love (rather than a different emotion) at a rate greater than chance.

6.2 Hostile emotions: Anger and contempt

People experience hostile emotions in their relationships, including anger, contempt, envy, and resentment. Of these, anger and contempt have received the most research attention. Although people can communicate angry feelings in constructive ways, it can be difficult to do so because of the attack action tendency associated with anger (Lazarus 1991). Body cues associated with anger include walking with stronger, heavier, and longer strides (Montepare, Goldstein, and Clausen 1987), standing with an erect posture, tilting one’s head backward, and raising one’s arms (Coulson 2004). Activity cues such as driving a car fast, clenching one’s fist, stomping, throwing things, and slamming a door are also associated with anger (Planalp et al. 1996; Shaver et al. 1987). Facially, anger is communicated via furrowed eyebrows, lips in a square shape or pressed together, and eyes bulging or in a fixed stare (see Burgoon et al. 2010 for a review). An angry voice typically sounds loud and fast, with a rising pitch if the person feels frustrated, or a low pitch if the person feels annoyed or threatened (Scherer and Wallbott 1994).

Anger is also expressed using touch. Hertenstein et al. (2006, 2009) found that, from behind a curtain, anger is communicated by squeezing, hitting, and trembling; in open space, anger is encoded and decoded through pushing and shaking that is strong and moderately intense, but of short duration. Violent behavior, including touch, is also related to anger. Among high school students involved in abusive relationships, over 70 percent of victims interpreted their partner’s violent behavior as reflecting anger (Henton et al. 1983). Nearly 54 percent of aggressors also saw their own violent behavior as motivated by anger. Anger is also associated with aggression in adult romantic relationships. Ellis and Malamuth (2000) found that anger/upset shared a .31 correlation with aggression. As of yet, however, little if any research has examined which specific forms of nonverbal aggression are fueled by anger.

Expressions of contempt are even more destructive to relationships than are expressions of anger (Gottman and Krokoff 1989). Anger tends to be associated with “short-term attack responses but long-term reconciliation, whereas contempt is characterized by rejection and social exclusion” (Fischer and Roseman 2007: 103). Contempt can be communicated nonverbally by raising and tightening the corner of one’s lip, looking down at someone, furrowing one’s brow as if confused, raising one’s eyebrows as if shocked or surprised, displaying facial expressions that show astonishment, rolling one’s eyes, and speaking in a sarcastic tone (e.g., Ekman and Friesen 1986; Gottman 1994; Scherer and Wallbott 1994). Sarcasm is communicated through speech that is louder, slower, and more low-pitched than normal (Rockwell 2000).

These types of nonverbal behaviors can escalate conflict. Indeed, subtle facial expressions that communicate disgust and contempt are among the best predictors of relational dissatisfaction and eventual decline (Gottman 1994). When husbands use contemptuous expressions, wives tend to display less positive affect and believe that relational problems are severe and will be difficult to solve (Gottman, Levenson, and Woodin 2001). Violent husbands use more contemptuous behavior than non-violent husbands (Holtzworth-Munroe et al. 1988). Studies of adolescents have also shown that girls perceive nonverbal expressions of contempt to be especially hurtful because they signal exclusion and imply that they are not as good as others are (Underwood 2004).

6.3 Dyadic patterns of emotional expression

In addition to communicating particular emotions, nonverbal expressions of emotion influence how both senders and receivers feel, as well as how they communicate. The facial feedback hypothesis provided the foundation for much of this work by showing that the movement of facial muscles sends signals to the brain, which then cause people to experience emotions consistent with the expressions formed by those muscles (Ekman, Levenson and Friesen 1983). Other researchers have proposed and supported similar hypotheses for vocal (Hatfield et al. 1995) and postural (Duclos et al. 1989) behaviors. Cappella (1993) advanced the interpersonal facial feedback hypothesis, which specifies that if a receiver mimics a sender’s expression, the receiver will likely feel some of the same affect that the sender is experiencing.

Work on motor mimicry and emotional contagion is based on similar reasoning. Motor mimicry occurs when a receiver reacts nonverbally to something that happens to a sender as if she or he is experiencing it, such as a mom wincing when she watches her son skin his knee (Bavelas et al. 1986). Bavelas and her colleagues found evidence of motor mimicry across emotions ranging from pain, embarrassment and disgust, to laughter, smiling, and affection. Other researchers have focused specifically on facial mimicry, which occurs when a receiver matches the facial expression of a sender (McIntosh 2006). Buck and Powers (2012) noted that neuroscientists emphasize the innate connections between motor mimicry, modeled behavior, and the acquired ability to feel empathy, whereas communication scholars focus on motor mimicry as a way of communicating support and similarity.

Motor mimicry is also theorized to promote larger patterns of emotional contagion. Work on the emotional contagion effect (Hatfield, Cacioppo, and Rapson 1994) suggests that people “catch” the emotions of those around them. This is in part because people often mimic the nonverbal cues of those around them. Mimicked facial expressions and body movements are theorized to trigger physiological changes that create a mood-contagion loop, with the changes in facial expression and body movement signaling the brain to feel whichever emotion is consistent with those changes.

Motor mimicry and emotional contagion are fairly common occurrences, especially in the context of social and personal relationships. Although some evidence suggests that people are less likely to mimic facial displays that convey low intensity positive emotion compared to high intensity positive emotion, high intensity negative emotion, or low intensity negative emotion (Fujimura, Sato, and Suzuki 2010), other research has demonstrated that people mimic even fleeting and relatively weak emotions similar to those occurring in casual interactions (Hess and Blairy 2001; Wild, Erb, and Bartels 2001).

Emotional contagion extends beyond face-to-face interaction. Hancock et al. (2008) induced some participants to feel sad and others to feel neutral before engaging in computer-mediated communication. The participants in the sad condition used more sad words, typed fewer words overall (similar to having less talk time) and engaged in slower message exchange compared to those in the neutral condition. Receivers in the sad condition experienced less positive affect than those in the neutral condition.

Emotional contagion and facial mimicry may be especially strong when people know and like one another (Kimura, Daibo, and Yogo 2008; McIntosh 2006). These patterns of emotional expression may help people feel comfortable with one another by increasing perceptions of similarity and helping people feel in sync. Hatfield and her colleagues (1994) argued that because emotional contagion leads to more synchronized interaction, it also fosters attraction, liking, and rapport. In group work settings, contagion of positive emotions can reduce conflict and increase cooperation (Barsade 2002).