Can you trust a customer's expression? Insights into nonverbal communication in the retail context†
The authors acknowledge the helpful suggestions of Jeffrey Nevid, Joseph G. Cunningham, and Anne Roggeveen on previous versions of this manuscript.
Synthesizing knowledge from psychology and marketing research, an understanding of nonverbal communication can help address when and how customers express their underlying feelings in retail interactions that are not evident in direct verbal expressions. Examining nonverbal behavior as an indirect measure of consumer response can enable retailers to better understand the needs of their customers. Nonverbal communication theory is used to develop a conceptual framework that builds on prior research on the situation, expressivity, social status, display rules, and their effects on customer expression. Lay wisdom suggests that customer expression should be revealing (e.g., “the eyes are the windows to the soul”). However, research reveals a myriad of situational factors that may lead customers to mask their true feelings. This paper offers nine theoretical propositions and summarizes research evidence related to these pro-positions from various substantive domains for marketing research. © 2010 Wiley Periodicals, Inc.
Imagine that, after months of Internet research and casual discussions with friends and family members, you finally decide to visit the dealership to purchase the car you want to buy. Imagine too that it is a little pricey, but it is your dream car and you're excited by the prospect of driving it home that very day. However, as you approach a salesperson, you realize that expressing your excitement would put you at a significant disadvantage in negotiating the best sales price. As a result, you make a conscious decision to restrain your expression of excitement by showing a sense of indifference about the car and its features, in the hope of driving down the price. Whether the salesperson takes your expression of indifference at face value or examines your behavior more closely to detect your underlying feelings may have a substantial impact on the outcome of the negotiation and the salesperson's ability to satisfy you as a customer. Retailers that can indirectly assess consumer mind-set by evaluating their nonverbal behavior stand to gain more insight into underlying consumer attitudes than what might be gleaned from explicit measures (e.g., a survey). This paper examines the complexities of customer nonverbal behaviors and their meanings to better understand how marketing professionals might use extant research on nonverbal behavior to inform retailers' and service providers' sales strategies, as well as to enhance customer satisfaction.
The challenge of understanding a customer's purchase intent seems well confirmed by the increase in number of in-house market research teams (Puccinelli et al., 2009). Firms now view this challenge as so formidable they are reluctant to outsource efforts to address it. This challenge becomes even greater when customers attempt to mask their true feelings. For example, customers may report an intention to buy a product but not actually make the purchase (Morwitz, Steckel, & Gupta, 2007). In other cases, customers may respond to demand characteristics that lead them to favor the brand sponsoring the research, when in fact they do not like the brand at all. Various research tools attempt to capture customers' true feelings and circumvent biased evaluations resulting from consumer attempts to conceal their attitudes (Puccinelli & Mast, 2002; Zaltman et al., 2001). For example, work on priming allows consumer researchers to study the effect of brands and logos on consumer perception in a nonconscious way (Puccinelli & Mast, 2002). Similarly, the Implicit Attitude Test (IAT) enables indirect measurement of consumer attitudes in a manner that is virtually impossible for consumers to consciously influence outcomes (Greenwald, McGhee, & Schwartz, 1998). By examining consumers' reaction times in responding to congruent versus incongruent stimuli, researchers can measure implicit consumer associations between different constructs. Consumers must rely on their unconscious associations to respond quickly, and even providing individuals with substantial incentives to respond differently shows no effect on their implicit responses.
Within this domain of indirect measures of consumer response, the use of consumer nonverbal behavior to assess attitudes has been largely ignored. The retail context provides a rich arena for the application of nonverbal assessment. Retailers have the relatively rare opportunity to interpersonally interact with consumers. Store managers can see how consumers navigate a store, how they respond to merchandise displays, and how they feel when they try on an item. Moreover, salespeople can interact one-on-one with customers to find out how the customer likes a new product line, any additional information the consumer may need before making a purchase, and what the deal breaker was for an item the customer left behind. The nonverbal behavior of consumers offers a wealth of information about consumer attitudes that can be accessed indirectly and is often guided by nonconscious processes. As a result, it represents a significant opportunity for marketing professionals to better understand consumer behavior in a retail context.
Industry recognition of the importance of attending to customer behavior in an interpersonal exchange is commonplace. Companies often emphasize the importance of service providers' intuition, as in the case of Ritz-Carlton managers who empower employees to identify ways that employees can exceed customers' expectations. The Four Seasons demands face-to-face interviews with property general managers for every prospective employee so as to enable the general manager to observe the interpersonal skills relevant to providing quality service. In turn, the company's best employees often describe “reading” customers' nonverbal behavior as an important part of their service strategy (Markos & Puccinelli, 2004). In this research, we found that the top service staff at a Four Seasons property described reading the nonverbal behavior of guests to determine their service needs. For example, a food server in the property's elegant dining restaurant described reading the nonverbal behavior of guests entering the restaurant in order to determine whether the occasion involved a business meeting or a casual meal among friends.
A broad range of companies, from MAC Cosmetics (L. Z. Williamson, personal communication, November 15, 2007) to Weight Watchers (Weight Watchers International, Inc., 2001), emphasizes the critical role of the interpersonal exchange between the service provider and customer. However, there exists a paucity of marketing research to guide these efforts.
NONVERBAL BEHAVIOR AS A KEY COMPONENT OF THE RETAIL ENVIRONMENT
The physical environment of the retail setting influences how customers feel (Baker et al., 2002). Studying consumer nonverbal behavior may be the most effective way to ascertain consumer reaction to the retail atmosphere. Research in the retail environment identifies three primary sets of cues: design, ambient, and social cues (Baker & Cameron, 1996). Research on design cues is the most abundant and suggests that the physical layout of retail space can influence the consumer's shopping experience and mood (Babin, Hardesty, & Suter, 2003). Ambient cues, such as music, also may influence customers' affective and subjective experiences, as well as their perceptions of merchandise value and store patronage intentions (Baker et al., 2002). Nonverbal behavior assessment can provide a way to assess the effects of these environmental cues on consumer attitude.
The third type of environmental cues, social cues, has been investigated less fully (Turley & Milliman, 2000). Most of the research in this area has focused on crowding and the consumer's attitude as a function of the number of people in the store (Barker, 1965). However, the role of retail salespeople would seem to be critical as well, given their critical role in determining customer experiences and satisfaction (Grewal & Sharma, 1991). Of the 60 studies on environmental cues reviewed by Baker et al. (2002), 14 consider customer perceptions of employees, and of those 14, only three are empirical. Recent retailing review articles have called for research to examine the role of retail employees in the customer–employee interactions (Grewal & Levy, 2007, 2009). Here we suggest that how customer nonverbal behavior is interpreted and responded to is an important predictor of consumer attitude in a retail environment.
Although customer nonverbal behavior can serve as a rich source of insight, it has gone largely unrecognized in marketing research. Researchers have investigated customer–service provider interactions (Yim, Tse, & Chan, 2008), the effects of physical characteristics of employees on customers' perceptions of employees (Gorn, Jiang, & Venkataramani, 2008; Naylor, 2007), tendencies for customers to mimic the behavior of others (Tanner et al., 2008), and the influence of nonverbal information on online customer interactions (Johnson, Bruner, & Kumar, 2006). However, there appears to be no prior research examining relationships between customer nonverbal expressions and underlying feelings and its use as a measurement device. The limited research in marketing that has considered nonverbal expressions centers on employees' expressions and their effects on customers' perceptions and behaviors (Hornik, 1992; Leigh & Summers, 2002; Wood, 2006). But the question remains: How and when do customers express their true feelings and when do they attempt to mask them?
Understanding the nonverbal behavior of customers is clearly critical to successful marketing (Bonoma & Felder, 1977). For example, nonverbal behavior appears to be equal to, if not more important than, verbal behavior (Stewart, Hecker, & Graham, 1987). Some researchers have speculated that 55% of interpersonal communication occurs through facial expressions (Mehrabian, 1972). Nonverbal behavior may serve several functions, including providing information, regulating interactions, expressing intimacy, and exerting social control (Patterson, 1982). More than 15 separate meta-analyses in psychology and related fields in the past decade have documented pertinent research (e.g., DePaulo et al., 2003; Elfenbein & Eisenkraft, 2010; Hall, Coats, & LeBeau, 2005; Riggio & Riggio, 2002).
Extending these ideas to marketing, Stewart, Hecker, and Graham (1987) emphasize the critical role of nonverbal behavior for advertising and personal selling. For example, in an advertising context, they show that nonverbal attitude measures may be more appropriate than verbal measures to gauge customer reactions to nonverbal stimuli such as the expression of a spokesperson in an advertisement.
This paper reviews nonverbal communication literature in an attempt to understand the relationship between customers' feelings and nonverbal emotional expressions. We have built a conceptual framework and identified several factors that may moderate this relationship (see Figure 1). To organize this literature review, Table 1 presents a series of nine propositions with supporting evidence. The next section of the review highlights connections between underlying or true feelings and nonverbal behaviors, followed by a discussion of the key moderators that may determine conditions in which customers express true feelings and those in which they are likely to mask them. Specifically, Figure 1 delineates four moderating factors: situation, customer expressivity, relative social status, and display rules. Situation comprises the effects of opportunistic situations and task involvement; customer expressivity involves age and gender effects; the social status category includes politeness and status imbalances; and display rules consist of normative masking and emotion dialects.
Table 1. Propositions Regarding Relationships Between True Feelings and Expression.1
|1||Customers express their true feelings in some contexts but not others.||Ekman, 1992a, 1992b; DePaulo, 1992; Ekman, Davidson, & Friesen, 1990.||(a, b)2 Nonverbal cues express emotion (happiness, sadness, anger, fear, disgust, surprise). (c) Individuals may mask their true feeling state due to self-presentation concerns. (d) Different facial muscles are used to create true and masked smiles.|
|2||Customer expression of true feeling reinforces true feeling, whereas customer masking leads true feeling to become more like masked feeling.||Davis, Senghas, & Ochsner, 2009; Hennenlotter et al., 2009; Mori & Mori, 2009; Tavassoli & Fitzsimons, 2006.||(a, b, c) Facial feedback hypothesis: People who express an emotion are likely to experience that emotion. (d) The way an individual expresses confidence also can have an impact in the marketing domain.|
|3a||Customers are more likely to mask their true feelings when faced with an opportunistic situation.||Fridlund, 1994; Fridlund, Russell, & Fernández-Dols, 1997; Russell, Bachorowski, & Fernández-Dols, 2003.||(a, b, c) Nonverbal behavior will often be used to deceive as it increases odds of positive outcomes (i.e., “opportunistic situations”).|
|3b||Customers are likely to mask their true feelings when they face more involving tasks or important purchase situations.||Ekman & Friesen, 1969; Petty, Cacioppo, & Schumann, 1983; Zaichowsky, 1985; Chandrashekaran & Grewal, 2003.||(a) Masked expressions carry the risk of the interpreter detecting the deception based on “leakage” of true feelings. (b, c) Involvement is the degree of per sonal relevance or importance of a product or purchase decision. (d) Consumers are more likely to use more detailed processing (possibly leading to masking) when involvement in a purchase situation is higher.|
|4||Customers who are more expressive are better at masking their true feelings.||Boone & Buck, 2003.||(a) More emotionally expressive people tend to be more accurate (i.e., more skilled) at nonverbal expression.|
|5||Older customers are better at masking negative emotions, whereas younger customers are better at masking positive emotions.||Coats & Blanchard-Fields, 2008; Blanchard-Fields & Coats, 2008.||(a) When interpersonal conflict arises, younger individuals are more likely to use “proactive emotion regulation,” whereas older adults are more likely to use “passive emotion regulation” strategies. (b) Older adults are more likely to regulate anger and thus experience less anger than younger adults.|
|6||Customers are better at masking true feelings when they use gender-stereotyped behavior.||Riggio & Friedman, 1986; Mast, Hall, Klakner, & Choi, 2008.||(a)2 Nonverbal cues associated with forming a good impression vary by gender. (b) Patient satisfaction was higher when clinicians used nonverbal cues associated with gender roles (i.e., females: more gazing, more forward lean, softer voice; males: loudervoice, more distance to patient).|
|7||Customers mask with politeness when the salesperson exhibits higher status, whereas customers do not mask with politeness when the salesperson exhibits lower status.||Brown & Levinson, 1987; Ambady, Koo, Lee, & Rosenthal, 1996; Hall, Coats, & LeBeau, 2005.||(a) Politeness theory examines the use of nonverbal behavior to smooth social interactions. (b) Politeness can be expressed nonverbally and is often done so without conscious knowledge. (c) Meta-analysis: High-status persons tend to have greater facial expressiveness, bodily openness, and use less interpersonal distance.|
|8||Customers express true feelings more if their culture has liberal normative masking display rules, whereas customers express true feelings less if their culture has restrictive normative masking display rules.||Diefendorff & Greguras, 2009; Fok et al., 2008; Saarni, 1979; Gross, 2002; Safdar et al., 2009.||(a, b, c) Each culture has specific rules dictating what is appropriate to express nonverbally and what is not. (d) Modulating one's expression to fit societal rules is referred to as masking (referred to here as normative masking). (e) Japanese display rules are more restrictive of intense displays (e.g., anger, disgust, contempt) than American display rules.|
|9||Customers' expressions will reflect their emotional dialect.||Elfenbein & Ambady, 2002; Marsh, Elfenbein, & Ambady, 2003; Chiao et al., 2008.||(a) Meta-analysis: Examines emotional dialects between cultures. (b) Findings indicate that facial expressions of emotion can contain sufficient differences in nonverbal expressions to identify an expresser's nationality or culture.|
CUSTOMERS' TRUE FEELINGS AND THEIR NONVERBAL EXPRESSIONS
Most scholars agree on a set of six basic emotional expressions that are universally recognized (happiness, sadness, anger, fear, disgust, and surprise) (Ekman, 1992a, 1992b). Ekman (1992b) uses the term “basic” in the sense that these emotions are biological in nature and have developed through evolutionary processes. Although basic emotions are universal and share core expressive components (Ekman, 1992a, 1992b; Elfenbein & Ambady, 2002), their expression can vary across cultures. The retail environment can foster emotional reactions in consumers. For example, customers may react negatively to a service failure (Menon & Dubé, 2004) or a salesperson's demeanor (Puccinelli, 2006) or be delighted when their expectations are exceeded (Oliver & Rust, 1997) or they have the pleasure of giving a gift (Vanhamme & de Bont, 2008). Customer feelings also may depend on feelings expressed by other customers (Howard & Gengler, 2001). Puccinelli (2006) found that consumers who reported they were in a bad mood responded negatively to a spokesperson in an ad expressing positive affect. If given a choice, consumers who are in a bad mood would choose a salesperson expressing either a neutral or a bad mood. If consumers in a bad mood were made to watch an ad featuring a positive, upbeat salesperson, the consumer reported feeling worse and indicated they were less willing to pay for the products endorsed by the salesperson (Puccinelli, 2006). Thus, if salespeople can read the mood of customers they may be able to use that information in their sales strategy to better serve them.
However, several factors likely affect the convergence between felt and expressed emotions, such as the situation, customer expressivity, relative social status of the consumer versus the salesperson, and display rules. Each moderating factor may play a role, as discussed in the following sections. A more detailed review of individual emotions (e.g., expression of happiness vs. sadness) and their differences lies beyond the scope of this paper.
The Nature of Nonverbal Behavior
As an illustration, consider one of the basic emotions, namely, happiness. In many cultures, smiling expresses happiness. Based on these universal behaviors associated with specific emotions, researchers developed various coding systems for the measurement of expression such as the Facial Action Coding System developed by Ekman and Rosenberg (1997). Following extensive training, researchers can recognize specific facial action units (FAU) that reflect the intensity of activation of facial muscles related to spontaneous expressions of particular emotions (e.g., contraction of the zygomatic major muscle of the face signifies a smile [FAU 12]; activation of the orbicularis oculi muscle to raise the cheeks and crinkle the skin around the eye is regarded as a genuine expression of happiness [FAU 6]; Ekman, Davidson, & Friesen, 1990). More complex messages, such as empathy and dominance, also seem to be associated with nonverbal behaviors (Knapp & Hall, 2006). Empathy, for example, may be expressed through forward leaning, close proximity, greater eye contact, more openness of the arms and body, more direct body orientation, and less eye movement (Hall, Harrigan, & Rosenthal, 1995).
Researchers have also suggested the value of making holistic judgments, in which coders observe videotaped behaviors and make overall ratings (e.g., how happy a person is), rather than using specific nonverbal behaviors in order to determine what someone may be feeling (Ambady & Rosenthal, 1993; Grahe & Bernieri, 2002). For example, a coder might use a rating scale anchored by “not at all” and “a great deal” to judge the level of happiness they believe a person to be experiencing.
For measures of nonverbal expressivity, it is important to assess the validity and appropriateness of the available measurement tools (Dobbs, Sloan, & Karpinski, 2007). Facial electromyography activity (EMG) is a common psychophysiological measure that provides an objective, behavioral method for measuring facial expressivity, which appears to be free of bias associated with self-reported measures. However, facial EMG measurement requires extensive training to administer correctly and cannot be employed easily outside a laboratory, which reduces its ecological utility.
Self-report measures provide an attractive alternative to behavioral measures, in part because they address issues of portability and do not incur a need for intrusive studies. However, self-reported expressivity falls prey to the common issues of bias and validity. Specifically, self-report bias raises the question of how accurate and valid a person's evaluation of his or her own behavior can be. Depending on the research objectives and context, different tools may be more or less appropriate. Additional research on expressivity would benefit from access to tools that are flexible, objective, ecologically valid, and practical.
Effective measures can capitalize on the ubiquity of customer expression from an early age. In research examining the development of expressions of interest, Camras et al. (2002) found that interested infants still their bodies, adopt a sober facial expression, and exhibit less fussing and distress. Studies of both children (Camras et al., 2002; Langsdorf et al., 1983) and adults (Reeve, 1993; Reeve & Nix, 1997) reveal consistent patterns of facial/body expression correlated with interest (e.g., reduced interest appears as eye blinks, closing of the eyes, and lowered heart rates; see Silvia, 2006). As people's interpersonal skills develop, they appear to cultivate behavioral repertoires that enable them to express a wide range of information about underlying emotional responses.
The Leakage Hierarchy
The behavioral repertoire people develop consists of highly controlled (e.g., facial expression), less controlled (e.g., tone of voice), and uncontrolled (e.g., pupil dilation) behaviors. This variation is partly due to biological automaticity and partly to relative levels of motivation, willingness, and cognitive capacity to exert control (Ekman & Friesen, 1969). Behaviors that tend to be less controlled may be more diagnostic of a person's genuine feelings than those that can be more easily controlled—a phenomenon referred to as the “leakage hierarchy” (Ekman & Friesen, 1969; Rosenthal & DePaulo, 1979a, 1979b) or the “unintentional display effect” (Bonoma & Felder, 1977). If a customer is trying to mask his or her excitement, it still might “leak” or be inadvertently expressed through the tone of voice, for example. The challenge that customers confront in their masking efforts is the irrepressibility of nonverbal behavior. That is, people cannot be completely expressionless (DePaulo, 1992). Masking efforts also may be more difficult because of processing impairment. Customers who experience true negative feelings often have difficulty processing incongruent information (Braun-LaTour, Puccinelli, & Mast, 2007). Investigators have used the IAT to examine such mood effects. For example, people in a bad mood were significantly slower to respond to incongruent stimuli (e.g., good words and a negative brand—Marlboro) as compared to those in a good mood (Braun-LaTour, Puccinelli, & Mast, 2007). More generally, the incongruence between a person's negative true feelings and positive feelings they are trying to express may make masking true negative feelings more difficult. Thus, if a customer seems ambivalent it may be that they are unsuccessfully masking a bad mood. That is, while they may be trying to appear happy, they may actually be feeling quite sad.
Nonverbal Behavior in Practice
Early marketing research that attempted to capitalize on the automaticity of nonverbal behavior used a measure of uncontrollable nonverbal behavior to assess attitude change in consumers (Halpern, 1967). Specifically, pupil responses indicated reactions to changes in product packaging without any influence of bias resulting from consumer recall of a pretest measure during the posttest measure. Measures of pupil dilation before and after a stimulus suffer no contamination from the initial attitude measure.
In other situations, customers may not be simply unintentionally biased but may actively seek to mask their true feelings, such as when they attempt to achieve self-presentation goals (DePaulo, 1992). They may want others to perceive them differently than the way they truly feel, especially if they are predisposed toward impression management. In this case, they may even choose retail contexts that support their impression management goals (Puccinelli, Deshpande, & Isen, 2007). In this research, people who were high self-monitors sought out mood-incongruent retail outlets ostensibly to regulate their mood. For example, if they were in a bad mood they sought out a store having a celebration to cheer them up. In contrast, low self-monitors chose mood-congruent retail outlets. Thus, salespeople should consider how goals may differ when the customer enters the store. For example, consumers inclined to impression management who experience true negative feelings may be looking to the salesperson to cheer them up.
First-time car buyers may feel nervous and excited but express confidence in order to ensure that salespeople view them as savvy consumers. DePaulo (1992) suggested that such impression management efforts are more difficult for masking certain emotions; because basic emotions are hard wired, they may be especially difficult to mask. Car buyers who see their preferred model in their preferred color displayed on the showroom floor may experience feelings of excitement or joy that may be nearly impossible to mask. From marketing research, we find that focus group participants' self-presentation concerns may influence the amount and nature of information they convey about themselves, threatening the validity of the focus group findings (Wooten & Reed, 2000). Understanding when and how customers mask their emotions also may be critical in personal selling contexts. A salesperson for a highly technical product (e.g., Apple's iPad) needs to be able to detect if a customer is trying to conceal technophobia or product-related anxiety. The salesperson might then recognize that he or she needs to spend more time explaining the easy-to-use features of the product.
At the extreme end of this spectrum, a considerable body of research exists that examines deception and the behaviors people tend to show when lying. Two recent meta-analyses revealed that liars nod less and move their hands, legs, and feet less than do people telling the truth (Sporer & Schwandt, 2007). They also appear less forthcoming (e.g., press lips together), tell less compelling stories (e.g., use fewer illustrators, gestures), and appear more tense (e.g., higher vocal pitch/frequency; DePaulo et al., 2003). A similar tendency might emerge among customers trying to express something other than what they truly feel. For example, the car buyer might press her lips together to suppress a smile when she sees the ideal car.
Interestingly, expressing a false emotional expression—such as putting on a “happy face” when feeling sad—may actually change a consumer's true feelings. Research on the “facial feedback hypothesis” indicates that people who form a facial expression of a particular emotion are more likely to experience that emotion (Davis, Senghas, & Ochsner, 2009; Hennenlotter et al, 2009; Mori & Mori, 2007). For example, when a happy person expresses happiness by smiling, she or he may become happier, and if a sad person masks the emotion by smiling, he or she may end up feeling less sad. In the marketing domain, expressions of feelings also may impact emotional experiences and reflect differences in the systems that control different modes of communication (e.g., cognitive, motor, perceptual; Tavassoli & Fitzsimons, 2006).
SITUATIONAL FACTORS: THE RELATIONSHIP IN CONTEXT
Expressions in Opportunistic Situations
In various situations, consumers likely act opportunistically and use nonverbal cues as a means to an end, such that they falsely portray their true feelings to a retailer or service provider. The behavioral ecology view of nonverbal behavior posits that nonverbal cues represent opportunistic social signals (Fridlund, 1994; Fridlund, Russell, & Fernández-Dols, 1997; Russell, Bachorowski, & Fernández-Dols, 2003). Nonverbal expression helps people navigate social situations to achieve the best possible outcomes (e.g., getting a great deal on a purchase). Taking this a step further, in opportunistic situations (i.e., in which deception is beneficial), consumers will likely mask their true feelings and use nonverbal cues to maximize their potential gains.
The social environment requires frequent and varied interpersonal exchanges every day. In many situations, a person's interests would be poorly served by expressing true feelings. For example, when placing a lunch order with an unkempt waiter, the consumer would not benefit from expressing disdain of this lack of concern about appearance. When negotiating the terms of a new cell phone contract, the buyer would not be well advised to express how desperate he or she is to close the deal. In such situations, intentionally miscommunicating feelings and intentions can result in a better outcome than revealing true feelings.
By contrast, it would be more beneficial to express true feelings in certain contexts. When patients need their doctors to take their symptoms seriously, they would benefit from nonverbally displaying some degree of fear and emotional concerns about the symptoms. In celebrating a friend's birthday at a restaurant, party guests can express pleasure in the meal to the waitstaff to add to the festive occasion. In these situations, expressing true feelings is likely to lead to better outcomes. From this perspective, the consumer's expression should vary according to situational demands (Fridlund, 1994; Fridlund, Russell, & Fernández-Dols, 1997; Mesquita, 2010).
Returning to the car shopping scenario, assume a customer has found the car of her dreams. Yet the customer faces a conflict: She loves the car and intends to buy it, but she cannot afford the sticker price. As previously discussed, this customer needs to make a conscious decision about whether to express excitement about finding the desired car or mask excitement by communicating indifference and reluctance.
Task Involvement or Importance
It is suggested above that in many instances consumers will mask their true feelings. The extent to which customers mask their true feelings may depend on their involvement in the task or its perceived importance. One customer negotiating on the price of a used car might mask her desperate need for the car, whereas another negotiating on the price of a book might express her sincere interest in the book. This difference in nonverbal behavior reflects the perceived importance of the situation and its outcome. Using nonverbal cues to persuade or deceive a salesperson demands a price, namely, the risk of being discovered. Despite one's best attempts to communicate a false feeling state, some true feelings almost invariably leak out for possible detection by the interpreter (Ekman & Friesen, 1969). The consequences of detection vary according to the situation but are important to consider. Detection might give a salesperson the upper hand in a negotiation or offend a food server who detects the customer's disgust with a soiled uniform, both of which likely lead to negative outcomes for the customer. If customers were to carefully weigh the risks associated with masking, they might employ it more selectively. Yet because there also can be benefits of masking, consumers may need to conduct implicit cost–benefit analyses based on situational factors.
The analysis may become more extensive when consumers are more directly involved in the decision. Involvement refers to the degree of personal relevance or importance of a product or purchase decision (Petty, Cacioppo, & Schumann, 1983; Zaichkowsky, 1985). Furthermore, the importance of the task or involvement in a purchase situation influences consumers' decision processes. Specifically, they are likely to engage in more detailed processing for more involving tasks (Chandrashekaran & Grewal, 2003).
High task involvement or importance likely encourages more misleading signals, as the consumer attempts to increase the odds of a positive outcome (e.g., communicating confidence despite experiencing anxiety during an interview). High task involvement also may provide the necessary motivation to mask true feelings, even through less controllable channels (Ekman & Friesen, 1969), in order to communicate the desired expression. In situations of low task involvement or importance, customers should be less motivated to exercise the effortful behavioral control necessary to mask true feelings (Ekman & Friesen, 1969), which in turn increases the risk of discovery. Therefore, it seems reasonable that people are less likely to mask their true feelings when they experience low task involvement.
EXPRESSIVITY: INTENSITY OF CUSTOMER DISPLAY
People vary in their degree of expressiveness. An existing self-reported measure of global expressivity assesses the individual's likelihood of nonverbal expression (Affective Communication Test—ACT; Friedman et al., 1980). It asks, for example, that respondents indicate their likelihood of tapping their foot to music, which correlates well with their general expressivity. Consumers who are more expressive tend to have better command of nonverbal language skills and greater ability to use these skills strategically. Boone and Buck (2003) showed that people who are more expressive tend to be more accurate in identifying behaviors that express their intended communication.
There may also be constraints of the situation that influence a person's expressivity. Research that was undertaken to assess the effect of the task a person was engaged in on an observer's interpretation of the person (Puccinelli, Tickle-Degnen, & Rosenthal, 2003) found that people are perceived to engage in more social behaviors when engaged in a discussion as compared to completing a puzzle. This effect may be due to a situational context that allows them to exercise their skill in nonverbal expression. However, it may also suggest that the situational context influences the observer's interpretation; that is, because the person is engaged in an interaction, the observer perceives more social behavior. For salespersons, it suggests that they need to be aware of how the retail environment may influence customer behavior as well as their own perception of the customer and how that may be independent of the customer's true feelings.
Effects of Age and Gender on Emotional Expressivity
As people grow older, their emotion regulation strategies change, which has a direct effect on expression and possibly an indirect effect on masking ability. Aging is a double-edged sword: It can be of considerable benefit to decision making (e.g., increased experience), but it also can be a great hindrance (e.g., decreased mobility). This dual influence appears as relevant for nonverbal expression as for facets of cognitive ability. Younger adults tend to be more expressive when interpersonal problems arise and are more willing to engage in confrontation (Coats & Blanchard-Fields, 2008). Therefore, when they feel anger, younger adults tend to vent their feelings, use gestures, and engage in conflict. Blanchard-Fields and Coats (2008) found that older adults are much more likely to regulate their moods to avoid experiencing hostility, so they experience hostility significantly less often than do young adults. The ability to regulate negative emotions in response to pain also leads older people to experience less pain (Paquet, Kergoat, & Dubé, 2005).
Age differences thus should moderate the likelihood that people mask their true feelings, though the effect may depend on the emotion being masked (e.g., masking hostility and expressing happiness by older adults). When distressed and engaged in an argument, confrontation, or negotiation, it may be beneficial for the consumer to communicate calmness (e.g., a “poker face”), and older adults are more likely to be successful in this attempt. On the other hand, younger adults may be better able to mask happiness to express hostility. Because they are more likely to engage in conflict and be more expressive when they do so (e.g., greater use of hand gestures), younger adults likely can better mask their positive emotions, although they have a harder time masking negative emotions and expressing positive ones (e.g., happiness). Perhaps this distinction accounts for the stereotype of the adolescent who appears impossible to please. These young people are masters at masking happiness with anger or indifference, which often prompts ever more fervent efforts to please them.
Applying this theory to the car shopping situation, in the case of a younger customer, false expression of hostility and indifference might be accepted as a reflection of a true feeling state (i.e., successfully masking excitement). If the customer is an older adult, though, masking of excitement by inciting conflict is likely to be detected because of the relative inability to express hostility.
Different gender roles or expectations apparently dictate the nonverbal behaviors deemed appropriate for each gender. Riggio and Friedman (1986) have found that the effective use of nonverbal signals to make a good impression varies between the sexes, such that men who make the best impression tend to use more physically expressive cues (e.g., gestures, body emphasis), whereas women who make the best impression tend to use more facial expressions. Such differences in successful impression management may reflect the interpreter's expectations. Society relies largely on gender stereotypes and predefined roles, at least in the realm of nonverbal communication. The findings are remarkably consistent across 25 years of research, and several excellent reviews summarize gender differences in nonverbal expression (e.g., Deaux, 1995; DePaulo, 1992).
In Briton and Hall's (1995a, 1995b) examination of beliefs about women's and men's nonverbal communication, they uncovered a common belief that women smile more than men, as measured by self-reports completed by 411 college students. It seems this belief is well founded, as female targets were in fact observed to smile more. Interestingly, the researchers found that priming different expectations of the frequency of smiles by gender had no effect on subsequent observations of smiles. Across a broader range of nonverbal communication, Briton and Hall (1995a) revealed that gender stereotypes extend beyond smiling. A survey of 441 men and women found that people believed that women were generally more expressive, more involved, and better at sending nonverbal cues. Again these expectations appear well founded, as they correlated positively with differences reported in observational studies (Briton & Hall, 1995a).
In the past five years, researchers continue to report similar results (e.g., Mast et al., 2008). The pervasiveness of these gender roles and expectations extends to the service setting. According to Mast et al., respondents expected female service providers to communicate with nonverbal behaviors that conform to a stereotype of how women are expected to act (e.g., more gazing, forward leaning, softer voice). Male service providers are also expected to follow a gender-stereotyped role (e.g., louder voices, more physical distance). Furthermore, these researchers found that both male and female physicians who treated patients using gender-specific stereotypical patterns of nonverbal behavior attained higher patient satisfaction ratings. As a result, it is not surprising that men's reactions to emotional advertising tend to follow gender stereotypes when they report their reactions in the presence of other people but are no different from those of women when they report their reactions privately (Fisher & Dubé, 2005). Ostensibly, men feel that they should express gender stereotype–consistent attitudes when in a public setting that appear to differ from their true attitude.
Despite this consistent reliance on gender stereotypes, there is evidence that as societal roles of women and men change, so too does the value of gender role–congruent nonverbal displays (Diekman & Goodfriend, 2006). Through a series of studies, these researchers found that as changes in social roles are identified, the value placed on gender role–consistent behavior changes as well. Thus, we would expect that as normative expectations of gender roles continue to relax, so too the differences in women's and men's nonverbal expressions will begin to shrink.
Taken collectively, this body of research illustrates the influence of gender roles. When a person uses nonverbal cues that are consistent with gender stereotypes, people tend to react more positively, which reinforces their display of such gender-typed displays. Over time, this reinforcement history should improve the person's ability to display gender-specific cues of different emotions consistently. Furthermore, when a person uses cues consistent with his or her gender stereotype to mask true feelings, it should be harder to detect, due to the increased skill with such cues and their conformance to the interpreter's expectation.
SOCIAL STATUS: POLITENESS AS AN ALTERNATE APPROACH TO PERSUASION
Although the preceding discussions depict the use of masked expressions in opportunistic situations as a fairly direct, intentional attempt to mask true feelings to achieve an optimal outcome, this is not always a person's intent. Instead, most people use an alternative, less deliberate method of influencing others in these situations, namely politeness, which varies depending on the social status of the interaction partner.
Nonverbal communication is not limited to expressing true feelings, but also can be motivated by the need to facilitate social interactions. Therefore, drawing on our car shopper example above, instead of expressing indifference, the shopper could attempt to ingratiate herself or himself with the salesperson to get a better price. Research on politeness theory examines the use of nonverbal behavior in the process of smoothing social interactions (Brown & Levinson, 1987) and reveals that people often engage in politeness without being aware of it as an influence tactic. Thus, instead of saying, “I'm here to buy a car,” the buyer asks politely, “Do you think you could help me find a car?” Moreover, the level of politeness likely depends on contextual cues. In interactions in a high-status context, say at a Jaguar dealership, the astute car buyer might be inclined to be especially polite and express politeness nonverbally (Ambady et al., 1996). Ambady et al. (1996) have shown that people use verbal and nonverbal strategies in combination. For example, when asking a Jaguar salesperson for help, the car buyer might offer a sheepish grin and shrug her or his shoulders to minimize the imposition associated with the request. Thus, even though nonverbal expression can communicate true feelings, it might be complemented by behavior designed to smooth social interactions independently of the person's true feelings.
Because interactions with higher-status individuals tend to provoke politeness norms and expressions of nonverbal behavior designed to smooth social interactions (Brown & Levinson, 1987; Ambady et al., 1996), politeness behaviors may also signal lower status. A distinctive pattern of nonverbal behaviors characterizes high-status persons. A recent meta-analysis of this research stream identified facial expressiveness, greater bodily openness, and smaller interpersonal distances as behaviors that are characteristic of higher status persons (Hall, Coats, & LeBeau, 2005). From a complementarity perspective, customers' politeness behavior should be associated with salespeople's high-status behaviors (or retail atmosphere cues), in which case politeness behavior by the customer might be obscuring his or her true feelings from the retailer or salesperson.
Display Rules: Effect of Cultural Norms on Customer Expression
Another consideration with regard to whether a person expresses an intention or true feeling state is cultural background. Each culture has specific rules, called display rules, which dictate what is appropriate to express (Diefendorff & Greguras, 2009; Fok et al., 2008; Saarni, 1979; Safdar et al, 2009). These rules consist of two subsets: normative masking and emotion dialects. The process and degree to which people modulate their expressions to conform to societal rules represent normative masking (Gross, 2002). The slight variations in actual behavior used to express feelings among cultural groups represent emotional dialects (Elfenbein & Ambady, 2002).
Puccinelli, Tickle-Degnen, and Rosenthal (2004) examined the effect of the task a person was performing on an observer's ability to accurately interpret the person's true feelings. They found that observers were able to more accurately interpret a person's true feelings when the person was engaged in a puzzle task as compared to a discussion. That is, observers were better able to discern what the person was truly feeling when the person completes a puzzle. They suggest that when a person engages in a discussion, he or she will conform to social norms that guide behavior in social interactions. However, puzzle interactions have less clear social norms guiding behavior. As a result, someone engaging in a puzzle interaction will be less likely to mask their behavior. For the salesperson, this means that how the customer interacts with a car may be more revealing than how the customer interacts with the salesperson. The interaction with the salesperson may be constrained by social norms and/or attempts to mask true feelings. This may be more difficult for the customer to do when simply looking at the car.
Display rules dictate the socially acceptable intensity and type of feelings to be expressed (Gross, 2002). These rules have the potential to create problematic misunderstandings in international interactions. For example, the politeness norms in Asian cultures lead people to express agreement even in situations where they may not agree. To an American counterpart who takes expression quite literally, he or she will understand that the parties have agreed even though they may not have. As a result of socialization, certain normative masking display rules are reinforced, which in turn shapes each person's view of the appropriateness of certain nonverbal behaviors (Brody & Fischer, 2000; Malatesta & Haviland, 1982; Mesquita, 2010). For example, Japanese display rules more strictly limit display of powerful emotions (anger, disgust, and contempt) than those in U.S. or Canadian cultures (Safdar et al., 2009). Although display rules in Eastern cultures tend to encourage masking true feelings, they also support the expression of politeness and social status deference through nonverbal behavior, which implies they may be more expressive in some contexts (Ambady et al., 1996). Some cultures may even seek to prevent the expression of certain emotions entirely; the remote Eskimo Utku tribe has such strong normative masking rules against expressions of anger that members will starve before they will express anger toward a fellow member of the tribe who has stolen their food (Briggs, 1970).
Although Ekman (1992a) argues for a universal recognition of basic emotions, it is important to discuss variations in the accuracy of interpretations of nonverbal cues across cultures. Different cultures have slightly different ways of expressing emotions. A recent meta-analysis of research on the influence of culture on the accuracy of interpretations of nonverbal behavior revealed that while people in different cultures can identify the same basic emotions (e.g., happy, sad, anger, fear) at above-chance levels, their accuracy significantly increases when making judgments about others from the same culture (Elfenbein & Ambady, 2002). This finding demonstrates an in-group effect, such that people who live in the same culture tend to be more accurate in identifying their compatriots' emotional expressions.
In further support of emotional dialects, Marsh, Elfenbein, and Ambady (2003) found that when respondents viewed photographs of emotional expressions of native Japanese and Japanese Americans, respondents accurately determined nationality at above-chance levels. This implies that differences in the way cultures express emotions (e.g., smile, frown) provides sufficient information to enable people to distinguish between emotional expressions of different cultures, despite facial similarities and matched intensity and type of expression.
Using an affective neuroscience approach, Chiao et al. (2008) used fMRI scans to show that native Japanese exhibit significantly more activation in the amygdala (i.e., vigilance center) when viewing pictures of people from their culture expressing fear (i.e., an in-group effect) as compared to viewing fearful faces of people from outside their culture (i.e., an out-group effect). This suggests that there is a real physiological difference in an observer's interpretation of an emotion in someone from their culture as compared to someone from another culture. The researchers replicated this effect with European Americans in the United States. This evidence reveals that even at a neurological level, people respond differently to emotional expressions of in-group versus out-group members. Thus, it may be these differences in physiological reactions to emotional expression by an in-group as compared to an out-group member that enable better recognition of emotion among in-group members.
Research on display rules has several clear implications for marketing. For example, through local hiring and selecting personnel from the same cultural background as customers, companies can enhance communication effectiveness and reduce risks of norm violations. The decision to relocate call centers to more cost-effective locations may thus have significant drawbacks in terms of customer satisfaction. In a high-stakes international negotiation, both parties need to understand their counterpart's display rules to recognize the meaning of the nonverbal behavior expressed by the foreign partner, as well as behavioral idiosyncrasies that characterize people in their own firm—that is, behaviors that people from the company's national culture exhibit that might be unique to people from that country. For example, among Indians, a distinctive head shake signals positive affirmation; among Americans, a similar gesture means “no.” The participants on both sides of a negotiation between an Indian and a U.S. firm would be well served to understand these differences in meaning.
Can a customer's expression be trusted? It depends. Nonverbal expressions can offer a wealth of information about customers that is likely to be more truthful than that which customers actually say. Yet several factors mitigate this effect. A customer can express his or her true feelings but also can mask them, and he or she may do so opportunistically. Certain behaviors (e.g., tone of voice) may be more diagnostic of true feelings; other expressions instead indicate what the consumer wants others to think (e.g., facial expression). This paper touches on key areas of research that inform this understanding, and the conceptualization offers a framework for examining the relationship between customers' true feelings and their emotional expression, while taking into account potential moderating effects of situation, expressivity, social status, and display rules. The theoretical propositions throughout Table 1 receive support from the illustrative empirical evidence in the text. However, they remain largely untested within marketing and thus represent diverse avenues for ongoing retail and consumer research.
To understand the implications of this framework for retail practice it is useful to consider the individual components of the model in more detail. First is the basic idea that customer feelings can be expressed through nonverbal behavior. These feelings can also be masked with the expression of a feeling the customer wants the other person to perceive. However, when a customer masks a feeling, they may unintentionally leak that feeling through specific nonverbal behavior. For the retail salesperson, this means that sometimes you can take a customer's nonverbal behavior at face value, but sometimes you cannot. Let's imagine you have a brand new product on the floor and you are trying to get customer feedback. A customer walks in and you ask her “What do you think of our new product?” She responds by smiling and saying “It's great.” However, you notice that her tone of voice is not enthusiastic and her body is oriented away from the product. For the salesperson, we might imagine one of three ways the salesperson could respond: (1) Take the customer at face value and respond by saying, “Yeah, isn't it great? Let me show you all the great features!”; (2) call the customer out on what you perceive as their true attitude and respond by saying, “You are just saying that. I can tell you don't like it. What can I show you instead?”; or (3) recognize that the customer doesn't like the new product but wants you to believe that she does and respond by saying, “What can I help you with today?” Based on our research, we would recommend option 3. This option both responds to the customer's true feeling and respects the customer's desire to be perceived as liking the new product.
The second basic idea is that the feeling customers express can affect the way they feel. If customers are feeling happy and they smile, the act of smiling may accentuate their feelings of happiness. The more interesting case is one in which the true feeling differs from the feeling they appear to express. In this instance a customer may not initially like a product, but because he or she has smiled while looking at the product, he or she feels better and may like the product more as a result. The challenge for the retailer is how to encourage the customer to smile when he may not feel like it. Our research suggests that the best strategy is a pleasant smile and display of moderately friendly behavior on the part of the salesperson, which is likely to invoke a social norm to smile back. Retailers, however, need to beware of encouraging salespeople to express extreme positive affect. Our research finds that customers in a bad mood are likely to respond negatively to an exuberant salesperson, which again underscores the value of observing the customer's behavior closely.
To better refine their strategy, retailers may also want to consider several factors that seem to moderate the customer's likelihood of expressing true feelings versus masking them. First, might the situational context provide an incentive to the customer to mask their feelings? For example, is there likely to be a negotiation around the price and is it an important purchase decision? A customer buying an expensive computer may be more likely to mask their feelings than if they are buying a T-shirt. In the computer sales context the customer may be keen to mask their satisfaction with the top-of-the-line model because she is anxious not to spend more than she needs to and believes the salesperson is likely to show her more alternatives if she conceals her satisfaction with the top-of-the-line model. For retailers operating in more expensive product domains or in ones where there is often a negotiation around price, the salesperson should be aware that customers are likely masking their true feelings. For example, customers may say they do not like a product and exhibit facial expressions of dissatisfaction with product features, but remain physically oriented toward the product and continue to examine its features.
Further refinement of retailer strategy can also be informed by the role of expressivity. In general, people who are more expressive are more likely to mask true feelings. In addition, specific customer characteristics may signal specific expressivity skill. Older customers are good at masking negative feelings, whereas younger consumers are better at masking positive feelings. Similarly, customers are likely to use gender-stereotyped behavior to mask their feelings. For retailers, such as videogame stores, who largely target young male consumers, customers may mask any satisfaction with a product and do so quite convincingly with loud, blunt behavior characteristic of men. Salespeople, do not be fooled!—These customers will likely make a purchase without added incentive.
The social status of the customer relative to the store may also increase customer masking efforts. If customers perceive that they are of a lower status than the retailer and/or its salespeople, they will likely mask feelings to be polite and show appropriate deference. High-end retailers, such as Tiffany, will need to be vigilant about this type of situation. A lower-status customer may mask their true feelings, making it difficult for the salesperson to know the customer's needs. For example, they may not want to impose on the salesperson to show them many different styles of a piece of jewelry, but as a result do not find what they are looking for.
Finally, the cultural background of the customer may influence both the masking efforts of the customer as well as the salesperson's ability to interpret the customer's behavior. Similar to the lower-status customer above, Asian customers may be similarly likely to mask their feelings so as to be perceived as polite. Further, the more culturally distant the customer is, the more likely the salesperson will be to get it wrong. Strategic retailers will seek to reduce this cultural distance. In foreign markets, retailers should hire from within local markets. In markets that attract an international clientele, a diverse staff combined with cultural awareness training should enable the sales staff to overcome any cultural barriers to recognizing customer needs.
A careful review of the nonverbal literature and the proposed organizing framework highlights the need for retailers and salespeople to exercise vigilance in their interactions with customers and attempt to detect opportunistic use of nonverbal signals. Paying careful attention to customers enables service providers to discern genuine signals from opportunistic ones, as well as help them build better rapport (Gremler & Gwinner, 2008) and better serve customers. Stronger nonverbal skills are likely to enhance customer perceptions of service quality and patronage intention. Strong interpersonal skills combined with an ability to adapt may even increase a salesperson's ability to complete a sale (Plouffe, Hulland, & Wachner, 2009).
These propositions also have important implications in the consumer advocacy domain. If retailers and salespeople understand customer expressions better, they may be able to help customers understand how their emotions affect their decision making. Some research suggests that consumers can benefit from identifying their feelings and recognizing the influence of these feelings on their decision making (Kidwell, Hardesty, & Childers, 2008). Consumers who are aware of how their true feelings affect their decision making tend to make choices that better support their long-term health. Retail employees who can identify consumers' affective expressions and use this knowledge to build rapport with them might have a distinct advantage over competitors and a better opportunity to help their customers (Gremler & Gwinner, 2008).
Customers are likely to use implicit cost–benefit analyses to determine whether to mask or express their true feelings. In turn, they may be more likely to mask their feelings for more expensive items that often require negotiation (e.g., buying a car, contracting for a house).
As summarized, prior research in nonverbal communication has dedicated considerable attention to detecting deception. In a retail context, identifying when a customer may be lying would seem critical, especially to support loss prevention efforts (retail losses due to theft and shrinkage account for more than $100 billion annually across 41 countries; Bamfield, 2009). The ability to detect when a customer is being untruthful could help retailers develop effective strategies for loss prevention through careful observation of the nonverbal behaviors exhibited by customers leaving a store with stolen items. However, retailers need to exercise caution in this domain, because many of the behaviors commonly thought to be associated with lying (e.g., gaze aversion) are not always reliable (Sporer & Schwandt, 2007).
In summary, the assessment of consumer nonverbal behavior represents an innovative way to indirectly measure customer reaction. Here, the retailing context is presented as an especially ripe domain for application. However, we can imagine myriad domains (e.g., business-to-business sales, advertising effectiveness, new product development, and charitable giving) where these lessons might be similarly relevant. This measure not only allows us to assess consumer attitude toward marketing but suggests ways to interpret and respond to this rich information source. Further, marketing provides a tremendous opportunity for further research and exploration that can inform nonverbal behavior theory with insights from a real-world context.