What do we get out of influencing others?

Abstract

Although a great deal of research has investigated the goals served by being influenced by others, virtually none has examined the goals served by having influence over others. The purpose of this paper is to draw attention to this understudied research area. Drawing on related research and theory, we suggest that having influence over others serves five needs and/or goals—for accuracy, belongingness, self‐esteem, control, and meaningful existence. Different types of social influence are compared and contrasted in the context of relevant dimensions of influence. We also make a number of suggestions for research to test this model.

Keywords:

• Conformity
• Compliance
• Obedience
• Persuasion
• Goals

A picture that appears frequently in social psychology textbooks was originally published in Asch's (1955) classic Scientific American article on conformity. It shows a puzzled participant in Asch's line judgment task struggling with the fact that his fellow participants are all agreed upon an answer that contradicts what his senses tell him. Over 50 years later, after hundreds of completed studies aimed at determining when and why people will conform, comply, be persuaded by, or obey a source of influence, we have a pretty good idea why the poor guy was confused. Research and theory have also provided considerable insight into why people such as Asch's befuddled participant often succumb to social influence.

Work examining social influence from the perspective of the source of influence, however, has been comparatively sparse. Consider the behavior of participants in another classic experiment in social psychology—Schachter's (1951) study on reactions to opinion deviates. This study revealed that when a dominant group opinion was challenged by a deviate, members of the group first attempted to persuade him to come over to the group's opinion, then ignored him when he failed to yield, and subsequently ended up rejecting him. Group members ultimately disliked those whom they failed to influence. What was going through the minds of group members when they realized the deviate could not be persuaded? Why did group members overwhelmingly prefer conforming as opposed to deviating confederates?

Social psychologists have conducted a multitude of studies (summarized by Cialdini & Goldstein, 2004) that attempt to determine when and why people are influenced by others. With few exceptions (e.g., Cialdini & Mirels, 1976), however, little work has examined why people seek to have an influence on others. Why do people care if others conform, obey, or change their opinions as a result of something they say or do? What needs or goals are served by having an impact on someone? And what are the emotional and interpersonal consequences of learning that one has failed to influence another?

The present paper draws attention to this neglected area of social influence and proposes a tentative, heuristic model to guide research on the functions of social influence. We propose that the goals served by being influenced—to be accurate, to belong, and to maintain a positive self‐concept (Cialdini & Goldstein, 2004)—are also often fulfilled by having influence. Specifically, we suggest that having an influence over others may lead to an increased sense of accuracy, an enhanced sense of belongingness, and higher feelings of self‐worth. We further propose that having influence over others serves two additional goals: a desire for control, and the perception that one's life is meaningful. We believe that these five goals, individually or in tandem, explain a great deal about why people strive to have an influence on others.

Throughout this paper we use the terms “need” and “goal” interchangeably. In the strict sense, a psychological need may be defined as something that an organism must satisfy in order to thrive, whereas a goal (or motive) is simply a desirable outcome (Kenrick, Neuberg, & Cialdini, 2007). To maintain consistency with the broader literature, however, we adopt the terminology commonly used in the field to describe each need and/or goal. For example, we refer to the need for control because control is typically described as a need rather than goal or motive, even though the empirical justification for considering this a need in the strict sense may be questioned by some.

We begin by elaborating each need and noting its role in other theoretical frameworks within social psychology. We then review the relevant empirical work linking the absence of influence with negative psychological outcomes and presence of influence with positive outcomes. Major forms of social influence (e.g., persuasion, compliance) are discussed with respect to their impact on each of the five needs, and several dimensions of social influence are proposed to have predictive value for the degree to which these needs might be thwarted or fulfilled. We end this paper with questions for future research.

GOALS SERVED BY INFLUENCING OTHERS

The definition of social influence (any instance in which one's thoughts, feelings, or behaviors are altered by the real, imagined, or implied presence of others) is practically synonymous with classic textbook definitions of social psychology (Nail, MacDonald, & Levy, 2000). We also define social influence broadly, borrowing Latané's (1981) definition of social impact: “By social impact, I mean any of the great variety of changes in physiological states and subjective feelings, motives and emotions, cognitions and beliefs, values and behavior that occur in an individual, human or animal, as a result of the real, implied or imagined presence or actions of other individuals” (p. 343). This definition incorporates the classic social influence topics of persuasion, conformity, compliance, and obedience as well as more recent work on behavior mimicry and emotional contagion.

Cialdini and Goldstein (2004) identified three core underlying human goals that determine how a target responds (or is vulnerable) to influence attempts: goals of accuracy, affiliation, and maintenance of a positive self‐concept. They argued that people have a desire to correctly interpret and react to incoming information, to create and maintain meaningful social relationships with others, and to behave in ways that allow them to sustain positive self‐ascribed traits. These three goals, which are conceptually distinct but undoubtedly related, are served by a broad spectrum of social influence, including conformity, compliance, obedience, persuasion, and behavioral mimicry.

Although much research has focused on the target of social influence, very little empirical work has examined influence from the perspective of the source. Perhaps the most obvious goals met by influencing others are instrumental or tangible ones, such as when a child obeys a parent, a salesperson successfully sells a car, or a defense attorney wins a case. In these examples the benefits to the source of influence are quite obvious. The work on dominance hierarchies in groups of animals (Chase, Tovey, Spangler‐Martin, & Manfredonia, 2002) and humans (Cummins, 1996) has taught us a great deal about the strategies dominant group members use to influence others in order to attain resources (summarized in Forsyth, 2005).

However, in this paper we are more concerned with the less obvious psychological benefits that might accrue as a result of having influence. We suggest that the three goals that explain why people succumb to influence—goals of accuracy, affiliation, and maintenance of a positive self‐concept—might also explain why people seek to have influence on others. To this list of needs we add needs for control and meaningful existence. Below we elaborate on each.

Accuracy

Accuracy pertains to the desire to be correct in one's beliefs and behaviors. Cialdini and Goldstein (2004) argue that people rely on descriptive norms to determine how to behave in a given situation. They look to others to define their “social realities,” particularly under conditions of ambiguity (Festinger, 1954; Sherif, 1936). Such an accuracy motive explains why people respond to informational influence (Deutsch & Gerrard, 1955). The goal to hold accurate attitudes has also been used to explain why we choose similar others and reject dissimilar others as friends (Pittman, 1998). A sense of accuracy may also serve the goal of reducing cognitive dissonance (Festinger, Riecken, & Schachter, 1956). We suggest that if people can influence others to think and act like them, then they may also feel more confident that their attitudes and behaviors are correct and appropriate.

Accuracy is similar to Baumeister's (1991) need for justification. Justification is the notion that people want to view their actions as good, just, and morally defensible, even when their behaviors are motivated by self‐interest. Influencing others to agree with or perform questionable behaviors may help to convince sources that their actions are justified.

Although we could find no empirical evidence that influencing others increases one's sense of accuracy, observational and anecdotal evidence underscores the importance of influence in one's life. Perhaps the most famous example of this is an event that occurred in the mid‐1950s. Members of a UFO cult predicted the end of the world on December 21st, 1955, by way of a catastrophic flood. The cult was headed by Chicago housewife Marian Keech, who purported to have received warnings of the flood from alien beings from the planet Clarion. True believers, she said, were to be rescued by flying saucers at midnight prior to the flood. To demonstrate their convictions, cult members parted with worldly possessions, quit their jobs, and broke ties with family and friends. Festinger et al. (1956) described the cult's behaviors in the days prior to and following December 21st. When preparing to be rescued the group made little effort to recruit new members, and it generally shunned publicity. After the predicted cataclysm failed to materialize the group campaigned actively to spread its message by seeking publicity from news reporters and recruiting new members. Festinger and colleagues concluded that the inconsistency between cult members' actions (e.g., quitting their jobs, breaking ties) and their beliefs (threatened by the failed prophecy) created a state of dissonance that could only be reduced by strengthening commitment to the original beliefs. In explaining the proselytizing that occurred after the failed prophecy, the authors noted: “The dissonance cannot be eliminated completely by denying or rationalizing the disconfirmation. There is, however, a way in which the remaining dissonance can be reduced. If more and more people can be persuaded that the system of belief is correct, then clearly it must after all, be correct.” (Festinger et al., 1956, p. 28). We suspect that having influence over others in our everyday lives (such as convincing a friend that Bob Dylan has a great voice) serves a similar accuracy goal.

Belongingness (affiliation)

The need for belongingness appears in various forms within a number of theoretical frameworks (e.g., Maslow, 1943; Murray, 1938). For example, Fiske (2003) maintains that a feeling of connectedness with others is a “core motive” underlying much of human behavior. Baumeister and Leary (1995) argue that people have a fundamental “need to belong,” defined as the desire for frequent interactions within ongoing relational bonds. Cialdini and Goldstein (2004) suggested that the desire to maintain relationships and to receive approval from others drives people to change their behaviors in response to pressures from others, which may form the basis of normative influence (Deutsch & Gerrard, 1955).

We suggest that influencing others may also sometimes serve belongingness needs by way of enhancing social connections and making people feel desired by others. Influence might increase one's perceived importance to relationship partners (such as when partners take our advice) and foster a sense of cohesiveness within groups. According to participative decision‐making theory (Yukl, 1989), the more influence group members have in the decision‐making process, the more positively they evaluate their position within the group. Successfully influencing group members might allow members to feel indispensable to their groups. Furthermore, influencing others may increase attitude similarity, which leads to increased liking (Byrne, 1971). Of course, the goal of belongingness would be less relevant in examples of social influence involving strangers, such as a salesperson–customer interaction (although Cialdini, 2000, has suggested that salespeople often capitalize on the belongingness need by treating customers as friends).

Self‐worth (positive self‐concept)

Nearly every model of needs or goals (e.g., Baumeister, 1991; Fiske, 2003; Maslow, 1943) includes the need for self‐worth (or a positive self‐concept). Cialdini and Goldstein (2004) argued that being influenced by others often serves this goal. We suggest that having influence over others enhances self‐worth because it provides people with a sense of accomplishment (in reaching one's influence goals) as well as a sense of acceptance or validation by others.

Self‐esteem derives both from perceived competence (Blaine & Crocker, 1993; Dutton & Brown, 1997; Tafarodi & Swann, 1995) and perceptions of acceptance by others (Leary & Baumeister, 2000; Leary, Haupt, Strausser, & Chokel, 1998; Leary, Tambor, Terdal, & Downs, 1995). Having influence might increase self‐esteem by way of the inclusion or acceptance that it conveys (see “belongingness” above), which is consistent with the sociometer model of self‐esteem (Leary et al., 1995, 1998). Harter, Waters, and Whitesell (1998) have linked self‐worth to a sense of “social validation” that comes about when others express interest in what one says, thinks, and feels. Influence allows one to feel confident that he or she has the respect and approval of others (Yukl, 2006). Perceiving that one is influential may also suggest that one is a capable, efficacious person (see “control” below).

Control

Control (or efficacy) is another need that appears in most contemporary models of needs (e.g., Baumeister, 1991; Fiske, 2003). Perceived control arises from a contingency between one's behaviors and desired outcomes. Fiske (2003) argued that a sense of control over one's environment is crucial for well‐being; we suggest that control is enhanced by having influence over another's thoughts, feelings, and behaviors. White (1959) posited that people have a basic “effectance motivation;” that is, people want to control important aspects of their environment. Successfully influencing others may provide such a sense of control over one's social life.

Control can be contrasted with the related concept of power, which appears in Murray's (1938) list of psychogenic needs. Power is defined as control over others' outcomes (Fiske, 2003). Although influence and power may sometimes overlap, prior work has established that they are conceptually and empirically distinct (Bennett, 1988). Power often involves altering the behavior of a target by using dominance, coercion, or force (Bennett, 1988). Other types of influence, such as persuasion or conformity, do not necessarily involve use of coercion or force. When people rely on power to induce change in others, they might achieve a sense of control but simultaneously undermine relationships with targets (Hollander & Offermann, 1990). Other forms of influence (such as persuasion) could enhance control while reinforcing relationships between sources and targets. Perhaps the main distinction between power and control is that the exertion of power is often seen as a means to the end of achieving tangible instrumental rewards (e.g., resources), whereas a feeling of control is typically seen as a desirable end in itself. Because our model focuses on the psychological needs satisfied by having influence, we include the need for control but not the related construct of power.

Meaningful existence

People need to feel that their lives have purpose or meaning (Frankl, 1959). According to Baumeister (1991), purpose is achieved through striving for and obtaining chosen goals and fulfillment‐states. Goals are desired possible future circumstances, whereas fulfillment states are desired moods. We suggest that influence satisfies both types of purpose. Having influence links one's behaviors to specific, short‐term outcomes (e.g., gaining compliance, getting children to obey) and thus serves an important instrumental function. Influence also provides an idealized state of fulfillment by way of convincing people that their very existence is meaningful.

According to terror management theory (TMT) (Greenberg, Pyszczynski, & Solomon, 1990; Pyszczynski, Greenberg, Solomon, Arndt, & Schimel, 2004; Pyszczynski, Solomon, & Greenberg, 2003; Rosenblatt, Greenberg, Solomon, Pyszczynski, 1989), people need to believe that their actions, creations, or ideologies are of lasting value. As a result of the frightening knowledge of their inability to achieve literal immortality, people develop a goal of preserving a symbolic sense of continuity so that they may continue to interpret their lives as meaningful despite their inevitable death (Drolet, 1990). Symbolic immortality, we suggest, may be achieved by having lasting influence on others. According to Dawkins (1986, p. 214):

When we die we can leave behind genes and/or memes. The difference between them being that our genes will be forgotten in just a few generations. As each generation passes, the contribution of one's unique genes is halved…. But if you contribute to the world's culture, if you have a good idea, compose a tune, invent a spark plug, write a poem, it may live on, intact, long after your genes have dissolved in the common pool.

To our knowledge, no research has directly tested the notion that social influence provides a path to symbolic immortality or existential meaning, although the idea was first raised by Aristotle and has also been discussed more recently (e.g., Bekhterev, in Tolman, 2001). The quest for meaning through influence, however, is similar to Erikson's (1963) concept of generativity. Generativity is typically achieved by contributing something of lasting value to later generations of people. Producing, nurturing, and guiding the next generation, raising offspring, transmitting values, mentoring younger workers, and contributing to the world through art or literature may lead to a sense of generativity (Lachman, 2004; Peterson & Stewart, 1996). Interestingly, these are also all mechanisms for ensuring lasting influence over others.

Although we are more interested in the psychological goals served by influencing others than the instrumental goals (e.g., gaining resources) explained by the research on dominance hierarchies, we note that it may often be the case that attaining instrumental goals satisfies psychological motives as well. Consider for example a successful car salesman. Reaching and even exceeding his monthly sales quota has the obvious benefits of attaining resources for his family. But the fact that people are buying the product he is selling may also keep him secure in the belief that he is serving a valuable role within his culture, thereby maintaining his sense of self‐worth and meaningful existence.

We note that the five goals we posit as fulfilled by having influence, while conceptually distinct, might often overlap to a large degree. Violating rules of good behavior can lead to being ignored, excluded, or ostracized, so the sense of accuracy that comes from influencing others may also serve the need to belong. Feeling that one's life has meaning could contribute to both high self‐esteem and a sense of control. Having influence over others may also affect some needs and not others. For example, getting others to obey probably increases the feeling of power or control but may not affect a sense of meaning or belongingness. In a later section of this paper we speculate about the goals most likely to be satisfied by various forms of influence.

EMPIRICAL SUPPORT

Many social phenomena, including obedience, persuasion, compliance, conformity, and social ostracism, reflect attempts by one person or group to have influence over another person or group. We suggest that influencing others in these and other circumstances results in an increased sense of meaningful existence, control, self‐esteem, accuracy, and/or belongingness, whereas failures to influence may lead to a sense that these needs are unmet. Because little work has examined need satisfaction directly, we focused our review on studies that link failed influence with frustration, malaise, or discontent, and successful attempts at influence with positive affect and well‐being. We also present work linking unsuccessful influence with interpersonal distancing and successful influence with interpersonal attraction, with the assumption that people will detach themselves from those who thwart the satisfaction of basic human needs (La Guardia, Ryan, Couchman, & Deci, 2000; Pinel, Long, & Landau, 2006).

Reactions to group deviates

The first line of evidence that loss of influence is aversive comes from research on responses to opinion deviates within groups. Schachter (1951) planted into all‐male discussion groups a group deviate, who consistently disagreed with the group's modal opinion; a slider, who initially disagreed with the other group members and then moved toward agreement; and a mode, who consistently agreed with the group. Groups tended to first direct communications toward the deviate more than toward the other two confederates, attempting to change their opinion, and then decreased communication over time when it became clear that the deviate would not be influenced. Results also showed that groups liked the mode and the slider much more than the deviate, and also rejected the deviate much more frequently than the other two confederates. Rejection of the group deviate was particularly likely when the group was more cohesive.

Later research using similar procedures (Levine, Saxe, & Harris, 1976) found that groups were most attracted to members who consistently agreed with them or those who moved from a neutral position toward agreement. Those who consistently disagreed or moved from a neutral position to disagreement were least well liked. Members who moved from one extreme to the other—agreement to disagreement or disagreement to agreement—were liked moderately and about the same. Although these studies do not provide direct evidence that having influence increases liking toward the target, they consistently show that failing to influence a target leads to increased dislike and rejection.

Ostracism

Studies in developmental and social psychology have revealed that lack of responsiveness or attention from others predicts reductions in need satisfaction and increases in frustration and distress. Developmentally, the earliest evidence for this proposition comes from research using the “still‐face” paradigm. In these studies an adult figure becomes verbally and nonverbally unresponsive to an infant during a brief period of time. The infant's response is compared to a baseline or normal play condition, and a reunion episode in which the adult resumes interaction with the child. Collectively, these studies show that infants between 1 and 12 months of age respond to still‐faces with decreased positive affect, increased negative affect, attempts to elicit the adult's attention by smiling, cooing, and crying, and self‐soothing behaviors such as thumb‐sucking and gaze‐aversion (see Adamson & Frick, 2003, for a review). Cooing and crying are interpreted as attempts by the infant to re‐establish interactions with the parent or other adult figure. Gaze aversion and withdrawal reflect attempts to cope with repeated nonresponsiveness. These findings follow directly from attachment theory (Bowlby, 1969, 1973), which predicts proximity seeking, and then later withdrawal, among individuals who experience loss of felt security with caregivers. This trajectory of behavior—first attempting to interact with the parent and then withdrawing—also shows an interesting parallel to the behavior of group members dealing with an opinion deviate in the Schachter (1951) study.

Distress reactions to being ignored do not end in infancy. Older children and adults display a host of negative reactions to being ostracized. These include an increase in anxiety, frustration, and shyness (Geller, Goodstein, Silver, & Sternberg, 1974), a loss of self‐esteem (e.g. Craighead, Kimball, & Rehak, 1979; Williams et al., 2002; Zadro, Williams, & Richardson, 2004), lower perceived control (Williams et al., 2002; Zadro et al., 2004), decreased feelings of belongingness (e.g., Williams, Cheung, & Choi, 2000; Zadro et al., 2004), and reduction in the sense that one's life is meaningful (e.g., Smith & Williams, 2004; Williams et al., 2002). Loss of felt belongingness and affiliation seeking are particularly likely when targets perceive that others are oblivious to their presence compared to when they perceive the ostracism as intentional (Sommer, Williams, Ciarocco, & Baumeister, 2001). Oblivious ostracism thus appears to be particularly threatening to felt belongingness, probably because it most closely resembles what it feels like to not exist (see also Case & Williams, 2004).

Williams (2001) argues that social ostracism threatens fundamental needs for belongingness, self‐esteem, control, and meaningful existence. We would suggest that social ostracism is conceptually quite similar to lacked or lost influence in relationships, thereby compromising the four needs identified by Williams plus the goal for accuracy. Targets of ostracism (compared to inclusion) also report greater dislike of sources (Hitlan, Kelly, Schepman, Schneider, & Zárate, 2006; Dotan‐Eliaz, Rubin, & Sommer, 2008; Williams et al., 2002) and a tendency to terminate relationships with sources if they are high in self‐esteem (Sommer et al., 2001). These findings fit with our suggestion that people will sever attachments with those whom they are incapable of influencing.

The effects of parental rejection on children's well‐being are also relevant here (Peretti, Clark, & Johnson, 1984; Peretti & McNair, 1987; Peretti, Early, & Chmura, 1998). Children who feel unable to affect their home environments are more likely to exhibit negative attention‐seeking behaviors (such as throwing things, temper tantrums, and pouting) at school (Peretti et al., 1984). In the study by Peretti et al., third‐grade students completed a survey assessing parental rejection (defined as indifference and a general atmosphere of unconcern for the child's welfare, or by active dominance and excessive demands by the parents). With the aid of an interview guide, teachers made a list of 60 students who were most likely to exhibit disruptive negative attention‐seeking behaviors. These 60 students were then compared to the non‐selected students. A strong relationship was found between parental rejection in the home and negative attention‐seeking behavior in the classroom. These results are consistent with the notion that a lack of influence in one environment may lead individuals to strive for it in another. Negative attention‐seeking behavior may reflect an attempt to regain a sense of control and attention deprived by parental rejection.

Influential leaders

The third stream of work attesting to the benefits of influence comes from the leadership literature. Hollander (1958, 1960) was the first to emphasize the importance of examining the influence exerted by followers. His idiosyncrasy credit model posits that leaders should allow themselves to be influenced by followers by initially complying and conforming to existing group norms. The idea is that followers whose leaders succumb to their influence will perceive their leaders as more competent and loyal, in turn leading them to be more welcoming of their leaders' influence, more responsive to their leaders, and more likely to afford their leaders latitude in making important decisions. Having influence on a leader might also allow followers to feel respected and valued (increasing self‐worth and belongingness), leading them to reciprocate this respect through increased tolerance of idiosyncratic leader behaviors (Hollander, 1960, 1992).

A connection between subordinates' perceived influence on supervisors and positive work outcomes can also be found in the leader–member exchange (LMX) theory of leadership originally proposed by Graen and his colleagues (Dansereau, Graen, & Haga, 1975; Graen, 1976; Graen & Cashman, 1975). LMX theory recognizes that a supervisor may form different relationships with different employees (Liden & Graen, 1980). A relationship high in LMX is defined as a high‐exchange relationship in which a select group of subordinates is given greater status, influence, and benefits. In turn, these subordinates are expected to work harder, be more committed to task objectives, and exhibit greater loyalty to the leader (Martin, Thomas, Charles, Epitropaki & McNamara, 2005; Scandura, Graen, & Novak, 1986; Yukl, 2006). A meta‐analysis conducted by Gerstner and Day (1997) found LMX to be significantly and positively related to job performance, satisfaction with supervision, overall satisfaction, organizational commitment, role clarity, and member competence, and negatively related to turnover intentions and role conflict. Additionally, research has revealed negative associations between LMX and job‐related depression and anxiety (Epitropaki & Martin, 1999) and perceptions of work stress (Nelson, Basu, & Purdie, 1998). Although studies in the leadership realm have not assessed need satisfaction per se, they are consistent with the view that perceptions of influence contribute to psychological well‐being.

A study by Morris, Hulbert, and Abrams (2000) was one of the few to manipulate perceived influence and examine its effects on relationships between leaders and group members. Participants were placed in groups consisting of three other participants and one confederate. They were instructed to make a recommendation to their group leader (confederate) regarding the financial settlement in a hypothetical lawsuit scenario. Influence was manipulated by providing each participant with false feedback indicating how much the group leader's final decision was influenced by the participant's recommendations as well as the recommendations made by the group as a whole. Members who perceived having personal influence over leader's decisions reported higher satisfaction (with the procedure, the decision, and one's role in the group) and commitment to the group. Perceptions of high (versus low) group influence predicted only commitment. These findings suggest that having influence on one's leader is desirable to group members and beneficial to relationships between followers and leaders, although the specific processes underlying these effects are not well understood. In our view these effects could be mediated by any one of the needs proposed to be impacted by social influence. That is, having an impact one's leader could enhance feelings of belongingness in the workgroup, increase feelings of self‐worth and control over important outcomes, confirm the accuracy of one's views or contribute to the perception of one's life as meaningful.

Interpersonal attraction

Work outside of the leadership literature has linked perceived influence with greater liking of targets (Gerstner & Day, 1997; Hollander, 1958; Yukl, 2006). In a recent experiment by Bruno, Sommer, Bourgeois, and Lo, 2008, participants playing the role of a manager in an organization were asked to make a recommendation to a colleague (played by a confederate) as to how to discipline an employee. In the no‐influence condition the confederate told the participant that the suggestion was good but that she had decided to take a different approach; in the influence condition the confederate said that she had planned to take a different approach but decided that the participant's recommendation was better. Participants were later asked to indicate how much they liked the confederate and to indicate their willingness to help her with an unrelated task following the experiment (hanging flyers around the campus). Participants in the influence condition reported greater attraction to the confederate and were more likely to comply with her request for help compared to the no‐influence condition. A control (baseline) condition in which participants received no feedback regarding the consequences of their recommendation resulted in attraction levels and helping rates falling directly between those in the influence and no‐influence conditions. These results suggest both that failure to influence is detrimental to relationships and that successful influence is beneficial to relationships.

Other studies suggest that even implicit processes may be explained in terms of achieved or failed influence. Chartrand and Bargh (1999) randomly assigned participants to interact with a confederate who either mimicked or failed to mimic the participant's behaviors (e.g., posture, touching one's face, crossing one's leg). Following the interaction, mimicked participants reported stronger feelings of affiliation with the confederate than participants whose behavior was not mimicked. This occurred despite the fact that participants were not consciously aware that their behaviors were mimicked. Similar research has linked naturally occurring synchrony and “affect matching” to increased liking and affiliation within dyads, even when the synchrony is not consciously perceived (Bernieri & Rosenthal, 1991).

Research suggesting a possible exception to the influence–attraction connection is that by Kipnis and colleagues (Kipnis, 1976; Kipnis, Castell, Gergen, & Mauch, 1976; O'Neal, Kipnis, & Craig, 1994; Rind & Kipnis, 1999). They argue that use of manipulative or controlling influence tactics leads sources to denigrate targets because they view targets as incapable of controlling their own behavior. Rind and Kipnis (1999) and O'Neal et al. (1994), for example, found that influence agents tended to devalue both themselves and their targets when forced to use threat, force, or deception (compared to rational arguments) to get others to perform a certain behavior. Most of the types of influence people exert on others in everyday life, however, are not overtly manipulative. Further, in the Kipnis et al. studies participants across conditions were led to believe that they had successfully influenced their peers. Thus it remains unclear whether the participants' perceptions of themselves and their targets would have compared unfavorably to a “no‐influence” condition. Sources who use unsavory tactics to influence others may still view themselves and their targets more positively than those whose attempts at influence are completely unsuccessful.

Collectively, the evidence generally points to a positive relationship between non‐manipulative forms of influence and interpersonal attraction. More research is needed to determine the extent to which increases in attraction might be mediated through need satisfaction. As within the work on leader–follower relationships, we would hypothesize that the link between perceived influence and general liking for others could be explained by fulfillment of needs for belongingness, self‐esteem, control, accuracy and/or meaningful existence.

Other evidence

Some ways of influencing others are not typically conceptualized as social influence in its traditional forms (e.g., conformity, compliance, obedience). For example, teaching is an important form of social influence, although we have never seen it described as such. The main purpose of teaching is to imbue others with knowledge. We suspect that when teachers, instructors, or professors alter the beliefs and abilities of their students, they may feel that several of the needs we have discussed (especially the need for meaningful existence) are met. Anyone who has graded an essay exam in which a student has really nailed an answer has undoubtedly experienced such perceived influence in a positive manner. Influencing the thoughts of one's students has often been given as an example of achieving Erikson's notion of generativity as described above. A recent, memorable subway ad intended to persuade riders to enroll in the New York City teaching fellowship program read: “You remember your first grade teacher's name. Who will remember yours?” According to the organization's website, advertising efforts to recruit high quality applicants to the program have been immensely successful (http://www.nyctf.org/about/history.html.

Another way to ensure lasting influence is to become a parent. By having a child, an individual can leave behind something that would not have otherwise existed; the child will go on to affect the world in the absence of the parent. If only through genetics, that child's life is shaped by his or her parents. Some parents may additionally hope that they will always be remembered by their children, that their children will go on to carry out their family name, their business, or perhaps even their ideology. In support of the notion that parenting provides a path to meaningful existence, research has linked the fear of death to an increased desire for children. In a series of three studies, Wiseman and Goldenberg (2005) found that men desired more children after being reminded of their death as compared to other aversive experiences (such as going to the dentist). The authors suggested that mortality salience might have intensified participants' goals of achieving a sense of purpose and symbolic immortality through reproduction.

TYPES OF SOCIAL INFLUENCE

Although there are no hard and fast lines between types of social influence (and there are underlying similarities among the types, see Latané, 1981) the field of social psychology has typically distinguished among the following: persuasion, behavior mimicry, conformity, compliance, and obedience. In this section we will address each type of social influence in terms of the relevant needs that we feel are most likely to be met by each.

Persuasion may be defined as change in a private attitude or belief as a result of receiving a message (Kenrick et al., 2007). Because one of the main goals of having attitudes is to express one's values (Katz, 1960), we believe that successfully persuading others will have the largest impact on a sense of accuracy and meaningful existence. As shown in Festinger et al.'s (1956) description of the UFO cult, convincing others that one's attitudes and beliefs are correct may serve to bolster those attitudes and beliefs.

Conformity can be defined as behavior change to match the actions or responses of others (Cialdini & Goldstein, 2004). Although the prototypical conformity scenario involves succumbing to majority influence (as in the classic Asch studies), the definition also applies to dyadic interactions between a single target and source of influence (for recent examples, see Castelli, Vanzetto, Sherman, & Arcuri, 2001; Quinn & Schlenker, 2002). Behavior mimicry may be defined as conformity to nonverbal behaviors and typically occurs below the level of conscious awareness (Chartrand & Bargh, 1999). Conformity has typically been explained in terms of both normative and informational influence (Deutsch & Gerrard, 1955). We expect that getting others to conform to one's actions will enhance belongingness, self‐esteem, and a sense of accuracy.

Compliance is behavior change that occurs as a result of a direct request (Kenrick et al., 2007). Probably the most salient benefits from getting another to comply are instrumental; for example, getting someone to buy a product or donate to a charity organization. However, it seems likely that at least some of the psychological needs discussed above, such as control and belongingness, would also be served by getting others to comply.

Obedience is behavior change that occurs in response to a directive from an authority figure (Kenrick et al., 2007). What usually differentiates obedience from compliance is that the former involves a command, whereas the latter involves a request. Another important distinction is that obedience typically involves a hierarchical relationship or status differential between the source and target, whereas compliance may or may not involve differences in status between the source and target. Obedience, we believe, will have the largest impact on control and the related yet conceptually distinct concept of power.

DIMENSIONS OF SOCIAL INFLUENCE

We thought it might be useful to define and contrast the different types of social influence by considering the dimensions on which influence may vary. (Note that the distinctions we will make among the different dimensions are not necessarily mutually exclusive, but we believe it is useful to draw out the differences.) Social influence may be more or less consciously attempted and perceived; intentional or unintentional; and directed at an ingroup or outgroup member. It may or may not involve imitation, and pressure to change may be ordered along a continuum from low to high. Finally, social influence may result in public or private psychological changes; and desirable or undesirable outcomes. These dimensions may impact the extent to which influence impacts one or more needs.

Imitation

Some types of influence (e.g., conformity, behavior mimicry) involve imitation (i.e., adopting the thoughts, feelings, or behaviors of another). Persuasion, or attitude change, may sometimes involve imitation (e.g., when someone persuades a friend to root for his beloved baseball team), and sometimes not (e.g., when an attorney convinces a jury that a defendant she believes to be guilty is not guilty). Another phenomenon, anticonformity (Nail et al., 2000) may be considered the opposite of imitation; it occurs when someone who already agrees with a source responds to influence attempts by moving away from the source's position. We would expect that successful influence involving imitation would be especially likely to enhance a sense of accuracy and belongingness, whereas failed influence would decrease perceived accuracy and belongingness. Influence situations not involving imitation (e.g., some persuasion situations, compliance, or obedience) should have their largest impact on perceived control.

Awareness

Social influence may take place either above or below the threshold of awareness, from the perspective of both the source and the target. In contrast to conformity, in which the target is likely aware of influence pressures, behavior mimicry often occurs below the level of conscious awareness of both the source and target. Targets in mimicry studies reported not knowing that they were influenced by the behavior of others, and sources failed to perceive that targets were mimicking their behaviors (Chartrand & Bargh, 1999). Persuasion may or may not occur above the level of consciousness on the target's part; there is some evidence that attitude change may occur subliminally (e.g., mere exposure effects, Zajonc, 1968). Compliance and obedience are typically thought of as conscious processes. One presumably must know that one is being asked or ordered to change before behavioral change will occur, although Cialdini (2000) emphasizes that being in a mindless state can enhance compliance. Although it seems intuitive to assume that increased awareness that one has influenced another would increase the positive effects of having influence, the work on behavior mimicry (Chartrand & Bargh, 1999) has revealed that increased feelings of affiliation with the target of influence did not require conscious awareness of having influenced the target.

Intention

Influence may be either intentional or unintentional. Compliance and obedience are typically conceptualized as intentional, whereas conformity and behavioral mimicry may occur in the absence of intent. Persuasion may be either intentional or unintentional; although some prototypical persuasion arenas such as courtrooms and debates clearly involve intent, much persuasion occurs in day‐to‐day communications among acquaintances (Latané, 1996; Latané & Bourgeois, 1996). It seems that failed attempts at influence would be especially detrimental to one's well‐being when the influence is intended; indeed, without intention it does not seem logical to even consider an influence attempt as failing. Again, however, the behavior mimicry research suggests that successful influence leads to a stronger sense of belonging with the target than does a lack of influence, even when no intent is involved.

Ingroup vs outgroup target

One dimension that may be critical in how a source might respond to the results of a social influence attempt is whether the target of an influence attempt is an ingroup or outgroup member. Research by Abrams, Wetherell, Cochrane, and Turner (1990) utilizing the autokinetic effect, Asch's (1955) conformity paradigm, and the group polarization paradigm suggests that influence is attenuated when the outgroup status of a source is made salient. This suggests that sources may be more satisfied with successful social influence and more bothered by failed social influence attempts when interacting with ingroup members than with outgroup members.

Public vs private change

A classic distinction in the field of social influence is whether or not public conformity also involves private attitude change. Although we could find no relevant research, we suspect that when a source of influence perceives that he/she has only effected public but not private change, the psychological benefits (with the possible exception of a sense of control) will be less apparent. Believing that a target is only paying lip service to one's attempts at influence will presumably be less satisfying. Although fictional, a clip from a recent movie The Break‐Up illustrates the idea. A woman (played by Jennifer Aniston) isn't satisfied that her partner (Vince Vaughn) is willing to do the dishes. She says, “I want you to want to do the dishes” (to which Vaughn's clueless character replies “Why would I want to do the dishes?”). Recent research (Patrick, 2007) shows that in fact people are happier when their partners do things for them because they genuinely want to.

Pressure to change

Several of the traditional types of social influence can be ordered along a continuum of how much overt pressure to change is perceived by the target. For example, conformity, compliance, and obedience can be ordered from relatively low to moderate to high pressure to change one's behavior. Persuasion attempts may be similarly conceptualized as relatively low or high in felt pressure to change one's attitudes. Because behavior mimicry occurs below the threshold of awareness, there is no perceived pressure to change. We would expect that as pressure to change increases, failed influence attempts would threaten the needs satisfied by having influence to a greater extent.

Desired vs undesired influence

Influence may occur in a manner that is desirable or undesirable to the influence source. For example, undesired influence occurs when a child adopts a parent's bad habits or a student models a mentor's tendency to procrastinate. We suspect that desired influence will always be more fulfilling than undesired influence, but that undesired influence may sometimes (or in some ways) be more rewarding than complete lack of influence. For example, compared to lack of influence, undesired influence may satisfy meaningful existence but threaten self‐esteem and control.

INDIVIDUAL DIFFERENCES

Although we are proposing that having influence over others serves a variety of needs for people in general, individual differences exist in the extent to which people desire to have influence over others, as well as the extent to which people perceive that they are influential (Bennett, 1988). Consistent with our theorizing, Bennett has shown that the perceived ability to influence is correlated quite strongly with measures of self‐esteem. Ability to influence is also positively related to measures of internal locus of control, mastery, and the enjoyment of one's work.

We suspect that people who are dispositionally high in the need to influence (Bennett, 1988) will be more strongly affected by the perception that they have or have not successfully influenced another. In turn, individual differences in the extent to which the various needs served by influence are met may also affect one's satisfaction with having influenced another. For example, Cialdini and Mirels (1976) showed that only those high in internal locus of control showed increased positive regard for a confederate (perceiving him to be more intelligent and attractive) after they thought they had influenced him.

GOAL‐BASED MODELS AND SOCIAL INFLUENCE

Because theirs was the only systematic goal‐based model of social influence in the literature, Cialdini and Goldstein's (2004) framework was adopted as a starting point for our tentative list of goals that might be served by having influence. In focusing on a phenomenon and then considering the potential goals attained or thwarted by the phenomenon, our approach is quite similar to Williams' (2001) work on social ostracism. Indeed, our model overlaps to a great extent with Williams' need–threat model of social ostracism, as well other psychological models of goals and/or needs, such as Maslow's (1943) self‐actualization theory, Deci and Ryan's (1985) self‐determination theory, Baumeister's (1991) needs for meaning, and Fiske's (2003) core needs model. Each of these models posits a small number of needs and/or goals and makes the assumption that satisfying these needs/goals leads to well‐being, and/or that having them unfulfilled leads to distress. We would not argue that the goals specified by our model are necessarily exhaustive; indeed, some of the related yet distinct needs/goals posited by these other models (e.g., Deci & Ryan's need for autonomy) might also be satisfied by exerting influence over others. We are simply proposing our framework as one that could guide future research on the functions of social influence.

We also acknowledge the need for future work to determine whether social influence might be better studied from the perspective of a deficit model (i.e., which assumes that lacking influence over others thwarts goals, causing distress), a growth model (i.e., which assumes that influencing others satisfies goals, enhancing well‐being), or both. The situation described above in which the doomsday cult members attempted to proselytize other after the failed prophecy (Festinger et al., 1956) provides an example of influence replenishing a deficit in perceived accuracy. Some of the related research discussed above—e.g., Schachter's (1951) research on group deviates and Williams' (2001) work on social ostracism—emphasizes the harmful consequences associated with a lack of influence, whereas other research—e.g., Chartrand and Bargh's (1999) behavior mimicry research, and emotional contagion work by Hsee, Hatfield, Carlson, and Chemtob, (1990)—focuses on the positive effects of having influence over others. Findings by Bruno et al. (2008) were consistent with both a deficit and growth model. Future research should include baseline conditions to assess whether having social influence increases well‐being and/or lacking influence decreases well‐being. We suspect that both may in fact be true.

FUTURE RESEARCH QUESTIONS

We have posited that influence from the perspective of the source has been a largely neglected area of study in social psychology. We have also drawn on theory and research to suggest various psychological needs that may be fulfilled by having influence. We conclude with some questions to be explored in future research.

1. Will successfully influencing another person lead to an increased sense of accuracy, belongingness, control, meaningful existence, and more positive self‐concept, when compared to failed attempts at influence? Will these effects generalize to all social influence domains, or will some types of social influence fulfill some needs better than others? To answer these questions, one could systematically pair a participant with a confederate who is trained to either yield to or resist social influence attempts (e.g., attempts at conformity, behavior mimicry, persuasion, compliance, obedience) and subsequently measure the extent to which each of the needs is being met.

2. Will systematically depriving people of each of these needs (i.e., belongingness, control, meaningful existence, positive self‐concept, and accuracy) lead to increased attempts at social influence? We expect that when these needs are threatened, people will put more effort into influencing others in order to restore their perceptions of whichever need was deprived. For example, one could deprive people of a sense of control or self‐worth using conventional experimental manipulations of these constructs (e.g., Baumeister, Heatherton, & Tice, 1993; Pittman, 1998) and then measure the effort or amount of time they put into influencing others.

3. Will successfully influencing others attenuate the effects of deleterious social stimuli? Tesser (2001) argues that the self‐concept is a fluid, dynamic entity, and that when one's self‐esteem is threatened in one domain it may be restored within another. To the extent that having influence over others increases self‐esteem, we may expect that having influence will counteract threats to self‐esteem. Similar logic may be employed for the remaining four needs. For example, prior work has shown that ostracism leads to increased conformity behavior, presumably to restore a sense of belongingness (Williams et al., 2000). Other research (summarized in Pyszczynski et al., 2003) has shown that being confronted with thoughts of one's own death threatens the self and leads to a wide variety of attempts to bolster one's belief in his/her cultural worldview. This includes engaging in more prosocial behavior, such as donating money to a valued charity (Jonas, Schimel, Greenberg, & Pyszczynski, 2002). One could cross an experimental manipulation of social ostracism or mortality salience with a social influence manipulation (i.e., having participants believe that their influence attempts on another were successful or unsuccessful). If influence restores or replenishes the needs threatened by social ostracism or reminders of one's own mortality, then behaviors typically associated with these threats (i.e., conformity and prosocial behavior) should decrease when people are provided with evidence that they have influenced others.

4. Is the relationship between influence and liking for targets mediated by need fulfillment? Perceived influence has been a neglected variable in the literature on interpersonal attraction and close relationships, yet this variable may be crucial to understanding when and why people form and break interpersonal bonds. Lost influence may explain low‐accommodation behaviors and other markers of relationship dissolution, even in the face of other features that make staying in the relationship necessary or desirable (such as high mate value or social/financial rewards/obligations). Having influence may also open people up to being influenced. Hollander's (1958) theorizing on idiosyncrasy credits is consistent with this notion.

5. Is having influence a possible path to good mental health? Interviews with long‐term targets of social ostracism (which again could be thought of as deprived influence) have revealed elevated levels of distress, alienation, and helplessness (Williams, 2001). On the flip side, markers of depression and anxiety include low self‐esteem, low perceived control, a sense of meaninglessness and belongingness—all needs we suggest as being met by having influence. If influencing others increases satisfaction of basic human needs, then mental health conditions characterized by low levels of these needs might be treated via cognitive interventions that emphasize the roles that people play in other people's lives. Research into the functions of social influence may be of interest to clinicians, social workers, and other mental health professionals whose goals are to better understand and improve subjective well‐being.

SUMMARY

We propose that the basic human needs for meaningful existence, control, accuracy, belongingness, and self‐esteem may be met by having influence over others. Studies on reactions to group deviates, social ostracism, leadership, and interpersonal attraction have linked failed influence to dissatisfaction and distress, and successful influence to relationship satisfaction and well‐being. Experimental studies that examine the causal impact of successful versus failed influence on need satisfaction, interpersonal attraction, and mental health are needed to provide additional tests of the theoretical propositions advanced here.

Notes

This material is based upon work supported by the National Science Foundation under Grant No. BCS-0744972.