National Academies Press: OpenBook

Enhancing Human Performance: Background Papers, Social Processes (1988)

Chapter: Substantive Matching as a Social Influence Strategy

« Previous: Front Matter
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 1
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 2
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 3
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 4
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 5
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 6
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 7
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 8
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 9
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 10
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 11
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 12
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 13
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 14
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 15
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 16
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 17
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 18
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 19
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 20
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 21
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 22
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 23
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 24
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 25
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 26
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 27
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 28
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 29
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 30
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 31
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 32
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 33
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 34
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 35
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 36
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 37
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 38
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 39
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 40
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 41
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 42
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 43
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 44
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 45
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 46
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 47
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 48
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 49
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 50
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 51
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 52
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 53
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 54
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 55
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 56
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 57
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 58
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 59
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 60
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 61
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 62
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 63
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 64
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 65
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 66
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 67
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 68
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 69
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 70
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 71
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 72
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 73
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 74
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 75
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 76
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 77
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 78
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 79
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 80
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 81
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 82
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 83
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 84
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 85
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 86
Suggested Citation:"Substantive Matching as a Social Influence Strategy." National Research Council. 1988. Enhancing Human Performance: Background Papers, Social Processes. Washington, DC: The National Academies Press. doi: 10.17226/786.
×
Page 87

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Bandler and Grinder (1975), founders of the neurolinguistic programming tradition, have argued that therapists who match their clients' preferred representational system (visual, auditory or kinesthetic) will "create an atmosphere of rapport and understanding" and, by implication, increase their influence over these clients. This is only one kind of matching that has been advocated as a means of achieving influence. Arguments can be made for the efficacy of matching attitudes, making eye contact, role reversal (in which one feeds back to the other her perspective in a quarrel), and the like. Furthermore, there is a long tradition of research and practice on the efficacy of reciprocity, that is, matching benefit for benefit and harm for harm. This paper presents an overview of theory and research on all kinds of matching...its impact on behavior and its origins. The paper contains two main sections. One deals with substantive matching, that is, similarity in appearance between one's behavior and characteristics and those of the other party. The other deals with reciprocity. - SUBSTANTIATE MATCHING AS A SOCIAL INFLUENCE STRATEGY In this section, we consider the rode of substantive matching in the social influence process. Included within this rubric are similarity of attitudes, appearance, and

2 personality characteristics, as well as nonverbal and verbal behaviors. The section will be divided into two subsections, one on matching of personal characteristics and the other on matching in communication. Part of the latter discussion will consider the role of substantive matching as a source of influence in psychotherapy. In reviewing these literatures, we will discuss research and theory that is mostly consistent with the following assertions. First, matching produces positive attitudes (by which we mean greater liking, rapport' or trust). Second, positive attitudes increase influence. Third, matching increases influence, a point that can be derived from the first two assertions. Matching of Personal Characteristics Similarity and Attraction A voluminous body of research has examined the link between similarity on dimensions such as attractiveness, attitudes, and personality characteristics and liking. Attitude Similarity Perceived similarity of attitudes appears to be a particularly important source of attraction. For example, in a field study, Newcomb (1961) found that the

3 extent of attitude similarity predicted the degree to which people were attracted to each other after several months of association. Laboratory experiments involving the manipulation of perceived attitudes also provide consistent support for the hypothesis that people tend to like others whose attitudes seem similar to theirs (see Berscheid and Walster, 1969; 1978; Byrne, 1971; for reviews). Some, and perhaps virtually all, people seem to be aware that similarity of attitudes increases liking, and use this principle to increase their attractiveness to another. In a study by Zanna and Pack (1975), female subjects learned that they would meet an attractive or an unattractive male, who held either traditional or liberal attitudes toward women. When the male was attractive, but not when he was unattractive, the subjects presented themselves as more liberal to the liberal male, and more traditional to the traditional male. Thus, the subjects deliberately matched (or gave the appearance of matching) the attitudes of the attractive male, presumably to increase their attractiveness to him. There are limitations to the strategy of attitude matching. There is no increase in liking when one perceives that another person is expressing similarity of attitudes with manipulative intent, for the purpose of ingratiation (Jones, 1964; Jones and Pittman, 1982; Jones and Wortman, 1973~. When one has a need for uniqueness (Snyder and Fronkin, 1980), a highly similar other will threaten this

4 need, and one will tend not to like the other. In addition, when similarity to another person has unpleasant implications, the other will be disliked. For example, attitudinal similarity leads to decreased willingness to interact with another if the other is believed to have a history of emotional disturbance (Novak and Lerner, 1968; see also Cooper and Jones, 1969; Taylor and Mettee, 1971; and see Berscheid, 1985 for a review). The effect of attitude similarity on attraction appears to be most important at the beginning of a relationship (Berscheid, 1985~. In dating couples, similarity of attitudes appears to decline as a determinant of the progress of the relationship (Hillel al., 1976; Levinger, 1972), whereas similarity of age, intelligence and physical appearance holds its own (Hill et al., 1976~. The reason for the declining significance of attitudes in these relationships is not clear. Conceivably, similarity of attitudes is an initial ''screening" criterion for involvement in a relationship, and similarity along other dimensions becomes more important once this criterion is satisfied. Similarity in Physical Attractiveness Couples in ~ong-term relationships tend to be similar in physical attractiveness. This is true of same-sex friends (cf. Cash and Deriega, 1978) as well as of dating couples (Berscheid et al., 1971; Murstein, 1972; Silverman, 1971) and married couples

5 (Murstein and Christy, 1976~. In contrast to attitude similarity, this effect does not seem to be due to the impact of similarity on attraction. In general, when judging strangers, people report more liking for more physically attractive others, regardless of how similar they are in attractiveness (see Adams and Crossman, 1978 ; Berscheid and Walster, 1974 ; Huston and Levinger, 1978, for reviews). However, it is inevitable that compromises must be made since the choice of a romantic partner must be mutual. Hence, in the long run, people tend to choose others who are similar in attractiveness because those individuals are more likely to choose them in return than are more attractive individuals. Similarity in Personality Popular intuition to the contrary, there is no convincing empirical support for either the hypothesis that similarity of personalities will increase liking or the hypothesis that dissimilarity (i.e., complementarily) of personalities will increase liking (e.g. t Ajzen' 1974; Hoffman and Mater/ 1966; Meyer and Pepper, 1977; Murstein, 1976). Theories of Similarity Effects on Attraction There are several theoretical explanations for the effects of similarity on liking. These tend to explain the effects of similarity on some dimensions better than others. Byrne (1971) has suggested a reinforcement interpretation, which

6 applies most clearly to attitude similarity. According to Byrne (1971:338), attitude similarity is directly reinforcing, because it satisfies the need to feel that one is "logical, consistent, and accurate in interpreting the stimulus world."' Heider's (1958) balance~theory also accounts for the effects of similarity of attitudes on liking. It suggests that if we share another person's attitude toward some object, then the tendency toward a balanced "P-O-X" (person-other-object) system will lead to a tendency to feel positively about the other. Furthermore, if we like another person, we will tend to assume that the other shares our attitudes (cf. Granberg and King, 1980; Levinger and Breedlove, 1966~. Berscheid (1985) notes that attitude similarity may have either positive or negative implications, although positive implications are probably more common. For example, attitude similarity may provide a person with consensual validation for her opinions, imply that the other will like her, permit greater prediction of the otherls behavior, and suggest that the person will enjoy interacting with the other in mutually enjoyable activities. Hence, Berscheid (1985:457) concludes that, "it is not so much similarity itself that is rewarding or dissimilarity alone that is punishing; rather it is the implications, presumed or anticipated, of those facts that are responsible for similarity-attraction effects and for their limitations and exceptions."

7 Matching of attributes that define or create social categories, such as religion or ethnicity, may increase liking by creating a sense of shared group membership. According to Heider (1958) "unit relations" lead to tt sentiment relations." That is, a sense of belonging together leads to liking, and a sense of not belonging together leads to disliking. For example, matching on religious or ethnic group membership tends to produce a perceived "unit relationship," which will tend to create positive sentiments. Research on "ingroup bias" effects has provided a considerable amount of evidence that perceived unit relations lead to greater liking for members of the unit (or group) and possibly Tess liking for those who are not members of the unit (cf. Brewer, 1979; Brewer and Kramer, 1985~. A striking finding of research on this "ingroup bias" effect is that the group boundaries may be based on trivial or even arbitrary criteria, and shill lead to ingroup bias. For example, an individual who has been (arbitrarily) characterized as an overestimator of dots will tend to rate other overestimators more positively than one who has been characterized as an underestimator (Tajfel, 1970~. Even when the group membership is randomly determined through a lottery procedure, individuals tend to rate ingroup members more favorably (Bil~ig, 1973; Bi1lig and Tajfel, 1973; Crocker et al., in press; Cracker and Schwartz, 1985; Locksley et al., 1980; Rabble and Horwitz,

8 1969). Thus, a sense of ingroup vs. outgroup, or "unit relation," may be created by matching people on virtually any attribute, however trivial. In summary, many types of similarity, but especially attitude similarity, lead to increased attraction under most circumstances. The relationship between similarity and attraction is probably multiply determined (Huston and Levinger, 1976~. Effects of Liking on Influence We have just reviewed evidence that matching of personal attributes, especially attitudes, leads to increased attraction for the person doing the matching. We now turn to the issue of whether this attraction increases that person's influence. Several theories predict that attractive others are more persuasive (McGuire, 1985~. For example, Kelman's (1961) analysis of social influence processes suggests two mechanisms. First, to the extent that one party likes another, the other has the power to punish the first by withdrawing from the relationship. This power may induce compliance in the first party--that is, he does what the other wants him to do (or what he thinks the other wants him to do) because he sees this as a way to elicit a desired response from the other. Second, to the extent that a party likes another, that party may identify

9 with the other and accept influence because of this identification. In other words, the individual may derive some satisfying part of his self-concept through his relationship with the other. This may lead him to accept influence as a way of maintaining this desired relationship, and consequently the desired self-concept. For example, whereas I may comply with the law because it has the power to punish me, I may accede to the influence of my mentor because my relationship to her comprises an important, and valued, part of my self-concept. Heider's (1958) balance theory also suggests that liked others will be more persuasive, because a system is balanced if we share the attitudes of those we like. Empirical evidence that liking increases influence has been provided in several studies (Eagly and Chaiken, 1975; Sampson and Insko, 1964; Schuler, 1982; Tannenbaum, 1956~. Attractive communicators are particularly influential when they advocate undesirable positions (Eagly and Chaiken, 1975~. Early and Chaiken have argued that this is because attractive communicators are not expected to advocate undesirable positions, giving them more credibility when they do. Attractive communicators are also more influential when the arguments in a message are weak (Brandstatter et al., 1982~. In other words, the attractiveness of the communicator matters more when the communicator's case is on shaky ground or has a relatively low chance of persuading the recipient.

lo When does the attractiveness of a communicator matter and what are its effects relative to the effects of the quality of the arguments themselves? In their elaboration likelihood model, Petty and Cacioppo (1981, 1986) have proposed a theory of the circumstances under which source characteristics may be more important than the quality of the arguments themselves. They suggest that persuasion may occur by one of two routes, which they call the central route and the peripheral route. Persuasion by the central route essentially means that recipients attend to the quality of the arguments presented. If the arguments are sound and convincing, then persuasion takes place. If they are weak, the recipient will not be persuaded. Persuasion by the peripheral route refers to the effects on persuasion by variables other than the quality of the arguments presented. For example, when message recipients are affected by the characteristics of the source of the message, then persuasion by the peripheral route is said to have occurred. According to Petty and Cacioppo (1981, 1986), persuasion occurs via the central route when recipients are involved in the the contents of the message, that is, when the message has personal consequences for them. Persuasion occurs by the peripheral route when they are uninvolved, that is, when the message does not have personal consequences for them. For example, in one study students were presented arguments supporting a comprehensive exam

11 requirement at their university (Petty et al., 1981). For subjects in the high involvement condition, this policy would be put in place the following year, and therefore would affect the subjects themselves. For subjects in the low involvement condition, the policy would be instituted several years after they graduated. For subjects in the high involvement condition, the quality and number of arguments presented determined their agreement with the message, but source characteristics had no effect. For subjects in the low involvement condition, the credibility of the source of the message, but not the quality of the arguments, determined agreement. In a similar study (Petty et al., 1983), under conditions of low involvement, the reputation of the source of a message affected persuasion; but under conditions of high involvement only argument quality affected persuasion. In a study directly relevant to the issue of source likeability, Chaiken (1980) found that a likeable source was more persuasive than an unlikeab~e source when subjects were low in involvement, but not when subjects were high in involvement. Chaiken's theoretical account of these effects differs in some ways from Petty and Cacioppo's elaboration likelihood model, but is generally consistent with the hypothesis that source characteristics influence persuasion only when the recipients of a message are low in involvement ~ see Chaiken and Stangor, 1986, for a discussion of the differences between the two models).

12 In summary, several theoretical perspectives suggest that attractive others are more influential than unattractive others. Furthermore, empirical evidence supports this hypothesis, at least under conditions of low involvement with the issue. To date, research suggests that liking for a communicator matters most when the message recipient is uninvolved in the issue, when the arguments are weak, or when the position advocated is undesirable. Effects of Similarity on Influence We have provided qualified evidence for two points regarding similarity in personal attributes: first, similar others are often liked more than dissimilar others, and second, liked others tend to be more influential than disliked others. Together these mechanisms imply one way in which similarity may affect interpersonal influence--by increasing liking which fosters greater influence. Similarity can have an effect on influence in a second way as well--by increasing or decreasing the credibility of the source of a communication (Simons et al., 1970~. This mechanism sometimes works in opposition to the attraction mechanism, when a similar communication source is better liked but regarded as lower in credibility than a dissimilar source.

13 Two types of similarity between communicator and recipient have been investigated in studies of the effects of source similarity on persuasion. One is similarity on the issue in question, that is, the extent to which the position advocated by the communicator is similar to the position already held by the message recipient. The second is similarity on other dimensions, such as socioeconomic status, attitudes on other issues, experience, etc. Turning first to studies of the second type of similarity (similarity on other dimensions), some research indicates that similar communicators are more persuasive (Berscheid, 1966; Brock, 1965; Mills and Jellison, 1968~. This is especially true when the similarity is relevant to the issue under discussion (Berscheid, 1966; Brock, 1965; see Simons et al. for a discussion), probably because issue- relevant similarity increases the credibility of the communicator. Furthermore, on matters of taste or preference, similar communicators may be more persuasive, because we believe we are more likely to agree with the preferences of similar than dissimilar others (Goethals and Nelson 1973). By contrast, communicators who are dissimilar by virtue of having more knowledge, training, or experience relevant to the issue than the recipient tend to be especially persuasive (Aronson et al., 1963; Boncheck, 1967; Haiman, 1949; Paulson, 1954~. Communicators who are dissimilar by virtue of having higher status or more knowledge may be more

14 persuasive on some types of issues but less persuasive on others. Such communicators are likely to be more credible and consequently more persuasive on matters of objective fact, because they are more likely to know the facts. Studies of opinion leaders suggest that they are similar to the people they influence in ways that increase liking and dissimilar in ways that increase credibility (see Simon et al. for a discussion). Opinion leaders have been found to share attitudinal and membership-group similarities with the persons they influence (Katz, 1957; 1963). At the same time, however, opinion leaders tend to be more competent, more interested, better informed, and to have access to wider sources of information than those they influence (Lazarsfeld and Menzel, 1963~. In summary, research suggests that, ho1 cling credibility constant, similar communicators are more persuasive than dissimilar communicators, probably because they are more attractive. The research reviewed in the previous section suggests that dissimilar communicators may be more ~ influential when their dissimilarity gives them credibility. Both of these effects are most likely to be true when the recipient of the communication is low in involvement (Petty et al.' 1981). - The second type of similarity is the match between the position advocated in a message and the position initially held by the recipient of the message. The social judgment model of Sherif and Hovland (1961) predicts that attitude

15 change (i.e., influence) will be greatest when the position advocated by the message neither precisely matches, nor extremely mismatches, the position of the recipient. Sherif and Hoviand invoke the concepts of assimilation and contrast to account for this effect. Specifically., they argue that positions that are close to one's own (i.e., are within the latitude of acceptance) will be assimilated to one's own position, and there is little room for attitude change. As a result, little attitude change Will take place. Positions that differ from one's own, (i.e., are within the latitude of rejection) will be contrasted with one's position, and will lead to little or no attitude change, or even change in a direction away from that advocated by the message (a "boomerang" effect). According to Sherif and Hoviand's social judgment model, then, neither matching nor mismatching of an opinion advocated by a message and the recipient's opinion is effective as a social influence strategy. A moderate degree of mismatching, which avoids the recipient's latitude of rejection, is likely to maximize social influence. Research on the effects of message-recipient match of positions on attitude change suggests that attitude change tends to increase as the message becomes more discrepant from the recipient's position, but with a negatively accelerated function (Bochner and Insko, 1966; White, 1975~. Boomerang effects occur only at extreme levels of discrepancy between the message position and recipient

16 position (Aronson et al., 1963; Bochner and Insko, 1966; Insko, Murashima and Saiyadain, 1966; McGinnies, 1973; White, 1975~. Both the assimilation and contrast model of Sherif and Hovland, and the finding that similarity of attitudes leads to greater influence, may~account for these effects. Summary and Conclusions The research reviewed in this section suggests that matching another's attitudes may be an effective strategy of social influence. Attitude matching results in increased liking which, in turn, increases influence. In addition, others with similar attitudes may have greater credibility. Matching on other dimensions, such as personality characteristics or appearance appears to be less effective as an influence strategy. One caveat regarding all of this literature concerns actual vs. perceived similarity or matching. The research in this area tends to manipulate or measure actual similarity, and examine its effects on liking. However, it is people's perceptions or judgments that they are similar or dissimilar that mediates most of-the effects. Although perceived similarity is often quite close to actual similarity, discrepancies between the two are sometimes

17 found. Such discrepancies sometimes reflect a strategic effort to curry favor by feigning similarity. Matching in Communication In this section, we consider the effects of matching behavior (either verbal or nonverbal) in communication. Matching of aspects of another person's communication behavior occurs in both informal (e.g., normal conversation) and formal (e.g., the therapeutic relationship) contexts. The literature on matching in communication will be reviewed in the following three areas: nonverbal studies, role reversal, and therapy studies. The literature on nonverbal behavior indicates that matching may occur in three major areas; eye contact, interactional synchrony, and other nonverbal conversational behavior (e.g., duration of pauses, interruptions). Eye Contact Eye contact, defined as meeting another's gaze, is a nonverbal behavior that can be used as an influence strategy. Engaging in eye contact has been shown to make communications more persuasive. Messages accompanied by eye contact are judged more authentic and favorable than

18 messages not accompanied by eye contact (Exline and Eldridge, 1967~. People are more likely to change their attitudes to agree with another person when the other makes eye contact than when he does not (Goldman, 1980). Finally, people comply more with requests from experimenters who engage in eye contact than experimenters who do not (see Kleinke, 1986, for a review). People seem to recognize the usefulness of eye contact in persuasion attempts. Subjects were found to engage in higher amounts of eye contact when instructed to be persuasive than when not given these instructions (Mehrabian and Williams, 1969~. Eye contact may increase influence in at least two ways. First, when a person makes eye contact with another, the other may like that person better, thus increasing that person's influence over the other. Second, eye contact indicates a desire to dominate. This increases influence to the extent that the other is dependent on the person who is making the eye contact. Evidence that eye contact can increase liking is provided in a study by Breed et al., (1972~. They found that students of college professors who engaged in eye contact liked their professors more (and performed better) than students of professors who did not make eye contact. Furthermore, the studies reviewed by Kleinke (1986) suggest that eye contact increases influence when it is interpreted by the recipient as an attempt to be friendly. In a review of the literature, Capella (1981) concluded that holding

19 i other variables constant, an increase in gaze during cooperative or neutral interactions leads to increased attraction. Evidence that eye contact is sometimes employed in an effort to dominate is provided by studies showing that people with a high need to control looked at others more while speaking than while listening (Exline , 1975~. In addition, research on negotiation (Carnevale et al., 1981; Lewis and Fry, 1977) has shown that competitive verbalizations are associated with eye contact and are much diminished when a barrier is placed between the bargainers so that they can talk to but not see each other. The latter finding suggests either that eye contact (staring) is an integral part of a dominance routine which cannot be easily enacted if staring is blocked, or that, in competitive settings, staring is interpreted as an effort at dominance, producing defensive counterdominance behavior. Recipients may interpret eye contact either as a manipulative strategy or an attempt at affiliation (Kendon, 1967~. Whether eye contact indicates attempts to be friendly or to dominate appears to depend on the nature of the interaction (Druckman et al., 1982~. Exline (1963) has - shown that in a competitive situation, those who are high in need for affiliation engage in less eye contact than those who are low in need for affiliation. This relationship is reversed in cooperative settings, with those who are high in need for affiliation engaging in more eye contact than those

20 who are low in this need. This suggests again that in competitive settings, eye contact may communicate attempts at dominance (and therefore is avoided by those who want to be liked), whereas in cooperative settings, it may communicate attempts to be friendly (and therefore is used by those who want to be liked). The effectiveness of eye contact as an influence strategy appears to be moderated by the characteristics of the recipient. Kleinke (1986) discusses several moderating variables, including the recipient Is age, sex, and culture. For example, many cultures such as the Navaho Indians and Japanese view looking while speaking as impolite, whereas infrequent eye contact in other cultures signals insecurity and dishonesty. As another example, before age six, children do not associate eye contact with affiliation. Taken together, these findings suggest that eye contact (matching gaze) may increase a partyts influence over another through two mechanisms--increased liking or increased dominance. Increased liking of one party for another may increase the other's influence in cooperative situations, whereas increased dominance of one party over another may increase the party's influence in competitive situations. Research showing that people engage in more eye contact when instructed to be persuasive (Mehrabian and Williams, 1969) suggests that they are able to use eye contact to increase their influence. Although Patterson (1982; 1983) has suggested that eye contact is used

21 intentionally to obtain social control, Kleinke (1979) found that people were unable to estimate accurately how much they were gazing at another, suggesting that they may not always be aware that they are using eye contact for this purpose. Interactional Synchrony A second type of matching of nonverbal behavior that may occur during interaction is interactional synchrony. This is defined as precise, harmonious changes in the direction of body movement between speaker and listener. Some research suggests that this phenomenon occurs during normal conversation. Condon and Ogston (1966; 1967) analyzed films of dyads interacting, and of groups of people listening to a speaker. Kendon (1970) filmed dyads interacting in a hotel lounge, and analyzed the interactions for evidence of nonverbal synchrony. In both cases, the researchers reported evidence of interactional synchrony, although in neither case were statistical analyses of the data reported. McDowall (1978) has argued, on the basis of evidence, that this phenomenon does not occur at levels greater than expected by chance. Even if interactional synchrony does not occur over the entire course of a conversation, however, it may be associated with particular events during the conversation, such as attempts to indicate understanding, agreement, liking, etc.

22 Both Kendon and Condon and Ogston propose that interactions synchrony results in increased rapport. According to Condon and Ogston (1966; 1967), synchrony provides constant feedback regarding the level of~attention and interest of a conversation partner. Synchrony appears to show the partner is "with'' us in movement as well as thought. Hence, it may produce a sense of being understood and enhanced rapport. If synchrony increases liking or rapport, it may alto increase influence, because of the effects of attraction on influence which we have noted above. However, there is as yet no empirical evidence on the effects of interactional synchrony on liking or influence. It is not clear whether participants in a conversation are aware of interactional synchrony when it occurs. It is possible that synchrony occurs unintentionally or unconsciously, perhaps as a result of a natural tendency to imitate. Alternatively, synchrony may be the result of both parties searching for a mutually accommodating pattern of body movements that is comfortable and acceptable to both of them. Can synchrony be used deliberately to increase ones influence over another? To the extent that synchrony ordinarily occurs outside of awareness, it may be difficult to control deliberately. However, it is possible that people can synchronize their behavior with another's

23 1.. deliberately, and thus increase their influence. Again, further research is needed to address this issue. Matching during Conversation Behavior A third type of nonverbal behavior that may be matched during conversation is response latency, intensity of response, and the incidence of pauses (Capella, 1981). In conversations, these nonverbal behaviors tend to correlate between two participants at a level of r=.39 to .62 (Capella, 1981~. Matarazzo and Weins (1972) found that people imitate the length of their partner's utterances, reaction latency, and percentage of interruptions. These results have been found both in the laboratory fallen et al., 1965) and in naturalistic settings (Matarazzo et a1., 1968). Research has not, to our knowledge, addressed the consequences of these types of nonverbal matching. As with interactional synchrony, they may produce increased rapport. Alternatively' as with eye contact! they may produce either increased liking or signals of dominance, depending on the situation in which they occur. They should also produce social influence if either of these hypotheses is supported, but-this issue has not been subjected to research. Matching as role reversal. Another type of matching that may occur during conversation is role reversal. In

24 role reversal, one presents the feelings and viewpoints of the other person while the other is present (Johnson, 1971a). Johnson (1967, 1971b) has studied the effects of role reversal in the context of negotiation. Accurate role reversal leads recipients to perceive their partner as understanding and themselves as understood (Johnson, 1971a). This enhances their willingness to make concessions, perhaps because of reduced defensive adherence to their position and a lessened sense of being threatened by the other's position (Johnson, 1971a). However, if the positions of the interactants are mutually exclusive, then role reversal results in less appreciation of the other's position than if each party simply presents his or her own position (Johnson, 1967~. Thus, the more incompatible the positions of interactants are, the less effective role reversal will be in encouraging rapport and resulting influence. The manner in which one role reverses also appears to affect the outcome. Wa~m-role reversal makes one perceive similarity between oneself and the party doing it and enhances trust in that party. However, it paradoxically also decreases that party's influence, probably because one concedes less when the other negotiator seems more sympathetic to one's position (Johnson, 1971c). The latter finding may apply only to competitive situations such as were studied in Johnson's experiment. Research has not investigated the effects of role reversal in noncompetitive

25 situations, but it seems likely that warm role reversal would lead to more influence than cold role reversal in such situations, because it results in increased attraction, which enhances influence in noncompetitive situations. Neurolinguistic Programming in Psychotherapy One domain in which the effects of matching in communication has received a great deal of attention recently is matching of therapist/client behaviors in psychotherapy. Bandler and Grinder (1975) have proposed a theory of neurolinguistic programming (NLP), to account for the success (or lack thereof) of psychotherapy. According to these authors, people can process reality though several representational systems, the three most important of which are the visual, auditory, and kinesthetic systems. Each individual is hypothesized to have one consistent system through which he typically processes reality, called the preferred representational system (PRS). The PRS of an individual is purported to be revealed both in the verbal content of his speech and in eye movements. For example, people who phrase statements using the words "I feel" and turn their eyes downward are hypothesized to process reality via the kinesthetic mode. Bandler and Grinder believe that their theory has many practical applications. They argue that assessing and

26 matching clients' PRS can be especially helpful in psychotherapy. Therapists are instructed to phrase their language to match the client's PRS and also to nonverbally match those eye movements that purportedly correspond to the client's PRS. Bandler and Grinder (1975~-suggest that "master" therapists match their clients' PRS and are effective because they match. They also indicate that NLP theory applies only to right-handed people. Matching clients' representational systems verbally and nonverbally is hypothesized, like accurate role reversal, to give the impression that the therapist is seeing the world the way their clients do (Grinder & Bandler, 1976). This should communicate to the clients that the therapist understands them thoroughly. Viewing one's partner as understanding is associated with increased levels of trust (Johnson, 1971a), as noted in our consideration of research on role reversal. Furthermore, trust in a counselor increases a counselor 's influence over a client (Strong and Schmidt, 1970~. Research on NLP has not generally been kind to this theory. First, with respect to whether individuals have preferred representational systems, (Gumm et al. (1982~) report that the three methods of determining a person's PRS (i.e., eye movement, verbalization, and self-report) are not significantly correlated. Eye movement, in particular, has been found to be an unreliable indicator of PRS (Thomason et al. 1980). PRS as indicated by subjects' verbalizations was

27 found to be stable over a one week period in one study (Birholtz, 1981). However, in this study all subjects were found to use the kinesthetic mode, which contradicts the theory's assumption that there -are individual differences in PRS. A study by Hammer (1983) suggested that people may have more than one representational system. Thus, the basic hypothesis that there is a PRS which differs across individuals is not yet established. Many of the studies cited as supporting the assumptions of NLP use the unreliable eye movement indicator of PRS (see Sharpley, 1984, for a review). In sum, the empirical foundation of NLP is quite weak. One study has investigated the consequences of matching the representational system (e.g., "I feel" or "I see") employed in an individual's most recent statement. This was found to lead to a perception of the therapist as empathic (Hammer, 1983). However, it should be noted that matching a person's most recently used representational system may or may not involve matching her preferred representational system (PRS). Other research (Paxton, 1981) has shown that both matching and mismatching a client's PRS as indicated by verbal statements led to more improvement than nonmatching. There are at least two possible explanations for this result. First, in light of Hammer's (1983) finding that people may have more than one representational system, it is possible that mismatching one representational system led to matching another. Alternatively, and more likely, it is

28 possible that how the client represents reality is irrelevant, and any statement by the therapist phrased in the form "I feel..." or "I see..." or "I hear..." is more effective than statements that are not phrased this way. The existing research on NLP theory suffers from several additional weaknesses (Sharpley, 1984~. Most important, research has not examined whether PRS matching is necessary or sufficient for successful therapy outcomes. Most studies to date have examined only a single session, involving experimenters and subjects rather than counselors and clients. Most studies have measured only the experimenter's judgments of the effectiveness of the matching, rather than assessing long-term changes in clients' behavior. Research has not addressed whether the PRS is actually a manifestation of an underlying neurolinguistic schema, or merely a J earned language pattern. Finally, research on NLP to date has almost exclusively used female undergraduates in American psychology laboratories, raising questions about the external validity and generalizability of the findings. It is not clear whether any positive effects of matching representational systems are due to the mechanisms proposed by the theory, or are merely due to reflecting the client's statements back to her, as advocated by client-centered therapists (cf. Rogers, 1951) for decades. Hence, it is yet to be determined whether NLP theory is something new, or merely a reinvention of the wheel.

29 In light of the research that is still needed on the fundamental assumptions of NLP theory, it might seem premature to call for studies of whether therapists (and others) can be trained to detect and match their clients' PRS. However, NLP is being taught at seminars for psychotherapists across the nation, suggesting that evaluation of the effectiveness of these seminars is needed. In the event that further research supports the assumptions and effectiveness of NLP theory, it would be interesting to determine the effects of matching PRS in relationships other than the client-therapist interaction. Other Therapy Studies Apart from matching a client's PRS, there are numerous other dimensions on which therapists might match their clients, including SES, gender, personality traits, and attitudes. Relatively little research has addressed the consequences of matching on these other dimensions of client-therapist interactions. In one large-scale study of matching, Berzins (1977) found that therapists who differed from their clients on the personality dimensions of autonomy and dependency are most likely to be successful. This may be because a dissimilar therapist provides a role model for clients along dimensions that are relevant to the clients' problems.

30 Research reviewed earlier suggests that matching client attitudes on topics of importance to the client may increase client liking for the therapist, and consequently increase the therapist's influence. Matching of behaviors other than NLP may also increase rapport in the therapist-client relationship, increasing the therapist's influence. However, some types of mismatching (e.g., mismatching of education or experience) may increase the therapist's credibility, and hence increase his or her influence. Clearly, the possibilities for using matching or mismatching to increase a therapist's influence over a client have not been adequately explored. Conclusions Research and theory on substantive matching suggests that matching on a wide variety of dimensions and behavior may increase influence over another person. One mechanism by which matching has this effect is by causing the other to like the matcher! consequently increasing influence. Research on matching of attitudes and nonverbal behaviors is especially consistent with this conclusion. Another mechanism by which matching may increase influence is by increasing the credibility of the matcher. However, whether matching or mismatching increases credibility may depend on whether the topic of the influence attempt is one of

31 objective fact or subjective preference. Research in persuasion contexts shows that the likeability and credibility of the communicator affect persuasion only if the recipient of the communication is relatively uninvolved in the issue. However, considerable attempts at social influence occur outside of direct persuasion contexts,~and the likeability and credibility of the communicator may have more impact in these other contexts. Research on role playing in competitive (negotiation) settings indicates that verbalizing the other's position in the other's presence increases one's influence by leading the other to feel understood and thus enhancing trust. Similarly, NLP theory proposes that matching a client's PRS increases that client's trust in the therapist, which may augment the therapist's influence. However, the role playing research indicates that warm role playing, while it augments liking and trust, may undermine influence by making the other think he is winning. This suggests that fostering liking and trust are not a sure road to influence in competitive interactions. One must also foster respect in the sense of a belief that one will not easily concede from onets essential position or allow oneself to be easily exploited. In the next section, we will consider the effects of matching in mixed-motive situations, in which both cooperation and competition are possible.

32 RECIPROCITY AND SOCIAL INFLUENCE This section deals with reciprocity, the matching of reward for reward and/or punishment for punishment. Reciprocity is found in relations between collectivities (groups, organizations, nations) as well as between people. Hence, the participants in the interactions discussed in this section will be called "parties" rather than "people." Reciprocity enters into the influence process in two ways: (a) Reciprocating another party's rewards and/or punishments tends to encourage that party to provide further rewards and avoid providing further punishments. (b) If the other party practices reciprocity, we can get that party to do favors for us by providing reward to that party. These points will be discussed in the two main subsections of this section. The mechanisms by which reciprocity leads to influence are quite different from those by which substantive matching leads to influence. Hence, there will be very little. overlap between the last section and this section. Reciprocating another Party's Behavior The bulk of the research on the effect of reciprocity is found in three kinds of studies: studies of instrumental learning, studies of concession matching in negotiation, and

33 studies of reactions to the other party's strategy in the prisoner's dilemma (PD). The instrumental learning paradigm is a simple case of reciprocity, with the trainer providing reward in exchange for cooperation and/or punishment in exchange for noncooperation. Much research has shown that this routine is highly effective in encouraging further cooperation from its target (see Hulse, Egeth and Deese, 1980). Negotiation can be defined as any communication sequence that is designed to reach agreement on an issue or issues for which there is an initial divergence of interest. Reciprocity of concession making has been shown to be an effective way to induce further concession making in this setting (Komorita and Esser, 1975; Wall, 1977, 1981). The PD is a reward structure that is usually presented in matrix form. An abstract version of the PD is presented below on the left, and a numerical example is presented below on the right. Column's Options C' D' Column's Options C' D' Row's C R,R' S,T' Row's C 6,8 -4,12 . Options D T,S' P,P' Options D 10,-2 3,3 The participants are viewed as having two options: to cooperate (C and C') or to fail to cooperate (i.e., to defect) (D and D'). The row player's outcomes for each pair

34 ,4 of options are given to the left of the comma and the column player's outcomes to the right of the comma.1 For example, when row chooses D and column chooses C', row gets T (10) and column gets S' (-2). The reward structure is a PD if T > R > P > S and Tl > Rl > pl > Sl. ~ The PD is paradoxical because actions that seem rational for a single party are self-defeating if both parties take them. Since T > R and P > S for both parties, it is rational for both parties to fail to cooperate (choose D and D' instead of C and C'). Yet if they do so, they are both worse off than if both cooperate, since R ~ P for both of them. It should also be noted that the parties are dependent on each other in the PD, that is, the other party's cooperation has value for both parties. This is because R > S and T > P for both parties. Two examples of the PD come to mind: (1) Two neighbors with different types of tools. Cooperative behavior (C) would be loaning a tool when needed, noncooperative behavior (D) would be failing to do so. Both parties are better off if they both loan tools (R,Rt) than if they both do not (P,P'). However, there is a temptation (T) to borrow but not to loan, which, if successful, gives the other the worst possible outcome (S). (2) Two nations each trying to decide whether to develop a new and more deadly weapons system (D and D') or not to develop the system (C and C'). The parties are better off if neither develops the system (R. R') than if both do so (P. P'). But there is a temptation

35 (T or T') to try to develop the system first before the other does so, and the danger that the other will develop it first (giving one an outcome of S or S') also motivates development of the system. The PD is a very common reward structure in real-life settings. Most ethical principles, e.g., sharing norms,.can be construed as producing mutual cooperation in what would otherwise be a troublesome PD. There is a link between the earlier section on substantive matching and work on the PD in that perceived attitude similarity has been found to predict cooperation (choice of C) in the PD (Tornatzky and Geiwitz, 1968; McNeel and Reid, 1975~. This is probably related to a broader finding that perceived ethnic or attitude similarity begets helpful behavior (Hornstein, 1976~. It seems reasonable to assume that attraction mediates these relationships in that perceived similarity produces attraction (Berscheid and Walster, 1969, 1978; Byrne, 1971; Newcomb, 1961) and attraction encourages cooperation in the PD (McC1intock et al., lg70; Swingle and Gillis, 1968; Oskamp and Perlman, 1966). We can define reciprocity.in the PD as cooperating (choosing C') in response to the other's cooperation (C) and failing to cooperate (choosing D') tin response to the other's noncooperation (D). Another term for reciprocity is the "tit-for-tat" strategy. This strategy has been shown to be very effective for eliciting cooperation from the other

36 party (Deutsch, 1973; Kuhlman and Marshello, 1975; Oskamp, 1971; Pilisuk et al., 1968; Pruitt, 1981; Wilson, 1971; Wrightsman et al., 1972). Explanations for the Success of Reciprocity There is considerable danger in trying to generalize across learning, negotiation and PD settings, because these settings differ on many dimensions. One dimension is that negotiation and PD settings usually involve mutually acknowledged divergence of interest, while learning settings do not. Another dimension is that the parties are trying to reach agreement in negotiation, while they are not in learning and PD settings. Nevertheless, these settings have many features in common, including the fact that reciprocity tends to encourage further cooperation. Hence, we will treat them together, while calling attention, from time to time, to possible differences. There are three types of explanation for the success of reciprocity in these settings, one in terms of automatic processes and the other two in terms of cognitions. The first explanation postulates some automatic reinforcement process that is outside cognitive awareness. This process/ if it occurs enhances the strength of responses that have been rewarded and diminishes the

37 strength of responses that have been nonrewarded or punished. The second explanation is in terms of expectations. In one version of this explanation, these expectations concern the behavior reciprocated. The expectations are: (a) the behavior for which ~ have been rewarded will continue to be rewarded and/or (b) the behavior for which I have been punished will continue to be punished. In another version, they concern the strategist who is engaging in reciprocity. They are: (a) the strategist will cooperate if I cooperate and (b) the strategist will not cooperate if I fail to cooperate. The latter expectations should lead people to cooperate whenever they see that there is more to be gained from the strategist's cooperation than to be lost by cooperating with the strategist. Another way to put this is that these expectations should lead people to cooperate whenever they regard mutual cooperation as more valuable than mutual noncooperation. If the participants are perceptive, the latter condition will obtain in the PD, because R ~ P for both parties. Hence reciprocity is a rational strategy in the PD because it encourages the other to cooperate.2 The third explanation (see Komorita and Esser' 1975; McGillicuddy et al., 1984) is that targets develop attributions about the strategist, seeing him as "strong" and "firm" (because he fails to cooperate when they fail to cooperate) but also as ''conciliatory" and "fair" (because he

38 cooperates when they cooperate). This leads them to cooperate with him in later settings because they expect to be rewarded if they do and not rewarded or punished if they do not. This explanation is similar to the second one in viewing expectations as the ultimate source of the target's cooperative behavior. It differs from the second explanation in its account of the origins of these expectations, postulating that they are produced by attributions about why the strategist was engaging in reciprocity rather than directly by his reciprocity. Two efforts have been made to test the three explanations just given in the realm of negotiation. Both are damaging to the automatic reinforcement explanation. Komorita and Esser (1975) examined the impact of four concession patterns: matching the other's concessions (reciprocity); making concessions on 100% of the opportunities; making concessions on 0% of the opportunities; and mismatching the other's concessions. In the latter strategy, the confederate conceded when the other did not concede and failed to concede when the other did. Concessions were most frequent in the matching condition, next most frequent in the 100% condition, next most frequent in the mismatching condition, and least frequent in the 0% condition. The authors rejected a simple conditioning explanation for these results, arguing that, if conditioning were at work, the mismatching strategy would have produced fewer concessions than the 0% strategy. There is some logic

39 to this argument, but it can be questioned on the grounds that the 0% strategy may have elicited a retaliatory reaction from the opponent which masked a conditioning effect that was indeed present. In a second study (McGillicuddy, Pruitt and Syna, 1984), more concessions were made by negotiators who had previously watched-their opponent follow a reciprocal strategy than those who had watched him concede rapidly or hardly at alla These results cannot be explained by automatic reinforcement, since the subjects observed rather than experienced their opponent's reciprocity. Some cognitive process that involved an image of the opponent almost certainly was involved. Auxiliary evidence suggests that this cognitive process may have involved attributions about the opponent. Opponents who followed a reciprocal strategy were rated as "stronger" than those who followed a soft strategy which involved making a concession on every trial. They were also rated as "fairer" than those who followed a tough strategy which involved infrequent concession making. In conclusion, evidence in negotiation settings is unsupportive of an automatic reinforcement explanation for the success of reciprocity and supportive of an attributional explanation. It should be acknowledged, however, that the latter explanation is only viable in settings where the subject can develop a conception of the source of the reciprocity. Such a conception is not

40 possible in traditional animal learning studies. Hence, one of the other explanations (e.g., automatic reinforcement or expectations tied to behavior) may be more appropriate for that setting. Conditions Affecting the Success of Reciprocity Reciprocity is more effective in some conditions than others. For example, research on the PD by Kuhlman and Marshello (1975) suggests that reciprocity is most effective when the other party has an individualistic orientation, i.e., is trying to maximize her outcomes. The alternatives to this orientation are a cooperative orientation, in which the other is concerned about the strategist's outcomes as well as her own, and a competitive orientation, in which the other is trying to do better than the strategist. Two groups of researchers have found that reciprocity is counterproductive when the other has a cooperative orientation. Kuhlman and Marshel~o (1975) showed that 100% cooperation elicited more cooperation from the other than did reciprocity. This is presumably because reciprocating a cooperative person Is occasional failure to cooperate amounts to overreacting to a temporary aberration. Evidence from a study by Clark and Mills (1979) suggests that reciprocity is resented in intimate relationships. A possible explanation for this finding is that the norm for intimate relationships

41 involves selfless generosity rather than an exchange of favors. Reciprocity is also ineffective when the other has a competitive orientation (Kuhlman and Marshello, 1975). This is presumably because mutual cooperation tends to be of little interest to a competitive person. Reciprocity seems tube most effective when it fits an a priori notion (a causal schema) held by the other party "about the types of causes that ought to be linked to particular types of effects' (Nisbett and Ross, 1980:115~. This is presumably because it is easier to perceive a relationship between one's behavior and the other's response when that relationship fits one's theories about cause and effect. The initial evidence for this assertion came from animal research showing that learning is much easier to establish for some connections between behavior and reward than for others. For example, Garcia, McGowan and Green (1972) have shown that rats can readily be trained to avoid a new tasting food that is followed by gastrointestinal illness but not to avoid a new shaped food that produces the same illness. The unfamiliar shape, on the other hand, is more readily conditioned to the avoidance of electric shock than is the unfamiliar taste. A similar phenomenon has been described by anthropologists who have found that primitive remedies often resemble in some way the disease being treated (Nisbett and Ross, 1980~. For example, Evans- Pritchard (1937) described the Azende practice of treating

42 epilepsy with the burnt skull of a red bush monkey that shakes like an epileptic. Presumably, this resemblance made the Azende much more alert to cases in which epileptic fits were relieved by this remedy than those in which they were not. Reciprocity is also more effective in producing cooperation in the PD when the parties move sequentially, with the strategist always moving second after seeing how the other has acted, than when they move simultaneously (Oskamp, 1971). When moves are sequential, there is no way for the other to temporarily exploit the strategist--failing to cooperate yet enjoying the benefits of the strategist's cooperation until the strategists detects what is happening. Hence it should be especially clear to the other that he faces a choice between mutual cooperation and mutual noncooperation. An example would be the familiar case where a tenant (the other) must pay the rent before the landlord (the strategist) allows him to move in. The landlord is less likely to be paid if the tenant moves in first or even if they take action simultaneously. In summary, reciprocity is especially effective in eliciting cooperative behavior from another party when that party is trying to maximize her outcomes, sees a logical link between the behavior in question and the reward or punishment that follows it, and takes action before that taken by the strategist rather than moving simultaneously with the strategist.

43 i Techniques for Making Reciprocity Work Learning theorists have developed a number of guidelines for the successful use of reciprocity (i.e., reinforcement) as an influence technique. Some of these are also supported in research on negotiation and the PD. Immediacy Reward or punishment should follow the target's behavior as quickly as possible until that behavior has been firmly established (Perin, 1943~. After that, it is useful to lengthen the interval between the desired response and the administration of reward or punishment. This will make the desired response more hardy, that is, more resistant to -extinction (Mackintosh, 1974~. If it is necessary to delay reinforcement at the outset of conditioning, it may be possible to substitute a symbol of the reward or punishment, for example, a verbal promise to give a child candy after dinner in exchange for straightening her room. For maximal effectiveness, such a symbol must be periodically followed by administration of the actual reward or punishment (Hulse et al., 1980~. Consistency Many learning studies have shown that consistent reciprocity is more effective for establishing a response than irregular or periodic reciprocity. For

44 maximum effectiveness, the other should be rewarded or punished every time he makes the desired or the undesired response (Hulse et al., 1980). This effect has also been shown in research on negotiation. A regular pattern of conceding every time the other concedes produces more concessions from the other than an irregular pattern, even if the same overall amount is conceded in the irregular pattern (Esser and Komorita, 1975~. Research on the PD (Kuhlman and Marshello, 1975; Oskamp, 1971; Tedeschi et al., 1968) and on bargaining (Komorita and Esser, 1975) also suggests that it is more important to be 100% consistent in rewarding cooperation than 100% consistent in punishing (or failing to reward) noncooperation. A possible explanation for this effect is that reward and punishment tend to be reciprocated by the target as well as to reinforce the target Is behavior. Hence, if the strategist is going to be inconsistent, it is better to be light on punishment (and thus to diminish provocation of the other) than to be light on reward (and thus fail to build up positive credit with the other). Though consistent reciprocity is best for establishing a response in others, inconsistent reciprocity ("partial reinforcements) paradoxically makes this response more hardy in instrumental learning settings, that is, less likely to disappear if reciprocity is discontinued (Skinner, 1938). Hence, some experts (e.g., Kazdin, 1975) recommend 100% reciprocity at first, to establish the desired response,

45 followed by a reduced rate of responding to encourage resistance to extinction. One needs to praise a child every time he cleans up his room, at first; but after a while, periodic praise will suffice. The finding that inconsistent reciprocity makes responses more hardy Inca standard result in learning studies, but it was not replicated in a PD study (Tedeschi et al., 1968). Size of the Reciprocal Response Other research suggests that larger rewards are more effective at strengthening cooperative behavior (Esser and Komorita, 1975; Logan, 1960). Again, the rule is different if hardiness is at issue. Provided that they are large enough to produce learning, smaller rewards will encourage greater persistence once reciprocity is discontinued (Hulse et al., 1980). One possible explanation for this latter effect is that smaller rewards make the other experience greater dissonance if the other enacts the response. Hence, the other is encouraged to develop arguments for the intrinsic value of this response, which in turn maintain the response (see Festinger, 1957~. The case for large incentives is less clear for punishment. Research on laboratory animals shows a simple relationship between intensity of punishment and extent of learning (Church, 1969). Yet punishment often encourages side effects, such as emotion, withdrawal and resentment, which may interfere with learning (Kaz~in, 1975~. The

46 stronger the punishment, the larger are these side effects. Findings on humans by Lindekold (1976) suggest that the most effective punishment is that which exactly "fits the crime," neither undermatching nor overmatching the other's level of noncooperation. Such a penalty is as large as possible without seeming unfair. There are arguments for and against gradually increasing punishment in response to noncooperation. In most laboratory studies, punishment that is introduced at full strength has proven most effective (Kazdin, 1975~. This is because people often habituate to punishment, learning to Jive with it. However, there are arguments for gradually increasing punishment which may sometimes mitigate this advice. For example, gradual increase allows one to explore for side effects. Gradual increase may also seem more legitimate to the target or to third party observers. Hence, gradual increase may make more sense when side effects or legitimacy are at issue. Variety Responses that are induced by a strategy of reciprocity are often quite specific to the setting in which they are learned/ the mode of behavior that was enacted at that time, and the presence of the person who is administering the incentives. To achieve greater generalization, reciprocity should take place in different situations, with different people, and with several variations of the desired response (Kazdin, 1975~.

Prior Experience with a Reward Reward will often work better if the other has experienced it in the past-and hence come to value it. In addition to the evidence in the conditioning literature (Kaz~in, 1975), this point is supported by a PD study by Komorita and Mechling (1967), in which preceding reciprocity by a period of unconditional cooperation enhanced the effectiveness of reciprocity. The more prior cooperation, the stronger was the effect (up to 10 trials in the condition that went the longest). Clear vs. Fuzzy Events For reciprocity to be effective in producing later cooperation, it is necessary for clearcut behavior by the target to be followed by a c~earcut response from the strategist. If one or both events are fuzzy, Jacking a definite time and place or hard to distinguish from everyday experience, the target may not notice the connection and, hence, learn nothing. Nonreward and nonpunishment tend to be fuzzy events, especially when there is little experience with their opposite. Hence, reciprocity may be ineffective when a desired target behavior is followed by nonpunishment or an undesired behavior by nonreward. It follows that a strategist who is trying to establish a new response should usually reward that response (and fail to reward or punish its nonoccurrence). If trying to suppress an old response, 47

48 he should usually punish that response (and reward or fail to punish its nonoccurrence).3 Problems with punishment The advantages of punishment are somewhat mitigated by the side effects that were mentioned earlier: emotion, withdrawal and resentment. These may lead to obduracy, aggression, escalation, and loss of positive influence with the other, making it necessary to continue relying on punishment in the future (Deutsch, 1973; Kazdin, 1975~. Hence punishment, if used, should be employed with some finesse, making sure that it seems legitimate and is employed in a context of overall concern for the other's interests (Pruitt and Rubin, 1986~. When possible, threats of punishment should usually be substituted for actual punishment, because if they work they do not require that the other party be hurt. Where a defensive response can be substituted for punishment, adverse side effects are also less likely (Deutsch, 19731. Summary Research on instrumental conditioning suggests the+ the prescription for establishing a new response by rewarding (reciprocating) this response is quite different from that for maintaining the response once established. For establishing a response, reward should be immediate, consistent and sizable. But quite the opposite is true if one wants a hardy response that will hold up in the absence of continuous reward. The usual solution to this conundrum

49 is to start out with immediate, consistent and sizable reward, to establish the response, and then gradually diminish al-1 three dimensions to increase its hardiness. Another useful technique is to give the individual a prior experience with the reward before beginning a reciprocal strategy. If one is trying to suppress a response r punishment is usually more effective than reward because it tends to link a clearcut response (the one to be suppressed) with a clearcut incentive (the penalty). Again, there is merit in employing immediate, consistent and sizable penalties, but this point must be qualified by the possibility of side effects. When side effects are likely, careful monitoring is needed. A policy of gradually increasing the size of the punishment may be needed in such a situation. Many of the same findings have been obtained in research on establishing cooperative behavior in negotiation and the PD. However, not all the findings on conditioning extend to this arena. Hence, we must be careful about generalizing too readily from the conditioning literature. Defects in the Strategy of Reciprocity A major defect in the strategy of reciprocity is that it tends to produce and perpetuate mutual noncooperation. If the other party fails to cooperate even once, the

50 strategist must immediately retaliate. The other will then often counter-retaliate, producing further retaliation by the strategist, etc. In this way, the parties become locked into a pattern of mutual noncooperation.- The problem is most insidious if the other party fails to cooperate at the beginning of the interaction, because the other then gains no experience whatsoever with the strategist's willingness to reciprocate cooperation and thus misses an essential part of the message. Moreover, there may still be a perceptual problem even if the other has initially cooperated before switching to noncooperation. This is because the other may not fully comprehend why the strategist then turns to noncooperation and may blame the strategist rather than herself (see Kelley and Stahelski, 1970~. The other may then retaliate, forcing the strategist to provide further punishment in what becomes a vicious circle. There are two kinds of remedies for these problems. One is to adopt some variant of reciprocity that is less likely to produce a locked in state. The other is to employ a "starting mechanism" once mutual noncooperation has set in. These will be discussed below. Variant Forms of Recinrocitv Several variants on reciprocity that are more effective than simple reciprocity can be identified. One requires the strategist to be slow to retaliate and alto slow to forgive. In a PD study by Bixenstine and

51 Gaebelein (1971), this "slow-slow" strategy was defined as follows: The strategist retaliated 60% of the time after the other's first noncooperation, 80% after the second noncooperation, and 100% after the third noncooperation. Also, if the other was not cooperating and switched to cooperation, the strategist cooperated 60% of the time after the other's first cooperative~response, 80% of the time after the other's second cooperative response, and 100% of the time after the other's third cooperative response. This strategy produced more cooperation than the usual "quick-quick" strategy, which involves immediate 100% · ~ reclproclty ~ A possible explanation for the success of the slow-slow strategy is as follows. If the strategist is slow to retaliate, this allows the other party to make an occasional error without running an undue risk of escalation. The other gets a second chance to correct such errors before becoming the victim of an escalative retaliation. Delaying retaliation may also make it clearer to the other party that she rather than the strategist initiated noncooperation. However, there is a problem with slow retaliation if it is coupled with fast forgiveness (in a "quick-slow" strategy), that is, if the other can redeem herself by a single cooperative action. The other may learn that it is possible to exploit the strategist's good will until he retaliates and then quickly get back into his good graces with a single act of cooperation so as to set him up for a new round of

52 exploitation. This leads to low outcomes for the strategist, as pointed out by Axeirod (1984) and documented by Bixenstine and Gaebelein (1971~. The solution, is to be slow to retaliate and slow to forgive. Once the other has antagonized the strategist to-the point of retaliation, the SiOW-SiQW strategy makes it hard for the other to redeem herself. The other must cooperate several times before the strategist will resume cooperation. A second approach would be for the strategist to be slow to punish and fast to forgive but to warn the other verbally that punishment will be faster the next time the other steps out of line. There is no research on this approach, which involves a mixture of actions and words, but it seems to make sense. A third strategy, which has received research support, is to precede reciprocity with a period of unconditional cooperation. Komorita and Mechling (1967) have shown that this encourages the other to cooperate. Earlier we interpreted this effect as due to the other party's becoming aware of how useful it is to have the strategist cooperate. A second possible interpretation is that the initial unconditional reward encourages the other to cooperate; hence, the other has a chance to learn that his cooperation will be reciprocated. A third possibility is that providing early reward gives the strategist idiosyncracy credits that permit her to punish the other for noncooperation on a later occasion without evoking the other's retaliation.

53 CLANG ~ no Merb~nisms Various auxiliary tactics are available if the other fails to cooperate from the beginning or if the situation has deteriorated into mutual noncooperation. Once these have been enacted, one can then turn to reciprocity in order to encourage the other to continue responding in the desired fashion. If the desired action is not in the other's repertoire, a shaping routine may be employed. This involves reinforcing a sequence of actions that look more and more like the behavior desired (Hulse et al., 1975~. Verbal instruction or modeling may also be used in such a circumstance. Alternatively, the desired action may be in the other's repertoire but be of low probability--a relatively rare response that will not be emitted in the normal course of events. If so, one can sometimes use prompts (Kaz~in, 1975), that is, stimuli that are known to elicit the response. For example, if one wants to reinforce smiling in an ordinarily sour person, one could show him cartoons from time to time. To avoid continued dependence on the prompt as an eliciting stimulus, one must then gradually "fade" the prompt out (Kaz~in, 1975~. If the parties can talk to each other, it may also be possible for them to coordinate mutual cooperation through negotiation, though this requires some element of trust. If trust is too low for genuine negotiation to take place, it

54 may be possible for a mediator to help them develop a deal. In cases of even greater distrust, it may be essential for one of the parties to take a dramatic unilateral initiative to enhance the others trust. Such initiatives will be discussed in the section on inducing reciprocity. Summary Strict reciprocity has a major defect as a strategy in that it often encourages the development of mutual noncooperation and it provides no way to escape from this state of affairs. A modified form of reciprocity, in which the strategist is both slow to retaliate and slow to reinstate after retaliation appears to be superior to strict reciprocity. Various starting mechanisms are also available once mutual noncooperation has set in, including shaping, the use of prompts, negotiation, mediation, and unilateral initiatives. Inducing Reciprocity by Taking Unilateral Initiatives If people are ready to reciprocate, one of the best ways to influence them is to provide them rewards. Reciprocity has been shown in a number of different types of behavior including helping (Depaulo et al., 1983; Greenberg, 1980; Greenberg and Frisch, 1972; Greenberg and Westcott, 1983; Stapleton et al., 1973), liking (e.g., Rosenfeld, 1967; see Berscheid and Walster, 1969, for a review), and

55 self-disclosure (Taylor et al., 1979~. Ciaidini (1985) has discussed at length how reciprocity can be used to influence others or obtain goods from them. Rewarding people in an effort to induce reciprocity often serves as a starting mechanism for escaping mutual noncooperation. Such actions are sometimes called unilateral initiatives in the tradition of Osgood's (1962, 1966) GRIT proposals. When a relationship is severely escalated, such initiatives are often more successful than negotiation. This is because words tend to be suspect since they can be used to trick and since each party is likely to question whether the other will actually follow through on its promises (Lebow and Stein, in press). Hence, the strategist must turn to actions, which often "speak louder than words." A prominent example would be Sadat's 1977 trip to Jerusalem, which ushered in the peace negotiations between Israel and Egypt. "President Sadat of Egypt, concluding that his country was involved in a lose-lose conflict process with Israel, made a gesture of good will in the form of a personal journey to Jerusalem. This started a de-escalatory spiral in relations between these countries that resulted in the eventual resumption of diplomatic relations"' (Pruitt and Rubin, 1986:109~.

56 Why People Reciprocate Reciprocity is by no means inevitable. When it occurs, it can be explained by a number of mechanisms, including the following: Normative Pressures A norm of reciprocity is found in all cultures and is probably essential to the survival of the social system. The norm states that "you should give benefits to one who gives you benefited (Gouldner' 1960:170). The norm of reciprocity may well be derived from the equity norm, which asserts that people should be rewarded in proportion to their merit. Encouraging a Positive Mood Receiving favors puts people in a good mood, encouraging them to be generous (Isen and Levin, 1972). Their generosity is not necessarily directed at the source of the favor; anyone can be the beneficiary. Encouraging Positive Feelings toward the Source of the Initiative People like those who reward them (Berscheid, 198S) and tend to do favors for those they like (Krebs and Miller, 1985). Encouraging Trust A belief that the source of the initiative is seeking a more positive relationship and

57 hence will cooperate further if we begin to cooperate is a crucial antecedent of cooperation in many situations. Encouraging Dependency The other discovers how valuable one's cooperative initiatives are and seeks to keep them coming by taking cooperative initiatives of his own. Mobilizing Third-partv Pressures Third parties who are trying to resolve a controversy tend to encourage reciprocity, because they believe it will move the controversy toward agreement and will encourage further concessions. Conditions Encouraging Reciprocity Reciprocity of unilateral initiatives is more likely and more extensive under some circumstances than others. Intimacy and Similarity Clark and Mills (1979) argue that reciprocity is normative in nonintimate relationships but nonnormative in intimate relationships. Hence, favors will tend to be reciprocated in the former but not the latter. Evidence of this effect can be seen in the finding that self-disclosure is more often reciprocated in interactions between strangers than those between members of married couples ( ).

58 A result that might seem contradictory to the one just described is Hattonts (1967) finding that black negotiators matched (reciprocated) the concessions of other black negotiators and mismatched those of white negotiators (making faster concessions when the other.conceded Jess). However, it seems likely that we are dealing with perceived similarity rather than intimacy in this study. The finding suggests that people who are similar to each other may be especially likely to reciprocate one another's favors, in an effort to develop a relationship. Attributions of Genuineness The likelihood and extent of reciprocity are also influenced by attributions about the motives underlying one's initiatives. Reciprocity is more likely if one is seen as genuinely trying to be helpful or conciliatory rather than seeking to induce indebtedness or create a good impression (Cialdini, 1985; Jones, 1964). This proposition is supported by the finding that unilateral initiatives are most likely to be reciprocated when the strategist is more powerful (has more threat capacity) than the target, next best when the two parties are equal in power, and worst when the target is more powerful than the strategist (Lindekold and Bennett, 1973; Michener, et al., 1975; Thibaut and Riecken, 1955~. The point is that more powerful people do not need to be generous in order to gain influence. Hence, their generous

59 behavior seems more genuine and less ingratiatory than that of less powerful people. The same proposition is also supported by the finding that generosity is more fully reciprocated the greater is the cost of being generous (Pruitt, 1968~; Presumably, greater cost makes generous behavior seem more genuine. This mechanism may have helped Sadat gain the trust of the people of Israel during his trip to Jerusalem. The costs he obviously bore (in terms of loss of support at home and in the Arab world) gave him particular credibility. A related source of credibility may have been the fact that his trip, once made, was irreversible (Lebow and Stein, in press). Salience and Clarity of the Initiative It also seems reasonable to assume that more noticeable unilateral initiatives will be more fully reciprocated. Unexpected actions, such as Sadat's trip to Jerusalem, tend to be especially salient, standing out as figure against ground. Furthermore, action is usually more salient than inaction. Hence, a party will usually get more credit for taking helpful actions than for failing to take harmful ones. (However, a villain seems to get more credit for not being villainous the more experience people have had with his villainy. This point has-been observed with hostage takers, who are apparently given more credit for being normally generous the longer their victim has been a hostage.)

60 Unilateral initiatives are especially effective at producing reciprocation when they are explained as efforts to generate mutual cooperation (Lindskold and Finch, 1981; Lindekold, Han and Betz, 19863. Such an explanation contributes to both salience and clarity of purpose, averting other interpretations such as the notion that the initiative is an effort at ingratiation or a sign of weakness. A Cognitive Model Other useful hypotheses are suggested by a model that is based on the two cognitive explanations for the success of reciprocity that were presented at the beginning of the prior section. According to this model, people will cooperate with us if three conditions are met: (1) They believe that we will cooperate with them if they cooperate with us, that is, they trust us. This may result from viewing us as "conciliatory" or "fair." (2) They believe that we will not cooperate with them if they do not cooperate with us, that is, that we cannot be exploited. This may result from seeing us as 'firm" or "strong." (3) They value our cooperation and/or believe that mutual cooperation is more valuable than mutual noncooperation.4 A unilateral initiative can be viewed as encouraging the first condition/ the belief that further cooperation will be repaid. Hence, the model implies that the second

61 i and third conditions will encourage reciprocity of unilateral initiatives. This may help explain why initiatives by more powerful parties tend to be more fully reciprocated (Lindskold and Bennett, 1973; Michener et al., 1975). These findings were interpreted earlier in terms of attributions. But it is also possible that people believe that they cannot exploit powerful others (condition 2) and hence are ready to reciprocate their cooperative initiatives. They realize that they must play ball with powerful parties because the latter cannot be pushed into unilateral cooperation. There is also evidence that people are more likely to reciprocate favors from a representative with tough, as opposed to soft, constituents (Wall, 1977). Perhaps they understand that a party with tough constituents is unlikely to cooperate if they fail to cooperate (condition 2). Hence it is necessary to reciprocate that party's initiatives in order to gain his cooperation. Reciprocity is also greater when people expect future interaction with the source of the unilateral initiative than when such interaction is not expected (Gruder, 1971; Marlowe et al., 1966). A possible explanation is that the expectation of future interaction causes them to value that person's cooperation (condition 3), making them more prone to reinforce this cooperation when it occurs. Their dependence on the other may also make it seem advisable to

62 follow the norm of reciprocity so as to make a good impression. PD's are often represented in consciousness in decomposed form. A value to self and other is placed on each party's possible actions, but no thought is given to the combined impact of the two parties' actions. When this is true, the same basic PD can sometimes be decomposed in two different ways: one in which each party appears to control the other's best outcomes and one in which the other appears to control them himself. Research (Pruitt, 1970) suggests that reciprocity is more common in the former than the latter condition. What the former decomposition presumably does is to persuade people that the other is strong (condition 2) and that they must seek the other's cooperation (condition 3~. Reciprocating the other's initiatives is a way to do this. Cost of Continued Conflict When people are dependent on us and believe that we will not cooperate unilaterally, they are likely to want to shift to mutual cooperation. This desire should be especially strong when conflict is unrewarding and there are costs of continued conflict. Hence, they should be particularly prone to reciprocate our unilateral initiatives. Another way to put this is that unilateral initiatives are most effective when the target is experiencing a "hurting stalemate" (the term comes from Touval and Zartman, 1985), unable to push us into

63 concessions and experiencing increased costs with the passage of time. Evidence for this assertion is anecdotal but interesting. Sadat's trip to Jerusalem followed a war between Egypt and Israel which must have created a sense that continued conflict was both futile and dangerous. Likewise, a successful series of unilateral initiatives aimed at the Soviet Union by President Kennedy in 1963 came soon after the Cuban Missile Crisis.5 Conclusions:- How to Encouraging Reciprocity The points made above suggest several guidelines for encouraging another party to reciprocate one's conciliatory initiatives. One guideline is to concede from strength. Political analysts often talk about-negotiating from strength, and there is some truth to the notion that people will yield to a party that has threat capacity. But there usually are limits to how much can be accomplished by threats. Further concessions beyond these limits can be achieved if the stronger party is willing to take conciliatory initiatives, that is to concede from strength. A second guideline is to arrange for one's constituents to adopt a tough line in public, in contrast to the conciliatory initiatives one is taking. This tactic is sometimes called the "black hat-white hat routine." A third guideline is to take dramatic and unexpected conciliatory

64 initiatives, such as Sadat's trip to Jerusalem. A fourth is to explain why one is taking conciliatory initiatives. A fifth is to choose initiatives that involve some cost to oneself. A sixth is to arrange for future interaction with the party to be influenced. A seventh into take one's initiatives at a point when the other party is in a hurting stalemate. Osgood (1962, 1966) and Lindskold (1986) have suggested four other elements of strategy that may increase the effectiveness of unilateral initiatives: (a) Take a series of unilateral initiatives over a period of time without requiring reciprocity from the other party. This makes the conciliatory message more credible and gives the other time to rethink his approach. (b) Use a diverse set of initiatives so as not to weaken oneself unduly in one area and so that the only possible common explanation for these initiatives is one's desire for cooperation. (c) Invite reciprocity with each initiative. (d) Reward the other for reciprocating initiatives and retaliate if the other tries to take advantage of them. In other words, adopt a strategy of reciprocity after the initial period of unilateral initiatives. These suggestions make sense, though research evidence for them is not substantial at this time-.

65 Conclusions Reciprocity contributes to influence in two ways. The first is that reciprocity can be a strategy for influencing other people. If we reward their cooperation and/or punish their lack of cooperation, they are likely to cooperate in the future. This is particularly true when the others have an individualistic orientation and a causal schema that links their behavior to our action, when their actions and ours are clearcut and sequential, and when our responses to their actions are immediate, consistent, and large. The second is that we can sometimes influence other people by taking positive, unilateral initiatives--that is, by doing them favors in the hope that they will reciprocate. Such initiatives are particularly likely to work when they are costly to ourselves, when they are salient and clearly explained, when we are seen as genuinely interested in the other people's welfare, when we seem powerful and firm in defense of our larger interests, when the others are dependent on us in the future, and when continued conflict is unrewarding and costly to them. A major problem with the strategy of reciprocity is that our retaliation in the face of failure to cooperate may foster a vicious circle. The strategy of reciprocity provides no method of escaping from such a circle. A way out of this circle is to take positive unilateral initiatives with the hope that the others will reciprocate

66 (Lindskold, 1986; Osgood, 1962, 1966). Such initiatives are especially likely to work under the conditions mentioned in the last paragraph. A cognitive model is useful for interpreting some of the findings on reciprocity. This holds that others will cooperate when they hold three expectations about us: (a) we will cooperate if they cooperate, (b) we will not cooperate unilaterally, and (c) our cooperation is desirable. Our reciprocity tends to produce the first and second expectations. In addition, favors from us tend to produce the first expectation, resulting in reciprocal favors when we are also perceived as strong (producing the second expectation) and when others are dependent on us (producing the third expectation). The three expectations just enumerated can be produced in other ways than by reciprocity and favor doing. For example, receipt of a message indicating cooperative intent from a powerful person should produce the first and second expectations, leading to cooperation. It follows that this cognitive model should have broad application. There is also a social nor- requiring that favors be reciprocated. Like any norm that benefits our interests, others are more likely to follow this norm when we are powerful and are attending to what they are doing. This provides an alternative interpretation for the finding that people tend to reciprocate our favors when they are dependent on us. The fact that such reciprocity is more

67 common when our generosity seems genuine suggests that the norm of reciprocity may derive, at least in part, from an equity norm which postulates that people should be rewarded in proportion to their merit. A VIEW TO THE FUTURE This has been a frustrating paper to write because, despite many efforts, we have been unable to find a link between our two main topics, substantive matching and reciprocity. Both contribute to influence but by very different intervening processes: substantive matching by its effect on attraction and communicator credibility, reciprocity by its effect on expectations and attributions. At some later time, it may be possible to state a relationship between these intervening processes and hence between our two topics (see point 6 below for some initial thoughts along these lines). But current theory and research are not very helpful for this purpose. We end with a discussion of research frontiers. In the realm of substantive matching, the following projects seem to have merit: (1) Studies by Chaiken (1980) and Petty et al. (1983) suggest that similarity in personal characteristics between communicator and target leads to attitude change only when subjects are uninvolved in the issues. Research is needed

68 to test this derivation and to examine whether it holds up in broader contexts. For example, are eye contact and role reversal only effective when involvement is low? (2) Research is needed on the conditions under which eye contact leads to influence through a mechanism of liking vs. One of dominance. We suggest that the former mechanism may be more common in cooperative settings and the latter in competitive settings. But is this really the case? (3) A1 most no research has been done on interactions synchrony, despite the intuitive appeal of this phenomenon. (4) The intriguing results on role reversal (Johnson) 1967, 1971) need to be extended. Does accurate role reversal lead to influence in noncompetitive settings? Does warm role reversal undermine influence (as found by Johnson) or enhance it (as we suspect) in such settings? (5) Much more research is needed on neurolinguistic programming (NLP) in light of the popularity of this approach to therapy. Do people have a preferred representational system (PRS)? If so, what are the best indicators of this system? Does matching the PRS contribute to rapport and success in therapy as indicated by Bandler and Grinder (1975~? Do these procedures only work with right-handed people? Can therapists learn to do this matching? If at all possible, these studies should be done in therapeutic settings rather than in the laboratory with undergraduates.

69 (6) Liking, respect and trust are all possible mediators of the effect of substantive matching on influence. But these states differ, and they need to be pulled apart in research. Doing so might help to forge a link between the material presented in the two parts of our paper, because respect is close to a view of the other as "strong" and trust is close to a view of the other as "fair," attributions that appear in the cognitive model presented in the last section and play different roles in that model. In the realm of reciprocity, the following projects seem worth doing: (7) Much more research is needed about the impact of reciprocity and favor doing on expectations and attributions about the actor. Most of the research on these phenomena in human settings has not been theory driven and has not attempted to measure intervening psychological states. An exception is the study of attributions in negotiation by McGillicuddy et al. (1984), which deserves to be extended. (8) Studies are needed concerning the impact of causal schemata on the likelihood of being influenced by a strategy of reciprocation. Do people who have had experience with particular linkages between behavior and reinforcement learn more readily when these linkages reappear? Animal research (e.g., Garcial McGowan and Green' 1972) suggests that they should, but the research on humans is only suggestive (see Nisbett and Ross, 1980~. -

70 (9) There have been many studies in instrumental learning settings of the importance of immediacy, consistency and size of reinforcement (reciprocity). This research needs to be extended into competitive situations such as those embodied in negotiation and the PD. In particular, the impact of these variables on hardiness of response should be examined in competitive settings. (10) Much more research is needed on the tradeoffs between influence and negative side-effects when reciprocity involves punishment. In particular, we need research on the conditions under which it is more effective to start punishment at full strength as opposed to increasing it gradually. (11) We need studies concerning the impact of a form of reciprocity that involves initial warnings about punishment rather than punishment itself. Such warnings, if credible, should have the desired impact on attributions and expectations without producing the negative side-effects often associated with actual punishment. Such warnings may also help the target understand her own role in provoking punishment if the strategist is provoked into using it. Hence, the target may be less likely to retaliate if punishment is used. (12) The notion that favors are less likely to be reciprocated in intimate than in nonintimate relationships (see Clark and Mills, 1979) needs further examination. This idea suggests that short-term balancing of accounts is less

71 likely in intimate relations. But is this also true for long-term balancing of accounts? If I do ten favors for you in an intimate relationship, are you as unlikely to reciprocate as if I do only one? Or will you reciprocate for fear that I may decide that I am making too much of a contribution to the relationship? (13) Several studies have shown that reciprocity is particularly likely when the source of a favor is powerful (Lindekold and Bennett, 1973; Michener et al., 1975) or has tough constituents (Wall, 1977). But the mechanism for this effect remains speculative. Is it because favors from a powerful other seem genuine? Or is it because such favors are unlikely to be interpreted as a sign of weakness and hence of readiness to continue unilateral cooperation? Studies are needed on these issues. (14) Finally, research is needed to test the seemingly reasonable hypothesis that reciprocity is especially likely in a hurting stalemate, when conflict is unrewarding and is producing escalatory costs. Clearly there is a great deal more to be learned in this area.

72 REFERENCES Adams, G. R., & Crossman, S. M. (1978). Physical attractiveness: A cultural imperative. Roslyn Heights, NY: Libra Publishing. Ajzen, I. (1974). Effects of information on interpersonal attraction: Similarity versus affective value. Journal of Personality and social Psychology, 29, 374 -380. Allen, B. V., Weins, A. N., Weitman, M., & Saslow, G. (1965~. Effects of warm-cold set on interviewee speech. Journal of Consulting Psychology, 29, 480-482. Aronson, E., Turner, J. A., & Carismith, J. M. (1963~. Communicator credibility and communication discrepancy as determinants of opinion change. Journal of Abnormal and Social Psychology, 67, 31-36. Axeirod, Re ( 1984 ~ ~ The evolution of cooperation. New York Basic Books. Bandler, Re , & Grinder, J. (1975~. The structure of ma-tic I. Palo alto, CA: Science and Behavior Books. Berscheid, E. (1966~. Opinion change and communicator -communicatee similarity and dissimilarity. Journal of Personality and Social Psychology, 4, 670-680. Berscheid, E. (1985~. Interpersonal attraction. In G. Lindzey and E. Aronson (Eds.), The handbook of social psychology (3rd ea., vol. 2, pp. 413-484~. New York: Random House.

73 Berscheid, E., & Walster, E. H. (1969). Interpersonal attraction. Reading, MA: Addison-Wesley. Berscheid, E., Dion, E., Walster, E., & Walster, G. W. (1971~. Physical attractiveness and dating choice: A test of the matching hypothesis. Journal of Experimental Social Psychology, 7, 173-189. Berzine, J. I. (1977~. Therapist-patient matching. In A. S. Gurman & A. M. Razin (Eds.), Effective psychotherapy: A handbook of research (pp. 222-251~. New York: Pergamon Press. Billig, M., & Tajfel, H. (1973~. Social categorization and similarity in intergroup behavior. European Journal of Social Psychology, 3, 27-52. Birholtz, L. (1981~. Neurolinguistic programming: Testing some basic assumptions. Dissertation Abstracts International, 42, 3565B. (University Microfilms No. 8118324). Bixenstine, V. E., and Gaebelein, J. W. (1971). Strategies of "real" opponents in eliciting cooperative choice in a prisoner's dilemma game. Journal of Conflict Resolution, 15, 157-166. Bochner, S., & Insko, C.;A. (1966~. Communicator discrepancy, source credibility, and opinion change Journal of Personality and Social Psychology, 4, 614-621. Boncheck, V. (1967~. Commitment, communicator credibility and attitude change. Dissertation Abstracts, 27, 3929A Dissertation Abstracts (universiLv Microfilms No i -3930A.

74 Brandstatter, H., Davis, J. H., & Stocker-Kreichgauer, G. (1982~. Group decision making. London: Academic Press. Breed, G., Christenson, E., & Larsen, D. (1972~. The effect of a lecturer's gaze direction upon his teaching effectiveness. Journal Supplement Abstract Service, 2, No. 226. Brewer, M. B. (1979). In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307-324. Brewer, M. B. & Kramer, R. M. (1985~. The psychology of intergroup attitudes and behavior. Annual Review of Psychology, 36, 219-244. Brock, T. E. (1965~. Communicator-recipient similarity and decision change. Journal of Personality and Social Psychology, 1, 650-654. Byrne, D. (1971~. The attraction paradigm. New York: Academic Press. Cappella, J. N. (1981~. Mutual influence in expressive behavior: Adult-adult and infant-adult dyadic interaction. Psychological Bulletin, 89, 101-132. Carnevale, P. J. D., Pruitt, D. G., and Carrington, P. (1982~. Effects of future dependence, liking and repeated requests for help on helping behavior. Social Psychology Quarterly, 45, 9-14. Cash, T. F., & Derlega, V. J. (1978). The matching hypothesis: Physical attractiveness among same-sexed friends. Personality and Social Psychology Bulletin, 4, The psychology of Annual Review of

75 240-243. Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 3g' 752-766. Chaiken, S., & Stangor, C. (1986). Attitudes and attitude change. Annual Review of Psychology' 381 575-630. Church/ R. M. (1969). Response suppression. In B. A. Campbell and R. M. Church (Eds.), Punishment and aversive behavior. New York: Appleton-Century-Crofts. Cialdini, R. (1984). Influence: How and why people a-tree to things. New York: William Morrow. Clark, M. S., and Mills, J. (1979). Interpersonal attraction in exchange and communal relationships. Journal of Personality and Social Psychology, 37, 12-24. Condon, W. S., & Ogston, W. D. (1966~. Sound film analysis of normal and pathological behavior patterns. The Journal of Nervous and Mental Disease' 143, 338-347. Condon, W. S., & Ogston, W. D. (1967). A segmentation of behavior. Journal of Psychiatric Research 51 221-235. Cooper, J., & Jones, E. E. (1969). Opinion divergence as a strategy to avoid being miscast. Journal of Personality and Social Psychology, 13, 23-40. Crocker, J., Thompson, L., McGraw, K. M., & Ingerman, C. (1987). Downward comparison, prejudice and evaluations of others: Effects of self-esteem and threat. Journal of Personality and Social Psychology.

76 Crocker, J., & Schwartz, I. (1985). Ingroup bias in the minimal intergroup situation: Effects of self-esteem. Personality and Social Psychology Bulletin. Depaulo, Brittingham and Kaiser (1983). Receiving competence-relevant help: Effects on reciprocity, affect, and sensitivity to the helper's nonverbally expressed needs. Journal of Personality Social Psychology' 45, 1045-1060. Deutsch, M. (1973). The resolution of conflict: Constructive and destructive Processes. New Haven, CT: Yale University Press. Druckman, D., Rozelle, R. M., & Baxter, J. C. (1982~. Nonverbal communication: Survey theory and research Beverly Hills: Sage. Eagly, A. H., & Chaiken, S. (1975). An attributional analysis of the effect of communicator characteristics on opinion change: The case of communicator attractiveness. Journal of Personality and Social Psychology, 32, 136-144. Esser, J. K., & Komorita, S. S. (1975). Reciprocity and concession making in bargaining. Journal of Personality and Social Psychology J 31' 864-872. Exline, R. V. (1963). Exploration in the process of person perception: Visual interaction in relation to competition, sex, and need for affiliation. Journal of Personality, al, 1-20. Exline, R. V. & Eldridge, C. (1967). Effects of two patterns of a speaker's visual behavior upon the

77 perception of the authenticity of his verbal message. Paper presented at the meetings of the Eastern Psychological Association, Boston, Mass. Exline, R. V., Ellyson' S. L., & Long, B. (1975~. Visual behavior as an aspect of power role relationships. In P. Pliner, L. Krames, & T. Alloway (Eds.), Nonverbal communication of aggression (pp. 21-52). New York: Plenum. Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press. Garcia, J., McGowan, B. K., and Green, K. F. (1972~. Sensory quality and integration: Constraints on conditioning. In A. H. Black and W. F. Prokasy (Eds.~. Classical conditioning II: Current research and theory. New York: Appleton-Century-Crofts. Gouldner, A. (1960~. The norm of reciprocity: A preliminary statement. American Sociological Review, 25, 161-179. Goethals, G. R., & Nelson, R. E. (1973~. Similarity in the influence process: The belief-value distinction. Journal of Personality and Social Psychology, 25, 117-122. Goldman, M. (1980~. Effect of eye contact and distance on the verbal reinforcement of attitude. Journal of Social Psychology, 111, 73-78. Gouldner, A. (1960~. The norm of reciprocity: A preliminary statement. American Sociological Review, 161-179. Granberg, D., & King, M. (1980~. Cross-lagged panel

78 analysis of the relation between attraction and perceived similarity. Journal of Experimental Social Psychology, 16, 573-581. Greenberg, M. S. (1980~. A theory of indebtedness. In K. J. Gergen, M. S. Greenberg, & R. H. Wills (Eds.~. Social exchange. New York: Plenum Press. ~ Greenberg, M. S., ~ Frisch, D. M. (1972~. Effect of intentionality on willingness to reciprocate a favor. Journal of Experimental Social Psychology, 8, 99-111. Greenberg, M. S. & Westcott, D. (1983~. Indebtedness as a mediator of reactions to aid. In J. D. Fisher, A. Nadler, & B. M. BePaulo (Eds.~. New directions in helping: Vow 1 Reactions to aid. New York: Academic Press. Grinder, J., & Bandler, R. (1976~. The structure of magic l I . Palo Alto, CA: Science and Behavior Books . Gruder, C. L. (1971) . Red ationship with opponent and partner in mixed-motive bargaining. Journal of Conflict Resolution, 15, 403-416. Gumm, W. B., Walker, M. K., & Day, H. D. (1982~. Neurolinguistic programming: Method or myth? Journal of Counseling Psychology, 29, 327-330. Haiman, F. S. (1949~. An experimental study of the effects of ethos in public speaking. Speech Monographs. 16, loo -202. Hammer, A. L. (1983~. Matching perceptual predicates: Effect on perceived empathy in a counseling analogue. Journal of Counseling Psycholgoy, 30, 172-179.

79 Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley. Hill, C. T., Rubin, Z., & Peplau, L. A. (1976~. Breakups before marriage: The end of 103 affairs. Journal of Social Issue, 32, 147-168. Hoffman, L. R., & Mater, N. R. F. (1966~. An experimental reexamination of the similarity-attraction hypothesis. Journal of Personality and Social Psychology, 13, 145-152. Hornstein, H. A. (1976~. Cruelty and kindness: A new look at aggression and altruism. Englewood Cliffs, NJ: Prentice- Hall. Hulse, S. H., Egeth, H., and Deese, J. (1980~. The psychology of learning. New York: McGraw-Hill. Huston, T. L., & Levinger, G. (1978~. Interpersonal attraction and relationships. Annual Review of Psychology, 129, 115-156. Insko, C. A., Murashima, F., & Saiyadain, M. (1966~. Communicator discrepancy, stimulus ambiguity, and influence. Journal of Personality, _, 262-27-4. Isen, A. M., & Levin, P. F. (1972). Effect of feeling good on helping: Cookies and kindness. Journal of Personality and Social Psychology, 2l, 384-388. Johnson, D. W. (1967~. The use of role reversal in intergroup competition. Journal of Personality and Social Psychology, 7, 135-141. Johnson, D. W. (1971a). Role reversal: A summary and review of the research. International Journal of Group

80 Tensions, 1, 318-344. Johnson, D. W. (197lb). The effectiveness of role reversal The actor or the listener. Psychological Reports, 28, 275-282. Johnson, D. W. (1971c). The effects of warmth of interaction, accuracy of understanding, and the proposal of compromises on the listener's behavior. Journal of Counseling Psychology, 18, 207-218. Jones, E. E. (1964~. Ingratiation: A social psychological analysis. New York: Appleton-Century-Crofts. Jones, E. E., & Pittman, T. S. (1982~. Toward a general theory of strategic self-presentation. In J. Suls (Ed.), Psychological perspectives on the self. Hillsda~e, NJ: Eribaum. Jones, E. E., & Wortman, C. B. (1973~. Ingratiation: An attributional approach. Morristown, NJ: General Learning Press. Katz, E. (1957~. The two-step flow of communication: An up-to-date report on an hypothesis. Public Opinion- Quarterly, 21, 61-78 Katz, E. (1963~. The diffusion of new ideas and practices. In W. Schramm (Ed.), The science of human communication. New York: Basic Books. Razzing A. E. (1975~. Behavior modification in applied settings. Homewood, IL: Dorsey. Kelley, H. H., & Stahelski, A. J. (1970~. Social interaction basis of cooperators' and competitors' beliefs

81 about others. Journal of Personality and Social Psychology, 16, 66-91. Kelman, H. C. (1958~. Compliance, identification and internalization: Three processes of attitude change. Journal of Conflict Resolution, 2, 51-60. Kendon, A. (1967). Some functions of gaze-direction in social interaction. Acta Psychologica, 26, 27-63. Kendon, A. (1970~. Movement coordination in social interaction: Some examples described. Acta Psychologica, 32, 100-125. Kleinke, C. L. (1979). Effects of personal evaluations. In G. J. Chelune and Associates, Self-disclosure: Origins, patterns. and implications of openness in interpersonal relations. (pp. 57-59) San Francisco: Jossey-Bass. Kleinke, C. L. (1986). Gaze and eye contact: A research review. Psycholocical Bulletin, 100, 78-100. Komorita, S. S., and Esser, J. K. (1975). Frequency of reciprocated concessions in bargaining. Journal of Personality and Social Psychology, 32, 699-705. Komorita, S. S., and Mechling, J. (1967~. Betrayal and reconciliation in a two-person game. Journal of Personality and Social PsYcholoqv, 6, 349-353. Krebs, D. L., and Miller, D. T. (1985). Altruism and aggression. In G. Lindzey and E. Aronson (Eds.), The handbook of social psychology (3rd ea., vol. 2, pp. 1-71~. New York: Random House. Kuhlman, D. M., and Marshello, A. F. J. (1975). Individual

82 differences in game motivation as moderators of preprogrammed strategy effects in prisoner's dilemma. Journal of Personality and Social Psychology, 32, 922-931. Lebow, R. N., and Stein, J. G. (in press). Beyond deterrence. Journal of Social Issues. Lindekold, S. (1986~. GRIT: Reducing distrust through carefully introduced conciliation. In S. Worche] and W. G. Austin (Eds.), Psychology of intergroup relations. Chicago: Nel son-Ha11. Lindekold, S., and Bennett, R. (1973). Attributing trust and conciliatory intent from coercive power capability. Journal of Personality and Social Psychology, 28, 180-186. Lindekold, S., Bennett, R., and Wayner, M. (1976~. Retaliation level as a foundation for subsequent conciliation. Behavioral Science, 21, 13-18. Logan, F. A. (1960~. Incentive. New Haven, CT: Yale University Press. Mackintosh, N. J. (1974). The psychology of animal learning. New York: Academic. Marlowe, D., Gergen, K. J., and Doob, A. N. (1966~. Opponents t personality, expectation of social interaction, and interpersonal bargaining. Journal of Personality and Social Psychology, 3, 206-213. McClintock, C. G., Nuttin, J. M., and McNeel, S. P. (1970~. Reward and score feedback as determinants of cooperative and competitive game behavior. Journal of Personality and Social Psychology, 4, 606-613.

83 McGillicuddy, N. B., Pruitt, D. G., and Syna, H. (1984). Perceptions of firmness and strength in negotiation. Personality and Social Psychology Bulletin, 10, 402-409. McNeel, S. P., and Reid, E. C. (1975~. Attitude similarity social goals, and cooperation. Journal Of Conflict Resolution, 19, 665-681. Michener, H. A., Vaske, J. J., Schleifer, S. L., Plazewsk' J. G., and Chapman, L. J. (1975~. Factors affecting concession rate and threat usage in bilateral conflict. Sociometry, 38, 62-80. Nisbett, R., and Ross, L. (1980~. Human inference: Strategies and shortcomings of social judgment. Englewood Cliffs, NJ: Prentice-Hall. Osgood, C. E. (1962~. An alternative to war and surrender. Urbana, IL: University of Illinois Press. Osgood, C. E. (1966~. Perspective in foreign policy (2nd em.. Palo Alto, CA: Pacific Books. Oskamp, S. (1971~. Effects of programmed strategies on cooperation in the prisoner's dilemma and other mixed- motive games. Journal of Conflict Resolution, 10, 221-226. Oskamp, S., and Periman, D. (1966~. Effects of friendship and disliking on cooperation in a mixed-motive game. Journal of Conflict Resolution, 10, 221-226. Perin, C. T. (1943~. A quantitative investigation of the delay-of-reinforcement gradient. Journal of Experimental Psychology, 32, 37-51. Pi~isuk, M., Skoinick, P., and Overstreet, E. (1968~.

85 Psychology, 8, 160-165. Tedeschi, J. T., Aranoff, D., Gahagan, J., and Hiester, D. (1968~. The partial reinforcement effect and the prisoner's dilemma. Journal of Social Psychology, 75, 209 215. Tornatzky, L., and Geiwitz, P. J. (1968~. The effects of threat and attraction in interpersonal bargaining. Psychonomic Science, 13, 125-126. Touval, S., and Zartman, I. W. (Eds.~. 1985. The man in the middle: International mediation in theory and practice. Boulder, CO: Westview. Wall, J. A., Jr. (1977~. Intergroup bargaining: Effects of opposing constitutent~s stance, opposing representative Is bargaining, and representatives' locus of control. Journal of Conflict Resolution, 21, 459-474. Wall, J. A., Jr. (1981). An investigation of reciprocity and reinforcement theories of bargaining behavior. Organizational Behavior and Human Performance, 27, 367- 385. Wilson, W. (1971). Reciprocation and other techniques for inducing cooperation in the prisoner's dilemma game. Journal of Conflict Resolution, Is, 167-196. Wrightsman, L. S., Jr., O'Connor, J., and Baker, N. J. (Eds.). (1972). Cooperation and competition: Readings in mixed-motive -tames. Belmont, CA: Brooks/Cole.

84 Predicting cooperation from the two sexes in a conflict situation. Journal of Personality and Social Psychology, _, 35-43. Pruitt, D. G. (1968). Reciprocity and credit building in a laboratory dyed. Journal of Personality and Social Psychology/ 8/ 143-147 Pruitt, D. G. (1970~. Motivational processes in the decomposed prisoner's dilemma game. Journal of Personality and Social Psychology, 14/ 227-238. Pruitt' D. G. tl981). Negotiation behavior. New York: Academic. Pruitt, D. G., and Rubin, J. Z. (1986~. Social conflict: Escalation stalemate, and settlement. New York: Random House. Rapoport, A., and Chammah, A. M. (1965~. Prisoner's dilemma: A study in conflict and cooperation. Ann Arbor, MI: University of Michigan Press. Rosenfeld (1967~. Skinner, B. F. (1938). The behavior of organisms. New York: Appleton-Century-Crofts. Stapleton, R. E.' Nacci' P./ and Tedeschi/ J. T. (1973). Interpersonal attraction and the reciprocation of benefits. Journal of Personality and Social Psychology, 28' 199-205. Swingle' P. G. J and Gillis/ J. S. (1968). Effects of the emotional relationship between protagonists in the prisoner Is dilemma. Journal of Personality and Social

86 FOOTNOTES 1The letters in the matrix on the right stand for the following words (Rapoport and Chammah, 1965): R = reward, S = sticker Is payoff, T = temptation' P = penalty. 2More precisely, reciprocity is a sensible strategy for party to adopt in the PD because of the following four features of the game (party is viewed as the column player and other as the row player in this analysis): (a) P > S; hence other cannot be expected to be content with a CD' outcome, which means that T' is only briefly attainable if at all. (b) R' > P' and R' > S'. (c) T > R; hence R' is not attainable by simply choosing C' . (d) R > P; hence other will prefer CC' over DD'. These features define a PD for other, since they imply T > R > P > S. They define a broader set of games for party (all games in which R' > PI and Rt > S') J including PD/ chicken, standard instrumental conditioning paradigms, and several others. 3Earlier notions that punishment is largely ineffective at suppressing responses have been shown to be erroneous (Hulse et al., 1980~.

87 4This-model is related to the goal/expectation model developed by Pruitt and Kimmel (1980). The latter model holds that people cooperate when they develop (a) the goal of achieving mutual cooperation and (b) trust that the other party will reciprocate cooperation. The link is that the goal of achieving mutual cooperation can be viewed as an outcome of valuing the opponent's cooperation (condition 3) and believing that the opponent cannot be exploited (condition 2~. Trust is condition 1. 5It is interesting to note that both actor and target were in the same hurting stalemate in these periods. Hence, it can be argued that a hurting stalemate encourages resort to unilateral initiatives as well as reciprocity to these initiatives. This suggests that hostile relationships are particularly likely to improve when both parties are experiencing a hurting stalemate at the same time.

Next: Culture and Group Cohesion »
Enhancing Human Performance: Background Papers, Social Processes Get This Book
×
Buy Paperback | $50.00
MyNAP members save 10% online.
Login or Register to save!
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!