Now that we have established the importance to people of forming social relationships, we can address how people choose others with whom to develop such relationships. Note that for the most part we’ll be focusing on social relationships, which include friendships as well as both opposite-sex and same-sex romantic connections, although the research on this topic has tended to focus primarily on heterosexual attraction and relationships.

One simple determinant of social relationships is known as the propinquity effect, propinquity meaning closeness in space. The original idea was that you can’t form a relationship with someone unless you meet them, and the closer you are physically to someone else, the more likely you are to meet and therefore form a relationship with him or her.

With the proliferation of Internet technology in many contemporary cultures, this is not so true anymore. Facebook, Twitter, blogs, message boards, and other apps make it more and more possible to form relationships with people we rarely, if ever, actually meet. This includes people from around the globe and in cyberspace with whom we share an interest. Before the Internet, people sometimes developed relationships with “pen pals,” friends known only through an exchange of letters. So for a long time, physical propinquity has not been necessary for developing a social relationship; nevertheless, it’s less important, at least for casual relationships, than ever before. At the same time, face-to-face social interactions are more important and beneficial to mental health and life satisfaction than cyber socializing (Green et al., 2005; Kraut et al., 1998; Latané et al., 1995).

512

With these caveats in mind, it’s still worth considering the role of proximity. The first empirical breakthrough in examining this factor was a study by Festinger and colleagues (1950). They interviewed residents in a new apartment complex. As in most apartment complexes, the apartment manager had placed residents in their particular apartments in an essentially random fashion. Festinger and colleagues saw this as a natural experiment that gave them the opportunity to study whether and how proximity influences friendship formation. They found that the physical location of one’s apartment within the complex had a large impact on who made friends with whom and on how many friendships one formed within the complex. For example, people were nearly twice as likely to form a friendship with the person in the next-door apartment as they were to form a friendship with the person who lived two doors away. In a similar study of classroom friendships, students were more likely to get to know a classmate who sat next to them than those who sat just a few seats away (Byrne, 1961). Many of us like to think we choose our friends carefully on the basis of their unique attributes, but these findings suggest that some of our choices are based largely on who happens to be next door or seated next to us in a class.

Festinger and his colleagues also found that among people who lived in first-floor apartments, those next to stairwells made more friends on the second floor, presumably because they were more likely to run into second-floor residents. Moreover, in a neighboring housing project, people whose houses faced the street made fewer friends in the area than people whose houses faced a common courtyard. This set of findings suggests that sometimes loneliness may develop not because of an individual’s attributes but simply because of the physical isolation inherent in where an individual lives or works.

A number of explanations have been offered for the surprising impact of physical location. One is based on familiarity. As you may recall from chapter 8, evidence supporting the mere exposure effect shows that we tend to like novel stimuli better the more we are exposed to them (Zajonc, 1968). The unfamiliar makes most people initially wary or even anxious. As a stimulus becomes more familiar, people feel more at ease, and that positive feeling becomes associated with the stimulus. In the case of interpersonal interactions, the more you see a new person, the more positive you are likely to feel about that person. Because you’ll see your next-door neighbor more often than the person a few doors down, the mere exposure effect works to the benefit of that nearest neighbor.

An Oregon State speech professor demonstrated the benefits of familiarity by having someone come to his class every day in a big black bag (Rubin, 1973). At first, the students were wary of the human black bag and sat as far away as possible. However, by the end of the course, the students were very friendly with the black bag and treated it like a beloved class mascot. Moreland and Beach (1992) assessed this phenomenon more systematically. They had typical college-age female confederates of similar appearance attend a class 0, 5, 10, or 15 times during a semester—without ever speaking to anyone in the class. At the end of the semester, they showed photos of each confederate to the students in the class and asked them to evaluate each confederate on dimensions such as honesty, popularity, and likeability. The more the confederate attended the class, the more positively she was rated. So we tend to like people who are more familiar.

513

More recently, Reis and colleagues (2011) systematically varied how many chats, ranging from 1 to 8, participants had with a same-sex fellow student. Generally, the more chats, the more the participants liked the other person. Additional measures in the studies showed that more conversation led to more liking because it increased comfort and satisfaction with the other person and a sense that the other person was responsive to them. These findings fit the mere exposure notion that by increasing familiarity, proximity leads to greater comfort and attraction. It also suggests another explanation for the proximity-liking effect: that the more interactions we have with people close by, the more attentive to us they seem. And it may not be just our perceptions of responsiveness that are influenced by repeated interaction. Chances are that the more we see people, the more they actually do respond to us and our needs.

A related explanation for the proximity effect is that, in general, casual interactions with other people are mildly pleasant. You exchange greetings, perhaps commiserate about the weather, or discuss the fortunes of the local sports teams. The more mildly pleasant conversations you have, the more positive feelings you will associate with the person with whom you are conversing.

Of course, there are important exceptions to the proximity effect—the roommates who grate on each other, the annoying neighbor, the cultural groups that share a border and can’t get along. Indeed, research has found that the mere exposure effect does not occur if the stimulus is initially disliked or is associated with negative outcomes (e.g., Brickman et al., 1972; Swap, 1977). If repeated exposure to others due to close proximity only reminds people of ways that they differ from one another, they can like others less rather than more (Norton et al., 2007). This brings us to our first broad theory of attraction, often known as the reward model of liking.

The core idea of the reward model of liking is simple: We like people we associate with positive feelings and dislike people we associate with negative feelings (e.g., Byrne & Clore, 1970; Lott & Lott, 1974). It is basically a classical conditioning model of liking, similar to the influences on persuasion we talked about in chapter 8. Recall that advertisers often will pair their product with an uplifting jingle or cute scene, trying to foster a positive association with the product. In the reward model, a new person begins as a relatively neutral stimulus. If exposure to the person is temporally linked to a second stimulus you already like, the positive feelings evoked by the second stimulus start to become evoked by the person. Conversely, if the second stimulus evokes negative feelings in you, some of those negative feelings start to become linked to the person. This raises the question, What are these second stimuli that influence our liking for others?

When we think of why we like someone, we usually talk about that person’s attributes, and that is certainly part of the total picture. But the reward model suggests that we could come to like (or dislike) someone not because of any attribute they have or behavior they engage in, but simply because they happened to be around when we were feeling good (or bad). The idea seems to fit the rather unfair practice of ancient Roman rulers who sometimes would kill messengers who brought bad news (prompting the old expression “Don’t kill the messenger”).

One early test of this idea had participants simply sit in a room for 45 minutes and fill out some questionnaires. One of the questionnaires described a stranger’s attitudes on various issues. The experimenter varied the temperature in the room so that it was either comfortable or unpleasantly hot. Participants were then asked to indicate how they felt about the stranger. As the reward model predicts, participants liked the stranger better if the room was comfortable than if it was not (Griffitt, 1970). Researchers have used a variety of other methods to make the same point. For example, one study had participants overhear bad or good news on a radio broadcast and then evaluate a stranger (Veitch & Griffitt, 1976). They liked the stranger better if the radio broadcast good rather than bad news. So sometimes, we may like or dislike someone because they just happen to be there when something pleasant or unpleasant happens to occur.

514

We probably overlook these sorts of influences most of the time in thinking about why we like or dislike someone, because we focus on that person’s attributes. On the other hand, most of us also probably have some intuition that feelings created by things we don’t cause can rub off on how people feel about us. For example, if you are going on a first date with someone—let’s say dinner and a movie—you probably hope the weather is nice, the food at the restaurant is good, and the movie is enjoyable, even though you have little or no control over those outcomes. Research suggests that intuition is correct: All else being equal, you will be better liked, and also like the other person better, if it’s a lovely evening, the food is delicious, and the movie is delightful. Of course, we often use this intuition to plan a date or meeting to our advantage. On their first date, one of your authors took his salsa-loving future wife to a salsa festival in the Southwest. Suffice it to say, the evening went quite well.

Having acknowledged the role of situational factors, we next ask, What attributes of people themselves evoke the positive feelings that increase our attraction to them? Most of the research on interpersonal attraction addresses this question in one way or another.

Transference

First, some attributes may evoke positive feelings because we associate them with people we like or positive experiences we had in the past. For example, Collins and Read (1990) found that people are often drawn to romantic partners who have a caregiving style similar to that of their opposite-sex parent. More generally, in a finding consistent with the Freudian concept of transference (Freud, 1912/1958), Susan Andersen and colleagues discovered that if a newly encountered individual resembles a significant other in your life whom you like or dislike, you will carry over those feelings to the new person (Andersen & Baum, 1994; Andersen et al., 1996). Often these associations can even be fairly subtle, such as sharing the same birthday or wearing a similar style of eyeglasses.

Culturally Valued Attributes

As cultural animals, we also are drawn to people who have talents or have achieved things that our culture values (e.g., Fletcher et al., 2000). Celebrities are an extreme example. Many if not most people are fascinated by celebrities because the culture has deemed them to be of great value—that’s what makes them celebrities. Even if the more cynical and reserved of us happen to run into Will Smith or Jennifer Lawrence at a bar in Los Angeles, we would be star struck. If we were lucky enough to have a conversation with them, undoubtedly we would tell virtually everyone we know about it. Why? Think back to BIRGing—basking in reflected glory (Cialdini et al., 1976). Being connected to, or simply being near, another person with culturally valued attributes can enhance our own self-esteem.

515

This form of attraction extends not only to the extremes of celebrity but also to any attributes highly valued. In the United States, these include such things as wealth, beauty, musical or athletic talent, and so forth. Acquiring a so-called trophy wife gives a sense of self-worth (a trophy) to an older, rich man who has a beautiful young wife by his side. In a less clichéd example, one of your authors had a roommate in college who was brilliantly talented in music. He read symphony scores for fun and on hearing a song once could play it on the piano. Your author recalls taking great pride in his friend’s ability, even though he himself entirely lacked that ability! This connects with our discussion of Tesser’s (1988) self-evaluation model in chapter 6: When someone else is talented in a domain that you don’t claim for yourself, it is easy to identify with him or her and gain self-worth from doing so.

Personality Traits

Next, let’s consider personality traits. Across a wide range of studies, people generally report preferring certain traits in their partners and friends. Not surprisingly, these include friendliness, honesty, warmth, kindness, intelligence, a good sense of humor, emotional stability, reliability, ambition, openness, and extraversion (e.g., Sprecher & Regan, 2002). The culture promotes the valuing of these traits. In addition, it’s easy to imagine how people with any of these desirable traits could evoke positive feelings in us, and, by the same token, how people with the opposite traits might evoke negative feelings in us. Which traits people desire in others depends to some extent on their relationship with the other person. For example, people report that agreeableness and emotional stability are more valued in a close friend than in a study group partner, whereas intelligence is reported to be more valued for a study group partner than for a close friend (Cottrell et al., 2007).

Although research on the traits we like in an ideal partner is valuable, we should note that the vast majority of this work assesses what traits people report or think they like, not those that they actually like (Eastwick et al., 2013). This is because the best test of what attributes people actually like would be very difficult, if not impossible, to conduct. You would have to assign people randomly and have them get to know other people who systematically vary in these traits to really sort out what attributes people like and dislike in others.

We make this point because the cultural worldview we learn as children teaches us that kindness, intelligence, honesty, and so forth are good qualities. Thus, our self-reports are likely to mimic these teachings. In fact, when we examine what traits people in different cultures claim they like, the traits they value mirror aspects of the culture. This was observed in a large Internet survey of participants from 53 nations. When they were asked to rate the importance of various attributes of a romantic partner, participants from modern, individualistic nations rated humor and kindness higher, and dependability and intelligence lower, than did participants from more traditional collectivistic nations (Lippa, 2007). Perhaps individuals in different cultures really do find different attributes attractive. But we can also interpret these findings to show that culture influences what qualities people think that they should like in others.

One method that tries to tease apart reported and actual preferences is to create a situation that resembles speed dating. In one set of studies, Eastwick and colleagues (Eastwick, Eagly et al., 2011; Eastwick, Finkel et al., 2011) had male and female participants sit at individual tables as a parade of potential partners (or “dates”) rotated through, spending about 4 minutes with each participant. Later, the participants were asked whom they would have liked to see again. The researchers found that the traits the participants reported caring about in a prospective romantic partner failed to predict how interested they were in others who had or did not have those qualities when they met them face to face.

516

Of course, speed-dating studies have their own limitations. One is that they don’t tell us which traits people like in others whom they have known for a long time. In fact, one study of middle-aged participants currently in relationships showed that the match between their reported ideal traits in a romantic partner and their perceptions of their current partners was in fact a good predictor of how positively they viewed their partner and the relationship (Eastwick, Finkel et al., 2011).

Once we acknowledge these caveats, Eastwick and colleagues’ findings may help explain why finding a date or friend online might not work (Finkel et al., 2012). You get all sorts of information about what the person is like before you even meet. It seems very handy. But this research suggests what we think we like doesn’t necessarily predict whether things will go smoothly when we meet that person face-to-face and learn more about each other. For example, you may think you’d like someone who is extraverted, the life of the party, and open to experiences, but on meeting such a person face to face, you may find him or her to be obnoxious or exhausting.

Beyond having talents, achievements, and desirable traits, people also can be attractive to us because they help to satisfy our psychological needs. One is the need to sustain faith in a worldview that gives meaning to life. Another is to maintain a strong sense of self-esteem. We like people who help validate these psychological resources. These include people who seem similar to us, who like us, and who flatter us. Let’s look closer at each.

Similarity in Attitudes

One of the strongest determinants of attraction is perceived similarity. As the old saying goes, “Birds of a feather flock together.” Similarity on several dimensions matters. People who become friends, lovers, and spouses tend to be similar in socioeconomic status, age, geographical location, ethnic identity, looks, and personality (Byrne et al., 1966; Caspi & Herbener, 1990; Hinsz, 1989). But particularly powerful is similarity in attitudes and overall worldview. Imagine Lana, a Republican Christian young woman who is against abortion and is a supporter of low taxes, a strong military, and gun rights. She likes country music and soft rock, skiing, tennis, action movies, designer clothes, fine wine, and Italian food. She meets Rosa and Erica at a party. Rosa is a Republican Christian who supports the same issues, who likes country music, tennis, action movies, fine clothes, and (OMG) Italian food. Erica is a Democratic agnostic who is prochoice and antigun, and likes hip-hop and metal, basketball, independent art movies, beer, and gluten-free vegan cuisine. Whom is Lana going to like better, and perhaps see as a potential new friend to hang out with?

Rosa, of course. She validates Lana’s beliefs about the way the world works, what good music sounds like, what makes for a good meal, and so forth. In having similar beliefs and preferences, she bolsters Lana’s worldview and self-esteem. Erica, in contrast, challenges the validity of Lana’s beliefs and preferences, threatening her worldview and self-esteem. In addition, Lana can imagine having good times shopping, eating, and going to concerts with Rosa; with Erica, not so much. Support for these intuitions comes from many studies showing that people with similar attitudes are more likely to be liked, become friends, and become romantic partners (Byrne, 1971; Griffitt & Veitch, 1974; Newcomb, 1956). In one early study, Newcomb (1956) examined attitudes of transfer students moving into a college dorm. When he tracked liking among those in the dorm, as more time passed, sharing similar attitudes became an increasingly stronger predictor of liking. Also, people who marry are likely to have more satisfying, longer-lasting marriages to the extent they perceive their attitudes to be similar (Bentler & Newcomb, 1978; Cattell & Nesselroade, 1967; Houts et al., 1996).

517

It is interesting to note that the causal arrow works both ways. Just as perceived similarity increases attraction, attraction increases perceived similarity. If we like someone, we also tend to assume he or she has similar attitudes (Miller & Marks, 1982). In addition, couples tend to think their attitudes are more similar than they actually are (Kenny & Acitelli, 2001; Murray et al., 2002).

Perceived Versus Actual Similarity

Several studies show that what is important for attraction and relationship commitment is how much people perceive that they are similar to another, and not necessarily how similar they are from an objective point of view (Montoya et al., 2008). For example, people’s initial attraction in a speed-dating context, and their satisfaction in long-term relationships, are better predicted by perceived similarity than actual similarity (Dyrenforth et al., 2010; Tidwell et al., 2013).

Why is actual personality similarity not very predictive of attraction and relationship satisfaction, whereas perceived similarity is? Perhaps the answer lies in that old trope that “opposites attract.” This idea has some intuitive appeal. Shouldn’t someone who likes to make decisions get along with someone who doesn’t? Shouldn’t someone given to emotional ups and downs fit with someone very stable and consistent? In other words, shouldn’t people who have complementary qualities get along (Winch, 1958)? Of course, dissimilar people sometimes will hit it off as friends or romantic partners, but most evidence suggests this is more the exception rather than the rule.

However, a few studies show ways in which opposites may attract. One way is that highly masculine men tend to be attracted to highly feminine women (Orlofsky, 1982). In addition, Dryer and Horowitz (1997) found in two studies that after a brief interaction, female students high in dominance preferred a submissive partner, and females high in submissiveness preferred a dominant partner. So it appears that for the traits of dominance and submission, complementarity contributes to attraction. The same applies to fiscal habits: People who tend to scrimp and save often marry people who like to spend. Still, their different spending styles contribute to conflicts over finances, which reduce marital well-being (Rick et al., 2011). Other studies show that people partner up with a dissimilar other if they are looking for a short-term, low-commitment relationship, presumably because they find their differences to be novel and exciting (Amodio & Showers, 2005).

Tesser’s (1988) self-evaluation maintenance model (described in chapter 6) suggests another way that dissimilarity can help a relationship. If friends or partners are both strong in the same domains of abilities and accomplishments, it can lead to threatening social comparisons and friction. This implies that people will get along if they are good at different things—if, for example, one person is great at math and the other is a great writer. That way, each person can take pride in the other’s accomplishments rather than experiencing self-esteem threat from them (Pilkington et al., 1991). To test this idea, Tesser (1980) studied the biographies of famous male scientists. He found that the scientists had had better relationships with their sons if the sons went into a field different from their own.

518

Similarity in Perceptions

So far, we’ve been talking about similarity (perceived or actual) in personality traits, demographic characteristics, behavioral preferences, or attitudes. These are all features of the self that William James called the “Me.” But recall that James distinguished what the self is in terms of the content of who we are (the Me) from our subjective point of view on the world around us (the I). According to Liz Pinel and colleagues (Pinel et al., 2004), we “Me-share” with others when we feel that we are the same kind of person, but we “I-share” with others when we believe that our subjective experiences of the world are the same, even if our respective “Me’s” seem very different.

By way of example, your current author once joined a crew tasked with painting the interior of non-air-conditioned dorms in one blazingly hot summer in Dallas, Texas. Being an introverted liberal from New York, he developed a strong dislike for a boisterous, politically conservative crew member from Georgia. However, about two weeks into the job, this odd couple came to realize that they shared a rare love of Italian Opera. They spent the rest of their time as fast friends, joyfully singing tunes from Verdi’s Aida while painting, probably high on paint fumes.

I-sharing can indeed create strong feelings of connection. In fact, as the paint crew story suggested, it can even lead people to look past the objective differences that normally keep them separated. In one study (Pinel & Long, 2012), heterosexual men were shown nonsensical associations between famous people and objects (e.g., “If Oprah Winfrey were a plant, what would she be?”) and four possible responses (in this case: Dried flower, Venus fly trap, Kudzu, or Red rose). They were instructed to “go with their gut” and select the response that made sense to them. Immediately after making their response, they learned how two other participants—one straight, one gay—responded to the same association. In truth, there were no other participants. The feedback was set up to make it appear that one of the two others I-shared with the participant, picking the same responses nearly every time. Finally, they were asked which other participant they wanted to interact with. Attesting to the power of I-sharing to bolster liking, heterosexual men preferred to interact with a gay man with whom they I-shared rather than another heterosexual man with whom they did not I-share.

If You Like Me, I’ll Like You!

Other people, particularly those we respect or care about, can bolster or threaten our self-esteem by liking or rejecting us. As we’ve discussed, social rejection and ostracism can be extremely distressing experiences. On the other hand, it’s hard not to feel a little better about someone who feeds your need for self-esteem by approving of and warmly embracing you.

In chapter 13 we discussed the norm of reciprocity. This idea extends to liking. All else being equal, if you find out that someone else likes you (more than he or she likes others), it makes you more likely to like them too (Condon & Crano, 1988; Curtis & Miller, 1986; Eastwick et al., 2007). In one study, people’s reports of how they fell in love or formed a friendship with a person indicated that a key factor was realizing that the other person liked them (Aron et al., 1989). In fact, when compared with attitude similarity, being liked by another is a stronger initial factor in attraction (Condon & Crano, 1988; Curtis & Miller, 1986). One obvious explanation for the reciprocity of liking effect is based on self-esteem. Our self-esteem benefits when others like us (Becker, 1962; James, 1890; Leary et al., 1995), so we like those who enhance our self-esteem. Another plausible explanation is that we expect someone who likes us to treat us well, so the anticipation of rewards enhances our liking for that person (Montoya & Insko, 2008).

519

Flattery

It’s no surprise that we also like people who compliment us, even to the point of flattery. Studies show that the more nice things someone says about us, the more we like them (Gordon, 1996; Jones, 1990). The benefits of flattery even extend to computers. When participants received randomly generated positive performance evaluations from a computer, they liked the computer better—even if they knew the positive evaluations were randomly generated and not contingent on their performance (Fogg & Nass, 1997)!

Flattery doesn’t always work, however. If it is clear the flatterer has an ulterior motive, the compliments are not quite so effective (Gordon, 1996; Matsumura & Ohtsubo, 2012). Still, we generally prefer someone who says nice things about us (even if that person’s motives are suspect) to someone who doesn’t have anything nice to say at all (Drachman et al., 1978). This is because any compliment will still make us feel good (Chan & Sengupta, 2010). We usually are more motivated to embrace positive feedback than to question its validity (Jones, 1964, 1990; Vonk, 2002).

Aronson and Linder’s gain-loss theory (1965) adds an interesting complexity to our tendency to like those who flatter us. They noted that in some contexts, a compliment from a stranger, or someone you know but never has complimented you, is more potent than a compliment from a friend or spouse. They proposed that this is because we have a long history of being complimented by a friend or romantic partner, so one additional compliment is expected and doesn’t affect us much: You already know the person likes you. But the unexpected compliment from a stranger or an acquaintance who has not expressed liking for you before is more unexpected and fresher, and thus may have a bigger impact on your self-esteem and your liking for the complimenter. On the flip side of the same coin, a criticism will have more impact if it comes from a friend or romantic partner because they usually say positive things to you.

Aronson and Linder tested this hypothesis by having participants overhear a series of evaluations of them by a discussion partner who was a confederate of the experimenter. There were four patterns of evaluations: consistently positive, consistently negative, initially negative and then becoming positive (gain), and initially positive and becoming negative (loss). The participants liked the confederate best in the gain condition, second best in the consistently positive condition, even less in the consistently negative condition, and least in the loss condition. Positive judgments from someone who was initially negative and negative judgments from someone who was initially positive polarized people’s attitudes for the evaluator. Aronson and Linder noted that this phenomenon may put a long-term spouse at a disadvantage relative to new people the spouse meets: The spouse’s compliments will have less impact, and his or her criticisms will have more impact. Married 10 years, Rondae may compliment his wife, Renée, on how nice she looks to little apparent effect as they head to a party, and then observe that she is quite overtly pleased when someone at the party says the same thing.

What other aspects of the person giving us compliments affect how much we like him or her? One saying floating around in our culture is that “playing hard to get” can increase one’s attractiveness. Evidence doesn’t generally support that idea, but it does support the idea that people are more attractive if they seem hard for others to “get” (Eastwick et al., 2007; Wright & Contrada, 1986). Some people are very free with compliments and seem to like everybody. Other people are very discriminating, doling out compliments only to the lucky few. Compliments and liking increase attraction more if the person giving them out seems discriminating rather than giving out compliments freely or seeming to like virtually everyone.

520