Psi man mascot


One of the first scholarly studies of friend­ship came from Theodore Newcomb in The Acquaintance Process (1961). Newcomb identified four factors that affect the probability of making an acquaintance.

1. Proximity. We are more likely to get to know somebody with whom we have regular contact.

2. Reciprocity. We like people who like us.

3. Similarity. We like people who share our values and beliefs.

4. Complementarity. We are attracted to people whose skills and abilities are complementary to our own. Complementary means different but compatible and mutually beneficial, like people with different skills who work together for a common purpose.

What four factors influencing friendship did Newcomb identify?

Of the four, the most prosaic (common­place, automatic) proved to be the most powerful: proximity. Obviously we cannot get to like someone if we never see that person. On the other hand, if we see someone frequently, it is convenient and helpful to establish a friendly relationship.

Which factor seems to be the most important predictor of friendship?

Accidental proximity is called propin­quity, literally "the coincidence of being near." In several studies, propinquity was a powerful predictor of friendship.

Priest and Sawyer (1967) studied stu­dents who moved into a newly construct­ed college dorm. After eight months, roommates were named as friends twice as often as floormates, floormates twice as often as students on other floors.

What did Segal do with police trainees?

Segal (1974) showed effects of propin­quity by assigning police trainees to seats in alphabetical order. At the end of the term he asked them to name their three closest friends. The result was an almost perfect correlation between friendship choice and seating order.

Could first impressions guide people to friendships? Numerous experiments show that people make a positive or negative evaluation of stimuli almost immediately (1/20 second) especially when the stimuli are faces. Would a positive response predict later friendship?

Chen, Whalen, Freeman, Taylor, and Heatherton (2015) decided to find out. They targeted 27 newly arrived grad­uate students from China. Previous researchers noticed "new arrivals vary greatly in their friendship formation patterns."

Some made friends outside their native culture. Others stayed with their in-group, associating mostly with other Chinese students.

After volunteering for the experiment, the newly arrived graduate students were scanned with an fMRI machine while being shown pictures of faces. First a face was flashed for 17 ms (milli­seconds), not enough time for conscious processing.

It could be a Chinese or Caucasian face and it could be expressing one of three emotions: happy, fearful, or surprised. Next came a neutral face, lasting 183 msec. It could be seen consciously.

All participants saw all six combina­tions, multiple times. Four of the 26 participants, questioned after scanning, knew they had seen some emotional faces. They were excluded from further analysis. The other 22 had not noticed any emotional expressions.

The biggest brain response was from the happy expressions. The only area of the brain showing major differences in response was the ventral striatum, an area associated with reinforcement and positive emotions.

Six months later, those who reacted immediately and positively to Chinese happy expressions, but not Caucasian, had mostly Chinese friends. Those who reacted to both groups equally had more balanced in-and-out-group friendship patterns.

The authors cautioned that this data was correlational. The greater likelihood of a positive response to non-Chinese faces could be a cause or an effect of greater likelihood of forming friendships with non-Chinese. There could be additionsl factors at work like greater prior exposure to Western faces while still in China.

What was discovered in a study of Chinese graduate students newly arrived in the U.S.? What are possible explanations?

We cannot conclude that initial positive responses caused friendships to form. However, the research did show that initial positive reactions, too fast to be consciously controlled, predicted the likelihood of later friendships.

Reciprocity, in studies of friendship, is liking someone who likes you. Reciprocity was manipulated in a well-known study by Aronson and Linder (1965).

Subjects "accidentally overheard" another subject, actually a confederate of the researcher, expressing liking or disliking of the subject. Then the subject was asked to fill out a questionnaire that expressed liking or disliking of the confederate.

As you might expect, a confederate who had made positive comments about the subject was liked; one who made negative comments was disliked. That, by itself, showed reciprocity.

A third group heard the confederate start by making bad comments and end by making good comments. These subjects, who thought they "won him over," gave the confederate the highest ratings of all.

What did Aronson and Linder discover in the "overheard comment" study?

Reciprocal friendships (relationships in which both parties like and admire each other) are important to the happiness and adjustment of school children. Berndt (2002) showed that children with high-quality friendships, characterized by low levels of conflict and rivalry, had more social success with all peers, not just close friends.

Maunder and Monks (2015) also found that having a best friend predicted generally positive social adjustment in children aged 9 to 11. "Children with a reciprocal best friend fared better in terms of friendship quality, school and peer identification and general self-worth compared to children without a best friend."

These are all correlational studies, so they cannot tell us about cause and effect. We do not know if having a best friend reassures a child and causes greater ease in other social relation­ships, or if the same children excel in relations both with friends and other acquaintances at school.

Newcomb, in his intensive study of students in a special rooming house, found similarity was a powerful factor in accounting for friendships. Similar roommates were much more likely to end up being friends.

Byrne (1961) showed the same thing in paper-and-pencil fashion. He gave subjects a description of another person and asked how much they thought they would like the person. The more closely the "other person" resembled the participant, the more the participant expected to like the other person.

How did Byrne demonstrate the power of similarity, using a paper and pencil test?

Similarity is not necessarily beneficial in the business world. Gompers, Mukharly­amov, and Xuan (2016) found that "venture capitalists who share the same ethnic, educational, or career back­ground are more likely to syndicate with each other" but this "reduces the probability of investment success."

Sometimes it might be preferable to align with people who have complemen­tary strengths, rather than similarities. In work teams and business partnerships, complementary skills allow a work team to deal with a wider range of situations.

Wagner (1975) found that camp coun­selors liked each other better if they had complementary needs. For example, one might need to be in control, while the other preferred to be directed.

In marriage literature, it is harder to find evidence of the benefits of complemen­tarity. Meyer and Pepper (1977) found similarity was a better predictor of marital happiness than complemen­tarity, in couples married up to five years.

White and Hatcher (1984) reviewed the literature and drew the same conclu­sion. "The few clinical studies available indicate that similarity is associated with marital success and is less asso­ciated with marital instability and divorce."

Definitions of similarity and comple­mentarity surely influence the results in such research. There are limitless numbers of dimensions on which two people can be compared. Long-married couples testify that comple­mentarity is important in some dimensions, similarity in others.

For example, similarity might be important when measuring sense of humor or religious outlook, or agree­ment on acceptable sexual practices. Complementary may be more important in areas like skills, temperament, or choice of domestic responsibilities.

How might the choice of dimensions affect research on similarity and complementarity?

Research shows a consistent advantage for good-looking people in establishing friendships. Beauty leads to better first impressions. Physically attractive people are also assumed to be superior in intelligence, health, wealth, and personality.

Dion, Berscheid, and Walster (1972) published a famous article about this, titled "What is beautiful is good." The results are found when subjects are asked to give quick impressions of people they do not know and probably reflects an evolutionary bias. Good health correlates with good looks, other things being equal.

Walster, Aronson, Abrahams, and Rottman (1966) randomly assigned couples at a "computer dance" (with partners chosen at random by a computer). They were expecting to find that people would be happiest when paired with "a partner of approximately his own social desirability."

The hypothesis was not confirmed. "Regardless of S's own attractiveness, by far the largest determinant of how much his partner was liked, how much he wanted to date the partner again, and how often he actually asked the partner out was simply how attractive the partner was."

Responding to criticisms that the computer dance was not typical of real dating situations, because it lacked risk, Berscheid, Dion, Walster, and Walster (1971) tried again. This time they varied the risk of rejection, greatly, by setting up special conditions where participants were warned of high levels of rejection and warned that no other partners would be provided if the first choice rejected an invitation.

Participants were asked to pick between 6 pictures "with whom they would actually like to out with on a date." This time the matching hypothesis was confirmed. However, it was confirmed at all levels of risk, which was unexpected.

In a more recent test, Taylor, Fiore, Mendelsohn, and Cheshire (2011) analyzed data from a popular online dating site to see if it confirmed the matching hypothesis. They found "evidence for matching based on self-worth, physical attractiveness, and popularity, but to different degrees and not always at the same stage of the dating process."

What was the "matching hypothesis" and how did research support it?

What are people looking for in a lasting friendship? Anderson (1968) presented students with a list of 555 adjectives and asked them to check off the ones that would be most important in a friendship.

What are characteristics of an ideal friend?

Of the top-rated eight adjectives, six formed a cluster that might be called sincerity or authenticity. An ideal friend is sincere, honest, loyal, truthful, trustworthy, and dependable. A person with these characteristics is predictable in a good way: steadfast and true.


Anderson, N. H. (1968). Likableness ratings of 555 personality-trait words. Journal of Personality and Social Psychology, 9, 272-279.

Aronson, E. & Linder, D. (1965) Gain and loss of esteem as determinants of interpersonal attractiveness. Journal of Experimental Social Psychology, 1, 156-171.

Berndt, T. J. (2002) Friendship quality and social development. Current Directions in Psychological Science, 11, 7-10. doi:10.1111/1467-8721.00157

Berscheid, E., Dion, K., 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.

Byrne, D. (1961) Interpersonal attraction and attitude similarity. Journal of Abnormal and Social Psychology, 62, 713-715.

Chen, P-H. A., Whalen, P. J., Freeman, J. B., Taylor, J. M., & Heatheron, T. F., (2015) Brain reward activity to masked in-group smiling faces predicts friendship development. Social Psychological and Personality Science, 6, 415-421. doi:10.1177/1948550614566093

Dion, K., Berscheid, E., & Walster, E. (1972) What is beautiful is good. Journal of Personality and Social Psychology, 24, 285-290.

Gompers, P. A., Mukharlyamov, V., & Xuan, Y. (2016) The cost of friendship. Journal of Financial Economics, 119, 626-644.

Maunder, R. & Monks, C. (2015) Are you my friend too? The role of reciprocal friendship: links to friendship quality, peer and school identification and general self-worth. Paper presented to: British Psychological Society, Psychology of Education Section Annual Conference, Liverpool, 22-24 October 2015. (Unpublished). Retrieved from:

Meyer, J. P. & Pepper, S. (1977) Need compatibility and marital adjustment in young married couples. Journal of Personality and Social Psychology, 35, 331-342.

Newcomb, T. M. (1961) The acquaintance process. New York: Holt, Rinehart & Winston.

Priest, R. F., Sawyer, J. (1967) Proximity and peership: Bases of balance in interpersonal attraction." American Journal of Sociology, 72,

Segal, M. W. Alphabet and attraction: An unobtrusive measure of the effect of propinquity in a field setting. Journal of Personality and Social Psychology, 30, 654-657.

Taylor, L. S., Fiore, A. T., Mendelsohn, G. A., & Cheshire, C. (2011) Out of my league: A real-world test of the matching hypothesis. Personality and Social Psychology Bulletin, 37, 942-954. doi:10.1177/0146167211409947

Wagner, R. V. (1975) Complementary needs, role expectations, interpersonal attraction, and the stability of working relationships. Journal of Personality and Social Psychology, 32, 116-124.

Walster, E., Aronson, V., Abrahams, D., & Rottman, L. (1966) Importance of physical attractiveness in dating behavior. Journal of Personality and Social Psychology, 4, 508-516.

Watson, P. (1975, December). Boss cows and sleepy, sexy sheep. Psychology Today, p.93.

White, S. G. & Hatcher, C. (1984) Couple complementarity and similarity: A review of the literature. American Journal Of Family Therapy, 12, 15-25.

Write to Dr. Dewey at

Don't see what you need? Psych Web has over 1,000 pages, so it may be elsewhere on the site. Do a site-specific Google search using the box below.