Flash mobs: Social influence in the 21st century

The author would like to thank Dr. Anthony Pratkanis for significant and helpful contributions to this article.

Abstract

In a flash mob, strangers come together in a public place, perform an unusual behavior, and randomly disperse. This article describes how the flash mob and a follow‐up discussion can be used as a class activity to learn about many social influence principles, including conformity, ostracism, and deindividuation. Data collected from two class sessions indicate that students felt the activity was useful, worthwhile, and should be repeated in future courses. This activity would be appropriate in a variety of courses, including introductory psychology, social psychology, organizational behavior, social influence and persuasion, and group dynamics.

The author would like to thank Dr. Anthony Pratkanis for significant and helpful contributions to this article.

The “flash mob” is a 21st‐century, technology‐driven phenomenon. A flash mob is a group of people, usually unknown to each other, who assemble suddenly in a public place, perform some unusual or notable activity according to predetermined instructions, and then quickly disperse. Flash mob instructions are usually distributed via digital communication networks, such as an Internet posting, e‐mail, or cell phones. In one example of a flash mob, “mobbers” gathered around a particular rug at Macy's department store in New York, saying that they were shopping for a “love rug” for their commune (Shmeuli, 2003). In another, mobbers brought umbrellas and walked two city blocks singing songs from the movie Mary Poppins (Bear, 2003).

In the classroom, examples of flash mobs can be used to demonstrate a wide range of social influence concepts, such as authority, compliance, conformity, and group norms. Rather than simply reading and talking about flash mobs, however, I suggest that the best use of this concept is to create a flash mob, with the students as mobbers. In this paper, I will describe how I have incorporated a flash mob activity in my social psychology classes, and how I guide students in relating their behavior to different concepts. Finally, I present student feedback from two flash mob activities.

PREPARING THE ACTIVITY

The guidelines for true flash mob activities are simple: the behavior must be silly, harmless, and peaceful, and should be short in duration (about 5 minutes). One behavior that I employ is walking around a public fountain, flapping and clucking like a chicken. Another is singing a popular song while walking in a circle in a public park. Other no‐equipment behaviors include sitting on a wall making bird calls, applauding in a synchronized pattern, and forming a conga line.

Some flash mobs incorporate the use of cell phones, books, or newspapers in their activity. For example, one flash mob activity consisted of walking on a particular stretch of public sidewalk, taking out a cell phone, and yelling, “Yes, yes!” (Shmeuli, 2003). In another flash mob, mobbers each brought a book and swapped with fellow mobbers for 10 minutes (Gardiner, 2003). A flash mob activity in Austin included slow‐motion sword fighting with rolled‐up newspapers (Bear, 2003).

Some flash mobs, such as the Macy's rug mob, have interfered with businesses; others, such as pillow‐fight flash mobs (pillowfightclub.org, 2006), include a level of violence that goes beyond the purpose of the flash mob. In both of these cases, the ethics of the behavior are in question. To determine an ethical flash mob activity, and one that is in keeping with the spirit of flash mobs, consider some of the general rules of a “good flash mob,” as described by Johnson (2003): Avoid places of business; don't get political; do not disrupt traffic; surprise the public, don't alarm them; and keep your shirt on.

Beyond the ethics of the flash mob, it is also important to consider the ethics for any classroom activity. Svinicki (1994) describes six principles of ethics for professors: autonomy, nonmaleficence, beneficence, justice, fidelity, and acting consciously. For the flash mob activity, the question of autonomy is of particular importance. Ethically, it is vital that students feel free to decline participation; additionally, as the students will discover in the post‐activity discussion, the perception of autonomy has a significant influence on some social influence phenomena.

The location of a flash mob activity should be carefully considered. Nonmaleficence is important to consider; flash mobs should not impact others, such as business owners or the public, in a negative way. Therefore, no place of business should be considered for a flash mob. Instead, public areas, such as parks, sidewalks, or common areas, would be appropriate. I typically use a public fountain near our campus or a common area on campus.

The instructions that are communicated to the students should include a very specific time in which to begin and end the activity; synchronizing watches is advisable. It should also be clear that students can participate in or observe the activity, or not participate at all, with no negative consequences, and that the professor will not be participating. Students should approach the location of the activity separately, and disperse in different directions at the conclusion of the activity.

There are many ways in which instructions can be communicated to the students. In the flash mob activities that I have used, I tell the students about the day, time, place, and behavior during the two class sessions prior to the activity. Alternatively, instructions can be shared electronically, as information about a true flash mob would be. For example, professors who use an online source, such as WEBCT or Blackboard, can post instructions. Using e‐mail is an option for some professors.

PERFORMING THE ACTIVITY

The flash mobs that I have organized were either at a nearby public fountain (a popular picnic spot) or on campus. I arranged the activity to begin 15 minutes after the scheduled class time, which allowed sufficient time for participants to walk to the location.

During the activity, the students should be completely autonomous. The professor should be an observer and should not give any guidance, especially regarding when to begin or end the activity. I typically have a location where I can watch the activity but not be seen by the participants.

The activity usually lasts between 2 and 5 minutes. At the conclusion of the activity, students converge back in the classroom for a post‐activity discussion. I begin the discussion 15 minutes after the conclusion of the activity.

POST‐ACTIVITY DISCUSSION

A post‐activity guided discussion can promote the understanding of the link between what the students did, felt, and observed and the related social influence concepts. Often, student responses to particular questions are excellent illustrations of topics, and reflect research findings. Some of the questions that I have used in the past, typical student responses, and suggested topics are detailed below (table 1 summarizes these discussion questions, related topics, and suggested readings).

Table 1. Suggested discussion questions, related topics, and suggested reading

How did you feel when you were waiting for the time to begin? Why?

Ostracism

• Eisenberger, N. I., Lieberman, M. D., & Williams, K. D. (2003). Does rejection hurt? An fMRI study of social exclusion. Science, 302, 290–292.

• Williams, K. D., & Gerber, J. (2004). Ostracism: The making of the ignored and excluded mind. Interaction Studies: Social Behaviour and Communication in Biological and Artificial Systems, 6, 359–374.

• Zadro, L., Williams, K. D., & Richardson, R. (2004). How low can you go? Ostracism by a computer is sufficient to lower self‐reported levels of belonging, control, self‐esteem, and meaningful existence. Journal of Experimental Social Psychology, 40, 560–567.

Did you wait alone or with other classmates?

Affiliation needs

• Schachter, S. (1959). The psychology of affiliation: Experimental studies of the sources of gregariousness. Stanford, CA: Stanford University Press.

• Bell, P. A. (1978). Affective state, attraction, and affiliation: Misery loves happy company, too. Personality and Social Psychology Bulletin, 4, 616–619.

What norms were in place during this activity?

Norms

• Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58, 1015–1026.

Did you participate in the activity? Why or why not?

Informational and normative social influence

• Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational social influences upon individual judgment. Journal of Abnormal and Social Psychology, 51, 629–636.

• Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgments. In H. Guietzkow (Ed.), Groups, leadership, and men. Pittsburgh, PA: Carnegie Press.

• Sherif, M. (1936). The psychology of social norms. New York: Harper.

Public compliance and private acceptance

• Festinger, L. (1953). An analysis of compliant behavior. In M. Sherif & M. O. Wilson (Eds.), Group relations at the crossroads. Oxford, UK: Harper.

Was it easy to perform the behavior?

Crowd behavior, emergent norms

• Aguirre, B. E., Quarantelli, E. L., & Mendoza, J. L. (1988). The collective behavior of fads: The characteristics, effects, and career of streaking. American Sociological Review, 53, 569–584.

Deindividuation

• Zimbardo, P. G. (1970). The human choice: Individuation, reason, and order versus deindividuation, impulse, and chaos. In W. J. Arnold & D. Levine (Eds.), Nebraska Symposium on Motivation (Vol. 17). Lincoln, NE: University of Nebraska Press.

Would participating in the flash mob be different if the professor was present?

Obedience to authority

• Milgram, S. (1963). Behavioral study of obedience. Journal of Abnormal and Social Psychology, 67, 371–378.

• Milgram, S. (1965). Some conditions of obedience and disobedience to authority. Human Relations, 18, 57–76.

What do you suppose the observers were thinking about you?

Correspondent inference theory

• Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process in person perception. Advances in Experimental Psychology, 2, 219–266.

Covariation theory

• Kelley, H. H. (1967). Attribution in social psychology. Nebraska Symposium on Motivation, 15, 192–238.

Fundamental attribution error

• Ross, L., & Nisbett, R. E. (1991). The person and the situation: Perspectives of social psychology. New York: McGraw‐Hill.

How might you increase the likelihood that observers would participate?

Minority influence

• Moscovici, S., & Faucheaux, C. (1972). Social influence, conformity bias, and the study of active minorities. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 6). New York: Academic Press.

• Moscovici, S., & Mugny, G. (1983). Minority influence II: Minority influence. In C. Nemeth (Ed.), Social psychology: Classic and contemporary integrations. Chicago: Rand McNally.

Was this activity worthwhile, even though you may have felt uncomfortable?

Cognitive dissonance

• Festinger, L. (1957). A theory of cognitive dissonance. Oxford, UK: Row, Peterson.

What are some of the ethical concerns for this activity?

Ethics of psychologists

• American Psychological Association (2002). Ethical principles of psychologists and code of conduct. American Psychologist, 57(12), 1060–1073.

How did you feel when you were waiting for the time to begin? Why?

Students typically report feeling high levels of anxiety while they were waiting to begin. They attributed their anxiety to the expectation of being perceived as deviant, and the fear that they may experience rejection from society; they understand how painful social ostracism can be. Their discomfort can be discussed as it relates to the research by Williams and colleagues (e.g., Eisenberger, Lieberman, & Williams, 2003; Williams & Gerber, 2004; Zadro, Williams, & Richardson, 2004), which reveals that ostracism is psychologically and physically painful. For example, in the Eisenberger et al. (2003) study, participants played a game of computerized keep‐away. When they were excluded from the game, their social pain was comparable to physical pain, as measured by an fMRI. After participating in a flash mob—or not participating, because of high levels of anxiety—students better understand some of the negative consequences of social ostracism.

Some students are surprised that they felt anxiety at the possibility that they would be ostracized by strangers (e.g., observers of the flash mob); it is easy for them to recognize that ostracism from their peer groups is unpleasant, but they do not realize the power of the threat of ostracism by society in general. Their experience is consistent with the Zadro et al. (2004) study. In this study, participants again played the computerized keep‐away game, but were given different information about their opponents. Even when participants were told that they were playing against a pre‐programmed computer, they reported feeling the pain of ostracism.

Did you wait alone or with other classmates?

Students typically tend to wait with fellow students. Their motivation to be with others who will also be participating in the flash mob is akin to research by Schachter (1959) and Bell (1978). Schachter and Bell found that when people are experiencing anxiety in expectation of a future event, they prefer to affiliate with others who will be sharing a similar fate or who were experiencing similar emotions.

What norms were in place during the activity?

After participating in the activity, norms were easily described. Some of the students discussed the norms of the location, such as, “People sit and eat lunch” and “You're not supposed to look at anyone else.” Others pointed out the norms of the group: “You're supposed to cluck,” “Most of us were flapping our arms.”

By pointing out these specific, conflicting norms, students gain an understanding of Cialdini's focus theory of norms (Cialdini, Reno, & Kallgren, 1990). According to Cialdini, norms fall into two categories: descriptive, or what is being done, and injunctive, or what should be done. Whether the norms influence an individual's behavior depends—at least in part—on the situation, including the salience of the norms. In the flash mob, norms of both the public and the mobbers were made salient, and the students were aware of these conflicting norms.

Did you participate in the activity? Why?

Students who did participate in the activity most often respond to this question with some form of “I did it because everyone else was doing it!” Clearly, conformity is one of the most salient concepts in the flash mob activity. Conformity refers to the way people change their behaviors (or attitudes, or beliefs) in order to fit in with others. Conformity can be discussed within the context of informational and normative social influence (Deutsch & Gerard, 1955) or in terms of public compliance and private acceptance (Festinger, 1953).

Normative social influence occurs when people look to others to determine what social behavior is appropriate in a given situation. The classic example of normative social influence is Asch's (1951) line study (although information influence was also present as described in Footnote 6 of Deutsch & Gerard, 1955). In this study, participants were asked to indicate which of three lines matched a target line. When confederates gave obviously incorrect answers, many participants went along with the wrong answer, apparently in order to get along with the group. Informational social influence, on the other hand, occurs when people look to others for “correct” answers. The classic example of informational social influence is Sherif's (1936) autokinetic study. The autokinetic effect occurs when a stationary point of light in a darkened room appears to move. In Sherif's research, participants were asked to form judgments about how far the point of light moved. When participants stated their judgments aloud, their responses converged; apparently, participants used others as sources of accurate information. This informational influence is readily seen when the participant left the group but still followed the norms set by the group in making judgments. Students who participate in the flash mob typically conclude that their behavior was more likely a result of normative social influence rather than informational social influence.

Students who participated in the flash mob are likely to have experienced public compliance and not private acceptance (Festinger, 1953). They changed their behavior to go along with the group, even though it is unlikely that they would perform the behavior on their own. This type of conformity is akin to the public compliance. In comparison, most students agree that they will not perform the flash mob behavior on their own in the future. In other words, they have not experienced private acceptance: although they changed their behavior for the activity, they have not changed their beliefs about the behavior.

Was it easy to perform the behavior?

Most students report that starting the behavior was difficult, but it was very easy once everyone was participating. This response can lead to a discussion on crowd behavior and deindividuation. The emergent norm perspective of crowd behavior proposes that a group forms its own norms that the members then conform to (Aguire, Quarantelli, & Mendoza, 1988). In comparison, deindividuation occurs when people in a group experience a decreased sense of individuality and self‐awareness, and are no longer controlled by societal norms (Zimbardo, 1970). Students can hypothesize which of these two theories better exemplifies their experience in the flash mob.

Would participating in the flash mob be different if the professor was present?

Students occasionally comment that the activity “would have been better” if I had told them when to begin the activity and when to end it. Students recognize that if I had told them that the activity was required, or if I had told them when to begin and end the behavior, they would have felt much less anxiety; it would be easy to attribute their behavior to an authority figure. This suggestion relates directly to a discussion of obedience: performing an activity as a result of a command from an authority figure. Milgram (1963, 1965) performed a series of studies in which a participant was commanded to administer electric shocks to a confederate. He found that obedience was inversely related to the proximity of the authority figure. When the experimenter was a few feet away from the participant, obedience was surprisingly high; however, when the authority figure gave orders via telephone or via tape recording, participants were much less likely to obey commands to shock the confederate.

What do you suppose the observers were thinking about you?

This question leads to a discussion of attribution theories, and whether observers attributed the students' behavior to dispositional or situational causes. An example of a dispositional attribution is “They probably thought we were stupid.” An example of a situational attribution is “They probably thought we were doing something for class.” Two popular attribution theories are Jones and Davis's correspondent inference theory (1965) and Kelley's covariation theory (1967). According to the correspondent inference theory (Jones & Davis, 1965), observers consider several factors before inferring whether a person's behavior is influenced more by the situation or the person's disposition: the social desirability of the behavior, whether the behavior was freely chosen, the intended consequences of the behavior, and previous knowledge of the person. Kelley's covariation theory (1967) proposes that observers make dispositional or situational attributions about a person based on the distinctiveness, consensus, and consistency of the behavior.

The discussion of dispositional and situational attributions can also lead to the topic of the fundamental attribution error (Ross & Nisbett, 1991): the tendency to attribute a person's behavior to dispositional causes, and underestimating the influence of situational factors. Many students appreciate that observers of the flash mob might assume that the mobbers are brave, radical, or rebellious, while the mobbers attribute their own behavior to the situation.

How might you increase the likelihood that observers would participate?

Students often have interesting ideas regarding this question. For example, one student proposed that the behavior should be singing a song that everyone knows; another proposed that the behavior should continue longer. This raised the topic of minority influence (Moscovici & Faucheaux, 1972; Moscovici & Mugny, 1983). It is possible for one dissenter to change the behavior of many within a group. To be effective, the dissenter needs to be consistent and confident. While the flash mob is not an individual, it is possible that the participants might inspire observers to join the group—if the group members are consistent and confident.

Was this activity worthwhile, even though you may have felt uncomfortable?

In post‐activity discussions, students often report having felt high anxiety and embarrassment. In post‐activity questionnaires, students rate this activity as worthwhile, useful, and enjoyable, and should be repeated in future classes (see Assessment, below). It is possible that the value of the activity outweighs the anxiety and embarrassment the students experienced; it is also possible that the students experience cognitive dissonance when asked to evaluate the activity. Cognitive dissonance (Festinger, 1957) occurs when a person experiences inconsistency between an attitude and a behavior. Some of the factors that increase the likelihood of cognitive dissonance include: the person must perceive that the behavior was a choice (Linder, Cooper, & Jones, 1967), the behavior was effortful (Wicklund, Cooper, & Linder, 1967), and the behavior had foreseeable, negative consequences (Goethals, Cooper, & Naficy, 1979). Therefore, students who participate in the flash mob are more likely to experience cognitive dissonance if their participation was completely voluntary, their embarrassment or anxiety was high, and they knew beforehand that they were going to be embarrassed or anxious. If they experienced cognitive dissonance, their ratings of the activity should be high.

What are some of the ethical concerns for this activity?

Students are quick to point out some of the ethical concerns in flash mobs as mentioned earlier (e.g., interfering with businesses, keeping participation optional). This question is a natural segue to a discussion of the American Psychological Association (APA) ethical guidelines (2002). These guidelines are in place to direct the ethics of all psychologists, whether they are working in a classroom, in a laboratory, or in a clinic. There are five general principles: “Beneficence and Nonmaleficence,” “Fidelity and Responsibility,” “Integrity,” “Justice,” and “Respect for People's Rights and Dignity” (pp. 3–4). Students can appreciate how each of these principles influences their participation in the flash mob.

POSSIBLE FOLLOW‐UP QUESTIONS

To encourage students to continue thoughtful deliberation of the flash mob and social influence, it may be appropriate to assign follow‐up activities. A follow‐up activity that I have found interesting is an investigation of the history of flash mobs and the current flash mob activity around the world.

Another follow‐up activity is to have students address questions related to their activity. Some of the questions that I have used include:

	When you're out with your friends and they do something you normally wouldn't do, do you go along with them? How does this compare to the flash mob?
	Would it be possible to form a flash mob website or group in our town? Would it be advisable?
	Why do people participate in flash mobs—if their professors aren't telling them to?
	How might you use a flash mob politically?
	Would people in collectivist cultures be more or less likely than people in individualist cultures to participate in a flash mob?

FLASH MOBS VS NORM‐BREAKING ACTIVITIES

Professors of social psychology often include an activity wherein students break or imagine breaking a social norm (cf. Pines & Maslach, 1993; Schneider, 2002). The flash mob is similar in that it includes a norm‐breaking component; however, the flash mob activity improves on the typical norm‐breaking activity because it incorporates so many social influence concepts. For example, the flash mob includes group dynamics, which can be used to demonstrate deindividuation, affiliation motivation, and the power of the in‐group. The flash mob has the added attraction of being technologically driven and contemporary, which brings social influence principles into the student's world. Finally, because the class participates (or observes) as a group, the activity serves as a bond among the students, which contributes to group cohesiveness.

EVALUATION OF THE FLASH MOB ACTIVITY

I have used the flash mob activity in several class sessions. In order to determine the effectiveness of the activity, I collected student feedback in closed‐ and open‐ended questions from two of these class sessions.

Participants

Two sets of data were collected. Of the 39 students enrolled in class for Group 1, 25 (64%) participated in the activity as mobbers and 5 (13%) observed the activity. Six students remained in the classroom, and three were absent on that day. Of the 44 students enrolled in the class for Group 2, 25 (57%) participated in the activity as mobbers and 6 (14%) observed. Eight students remained in the classroom, and five were absent on that day. One participant (Group 2) had limited mobility and used a wheelchair, and had no difficulty participating.

Measures

After the activity and the post‐activity discussion, participants completed a questionnaire with closed‐ and open‐ended items. Students responded on a 1 (strongly agree) to 7 (strongly disagree) scale (see table 2). There were eight closed‐ended items. Three were used to evaluate how much students felt they learned about conformity, social norms, and social influence. Four items were used to determine whether this activity was a waste of time, useful, enjoyable, and whether it should be used in the future. One item was used to compare their learning to the amount of effort they expended. Responses to this item ranged from 1 (more learning than effort) to 7 (more effort than learning).

Table 2. Closed‐ended evaluation items, means, and SD, on a 7‐point strongly agree to strongly disagree scale

Questionnaire item	Mean	SD
I learned a lot about conformity in this activity	2.50	1.24
I learned a lot about social norms in this activity.	2.34	1.10
This activity was a waste of time.	6.22	1.17
I learned more about social influence than if I had not participated in this activity.	3.35	1.67
The instructor should use this exercise again in future quarters.	1.54	1.13
This activity was useful.	1.72	1.05
This activity was enjoyable.	2.08	1.53
Comparing how much I learned to how much effort I put into this activity, I experienced∗	3.18	1.59
∗Responses to this item ranged from 1 (more learning than effort) to 7 (more effort than learning); the remaining items were rated on a 1 (strongly agree) to 7 (strongly disagree) scale.

Four open‐ended questions were included. Students were asked what they learned, what was the best part of the exercise, what was the worst part of the exercise, and how the exercise should be changed in the future.

Results

In the closed‐ended items, students tended to agree that they learned a lot about conformity and social norms in this activity (M = 2.50 and 2.34, respectively). They indicated that their learning was about equal to their effort (M = 3.18). They found the activity useful (M = 1.72) and enjoyable (M = 2.08), disagreed that this activity was a waste of time (M = 6.22), and agreed that the exercise should be used in future quarters (M = 1.54).

The first open‐ended question asked students what they learned in this exercise. Almost every student wrote about the dynamics of social influence and norms; some students used the appropriate terminology, while others wrote about their feelings while participating. Common responses included, “I learned that I was very concerned with breaking the norms and how people would react to my actions” and “That to do something outside the norm, people need to have some sort of ally, whether the other person participates or not.” One student wrote, “I also learned you can motivate people to join you in breaking social norms.”

The second and third questions asked students about the best and worst parts of the exercise. Most students wrote that the activity was fun and/or funny, and that they enjoyed watching the reactions of the public. Several students wrote about how they enjoyed being a part of the group, and three wrote that their favorite part was when the activity was over. The most commonly cited “worst part” of the activity was anxiety: five students wrote about the anxiety they felt on the walk over to the destination, and seven wrote about the anxiety they felt as they began performing the behavior.

The final question asked how the exercise should be changed in the future. A total of 19 students left this item blank or indicated that no changes should be made. Feasible suggestions include performing the activity in a more populated area, singing a familiar song and see if others join in, and having a student confederate who walks in the opposite direction compared to the other students.

Discussion

Overall, students tended to enjoy this activity. More importantly, in their participation in the post‐activity discussion as well as their responses on the survey, they demonstrated that they better understood many social influence principles.

This activity continued its educational value beyond the section on social influence. For example, later in both quarters, when the topic of discussion was attraction, students realized that their friendly feelings toward classmates might have been influenced by the misattribution of emotion (Dutton & Aron, 1974); that is, they might have attributed the physiological arousal caused by their imminent embarrassment to an increased attraction toward each other. When the topic was the self, students clearly understood the concept of the “spotlight effect” (Gilovich & Savitsky, 1999): they each thought that the observers were watching them perform the activity, when in reality the observers may not have watched any one of the mobbers individually.

There are many benefits of using this modified flash mob as a source of active learning. For example, students do not need to have a lot of background information in psychology or in social psychology in order to learn from this activity; at the same time, a student who has a more extensive background in social influence principles can still benefit from the dynamics of the activity. Additionally, the modified flash mob is effective for students who participate and for students who do not participate: even the students who declined to participate gained a new understanding regarding the power of social norms, and how social norms impacted their choice of behavior.

CONCLUSION

Flash mobs are intriguing social phenomena. Organizers create a social group out of strangers, and the mobbers are both group members and deviants at the same time. These overlapping roles afford opportunities to discuss many terms and concepts that are relevant to social influence, including one of the most important: across time, across cultures, and across nations, humans are social beings.

Notes

The author would like to thank Dr. Anthony Pratkanis for significant and helpful contributions to this article.