For Spring 2018: Totally online!


SYP 5105-01           FALL 2017



  • Both cognitive and reinforcement approaches are still important, but are less obviously cited in the groups research literature.
  • Processes such as social facilitation (audience and co-actor effects) are important, but don't meet the criteria of "a group".
  • Cohort membership (people possessing at least one common characteristic) probably contribute to diverse social identities; the connection to a group is not always explicit.
  • Aggregates refer to physical proximity; they have been implicated in fads, panics, riots.
    • Social facilitation can relate to learning and performance in many situations; cohort studies are often used in market research.
  • Key issues in groups address self and other definition as a member, and normative expectations about behavior.
  • Common and interdependent goals are critical to define a collectivity as a group.
  • While direct face-to-face interaction and group cohesion are often studied, they are somewhat less important--examine online groups, for example.
  • Groups have a composition (e.g., by age), a structure (e.g., social roles), and internal processes (e.g., communication patterns).
  • Groups vary in importance to the individual, e.g. primary versus secondary groups.
  • Online groups almost certainly have effects; their study is still iyoung, but growing.



Midsemester, we are now into our second major section of the Theories of Social Psychology course. The first section had a heavily structural determinism  focus, or how a variety of social factors influence the individual. Throughout this first section, we often compared more cognitive approaches with more reinforcement approaches. In the Socialization section, we also examined more developmental approaches, with a more explicitly biological, evolutionary or maturational perspective, contrasting these with more reinforcement approaches.

The second major portion of our course addresses group and interaction issues. Although it does not have the depth of our EPLS Group Processes course (offered next semester), this second section introduces major ideas about groups. We will also examine basic issues in communication and interaction. Cognitive and reinforcement approaches are not as obvious in this section, but they are nonetheless present.

Consider the concept of "social role," or a social position associated with a set of position-specific behaviors and with position-specific requirements and privileges. Social roles emerge from the patterned interactions associated with a specialized division of labor in a particular group. Clearly, a strong cognitive component is present. Individuals must:

A certain level of social competence is assumed and required to enact a successful role performance. Individuals must also know how to segment or segregate roles because behavior appropriate for one type of role may be wildly inappropriate for another. For example, it is typically expected and desirable for family members to kiss and hug one another, but you would be careful about this role enactment at the office. Role performers must be able to separate different types of scripts for the same role to enact with different complementary roles. Thus, at work you might criticize a role subordinate, but not a role superior. A lot depends on an individual being able to learn role components, and then figuring out the appropriate times to perform them.

Similarly, a certain level of social competence, or emotional intelligence  ("EI" or "EQ"), is required to receive and dechipher verbal and nonverbal messages. A person with high EI/EQ is sensitive to the messages conveyed in a glance, a shrug, or vocal pitch. They can quickly assimilate information, attach it to a particular social schema, then select the best script to respond.

At the same time, reinforcement is important. Picture a child watching parents, teachers, nurses, or characters in television situation comedies enact adult roles. Certain of their behaviors will be rewarded--mommy gets a raise at work, everybody tells daddy how terrific his barbeque is, the romantic lover on TV is rewarded with a kiss, and the villain goes to jail for robbing a bank. Supporting the notion of cognition in learning, the child understands that s/he doesn't yet have a job, "man" the barbeque grill, or kiss the wrong person (most of the time, anyway). The behavior is stored, to be trotted out and performed under role-appropriate conditions. Role appropriateness is one of the sometimes submerged dimensions that can determine the difference between learning and performance that Albert Bandura has described.

Extending modeling approaches, we can see how filmed portrayals can contribute to role behavior. Many people have three major points of reference for family behavior: their own family of orientation (the one you grew up with), their own family of procreation (that's the family you create)--and fictitious portrayals in plays, television, movies and so forth. Filmed portrayals not only inform you about appropriate ways for you to act, they also inform you about how others could be expected to act in response in similar situations. You may base your decision about courses of action on how fictitious characters respond to a lead character's action. Considering that some families appear to engage in continual mutual insults, that The Three Stooges is still present in reruns, that children treat adults like retarded cretins on the Disney Channel, and the worst excesses against family members appear to be excusable with a "sorry, Babe," clearly the American public is being treated to a terrible series of scripts for familial interaction. Perhaps we are laughing or sighing with relief because our own families look so good in comparison.

Depending on circumstances, life in other families may be partly submerged and not totally observable, even to relatives, making media portrayals more salient. An early national book on domestic violence was titled Behind Closed Doors for good reason.

Both reinforcement and cognition play important roles in exchange processes. In exchanges,  people decide about changing or remaining in a particular situation (e.g., a job or a marriage) or about whom they will interact with in part based upon the rewards associated with the particular situation, social role or the interactional partner. In virtually all cases, when rewards are below those expected by the individual's assessment of the costs and efforts that s/he has put into the situation, the individual becomes angry. What happens in some cases, however, is that when rewards are disproportionately "too high," individuals are not satisfied either. Equity experiments demonstrated that when rewards are too high, if people are able, they often increase either the quantity or the quality of their inputs (instead of laughing all the way to the bank). Some report feeling guilty. These illustrations of cognitive consistency in a reward-punishment situation should remind us of the research on oversufficient justification, in which "too much" reward can negatively impact on internal motivation. Cognitive consistency appears active in interactional reinforcement situations too.

So, just because cognitive and reinforcement approaches are not quite as obvious in the second section of the course doesn't mean that they are unimportant. I will try to make the connections more explicit as new material is introduced.


Social Facilitation has become a "catchall" phrase that refers to the effects of having other people* in one's immediate environment. It is one of the oldest research traditions in Social Psychology, dating back to 1898.The work of Robert Zajonc (pronounced "Zy-ounce"), among others, indicated that the  mere presence of other people is physically arousing. Borrowing from several learning theories (especially that of the late Clark Hull), many behavioral scientists assert that moderate amounts of physical arousal enhance performance. Many studies have found that moderate amounts of anxiety increase the drive or motivational component of the learning/performance equation, thus enhancing the performance of the best-learned response. Too little arousal, and the organism is not motivated, too much and it can be "flooded" by anxiety.

Thus, we have studied:

(1) audience effects, where the actor performs and the others observe. This is perhaps the most common use of the term "social facilitation".

(2) There are also co-action effects. Co-actors individually perform the same actions at the same time, typically in the same place. Think: group of students in a room writing an exam. Co-action experiments were often used in earlier studies of social facilitation.

In both audience effects and co-action, there is no real division of labor, no coordinated cooperation, in short, neither really resembles what is usually thought of as a group. Sometimes co-actors are also in aggregates, however (see below). The arousal effects of having others present may interfere with learning although they can benefit performance. For example, as a consequence of his series of studies, Zajonc advised students to study alone (rather than in the library or in a study hall) but to take an exam in a group situation, where the arousal effects of co-actors will facilitate the performance of well-learned responses perhaps by increasing motivation or drive.

*SOCIAL FACILITATION WITH 'ROACHES! (our local Palmetto Bugs the size of mice)
(yes, you read that right)

It doesn't have to just be other people. My favorite example of Robert Zajonc's work on social facilitation involved audience effects with our old local friends: cockroaches. I was privileged to observe first-hand several trials in his studies with Palmetto Bugs. You will be interested to know that the researchers had great difficulties keeping our Palmetto Bugs alive in their travels to chilly Michigan. Zajonc and his students trained cockroaches to "run" toward a light in a clear lucite t-maze. Those of us who have been fortunate enough to share a residence with 'roaches can testify that their normal mode is to AVOID light. An "audience box" was located beneath the t-maze where the 'roach marathoners could see, and presumably smell, their colleagues. 'Roaches ran their t-mazes faster and more accurately with an audience of their fellow 'roaches than when no audience was present.


More seriously, we can apply some of the social facilitation material to anger management situations such as domestic violence. We know that domestic violence runs in families, and that abused children more often become abusers themselves as adults. This is most often interpreted as a case of modeling or imitation learning. However, issues in reward are clearly questionable; many individuals who grew up in homes where spousal violence (and often child abuse) occurred rarely approve and it is unclear exactly what is rewarding about such a situation to anyone. It is not even clear if such a family of orientation situation contributes to disinhibition in family situations later. But we can take our application of social facilitation a step further:

Recall from above that much family behavior is typically private behavior, taking place "behind closed doors." Our widest repertoire of family behavior and our best-learned responses almost certainly come from what each of us repeatedly observed in our own families. Because of privacy, we know little about the intimate lives of other families and the media present glossy, over dramatized, or ludicrously unbelievable versions of family life. When disagreements and conflicts erupt in families, emotions run high and everyone is in a state of physiological arousal. That is exactly the time when a well-learned response is most likely to be performed. If that response is violent, its probability of occurrence is enhanced. Thus, I suspect that social facilitation adds a necessary explanation for the transmission of family violence across generations to modeling approaches. While modeling may explain how the behavior is acquired, social facilitation may explain under what circumstances it is performed.

One solution is to expand work on anger management currently introduced into the schools as early as elementary school. Children learn to self-monitor to identify angry feelings and how to express or handle them in ways that do not seriously harm others. I would suggest going a step or two further (and some teachers do), introducing role-playing and rehearsal. Simply reading about alternative behaviors is likely to be insufficient because this will keep such behaviors 'way down in the response habit hierarchy. Role-playing and rehearsal can provide the practice to raise more peacable behaviors in the response repertoire.


Members of a cohort share at least one common characteristic. Demographers usually reserve the term for people who have a common experience at a singular point in time, such as a birth cohort ("generation") or students who entered Florida State University as freshmen in the Fall of 2017. But other examples include the  "soccer mom" or "angry white male" vote, i.e., collections of individuals with a common characteristic. Members of a cohort may never meet one another, may never even gather in the same place, and generally do not consider themselves "group members."  In fact, members of a cohort may be quite diverse with respect to other characteristics (are all soccer moms the same?) because these other characteristics or experiences cross-cut what they have in common.However, cohort members may self identify with their cohort ("I'm a Generation Xer" or "I'm the hockey mom"). It is likely that our memories of Septemer 11, 2001 or the economic "meltdown" of 2008 will in part set us aside from future generations just as memories of the Great Depression and World War Two have defined many people who are now senior citizens. Self-identification with a cohort can influence behavior through collective identification. Politicians and market researchers are well aware of cohorts, and use them to target segments of the population for advertising and other persuasion campaigns. Cohort experiences may also make it easier for people to get to know each other when they first meet, by providing common experiences.

In addition in the United States, political party is now occupying an important self social identity, which it did not do 50+ years ago (when survey respondents in the original Campbell et al The American Voter often said "I vote for the man [sic] not the party). Recently I found that political party identification is the leading predictor of 2010 perceptions of scientist unity on global warming, more important than educational variables, and more important than it was in 2006.

Aggregates are typically collections of individuals who are physically proximate at the same time. These could be an aggregate in the Stone Building elevator or a crowd at an FSU football game. Examples of aggregate effects include:

Audience effects (e.g., cheering on a performer)
Crowd effects
Co-action effects (crowds and co-action both can be physically arousing)
Fads and/or "crazes" (e.g., "flash mobs")

Historical studies of riots and revolutions show that crowd behavior can be the spark that topples governments or creates a night of terror. There is at least the suggestion that rape, which is often a group perpetrated event, may be an example of crowd behavior. Studies of these events often reference the arousing effects of others, the apparent normlessness of crowds, and the "confrontational nature" of the events (i.e., spontaneous, unplanned, and immediate).

While crowd effects or fads may appear spontaneous and "unthinking," social comparison factors are likely to be a factor in people adopting crowd behaviors or fads. The more similar that others are to us, probably the more likely it is that behaviors are to be adopted. These are yet more examples of the importance of cognitive consistency factors in guiding our behavior.

Plus, a crowd or a cohort may turn into a group if the right circumstances are in place. If you are with a crowd of strangers and the Stone elevator begins to play games, you will probably turn into a group quickly. Members of a crowd coalesce into revolutionaries. The attacks of September 11 brought many New Yorkers togethers, and symbols of American patriotism (flags, bumper stickers, lapel pins) spontaneously and quickly became common.

But it takes more than just a bunch of people in the same place to be a group. Aggregates and cohorts do influence behavior . They are phenomena clearly worthy of study--but they aren't groups, although they are sometimes confused with groups. We need to separate these concepts carefully so that we don't misapply them.


Here are some keys to defining a group. Some of the most important factors are relatively intangible:

(1) Self-definition as a group member

(2) Other-definition as a group member

Without self-definition, we may need surveillance to keep the person in the situation (think prison, some students, or another "captive audience"). Without self-definition, the individual will not be motivated to act in the group's best interests, and, in fact, may think it silly or futile to even conform with group norms. Individuals who are compelled to be in a group without self-definition may even act to undermine the group.

Without other-definition, we may perceive the person as a fraud and expel them. At the least, we may create many distance mechanisms (such as ignoring the person, "rewriting history" to exclude the individual from the group, or literally "cutting them out of the picture".) Someone who is "in, but not of" the group is not seen as entitled to group privileges or group rewards.

(3) Direct interaction (usual but not necessary!)

When many people think of a group, the first idea that comes to mind is face-to-face interaction. Yet, many groups, particularly secondary or tertiary groups, survive without it. People may, in fact, have face-to-face interaction on a daily basis, yet be an aggregate rather than a group ("the elevator crowd," for example). Online groups may have delayed interaction, which is written and somewhat stilted, depending on server speed and other factors, yet individuals may strongly identify with membership. For example, I have been part of the "AAPOR-net" listserv for over fifteen years and have never directly met many of the members. Yet after literally thousands of emails, I "feel" that I "know" many persons on it.

(4) Cohesiveness

Cohesion has so many definitions that they would fill a book by themselves. The ancient Cartwright and Zander definition was "the total forces attracting members to a group" (and presumably keeping them there). Beatrice Lott defined it as the number of ties among [individual] group members. Lott's has been the most widely used definition since the mid1960s. I define cohesiveness as a sense or spirit of group unity collectively held by the membership, i.e., as a group property rather than some mathematical function of individual scores.

(5) "Normative expectations", i.e., rules regulating member behavior

Such norms provide predictable, stable group interaction. Groups that provide norms and evaluate behavior are often called normative reference groups or exerting normative influence. It is the absence of such normative expectations that can produce the erratic, often frightening behavior we see in crowds. Under such circumstances, no one is quite sure what to expect.

However, some behavior that appears idiosyncratic or spontaneous to outsiders may, in fact, be normative for that particular group. Several years ago, when I began an ethnographic study of religious congregations, I was startled the first time someone next to me rose and began speaking in tongues. I soon learned, however, that this individual enacted this behavior during practically every service and another congregant would then translate for her into English. In another congregation, at one part of the service, an elderly gentleman would rise, toss his cane into the aisle and begin dancing in the aisle. This happened practically every Sunday at the same part of the service. Both behaviors were clearly normative, definitely not spontaneous, for those congregations and a designated person enacted them in each case at a predictable point in the service.

Probably the most critical defining characteristic of a group is:

(6) Interdependent and common goals that require coordination among the membership

Such goals can create a division of labor and specialized roles which may then later cause role conflict.

If I had to select just one defining characteristic of a group, interdependence would be it. Interdependent goals require coordination among the membership to achieve them and they require the coordinated efforts of at least two people working together. Interdependent goals imply a division of labor so that each member has a unique and specific task. It is the interdependence and interlocking roles that are either created by the group (in more informal groups) or that exist as a pre-existing structure (in formal groups) that melds people together into a unity.

Thus, a group is a collection of individuals who both define themselves and are defined by others as members of the collectivity, who share common goals which require a coordinated interdependent effort to achieve.

Why compare groups, aggregates, cohorts and social facilitation?
Because (1) they are often mistaken for one another but (2) the processes and consequences in each are distinct. Some of the contradictory findings in group research occurs because of the failure to consistently define the term "group".
If you want to study groups, don't study aggregates.


Groups exist in environments that provide inputs and feedback, and receive outputs (open system model).

Groups have:

a composition (e.g., size, age or gender),

a structure (the roles and rules--norms--that hold a group together),

and internal processes (e.g., leadership or conformity).

Membership groups, which you belong to, have a stronger influence over you than nonmembership groups. Remember how often roles and identities from membership groups were mentioned in the Twenty Statements Test!

Primary groups are the most central to you. They are usually smaller, interaction is usually direct, typically face to face, and members are strongly interconnected, such as in a family. Friendship groups, some work groups, and often military groups will serve as primary groups for the individual. Because primary groups are so important to the individual, they tend to have the most influence. Socialization at all ages occurs within primary groups.

Most of us belong to a plethora of other groups: work groups, religious groups, political groups, hobby groups, neighborhood groups, fraternal organizations, and so forth. These secondary, or even tertiary groups, are important to us, even if they are not as central to us as primary groups. Interaction tends to be more infrequent and/or indirect, perhaps only rarely face-to-face.

Informal groups dissolve when all the original members leave. Friendship groups often are informal groups because members may move away, create new roles that conflict with group behaviors  (parenthood?) or positive ties may dissolve. (If people stay in one place friendship groups can become relatively permanent.)

Formal groups survive at least one complete membership turnover. Because of this survival factor, formal groups often have more resources and advantages than informal groups. For example, formal groups more often have a structured and even pre-existing division of labor and recruit members to fill specialized roles. They often have a written history or colorful symbols, such as flags. Of course, informal groups often exist within formal ones, such as the friendship groups that develop in a work setting.

The status that a group holds in the larger society can be an attractive feature that induces a cohesive group.  Most societies rank groups from higher to lower, often along more than one dimension. And, within a particular institution (such as religious denominations), groups will be ranked again. The dimensions of status rankings vary enormously: some dimensions stress wealth or knowledge or physical attractiveness, others stress piety. Even if the members can't stand one another, if the group is prestigious enough, people will want to join and remain .

Cohesive groups, with a strong sense or self-perception of group unity, are typically described in positive ways: as more productive and more satisfying. What is really going on is that cohesive groups can better set and enforce norms and align productivity with group norms. BUT these group norms may or may not coincide with "productivity" as this is defined by the outside world or by the larger organization. Thus, cohesive groups can actually be "less productive" if, for example, the group goals are socializing rather than producing material goods in an organization.

Conformity processes are more concentrated and effective in cohesive groups. This means that such groups can be cruel to deviants, see the world in polarized terms ("us versus them"), and make very bad decisions--none of which typically gets defined as "productive." "Groupthink" is one example of cohesion plus group insularity carried to extremes (plus overly directive leadership).

We will examine cohesion, its causes and consequences more in a later guide on persuasion, conformity, and compliance. Cohesion is probably the single most commonly referenced and studied property of groups. And, perhaps, the most misunderstood.


One modern innovation is "groups" that only communicate via the Internet. These may be members of various chat rooms, listservs associated with professional societies, a group of Facebook "friends," or students who are all enrolled in a Distance Learning course. As you know, "distance learning" has become more and more common.

Experimentally, such "groups" are actually an old story. A common experimental paradigm to assess the effects of a group is to create a bogus group that is accessed only through computers. In this way, feedback and other information from "group members" to the naive participant can be precisely arranged, manipulated, and standardized by the experimenters. Many studies show that "membership" in and feedback from these contrived groups influence behavior, attitudes, and identity. Thus, we would expect that "real" online groups also can influence their members, probably even more so, because interaction is more naturalistic and group members typically have characteristics in common.

But are online groups really groups in the sense of the defining characteristics listed above?

For obvious reasons, all research on online issues is relatively new. However, here are some factors to mention:

However: Earlier studies found that online time eats heavily into "TV time" although later studies do suggest that Internet activity may slightly lessen interaction with relatives and friends. Thus initially among Internet innovators, one potentially solitary activity was exchanged for another rather than previously social beings "morphing" into recluses as has happened to some degree as the Internet has become a "mass market" activity. And, while it is true that families or friends may watch TV together, online time is often more active than passively watching a TV screen.

Stay tuned for further developments!

OPTIONAL:See the Sage journals, Social Science Computer Review, or Bulletin of Science, Technology & Society for more issues with online behavior and society (these journals can be accessed from FSU computers and...sometimes...the library proxy).

This page was built with Netscape Composer.
Susan Carol Losh October 30 2017