Printer Friendly

The analysis of rule-governed behavior in social psychology.

For some time now, behavior analysts have been admonished to cooperate with researchers in other fields (e.g., Czubaroff, 1993; Glenn, 1993). If such efforts are to be successful, it is essential to find common grounds for mutually useful dialogues. In this article I outline a shared interest that can bring behavior analysts and social psychologists together. Because researchers in different fields belong to separate verbal communities, shared interests are often obscured by specialized concepts and procedures. But when we pierce the verbal thickets, we discover an excellent example of common ground: rule-governed behavior.

In recent years, behavior analysts have paid increasing attention to rule-governed behavior, and by now the analysis of rule-governed behavior is as useful as the study of contingency-shaped behavior has always been (e.g., Baum, 1995; Hayes, 1989; Malott, 1989; Zettle, 1990). On the following pages, I suggest that a considerable portion of experimental and theoretical work in contemporary mainstream social psychology is also concerned with rule-governed behavior, even though that term is not used. I intend to show that effective cooperation of behavior analysts and social psychologists is made possible by their shared interest in rule-governed behavior, and that together they contribute to a better understanding of the activities that constitute daily life.

Behavior Analysis and Rule-Governed Behavior

As Vaughan (1989) and Zettle (1990), among others, have pointed out, the study of rule-governed behavior became necessary when behavior analysts began to seriously confront questions about the role of 'thought" in the analysis of human behavior, and especially in the complex activities that are enmeshed in the intricacies of daily life. Rules are generally considered to be verbal descriptions of contingencies, or contingency-specifying stimuli that are part of a person's context. According to Baum (1995), "a rule is a verbal discriminative stimulus... [which] signals a short-term contingency... [and] is occasioned by a long-term contingency on the same behavior" (p. 3). Typical rules are admonitions and instructions; for example, a parent's statement: "Do your homework." The child's behavior is reinforced by parental approval, while the parent's behavior is reinforced by the child's compliance. As Baum suggests, however, the more important reinforcers for both participants in the interaction are temporally distant: for example, the child's education. Today, there seems to be general agreement that (a) "rules provide a behavioral mechanism for understanding how thoughts or self-talk might control goal-oriented behavior" (Malott, 1989, p. 274) and (b) "rule-governed behavior may be regarded as behavior under the control of verbal contingency-specifying stimuli" (Zettle, 1990, p. 44).

The most significant difference between a person's environment and rules is that past and present contextual events are the primary determinants of behavior, while rules serve as controlling variables for human activities (Hayes & Brownstein, 1986). For example, Zettle (1990) points out that 'thoughts, beliefs, and related cognitive phenomen[a] are not viewed as initiating causes for other behaviors. ... [I]nitiating causal factors are reserved for directly manipulable environmental events that can both predict and control behavior" (p. 42).

Although many rule-governed activities in daily life involve rules presented by other people (e.g., following instructions, obeying laws), numerous rules followed by individuals are formulated by themselves. Such self-rules are "contingency-specifying stimuli produced by one's own verbal behavior" (Zettle, 1990, p. 44) that develop from a person's own or observed experiences with contingencies, usually in interactions with other people. Self-rules may be expressed verbally, to oneself or to other people, but "self-rules that are only verbalized covertly may exert at least some control over behavior" as well (Zettle, 1990, p. 45).

As descriptions of contingencies, rules serve at least two major functions that are especially significant in the complex activities of daily life. First, "when a rule is stated to or by [a] person, that rule-statement prevents the specified outcome from reinforcing or punishing response classes not specified in the rule, thereby leaving the specified response class as the only one to be affected" (Hayes & Meyerson, quoted by Malott, 1989, p. 274). Thus, a rule connects one or more discriminative stimuli, the associated activities, and the relevant consequences - regardless of the numerous other events that are likely to occur during the intervals between any two components of the contingency. Second, when a behavior's consequences are temporally far removed from the present, "human beings optimize outcomes by following instructions or rules that specify the outcomes of their actions; in other words, it is not the delayed outcomes but rather the rules stating those delayed outcomes that more directly control the actions" (Malott, 1989, p. 283).

Both functions are especially important for numerous activities of daily life, especially when the several elements of contingencies are enmeshed in the complexities of a heterogeneous society. For example, rules such as "honest people go to heaven," describe an outcome for specific classes of behavior that, overwhelmingly important though it might be to some people, lies years and perhaps decades in the future. Behavior analysts would argue that, in much of daily life, it is not an activity's distant consequences, but rather the rules describing them, that control today's and tomorrow's activities. However, only those individuals who subscribe to that rule are likely to behave accordingly; those who say that they "do not believe in heaven" presumably do not subscribe to that rule and their actions will not be affected by it.

A major shortcoming of most behavior-analytic discussions of rule-governed behavior is their emphasis on verbal interactions; most analyses and examples of rules involve "speakers" who formulate rules and "listeners" who then behave accordingly. In daily life, however, self-rules are more common and important than speaker-listener relations.

Social Psychology and Rule-Governed Behavior

Although social psychologists do not explicitly talk about rules or rule-governed behavior, much of their experimental work analyzes behavior that is, in fact, governed by rules that normally remain largely unarticulated. The significance of rules for social psychology as well as behavior analysis arises from the fact that so many aspects of daily life require individuals to "deal with delayed, improbable, and small, cumulating outcomes" (Malott, 1989, p. 289). Most social psychologists, however, do not work with rules that emanate from "speakers" or take the form of instructions; instead, researchers are interested in self-rules and their implications for behaviors in various situations.

In numerous studies, a significant portion of the experimental procedures focuses on the discovery of the self-rules that people use in various aspects of daily life. The most common instrument for discovering people's self-rules is the questionnaire, which is typically administered at the beginning of an experimental session. On rare occasions, the several possible rules are explicitly stated, and individuals are asked to indicate the extent to which they agree with each. Hence the formal correspondence of self-rules in behavior analysis and social psychology can be easily established. More frequently, people are asked to what extent they accept various general statements, each of which summarizes several related rules. Researchers then infer a person's specific self-rules from the way these items are answered.

The functional correspondence of both self-rules and rule-governed activities in behavior analysis and social psychology becomes evident when the situations in which self-rules operate are examined. For example, self-rules associated with people's time horizons describe relationships of present activities with either proximate or distant consequences. Self-rules used by "optimists" and "pessimists" describe linkages between present activities and positive or negative consequences that vary in salience and probability of occurrence. Whether or not a person describes linkages between various activities and their consequences in the first place, or denies such contingencies, reflects the individual's attribution style. In social interactions encompassed by equity theory, people treat each other in large part on the basis of behavior-consequence linkages that express cultural definitions of fairness.

In the following sections, I consider recent work on rule-governed behavior in these four areas, in order to illustrate the range of topics and experimental procedures. In each area, I focus on researchers' discovery of people's unarticulated self-rules and their implications for behavior. Vernacular statements of rules reflect individuals' time horizons; the rules of optimists can be easily inferred from "life orientation" questionnaires; rules must be abstracted from several questions regarding individuals' attributions; and studies associated with equity theory require the deduction of rules from observed behavior.

Social psychologists usually consider research in these four areas as focusing on "personal characteristics" - for example, being optimistic or having a sense of external control. Yet the experiments themselves indicate that each of these "personal characteristics" is essentially a set of self-rules that (a) summarizes an individual's experiences with certain behavior-consequence linkages (Kunkel, 1997), and (b) controls relevant activities. Some investigators speak of these characteristics as if they were independent variables. Once again, however, the actual procedures in numerous experiments indicate that situational factors are the primary determinants of activities; self-rules operate as subsidiary controlling factors.

Time Horizons

Most people are aware that many of their activities are followed by several consequences spread over a period of time, but social psychologists have discovered that individuals differ in their descriptions of the future: some people look only a short time ahead, others look far ahead. Individuals with short time horizons are likely to describe fewer consequences for their actions, and proximate events are likely to loom large, mainly because there are few if any distant ones. Individuals with long time horizons are likely to describe more consequences for their actions, including quite distant ones, and proximate events are likely to be comparatively less significant. Thus, different time horizons lead to different descriptions of contingencies, which are reflected in the self-rules that people use in selecting their behaviors.

In this area, as in so many others, social psychologists have constructed questionnaires to discover people's usually unarticulated self-rules. A recent example is the "Consideration of Future Consequences" scale (Strathman, Gleicher, Boninger, & Edwards, 1994, p. 752). Individuals are asked to indicate how characteristic each of twelve items is for them: 1 extremely uncharacteristic, 2 - somewhat uncharacteristic, 3 - uncertain, 4 somewhat characteristic, 5 - extremely uncharacteristic.

The most significant of the twelve items are:

1. Often I engage in a particular behavior in order to achieve outcomes that may not result for many years.

2. I only act to satisfy immediate concerns, figuring the future will take care of itself.

3. My behavior is influenced only by the immediate (i.e., a matter of days or weeks) outcomes of my actions.

4. I think it is more important to perform a behavior with important distant consequences than a behavior with less-important immediate consequences.

5. I think that sacrificing now is usually unnecessary since future outcomes can be dealt with at a later time.

6. Since my day to day work has specific outcomes, it is more important to me than behavior that has distant outcomes.

A summary measure of responses places an individual somewhere along the time horizon continuum. At one extreme, people prefer activities with immediate consequences and disregard distant events (short time horizon), while at the other end individuals prefer activities with distant consequences and disregard immediate events (long time horizon).

For behavior analysts, most of the dozen items are vernacular statements of general self-rules that describe the linkage between present activities and proximate or distant consequences. However, what matters is not so much a person's overall position on the continuum as the specific self-rules contained in the set. From an individual's pattern of responses one can easily deduce the general self-rules that constitute the typical long and short time horizon: "I select behaviors primarily on the basis of their temporally distant consequences," and "1 select behaviors primarily on the basis of their immediate consequences***:'

Each item is only part of a self-rule which individuals may use to define and assess the consequences of various actions. The self-rule is completed by an individual's addition of a statement regarding the item being characteristic to some degree, that is, how important or frequently used that contingency is. Each of these general self-rules may be expressed in more specific terms. For example, Item 2 reflects a self-rule that one might express as: "activities with proximate events matter most," and implies the further self-rule: "activities with distant consequences are unimportant***;' The several implications of each general self-rule have to be stated in subsidiary rules, for example, about the relative weight of immediate and distant consequences for particular classes of activities.

Some of the above items are almost direct expressions of temporal self-rules (e.g., Items 1, 2, and 3). Other items serve as the basis from which researchers can easily abstract time-horizon rules, and at the same time check respondents' consistency. Item 6, for example, reflects the self-rules: "Behaviors with immediate consequences are very important," and "activities with distant consequences are unimportant."

Both social psychologists and behavior analysts would expect that people's activities will be congruent with the relevant self-rules. Thus, when researchers know an individual's position on the time-horizon continuum, they can predict the likely activities in various situations (e.g., Nuttin & Lens, 1985). For example, people with long time horizons are less likely to have regrets about current negative experiences (e.g., what they could have done differently), and are more positive about their future plans, than are people with short time horizons (Boninger, Gleicher, & Stratham, 1994).

Life Orientations

The several consequences that follow a behavior may be positive or negative, and they occur on various schedules - or, as social psychologists say, they occur with varying probabilities. In the laboratory, experimenters arrange these elements; hence they are known, at least to the researcher. In daily life, however, the several elements of a behavior's total outcome are often unknown to both the subject and the researcher - specially when so many events are unpredictable or the consequences lie far in the future.

Most people's descriptions of contingencies and schedules are based primarily on their own experiences. These descriptions, however, may include faulty memories and erroneous inferences (e.g., Nisbett & Ross, 1980). Hence people's assessments of the future are likely to differ somewhat, even when they appear to have had similar experiences. A person's descriptions of future events include numerous self-rules about important behavior-consequence linkages that function as controlling variables for various behaviors in relevant situations. The most significant dimension of these self-rules is the proportion of positive and negative consequences, and the probabilities of the occurrence of each. Here social psychologists usually speak of the optimism (and pessimism) dimension.

An individual's position on this dimension is usually determined by responses in the "Life Orientation Test" (LOT) (Scheier, Carver, & Bridges, 1994, p. 1073). Subjects are asked to indicate the extent to which they agree with ten items (of which four are fillers): 0 - strongly disagree, 1 - disagree, 2 - neutral, 3 - agree, 4 - strongly agree.

The six significant items are:

1. In uncertain times, I usually expect the best.

2. If something can go wrong for me, it will.

3. I'm always optimistic about my future.

4. I hardly ever expect things to go my way.

5. I rarely count on good things happening to me.

6. Overall, I expect more good things to happen to me than bad.

Each of these statements describes in very general terms the complex relationships between a person's more significant activities and their likely outcomes, in terms of both value and probability. Social psychologists combine individuals' responses to these items into measures of optimism and pessimism, and then speak of people as being optimists or pessimists to some degree.

Behavior analysts would consider each item as a vernacular description of the linkages between various important activities and the probabilities and proportions of positive and negative consequences. Each item, therefore, when it is combined with a person's statement regarding his or her agreement with it, may be considered as a summary of several specific self-rules.

For example, Item I includes a rule that might be phrased as: "There is a high probability that my activities lead to positive events." It also reflects another rule: "I can perform the required activities reasonably well." Item 4 contains the opposite rule: "There is a high probability that my activities produce negative events." But it also implies another rule: "There is little connection between my actions and later events." The LOT contains near-duplicate items, for example, 2 and 4, and opposite items, such as 1 and 5. Although these serve primarily to assess respondents' consistency, they are also quite useful in tapping slight variations in the self-rules people use to describe (and deny) linkages between their activities and later events.

Social psychologists are wont to speak of optimism and of pessimists, but these terms do not refer to internal processes or to individuals with certain characteristics. The experimental procedures that are commonly used clearly indicate that "optimism" is a label for a particular set of self-rules that link numerous important behaviors to highly likely positive consequences and very unlikely negative events. A "pessimist" is a person whose important behaviors are controlled by a quite different set of self-rules that link significant activities to highly likely negative consequences and unlikely positive events, or deny the linkage altogether.

Although a person's behaviors are determined primarily by situational factors, the controlling function of self-rules becomes evident in studies that compare the behaviors of optimists and pessimists. For example, in situations where the probability of solving a problem is low, optimists are more likely than pessimists to try various solutions. In addition, optimists typically use effective procedures to overcome difficulties, whereas pessimists typically engage in activities that are likely to be unsuccessful (e.g., Scheier, Weintraub, & Carver, 1986). Furthermore, the different sets of self-rules used by optimists and pessimists have significant implications for health (e.g., Scheier & Carver, 1992).


Many events of daily life have self-evident causes and generally agreed-upon determinants, hence the explanations are clear and straightforward. For example, when I carry too many dishes from the table to the sink and some slide off the top, it is obvious to everyone that I piled them too high or carelessly. But in ambiguous situations, and especially when events can be explained in several plausible ways, causal uncertainty is rather common (e.g., Weary & Edwards, 1994). Under such circumstances, most people then select one or another possible explanation (i.e., make an attribution) which does not necessarily reflect the actually operating determinants.

For more than 30 years, social psychologists have analyzed the several ways in which people select explanations of ambiguous and "difficult to understand" events that happen to them and to others, and the implications of the different explanations that individuals use (e.g., Fiske & Taylor, 1991, pp. 22-95). For example, one student might ascribe her good grade to diligent study, while another talks about good luck (e. g., "They asked the right questions"). Plausible explanations of negative events, such as low exam grades, are usually more numerous: One person might ascribe a poor score to inadequate study, another to a hard test, another to not feeling well that day, another to an unfair teacher, and still another to bad luck (e.g., "the test did not cover the material I studied.") The explanation an individual selects, especially for a negative event, affects subsequent behavior. In the above example, only students who admit that they did not study enough are likely to study more for the next test. The other students are likely to hope that the presumed causal factors will not operate next time the test will be easier, I'll feel better, or the "right" questions will be asked.

Social psychologists have discovered that people follow distinct patterns in their descriptions of causal factors, especially in ambiguous situations or when novel events happen to them and other people. Most causal factors which individuals describe to explain ambiguous events can be placed along three dimensions, in descending order of importance: personal (internal) - external; global - specific; and stable - unstable. The latter two dimensions are highly correlated, as one might expect, and both are quite independent of the first (Peterson & Villanova, 1988).

If rules are contingency-specifying stimuli, then attribution is the most relevant area in social psychology. Indeed, a considerable portion of attribution research focuses on the analysis of self-rule-governed behavior. During the last 15 years, many experimenters have given subjects the "Attributional Style Questionnaire" (ASQ), which asks people to describe the causes that they believe determine some plausible events in their lives (Peterson et al., 1982). Subsequently, Peterson and Villanova (1988) developed an expanded version (EASQ), which consists of 24 negative items, compared to the present form that contains 6 positive and 6 negative items. Later, Whitley (1991) presented a short form of this expanded version. Here, I discuss only the original version, because so many experiments have used it; moreover, the three versions do not differ in structure.

Individuals are asked to describe the causes which they believe determine 12 hypothetical events (Peterson et al., 1982, pp. 291-292). For example:

1. "You have been looking for a job unsuccessfully for some time."

After individuals have written down what they consider to be the one major cause of this event, they answer four questions about that presumed cause:

a. Is the cause of your unsuccessful job search due to something about you or to something about other people or circumstances?

due to other people 1 2 3 4 5 6 7 due to me

b. In the future when you look for a job, will this cause again be present?

never 1 2 3 4 5 6 7 always

c. Is the cause something that just influences looking for a job, or does it also influence other areas of your life?

this situation 1 2 3 4 5 6 7 all situations

d. How important would this situation be if it happened to you?

unimportant 1 2 3 4 5 6 7 very important

Some of the other events in the questionnaire are:

2. You give an important talk in front of a group, and the audience reacts negatively.

3. You do a project that is highly praised.

4. You can't get all the work done that others expect of you.

5. You apply for a position that you want very badly (e.g., important job, graduate school admission) and you get it.

6. You go out on a date and it goes badly.

On the basis of subjects' responses to these 48 questions (4 questions for each event), social psychologists place individuals somewhere along the three dimensions of presumed causes described earlier. The combination of positions is the foundation of a person's "attribution style."

Behavior analysts would look at attribution in terms of various self-rules by which individuals describe (or deny) the linkages between their own activities and other events. From this ASQ questionnaire, we can infer several rules which an individual may use to account for various positive and negative events. For example, people who place high on the "personal (internal)" end of the control dimension (i.e., high numbers on the "a" questions), in effect describe self-rules that establish definite linkages between their own activities and various positive and negative consequences. A person who places high on the "global" end of the control dimension (i.e., high numbers on the "c" questions), describes self-rules about the ubiquity of the factors that are linked with one's behavior, regardless of what an observer might know. These presumed factors can be positive or negative, which again reflect the self-rules a person uses; for example, a personal global factor might be high intelligence, or it could be low intelligence. Thus, what social psychologists call a person's "attribution style" is actually the set of self-rules an individual habitually uses.

It is interesting that, especially in ambiguous situations, most people are quite consistent in selecting and combining presumed causal factors into an overall explanation, or set of self-rules. For example, most people describe positive events as being the consequences of their own (personal) activities, whereas negative events are considered to be the results of other (external) factors, such as bad luck. In addition, some individuals attribute events to (global) factors such as high or low intelligence, and others attribute events to (specific) factors such as a hangover. Finally, some persons attribute events to (stable) factors, such as low or high intelligence, whereas others attribute events to temporary (unstable) factors, such as being overtired.

Several experiments (e.g., Alloy, Peterson, Abramson, & Seligman, 1984) have shown that the most common rule set, especially among normal individuals, consists of self-rules one might phrase as: "positive events are related to my preceding behavior," and "negative events are independent of my actions***:' But there are numerous other possible self-rules, for example: "positive events are related to luck," and "negative events are due to my low intelligence," "I don't have a math brain," or "I'm clumsy and can't do anything right***:'

Every individual's attributional style or self-rule set has implications for much of daily life, ranging from simple activities to larger aspects of a person's life, such as academic success, pessimism, and even depression (e.g., Marsh & Weary, 1989; Metalsky, Abramson, Seligman, Semmel, & Peterson, 1982; Peterson & Barrett, 1987; Strickland, 1989). In terms of negative implications, the most detrimental self-rule set links negative experiences to personal, global, and stable factors, such as low intelligence. There are several effective rule sets, whose components vary with circumstances. For example, in order to learn from experience, a person should rank high on the "personal" end of the control dimension, that is, employ rules that establish contingencies. But those self-rules should be combined with others that emphasize specific and unstable factors, such as a temporary lack of the required skills, Together, effective self-rules should enable the individual to say: "I did not succeed today because I didn't do this right - but I'll do it differently and better next time."

Equity Theory

Over the years, social psychologists have examined numerous aspects of social interaction, ranging from the productivity of work teams to the happiness of lovers. Laboratory research covers an equally wide range, from the variables that affect the operation of groups to the factors that weaken amorous dyads. In this section, I sketch one aspect of social interaction now known as equity theory (e.g., Walster, Walster, & Berscheid, 1978).

Some years ago social psychologists discovered that when people work in small groups, individuals tend to distribute rewards and blame to themselves and the other members on the basis of two quite different principles. When people allocate rewards (and blame) on the basis of a person's behavior, they are said to employ the equity norm: The better a person's performance, or the greater an individual's contribution to the group, the greater will be the reward; individuals who contribute least receive the lowest rewards. When people divide the available rewards (and blame) equally among the group members, they are said to use the equality norm: Performance is irrelevant, any individual's contribution to the group does not matter.

Behavior analysts would say that in small groups, members use various self-rules to distribute rewards and blame among themselves. The essential difference between the two allocation norms is the existence or absence of behavior-consequence linkages. Hence there are at least three sets of rules that group members use: one set of self-rules establishes contingencies (i.e., the equity norm); another set of self-rules denies contingencies (i.e., the equality norm); and a third determines which of the first two sets will be used in various circumstances. Considerable research over three decades has revealed several interesting patterns in the way people use one or the other allocation norm (e.g., Walster, Walster, & Berscheid, 1978). Here I mention only two.

Gender or Familiarity?

Laboratory studies during the 1970s demonstrated, among other things, that in group settings males tend to emphasize the equity norm, and females seem to prefer the equality norm (e.g., Reis & Jackson, 1981). This result was so robust that several quite plausible explanations were offered, such as males' concern with status and females' concern with affiliation, at least in our culture (Reis & Jackson, 1981). Behavior analysts would describe this kind of research as the study of apparently gender-linked rules that people use when they decide whether or not to establish behavioral contingencies for group members (including themselves).

In their path-breaking research, however, Reis and Jackson (1981) demonstrated that the choice of one or the other allocation norm is determined by a quite different factor: individuals' presumed familiarity with the required behavior. In this experiment, the laboratory task consisted of precisely naming several objects used in the home, and of precisely describing the functions of each. A person's score was based on the number of correct names and descriptions within a time frame, and obviously depended on a thorough knowledge of the objects. The results did not fit the previous pattern but were internally consistent. Both men and women used the equity norm to allocate rewards among group members when the experimental tasks were sex-appropriate, and the equality norm when the tasks were sex-inappropriate.

For example, when the task was to identify a colander by name and specify its functions, women used the equity norm when they had female partners, and the equality norm when they had male partners. Conversely, when the task was to identify a hacksaw by name and specify its functions, men used the equity norm when they had male partners, and the equality norm when they had female partners.

As this study demonstrated, there are no gender differences in the use of allocation norms when individuals take into account a culture's definition of gender-appropriate tasks and the resulting differences in presumed experiences. How then does one account for the numerous contrary findings in previous work? Reis and Jackson suggest that the earlier researchers (most of whom were males) inadvertently had used primarily male-appropriate tasks. This situation then led men (in male groups) to use the equity norm, and women tended to use the equality norm for themselves and their male and female partners.

The disappearance of gender differences in the selection of allocation norms, and the significance of task familiarity, however, do not eliminate another question: Why do individuals, regardless of gender, make the same differential use of allocation norms, and do it consistently? The question gains significance because this differential use is now more robust than before.

When we inspect the various situations in which the two allocation norms are used, it quickly becomes apparent that the differential use of the two norms occurs in the service of another, overriding, self-rule: "rewards (and blame) should be distributed fairly." The questionnaires that subjects filled out after the experiment suggested that fairness had been a concern.

This and other laboratory data suggest that most people, at least in our culture, tend to use one fundamental self-rule for the distribution of rewards: "One Must Be Fair? In other cultures, other basic rules are used. Individuals use whichever specific self-rule seems to them to be most fair in various circumstances. Although the experimenters did not ask direct questions about allocation norms or the fairness principle, both of these can be inferred from either the subjects' behaviors or the postexperimental questionnaire. Thus, when people can reasonably assume that they and their partners are equally familiar with the task (e.g., two women describing a colander, or two men describing a hacksaw), both men and women use the self-rule that establishes behavior contingencies (i.e., the equity norm): Both partners compete on equal terms (with presumably equal experience and knowledge), and "the better person wins." But when it makes sense for people to assume that one partner (they themselves or the other person) is more experienced or knowledgeable than the other (e.g., in our culture, women assume they know more about colanders and men assume they know more about hacksaws), then the self-rule that denies behavior contingencies (i.e., the equality norm) is used.

Before we conclude that such fundamental self.rules as "Be fair!" govern substantial portions of human interactions, we must recognize the somewhat restricted situation of the laboratory. What happens in an experiment with its trivial outcome may not occur in real-life situations, especially when much of one's future is involved. In the above experiment, for example, the subjects divided $2.95 between themselves, understood that they would not meet again, and knew that no one cared about their scores on this trivial task of identifying a dozen household items. But what would happen in real life? What self-rules will be employed when the tasks are significant, the rewards are substantial, and the other people are strangers, competitors, colleagues, or friends? For the most part, such field studies remain to be done.


The rules that control the selection of allocation norms depend also on the factors that people assume determine an individual's eventual success or failure. Greenberg (1980) performed an experiment in which subjects and their partners were either responsible or not for success in a rather simple task. As behavior analysts would say, half the subjects faced behavioral contingencies, the other half did not. Two individuals in separate cubicles "competed" by throwing five dice from a cup and adding up the points. The rewards that were to be allocated by losers and winners (arranged by the experimenter) were hypothetical, consisting only of poker chips. In the "nonresponsible" condition, the subjects were told to simply add the points shown on ten throws of the dice. Clearly, the total points a person received were entirely a matter of chance or luck. In this case, individuals were not responsible for winning or losing (i.e., there were no contingencies), and both winners and losers tended to allocate the rewards (poker chips) by following the equality norm. In the "responsible" condition, individuals were told that they were competing not only for points on the dice but also in terms of time and accuracy in adding up these points. To a considerable extent, the competitors were responsible for their performance (i.e., there were contingencies); now, both winners and losers tended to allocate the rewards by following the equity norm. There were no gender differences in the selection of allocation norms.

After the experiment, subjects filled out questionnaires which, among other things, asked directly about the fairness of the allocation norm they had used. Both women and men apparently subscribed to the fairness rule as it exists in our culture and chose the allocation norm they deemed appropriate - that is, considered to be fair - under the circumstances.

Behavior analysts would say that in this experiment individuals' rules about the selection of allocation norms (i.e., self-rules that establish or deny contingencies) took into account the differential contingencies inherent in the two tasks. The choice of norms, in turn, was a manifestation of the fundamental self-rule: "Be fair!"

Again one wonders, however, what people would do in situations where the rewards are materially and/or socially significant. Furthermore, one wonders which self-rules people are likely to use in daily life, where circumstances are often ambiguous and individuals do not know much about other people's familiarity with the tasks at hand or the determinants of outcomes. In ambiguous situations and in the company of strangers or among friends, who can tell what is fair?

An Application

Equity theory is a powerful tool for understanding the dynamics of social interaction in large groups and small. I provide only one illustration: studies of liking and long-term relationships. Social interactions will continue as long as the participants benefit from each other (e.g., Walster, Nalster, & Berscheid, 1978). This is particularly true when two people like each other; they will continue to do so, and their relationship will last, as long as they believe that each is being treated fairly by the other. For example, Walster, Walster, and Traupmann (1978) concluded that partners are happiest, and amorous relationships are likely to continue, when each believes that he or she receives benefits appropriate to his or her contribution to the relationship. Here the self-rules apparently are: "My partner should treat me fairly" and "I should treat my partner fairly." Problems arise, of course, in defining the important variables for assessing one's own and the partner's contribution to and benefit from the relationship. When a person defines the relationship as one in which he or she consistently receives inadequate benefits (regardless of what actually occurs), it is likely to dissolve.


Guerin (1992) believes that so far, at least, behavior analysis has had little direct impact on social psychology. Although this is an arguable position (e.g., Kunkel, 1996), there are good reasons for being optimistic about the future of cooperation. Chief among these is the existence of common ground on which behavior analysts and social psychologists can meet to discuss shared interests; those discussions, in turn, may well lead to joint efforts to increase our understanding of various phenomena in daily life.

In this article, I examine four significant research areas in mainstream social psychology that are part of the common ground. At first glance one might think that these areas - time horizons, life orientation, attribution, and equity theory - are of little interest to behavior analysts. After all, the major variables under investigation seem to be so different from behavior analysis and so diverse: a person's time scale, optimism and pessimism, personal and external control, and choice of allocation norms. But when one examines the major research questions and the actual procedures of experiments, it becomes apparent that investigators share a familiar common interest: the analysis of rule-governed behavior.

In particular, social psychologists are trying to discover and understand the normally unarticulated self-rules that people use in their daily activities. The functional correspondence of self-rules (or private talk) in behavior analysis and social psychology is demonstrated by the fact that in both fields self-rules are verbal descriptions of behavior-consequence linkages that control activities in relevant situations.

Self-rules are especially necessary when people face the complexities of daily life: when the consequences of numerous activities are spread over a broad temporal range (time horizons), when these consequences are uncertain and could be positive or negative (life orientation), when the linkages between many activities and later events are debatable (attribution), and when all of these variables are part of social interaction (equity).

I have selected these four topics in large part to illustrate the variety of procedures commonly used by social psychologists to discover and elaborate some of the rules commonly used by people in various situations. The self-rules contained in a person's time horizon are delineated in ways that require few inferences, whereas the discovery of self-rules contained in optimism and attribution require progressively more inferences from questionnaire responses. The self-rules contained in allocation norms, finally, must be abstracted from the activities themselves.

Vaughan (1989), Zettle (1990), and others have pointed out that private talk or "self-talk" controls much of adult behavior. If we consider self-rules to be important examples of self-talk, then questionnaire items such as those mentioned in earlier sections are good examples of both. However, most of these items are only partially stated rules, which must be completed by an individual's responses. For example, the statement "if something can go wrong for me, it will" is completed by a person's expression of agreement. If "strongly agree" is added, researchers can derive one set of self-rules for the individual, that includes: "Many of my actions are followed by negative events" and "there isn't much I can do to avoid negative events***:' If a person adds: "strongly disagree," a quite different set of self-rules becomes apparent, which includes: "Many of my actions are successful," and "I can easily avoid negative events." Both completed questionnaire items are summaries of self-rules because each describes the existence or absence of linkages between a person's behaviors and various positive or negative subsequent events.

So far, most social psychologists have not paid much attention to the origins of the self-rules they study. However; I have recently pointed out that the questionnaires mentioned earlier, among some others used by social psychologists, provide information about individuals' learning histories (Kunkel, 1997). Most questionnaire items summarize significant past behavior-consequence linkages that people experienced in various situations. For example, when a person agrees with the item "1 rarely count on good things happening to me? that statement is evidently a summary of one's own past experiences with numerous linkages between various activities and negative consequences. Social psychologists rely on their subjects' descriptions of the past because they have no way of directly observing those events. Researchers would not need such questionnaires if individuals' relevant learning histories were known or part of the experiment itself.

At the same time, the questionnaires mentioned above provide the basic information from which experimenters derive the self-rules that control their subjects' present and future activities. In social psychology, then, self-rules are more than "contingency-specifying stimuli produced by one's own verbal behavior" (Zettle, 1990, p. 44). In addition, a person's self-rules are summaries of relevant experiences combined with the assumption that future contingencies will be much like past contingencies.

Hayes (1992) has described some of the ways by which past organism-environment interactions can be brought into the present, so to speak, where these determine an individual's behavior. If questionnaire responses describe an individual's past experiences, as I have suggested elsewhere (Kunkel, 1997), and if those reponses are also indicative of self-rules, as I have argued in this paper, then these instruments provide snapshots of the continuous stream of individual-context interactions that constitutes daily life. A self-rule (or private talk), derived from important dimensions of a person's experiences, brings an individual's past into the present.

Social psychologists would agree with behavior analysts that an individual's past experiences with behavior-consequence linkages are the major determinants of present and future activities. In their daily lives, however, most people do not articulate those experiences; indeed, there is little reason why they should do so. Hence researchers must develop instruments to discover this information. When social psychologists want to predict a person's behavior but know nothing about the relevant learning history, they typically use one of two common procedures: (a) Individuals are asked to describe their past experiences with relevant contingencies, or (b) individuals are asked to describe anticipated contingencies, which reflect self-rules.

Hayes and Brownstein (1986) suggest that behavior analysts "must explain both how external events gave rise to the private talk, and how the private talk came to control the behavior of interest. When this is done, 'self-rules' may participate in an overall causal relation, but they should not themselves be seen as causes" (p. 188). In social psychology, the two conditions are satisfied by (a) the assumption that a self-rule (or private talk) summarizes past experiences, and (b) the fact that a particular set of self-rules is relevant only in specific circumstances. For example, self-rules summarized as "pessimistic" do not matter much in situations where time horizons are the significant variables.

Social psychologists would agree that when they know both an individual's present context and the relevant self-rules that are employed, they can predict activities more accurately than if they know only the situation. Researchers also agree that if they know only a person's self-rules - summarized, for example, by a high degree of optimism - and nothing about the context, they can not predict any behaviors. For example, in order to predict whether a person with an unknown history will buy a lottery ticket, we must know the saliant components of the present situation, such as the opportunity to buy a lottery ticket, the money available, the amounts of prizes, and the stated probabilities of winning. These major determinants of subsequent behavior are augmented by self-rules as controlling variables. If we also know the relevant set - that is, whether the person uses the "optimistic" or "pessimistic" self-rules mentioned earlier - we are in a better position to predict whether or not a lottery ticket will be purchased than if we had no information about the relevant self-rules. Conversely, if we know the individual's relevant reinforcement history, we can predict whether a ticket will be purchased, without knowing the person's self-rules.

According to Vaughan (1989), "[d]iscovering the conditions that produce correspondence between saying and doing may be the most profound contribution behavior analysts make in advancing a science of behavior" (p. 110). In this article, I suggest that by teaming up with social psychologists who analyze self-rule-governed activities, behavior analysts are more likely to discover these conditions than by working alone.


ALLOY, L. B., PETERSON, C., ABRAMSON, L, Y., & SELIGMAN, M. E. R (1984). Attributional style and the generality of learned helplessness. Journal of Personality and Social Psychology, 46, 681-687.

BAUM, W. M. (1995). Rules, culture, and fitness. The Behavior Analyst, 18, 1-21.

BONINGER, D. S., GLEICHER, F., & STRATHMAN, A. (1994). Counterfactual thinking: From what might have been to what may be. Journal of Personality and Social Psychology, 67, 297-307.

CZUBAROFF, J. (1993). Convergences with behavior analysis: Recommendations from the rhetoric of inquiry. The Behavior Analyst, 16, 1-8.

FISKE, S. T., & TAYLOR, S. E. (1991). Social cognition (2nd ed.). New York: McGraw-Hill.

GLENN, S. G. (1993). Windows on the 21st century. The Behavior Analyst, 16, 133-151.

GREENBERG, J. (1980). Attentional focus and locus of performance causality as determinants of equity behavior. Journal of Personality and Social Psychology, 38, 579-585.

GUERIN, B. (1992). Behavior analysis and social psychology: A review of Lana's Assumptions of social psychology. Journal of the Experimental Analysis of Behavior, 58, 587-604.

HAYES, L. J. (1992). The psychological present. The Behavior Analyst, 15, 139-145.

HAYES, S.C. (Ed.). (1989). Rule-governed behavior: Cognition, contingencies, and instructional control. New York: Plenum.

HAYES, S.C., & BROWNSTEIN, A. J. (1986). Mentalism, behavior-behavior relations, and a behavior-analytic view of the purpose of science. The Behavior Analyst, 9, 175-190.

KUNKEL, J. H. (1996). What have behaviorists accomplished - and what more can they do? The Psychological Record, 46, 21-37.

KUNKEL, J. H. (1997). Three contributions of social psychology to the analysis of the behavior-consequence linkage. The Psychological Record, 47, 201-220.

MALOTT, R. W. (1989). The achievement of evasive goals: Control by rules describing contingencies that are not direct acting. In S. C. Hayes (Ed.), Rule-governed behavior: Cognition, contingencies, and instructional control (pp. 269-321). New York: Plenum.

MARSH, K. L., & WEARY, G. (1989). Depression and attributional complexity. Personality and Social Psychology Bulletin, 15, 325-336.

METALSKY, G. I., ABRAMSON, L. Y., SELIGMAN, M. E. R, SEMMEL, A., & PETERSON, C. (1982). Attributional styles and life events in the classroom: Vulnerability and invulnerability to depressive mood reactions. Journal of Personality and Social Psychology, 43, 612-617.

MISBETT, R. E., & ROSS, L. (1980). Human inference: Strategies and shortcomings of social judgment. Englewood Cliffs, N J: Prentice-Hall.

NUTTIN, J., & LENS, W. (1985). Future time perspective and motivation: Theory and research method. Hillsdale, NJ: Erlbaum.

PETERSON, C., & BARRETT, L. C. (1987). Explanatory style and academic performance among university freshmen. Journal of Personality and Social Psychology, 53, 603-607.

PETERSON, C., SEMMEL, A., v. BAEYER, C., ABRAMSON, L. Y., METALSKY, G. I., & SELIGMAN, M. E. P. (1982). The attributional style questionnaire. Cognitive Therapy and Research, 6, 287-300.

PETERSON, C., & VILLANOVA, P. (1988). An expanded attributional style questionnaire. Journal of Abnormal Psychology, 97, 87-89.

REIS, H. T., & JACKSON, L. A. (1981). Sex differences in reward allocation: Subjects, partners, and tasks. Journal of Personality and Social Psychology, 40, 465-478.

SCHEIER, M. F., & CARVER, C. S. (1992). Effects of optimism on psychological and physical well-being: Theoretical overview and empirical update. Cognitive Therapy and Research, 16, 201-228.

SCHEIER, M. F., CARVER, C. S., & BRIDGES, M. W. (1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the life orientation test. Journal of Personality and Social Psychology, 67, 1063-1078.

SCHEIER, M. F., WEINTRAUB, J. K., & CARVER, C. S. (1986). Coping with stress: Divergent strategies of optimists and pessimists. Journal of Personality and Social Psychology, 51, 1257-1264.

STRATHMAN, A., GLEICHER, F., BONINGER, D. S., & EDWARDS, C. S. (1994). The consideration of future consequences: Weighing immediate and distant outcomes of behavior. Journal of Personality and Social Psychology, 66, 742752.

STRICKLAND, B. R. (1989). Internal-external control expectancies: From contingency to creativity. American Psychologist, 44, 1-12.

VAUGHAN, M. (1989). Rule-governed behavior in behavior analysis: A theoretical and experimental history. In S.C. Hayes (Ed.), Rule-governed behavior: Cognition, contingencies, and instructional control (pp. 97-117). New York: Plenum.

WALSTER, E., WALSTER, G. W., & BERSCHEID, E. (1978). Equity: Theory and Research. Boston: Allyn & Bacon.

WALSTER, E., WALSTER, G. W., & TRAUPMANN, J. (1978). Equity and premarital sex. Journal of Personality and Social Psychology, 36, 82-92.

WEARY, G., & EDWARDS, J. A. (1994). Individual differences in causal uncertainty. Journal of Personality and Social Psychology, 67, 308-318.

WHITLEY, B. E. (1991). A short form of the expanded attributional style questionnaire. Journal of Personality Assessment, 56, 365-369.

ZETTLE, R. D. (1990). Rule-governed behavior: A radical behavioral answer to the cognitive challenge. The Psychological Record, 40, 41-49.
COPYRIGHT 1997 The Psychological Record
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kunkel, John H.
Publication:The Psychological Record
Date:Sep 22, 1997
Previous Article:Equivalence class formation using stimulus-pairing and yes-no responding.
Next Article:Chimpanzee (Pan troglodytes) counting in a computerized testing paradigm.

Related Articles
The old college try falls short.
Cognitions, thoughts, private events, etc., are never initiating causes of behavior: reply to Overskeid.
Toward a socially valid understanding of problem behavior.
Enhancing accountability in behavioral consultation through the use of single-case designs.
Comment on "rule-governed behavior and self-control in children with ADHD: a theoretical interpretation".
The long good-bye: why B.F. Skinner's Verbal Behavior is alive and well on the 50th anniversary of its publication.
Peter Harzem (1930-2008): bonds and remembrances in Mexico.
In memoriam.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters