Towards a Cooperative Social-Personality Psychology

The most useful way to consider situational and personal variables is as interactional partners. This view was operationalized in Lewin’s (1936) well-known formula: B = f (P,S), which explicitly defines behavior as a function of the person and the situation. This equation implies that if we knew all of the relevant psychological properties of a person and all of the relevant properties of his or her situation, we could predict with high precision what the person would do.

Lewin’s equation can be further arranged to illuminate other associations between behavior, personality, and situations (Bandura, 1978). P = f (B,S) implies that persons can be conceptualized in terms of their behaviors in every situation of their lives. This arrangement of the equation exemplifies classical Watsonian behaviorism (Watson, 1930) and also characterizes Mischel and Shoda’s (1995) “if… then” statements. Another way to arrange the equation, S = f (B, P), suggests that a situation can be understood in terms of the behaviors that different people might perform in it. This idea formed the basis for the template-matching technique introduced by Bem and Funder (1978). For example, it is possible to describe a college campus in terms of the kind of people who commonly attend it and do well there. A college campus in which students who are introverted, philosophical, and intellectual do well is a very different situation than a campus where better outcomes are attained by students who are extraverted, athletic, and rambunctious.

The psychological triad represented by these three formulae suggests that persons, situations, and behaviors should be studied in unison (Funder, 2006). A serious obstacle to achieving this goal concerns the uneven development of conceptualizations and measurement technologies. Although many methods are available for assessing personality, relatively fewer methods have been established for studying behavior and almost no methods for describing situations.

The conceptualization and measurement of personality traits is well developed and ranges from the assessment of a small number of  essential global traits (e.g. The Big Five), to large, comprehensive sets of mid-level characteristics that describe many ways in which individuals differ (e.g. The California Adult Q-set). Moreover, a large number of trait measures come packaged with a theory that explains the behaviors and outcomes to which the trait is purportedly related and an adequate validity literature that addresses psychometric properties and observed external correlates.

In contrast, true behavioral measurement (i.e., direct observations of behavior by independent observers who describe a behavior that they have actually seen someone do) is rarer than one might think (Baumeister, Vohs, & Funder, 2007).  Systematic attempts to develop a taxonomy of behavior within a theoretical framework are even rarer. Behavior, when it is actually observed, is almost always chosen to illustrate a particular theoretical prediction and the typical study includes one behavior, which might be something as detached from real-life action as a button press on a computer keyboard or a questionnaire response. Although classic social psychology studies in the 1960’s and 1970’s sometimes directly observed single behaviors that were important and consequential (e.g., bystander intervention and obedience), focusing on a single behavior provides a narrow view of the many different things that people might be doing at the same time. For example, when people obey commands by an authority figure to shock a victim, do they plead with the authority figure that shocking the victim seems wrong? Do they ask the authority figure how dangerous the shocks are? Do they try to communicate with the victim? (Milgram, 1974/2004, informally reported that his subjects did all of these things, but did not provide any direct measurements.) In short, broader conceptualization and measurement of behavior is sorely needed.

The Riverside Behavioral Q-set (RBQ; Funder, Furr & Colvin, 2000; RBQ 3.0) is one possible, partial solution to this problem. The RBQ is a comprehensive set of 67 items that describe a broad range of socially relevant behaviors. RBQ items describe behavior at a mid-level of generality so that the behaviors that are coded are not too microscopic (e.g. eye blinks) or too macroscopic (e.g. socially successful). The items were originally derived from the items of the California Adult Q-Set (CAQ: Block, 1961; Bem & Funder, 1978) and were re-worded to emphasize behavioral display rather than trait inference. For example, an item in the CAQ reads “is critical, skeptical, not easily impressed” and the associated RBQ item reads “expresses criticism.” The RBQ is a valuable tool for a researcher who is interested in measuring a variety of behaviors that are relevant to a wide range of personality constructs and social situations. It has been used to illustrate the independence of behavioral change and consistency (Funder & Colvin, 1991), the behavioral correlates of various personality traits (e.g. Fast, Reimer, & Funder, 2007), and for many other purposes.[1]

Numerous researchers have complained about the lack of methods for describing features of situations (e.g., Funder, 2000; Hogan & Roberts, 2000; Reis, 2008; Swann & Seyle, 2005). In social psychological experiments, a situational manipulation is typically chosen to test a specific theoretical prediction, not because it is necessarily viewed as an important dimension of situations in general. Social psychology could be said to contain a huge amount of information about how narrowly defined situations affect behavior, but this knowledge is fragmented because there exists no method for organizing findings into a coherent summary. What is needed is a way to conceptualize and measure the active ingredients of situations. This goal requires identifying attributes that can be used to describe all situations, a daunting task. Gilbert and Malone (1995) observed that “when one tries to point to a situation, one often stabs empty air. Indeed, the constructs to which the word situation refers often have little or no physical manifestation” (p. 25).

Thus far, researchers have suggested that situations can be described along three conceptual levels (Block & Block, 1981; Saucier, Bel-Bahar & Fernandez, 2007). Level 1, the broadest level, involves objective aspects of situations that are relatively resistant to differences in perception. According to Saucier et al. (2007), this includes factors such as temperature and the number of people present. Level 2 involves describing situations in terms of an over-arching characterization that most people in the situation would agree upon, such as a research seminar, a funeral, a party, and so on. Level 3 is comparatively subjective and involves properties of situations that are psychologically provoking and for each individual the specific provoking properties may be different.

The problem with Level 1 description is that it is unlikely to capture the psychologically active features of situations. Gilbert and Malone (1995) suggest that more subtle aspects, such as another person smiling or making eye contact with a person, are likely to provoke psychological reactions. The problem with Level 3 is that it is too solipsistic and renders the study of persons in situations impossible (Reis, 2008). Indeed, it absorbs the analysis of situations into the analysis of persons, because it would require, for example, that a “noisy party” be described as attractive for an extravert and aversive for an introvert.  Redescribed in that way, the noisy party disappears.  In contrast, Level 2 is at a mid-level of analysis that has the potential to reveal psychologically active features of situations in an empirical, reasonably consensual manner.

The Riverside Situational Q-set (RSQ: Funder & Wagerman, 2008; Wagerman & Funder, in press) has recently been offered as a new method to describe situations at Level 2. Similar to The Riverside Behavioral Q-set previously discussed, the items of the RSQ derive from the items of the California Adult Q-Set (CAQ: Block, 1961; Bem & Funder, 1978). The 100 CAQ items were examined for situational relevance and re-worded to describe characteristics of situations that afford the opportunity to express each of the corresponding personality characteristics.  For example, the CAQ item, “is critical, skeptical, not easily impressed” has an associated RSQ item that reads, “Someone is trying to impress someone or convince someone of something.” The idea is that the degree to which an individual is critical or skeptical might be revealed in a situation where another person is trying to be impressive or convincing.  The advantage of basing the RSQ on the CAQ is that items are specifically intended to describe aspects of situations that are personologically relevant.

It is possible and in fact likely that the particular items of the RBQ and the RSQ fail to address all the essential attributes of behavior and of situations, and it is certain that their representation of these domains is incomplete.  The development of theoretical conceptions of behaviors and situations, and of the assessment tools to make these conceptions addressable through empirical research, is a long-term project that will take long years of work by many different investigators before it comes to fruition.  The intention of the present section of this chapter is simply to suggest that it is time to begin.