How do psychologists use if/then contingencies as a way of describing someone personality?

One of the goals of personality psychology is to be able to describe and predict individuals' behavior with maximum accuracy. Historically, and across a wide variety of linguistic and cultural contexts, one of the most popular and widely accepted methods for attempting to do so has been through traits (John, Naumann, & Soto, 2008). Traits are long-lasting attributes of personality that presumably involve stability across different situations (John & Gosling, 2000). Under an extreme version of a trait theory, there would be no behavioral variation attributable to situations.

It is, however, a commonplace observation that individuals do not behave consistently regardless of where they are or whom they are with; people often vary their behavior to fit the situational context. Therefore, to understand the whole person, or to capture the patterns accurately, both situations and behavioral attributes must be accounted for (Bem and Allen, 1974, Fleeson, 2004, Funder, 2006, Mischel and Shoda, 2000). This paper will examine the relationship between situations and trait-variation, in the interest of developing a fuller understanding of how individual personality is exposed.

Trait models of personality assume cross-situational consistency (John & Gosling, 2000). That is, the personality terms are used in a decontextualized way and instructions for the questionnaires often ask individuals to generalize or average across situations; at issue is whether the person is outgoing, kind, etc., in general, as opposed to in one or another specific situation. Using trait factors to describe personality provides a summary of an individual's behavior more or less averaged across situations. It assumes consistent trait-behavior, e.g., individuals who are extraverted should be outgoing both at a party and if they are working at a job. It also assumes stability over time, that an individual's trait scores would remain the same at one time point as at any other time point. The interaction of situations and stability is not easily checked in studies of retest stability because the retest may involve the same averaging across situations, rather than the individual being described in a new or different situation.

In his 1968 book, Personality and Assessment, Walter Mischel expressed the belief that the predictive ability of traits is severely limited; that individuals vary so much from situation to situation and over time that there is no such thing as stable personality. Mischel (2004) cited studies that found that the average correlation coefficient for daily behavior across situations was about 0.14 (Newcomb, 1929), and that the assumption that rank-ordering of individuals on trait behavior would remain stable across situations was unsupported (e.g., Hartshorne & May, 1928). This perspective generated a new line of research that focused on personality as dynamic and variable rather than stable over time and place. In this perspective personality is not consistent across situations, thus behavior is contingent on situational context. It was not made clear, however, which types of situations have the greatest effect on behavior overall, or on particular behavioral tendencies, or even which situation variables were most worthy of attention. (Leaving aside the point that Mischel's approach underestimated the power of aggregation.)

Much has changed, of course, in personality psychology since Mischel's 1968 book. In particular, there has been a revolution in understanding of which traits should have most priority in measurement. From the early years of psychology, a rationale for identifying the most important personality variables has been advanced: take all the words that can be used to describe other people (Allport and Odbert, 1936, Galton, 1884, Norman, 1963) and reduce them down into a few broad variables, called factors (Saucier & Goldberg, 2001). Large samples of individuals would rate themselves or others on all these personality descriptors, and then factor analysis would be used to identify which descriptors have something in common, would be given a common label, and used henceforth as key measured variables of personality.

It was out of this approach that the Big Five personality factors were developed, including Extraversion, Agreeableness, Conscientiousness, and Emotional Stability, as well as Intellect (Goldberg, 1990), for which prominent questionnaire versions are labeled Openness or Open-mindedness (Soto & John, 2017). The Big Five sprung from early studies of personality language in English, Dutch, and German. Ashton et al. (2004) later found that across seven different languages, a six-factor model emerged. Five of the factors resembled those of the Big Five (albeit with some moderate differences in agreeableness and in emotionality vs. emotional stability), with an additional factor they labeled Honesty-Humility. However, a sizable group of lexical studies was omitted from their analyses. Saucier (2009) conducted a study that also included languages with more inclusive variable selections and likewise found the Big Five to have one factor too few. A slightly variant sixth factor, called Honesty/Propriety, helped to account more fully for the content in common between personality terms. An up-to-date understanding of the relations of situations with trait variation should take account of the Big Five (and Big Six) model.

Fleeson (2001) found evidence partly supporting a situational perspective in the spirit of Mischel, but also partly supporting the validity of trait models. He examined Big Five-relevant behavior over 2–3 weeks and found that, while there was very low predictability hour-to-hour, when scores were averaged for the 2–3 week duration, there was very strong predictability for behavior. The Big Five trait model still explained behavior, especially when aggregated over time. Borkenau and Ostendorf (1998) suggested that perhaps the Big Five could also be used to account for intra-individual differences in behavior (i.e., differences across situations for the same individual), though this has not yet become common practice beyond the work of Fleeson.

The evidence thus far suggests that both situational context and trait factors are important when describing personality. This idea, that personality includes the interaction between traits and situations, has been increasingly salient within personality psychology in more recent years (Eigenhuis, 2010, Fleeson, 2004, Funder, 2006, Sherman et al., 2015). In fact, this idea had been suggested before by researchers who worked from the theory of interactionism (Endler & Magnusson, 1976). Interactionism is the belief that personality is contingent on individual differences in reaction to situations rather than general trait behaviors. As an illustration, Shoda, Mischel, and Wright (1994), had counselors at a children's summer camp rate campers on several trait-like behaviors while involved in different camp activities. They found that the children's behavior patterns were best depicted as individual ‘profiles,’ described as ‘if…then’ contingencies. For example, for one child, being teased by a peer led to acting aggressively, but being praised by an adult led to acting friendly. For another, being praised by an adult led to acting aggressively, but being warned by an adult led to acting friendly. These patterns varied between children.

These findings from Shoda et al. (1994) indicate that there may be specific patterns to the variation in individual behavior across different situations. In other words, an individual will tend to have a consistent pattern in behavior for one specific situation versus another, and individuals will vary in this pattern. The research that follows builds on a similar assumption, as well as an if-then-contingency rationale but relates a wide range of situations to contemporary models of personality structure. The major questions addressed in the research are the following. To what degree do various situations have an impact on perceived variation in various traits? Are some traits more versus less susceptible to situational influence? For a given trait, are particular situations especially prone to yield dispositional variance? And, related to the last question, is behavior variation within certain situations (but not all situations) diagnostic of individual differences in a trait dimension, as measured by a commonly used personality questionnaire?

Participants

Participants were recruited through a university human-subjects pool. This pool consists of students from introductory psychology and linguistics classes. Participation was voluntary and for course credit. Students were given the option of choosing from several studies in which to participate or writing a paper instead.

The total original number of participants was 569. Data from 69 participants was removed (N = 500) because self-reports contained mostly the same repeated response or the time

Analyses and results

The original dataset was an “attributes/situations across persons” (ASxP) data matrix that included each of the 15 attribute items for each of the 29 situations (e.g., “Is outgoing while alone,” “Is quiet while alone,”…) as the 435 columns, with one row for each participant (i.e., 500 rows). For all analyses, each pair (or triad) of items that measured an attribute were aggregated so that there was a single score for each of the six trait-dimensions for each situation (e.g., the prior examples

Discussion

This study was aimed at assessing whether self-ratings on the BFI-VI (intended to measure stable personality traits as on the typical questionnaire) were indicative of the salience of certain situations in terms of how they might inform individuals' ratings of their own overall personality. This relationship between situations and personality was also tested by examining the influence (or lack thereof) that situations have on trait-behavior variance. Here we discuss some of the most noteworthy

Conclusion

In order to understand personality comprehensively, effects of situational context need to be taken into account. The present study examined how situational context influences the diagnosis of an individual's trait levels. This led to several crucial findings. Personality traits seem to differ in how much they vary by situation; particularly, Honesty/Propriety tends to vary the least across situations. Also, generally, Big Six traits may in fact be somewhat situation contingent (e.g.,

Recommended articles (6)