Accepting personality test feedback: A review of the Barnum effect

This article attempts a comprehensive and critical review of the by-now fairly extensive literature on the Barnum effect—the approval/acceptance by subjects of bogus personality interpretations supposedly derived from standard tests. Since the last major review eight years ago various methodological extensions have occurred and various rival hypotheses for established findings have been proposed. The present review is divided into three major sections: client and clinician characteristics; feedback statements and test format; and implications for personality assessment and measurement. Nearly 50 studies on the acceptance of personality interpretations are systematically reviewed and criticized.     

The Barnum effect in a group setting

The present study expands the typical Barnum effect paradigm by investigating reactions to feedback in a group setting. People initially participated in a group experience (8 per group), and then were given bogus positive or negative feedback (the favorability manipulation) purportedly prepared by either the group leader or another group member (the source status manipulation). The answerability manipulation led participants to believe that they either would or would not have to share their reactions to the feedback with the person who prepared it. Positive feedback was rated as more accurate and accepted more highly than the negative feedback, though no differences in recall of either the positive or negative feedback emerged. Additionally, a pattern of results revealed that the feedback from the group leader generated greater perceived accuracy, acceptance, and recall than did feedback from another group member. Implications for the impact of feedback given by a high status person in group settings are discussed.     

Paranormal Beliefs and the Barnum Effect

The acceptance phenomenon (i.e., the general tendency for persons to accept almost any bogus personality feedback; Layne, 1979) may be related to certain paranormal beliefs (Alcock, 1981; Hyman, 1981). Paranormal beliefs are beliefs in phenomena that, if authentic, violate basic limiting principles of science (Broad, 1953; Tobacyk & Milford, 1983). One specific illustration of the acceptance phenomenon is the "Barnum effect," which refers to acceptance of bogus personality feedback consisting of relatively trivial statements with a high base rate (Layne, 1979; Snyder, Shenkel, & Lowery, 1977). The Barnum effect has been implicated in personal validation and in "cold reading," both of which are used routinely by many astrologers, clairvoyants, faith healers, fortune tellers, graphologists, mediums, Tarot Card readers, and others (Hyman, 1981). It was hypothesized that paranormal beliefs emphasizing divinatory procedures that produce personalized feedback (i.e., precognition, spiritualism, superstition, and witchcraft) are associated with greater susceptibility to the Barnum effect.     

The Barnum effect in a computerized Rorschach interpretation system

Twelve psychiatric outpatients were administered the Rorschach test, and results were interpreted using the Exner (1983, 1986) Report for the Comprehensive System computer-based test interpretation (CBTI) program. Four psychiatrists made accuracy ratings for both real and bogus reports for each of their patients. Data were analyzed using two-way analyses of covariance (ANCOVAs), where report type was a repeated main effect, psychiatrist was a random main effect, and the number of statements in the report was the covariate. Results indicated that this CBTI provided only 5% discriminating power for any one patient, with 60% of the interpretive statements merely describing typical characteristics of the outpatient population. No significant psychiatrist, interaction, or covariate effects were encountered.     

Scaling the helpfulness of antidepressive activities

This study examined the feasibility of obtaining quantitative estimates of the subjective helpfulness of antidepressive activities using an interval scale that had previously been found suitable for rating unpleasant qualities. The accuracy with which psychology students could predict the mean item ratings for their group was also investigated. Findings suggested that the scale is suitable for rating the positive quality of helpfulness and that on the whole students had an accurate idea of how helpful their peers would rate the activities; their idea appeared to be independent of how helpful they themselves found an activity. Some implications of the findings for cognitive dimensions of antidepressive behaviour, analogue research, and clinical practice are discussed.     

Note on the rationality of rule-based versus exemplar-based processing in human judgment

This paper reports a study of the relationship between rule- versus exemplar-based processing and criteria for rationality of judgment. Participants made probability judgments in a classification task devised by S. W. Allen and L. R. Brooks (1991). In the exemplar condition, the miscalibration was accounted for by stochastic components of the judgment with a format-dependence effect, implying simultaneous over- and underconfidence depending on the response scale. In the rule condition, there was an overconfidence bias not accounted for by the stochastic components of judgment. In both conditions the participants were additive on average and reasonably transitive, but the larger stochastic component in the exemplar condition produced somewhat larger absolute deviations. The results suggest that exemplar processes are unbiased but more perturbed by stochastic components, while rule-based processes may be more prone to bias.     

Assessment of the Barnum Effect in Computer-Based Test Interpretations

A prototypic experiment for validating computer-based test interpretations (CBTIs) was conducted. Undergraduates (N = 63) completed the Comprehensive Personality Profile Compatibility Questionnaire (CPPCQ; Craft, 1987). One treatment group rated real CBTIs for relative accuracy, and another group rated bogus CBTIs. A significant main effect for differences in ratings indicated that the real CPPCQ profiles were rated as 74.5% accurate whereas the bogus CBTIs were rated as 57.9% accurate. Several covariate effects were tested, but none were significant.     

Reconceiving rationality: situating rationality into radically enactive cognition

Rational beliefs and actions are typically evaluated against certain benchmarks, e.g., those of classical logic or probability theory. Rationality therefore is traditionally taken to involve some sort of reasoning, which in turn implies contentful cognition. Radically Enactive (and Embodied) views of Cognition (REC), on the other hand, claim that not all cognition is contentful. In order to show that rationality does not need to lie outside of REC’s scope of radicalizing cognition, I develop a Radically Enactive notion of Rationality (RER), according to which rationality is embodied, situated and contentless. For RER, an organism acts rationally insofar as it sustains a proficient interaction with its environment, which in turn requires the coordination of cognitive abilities in accordance with environmental constraints. Rationality is thus distinguished from reasoning, for reasoning is understood as a capacity to coordinate representational cognitive abilities.

     

Perceptions of helpfulness and behavior in groups

Before, midway, and late in a 50-hour small group experience, 28 persons separately and subjectively appraised the helpfulness of 60 items representing Yalom's (1975) 12 curative factors of group psychotherapy. Items from the catharsis and interpersonal learning factors proved highly and increasingly helpful, despite their poor internal consistency. Item helpfulness also linked to members' ingroup behavior, as measured by peer ratings of acceptance-rejection of self and others. (Hurley, 1976). Persons more accepting of self and others regarded family reenactment items as more helpful, and advice and guidance items as less helpful, than did members who were more rejecting of self and others.     

Solo pioneers: Single-minded rationality

There is an increasing number of single women in the United States. These women are having to confront new emotional problems as well as previously unknown opportunities for personal growth and development. The current article identifies two polarities of female singlehood, i.e., the desperate, dependent woman and the solo pioneer woman. The cognitive and behavioral characteristics of the two polar types are delineated.Desperate, dependent women endorse irrational beliefs which involve their self-worth being contingent upon the love and approval of men. Solo pioneers are autonomous in their personal identity and are able to construct meaningful lives for themselves outside of the context of marriage. Single mothers by choice are presented as being a prototype of solo pioneers.Laura Primakoff, Ph.D. has developed a cognitive-behavioral treatment program for individuals living alone (Primakoff, 1983).     

Ambiguity and rationality

Recently, several theories of decision making and probability judgment have been proposed that take into account ambiguity (Einhorn and Hogarth, 1985; Gardenfors and Sahlin, 1982). However, none of these theories explains exactly what the psychological causes of ambiguity are or addresses the issue of whether ambiguity effects are rational. In this paper, we define ambiguity as the subjective experience of missing information relevant to a prediction. We show how this definition can explain why ambiguity affects decisions in the ways it does. We argue that there are a variety of rational reasons ambiguity affects probability judgments and choices in the ways it does. However, we argue that the ambiguity effect does not cast doubt on the claim that utility theory is a standard of rational choice. Rather, we suggest that the effect of ambiguity on decisions highlights the fact that utility theory, like any normative model of decision making only prescribes the optimal decision, given what one knows.     

Representation and rationality

David Lewis (1974, 1994/1999) proposed to reduce the facts about mental representation to facts about sensory evidence, dispositions to act, and rationality. Recently, Robert Williams (2020) and Adam Pautz (2021) have taken up and developed Lewis's project in sophisticated and novel ways. In this paper, we aim to present, clarify, and ultimately object to the core thesis that they all build their own views around. The different sophisticated developments and defenses notwithstanding, we think the core thesis is vulnerable. We pose a dilemma by considering the two sides of a current epistemological controversy over the relation between evidence and rational belief: permissivism vs. uniqueness. As we argue, the prospects for the Lewisian project look dim when either supposition is clearly made.