Do people believe mental disorders are real and possess underlying essences? The current study found that both novices and practicing clinicians held weaker essentialist beliefs about mental disorders than about medical disorders. They were also unwilling to endorse the idea that mental disorders are real and natural. Furthermore, compared with novices, mental health clinicians were less likely to endorse the view that there is a shared cause underlying a mental disorder and that one needs to remove the cause to get rid of the mental disorder. Clinicians were polarized on their views about whether mental disorders are categorical or dimensional. These findings reflect current controversies about mental disorders in the field at large.Prior to the understanding of sex chromosomes, gender was distinguished on the basis of cooccurrence of surface features (e.g., voice, height; Rosch & Mervis, 1975). Even so, gender categories were almost certainly treated as real and natural: People most likely believed that an underlying, hidden essence made a man a man or a woman a woman, although they did not know what that essence was (Medin & Ortony, 1989). The current understanding of mental disorders presents an analogous situation. That is, knowledge about the etiology of most mental disorders in the current taxonomy is incomplete (e.g., American Psychiatric Association, APA, 2000). Do people nevertheless believe that mental disorders are real, exist naturally, and possess hidden essences?There are reasons to predict that people believe mental disorders are real and possess essences (e.g., Kendell, 1986). For instance, past research suggests that laypeople have a tendency to essentialize many forms of human groupings, such as racial and ethnic categories (e.g., Hirschfeld, 1995). Furthermore, proponents of the disease (or medical) models for mental disorders maintain that each disorder is universal and has a biologically based etiology with discrete boundaries (see Haslam, 2000, andKiesler, 1999, for reviews). If people hold disease models for mental disorders, they would view disorders as real and as having essences.Clinicians, in particular, may have such a view because of their experience with and knowledge about the domain. For instance, the symptom-level descriptions adopted in the contemporary versions of the Diagnostic and Statistical Manual of Mental Disorders (DSM; APA, 1980APA, , 1987APA, , 1994 are often deemed (e.g., Kihlstrom, 2002;Kupfer, First, & Regier, 2002) to have been built on the Kraepelinian assumption: "Cases of mental disease originating in the same causes must also present the same symptoms, and the same pathological findings" (Kraepelin & Diefendorf, 1904/1907 However, mental disorders instead might be seen as nominal and constructed by culture (e.g., Cohen, 1981;Sedgwick, 1982). The relatively well-known fact that the etiologies of most mental disorders are unknown might make both experts and novices hesitant to endorse even the possibility of essences or reality for these disorders. Fur...

The purpose of this study was to examine whether contextual information about patients' clinical presentations affected clinicians' judgments of conduct disorder symptoms. Forty-five clinicians read vignettes describing hypothetical patients who displayed one conduct disorder symptom alongside information about the patients' home, school, and peer contexts. Clinicians judged the likelihood of patients meeting conduct disorder criteria. Contextual information highly affected judgments and these effects varied across the 15 conduct disorder symptoms. It is important to note that clinical judgments were not in agreement on the symptoms affected by context.

It is generally agreed that our concepts are formed around clusters of correlated features (Rosch, 1978). For instance, having wings is more likely to occur with being able to fly and having feathers than with being able to swim and having gills. Thus, we give the name bird to creatures that have wings and feathers and are able to fly. Within these correlation clusters, there are specific feature correlations that are more important and explicit in our conceptual representations.In our bird concept, having wings and being able to fly would be more closely tied to each other than having feathers and chirping. The goal of the present study is to examine which feature correlations, among all possible pairs in a cluster, are explicitly represented in concepts and affect typicality ratings of exemplars beyond the contribution of their individual features.The work on people's naive theories of natural categories (Carey, 1985;Keil, 1989;Murphy & Medin, 1985) sheds some light on our goals. According to this approach, people do not simply notice and record statistically correlated propertieswhen learning categories. Instead, people's knowledge of a domain emphasizes only a subset of the statistical correlations that occur. We believe that this possibility has yet to be tested, especially with natural categories, as will be elaborated upon below. In our studies we will strive to (1) demonstrate that explicitly represented feature correlations in concepts are theory based, (2) examine the content of these naive theories, and (3) demonstrate that these theory-based, explicitly represented feature correlations are more determinativeof category members' typicalitythan are theory-neutral feature correlations. To achieve these goals, we must first review previous studies on the effect of correlated features, particularly noting studies that involved correlated features in real-life categories. Effect of Correlated Features Found in Previous StudiesAre people sensitive to all correlations among features when learning novel categories? In an attempt to answer this, Medin, Altom, Edelson, and Freko (1982) had participants learn artificial categories. When asked to judge category membership for new items, those that preserved the study phase correlation were more likely to be selected as members, even when the item with the preserved correlation contained fewer typical features (see also Wattenmaker, 1991Wattenmaker, , 1993. However, these results might not generalize to realworld cases, because the correlation between features was perfect, unlike in natural concepts (e.g., having wings and being able to fly). Furthermore, in these studies, a rule involving the correlated features was more diagnostic and 107Copyright 2002 Psychonomic Society, Inc.This project was supported by National Institute of Mental Health Grant RO1 MH57737, awarded to the first author. We thank Kathy Johnson, Barbara Malt, Thomas Palmeri, and Steven Sloman for their extremely helpful comments on earlier drafts of this paper. We also thank Barbara Malt for pro...

The average person possesses superficial understanding of complex causal relations and, consequently, tends to overestimate the quality and depth of their explanatory knowledge. In this study, we examined the role of this illusion of explanatory depth (IOED) in politics—inflated confidence in one's causal understanding of political phenomena—for endorsement of conspiracy beliefs. Utilizing a pre‐/post‐election panel design and a large sample of U.S. Citizens (N = 394) recruited in the context of the 2016 presidential election, we provide evidence that political IOED, but not a non‐political IOED, was associated with increased support for general and election‐specific conspiracy beliefs, particularly among political novices and supporters of the losing candidate. We find this pattern of results net the influence of a broad range of variables known to covary with conspiracy beliefs. Implications for theory and the need for future research are discussed.

Dennis and Ahn (2001) found that during contingency learning, initial evidence influences causal judgments more than does later evidence (a primacy effect), whereas López, Shanks, Almaraz, and Fernández (1998) found the opposite (a recency effect). We propose that in contingency learning, people use initial evidence to develop an anchoring hypothesis that tends to be underadjusted by later evidence, resulting in a primacy effect. Thus, factors interfering with initial hypothesis development, such as simultaneously learning too many contingencies, as in López et al., would reduce the primacy effect. Experiment 1 showed a primacy effect with learning contingencies involving only one outcome but no primacy effect with two outcomes. Experiment 2 demonstrated that the magnitude of the primacy effect correlated with participants' verbal working memory capacity. It is concluded that a critical moderator for exhibition of the primacy effect is task complexity, presumably because it interferes with initial hypothesis development.

Existing models of causal induction primarily rely on the contingency between the presence and the absence of a causal candidate and an effect. Yet, classification of observations into these four types of covariation data may not be straightforward because (a) most causal candidates, in real life, are continuous with ambiguous, intermediate values and because (b) effects may unfold after some temporal lag, providing ambiguous contingency information. Although past studies suggested various reasons why ambiguous information may not be used during causal induction, the authors examined whether learners spontaneously use ambiguous information through a process called causal assimilation. In particular, the authors examined whether learners willingly place ambiguous observations into one of the categories relevant to the causal hypothesis, in accordance with their current causal beliefs. In Experiment 1, people's frequency estimates of contingency data reflected that information ambiguous along a continuous quantity dimension was spontaneously categorized and assimilated in a causal induction task. This assimilation process was moderated by the strength of the upheld causal hypothesis (Experiment 2), could alter the overall perception of a causal relationship (Experiment 3), and could occur over temporal sequences (Experiment 4). Keywordscausal reasoning; categorization; ambiguity interpretation; belief assimilation People appear to learn many novel, causal relations without much difficulty: They do not need intensive training to learn that flipping a switch causes a light to be turned on or that eating spoiled food causes stomachaches. Such apparently straightforward causal learning is actually impressive, given that the environment does not always present a learner with neatly organized data. In the current set of studies, we examine one of the difficulties people have to face as a result; namely, that observations may not come readily classified as evidence for or against a causal hypothesis.Let us illustrate this problem in the context of covariation-based models of causal induction. The focus of these models has been on how people use covariation data to learn causal relations (e.g., Cheng, 1997;Cheng & Novick, 1990;Collins & Shanks, 2002;Jenkins & Ward, 1965;Rescorla & Wagner, 1972;Shanks, Lopez, Darby, & Dickinson, 1996). In these models, Correspondence concerning this article should be addressed to Jessecae K. Marsh, Department of Psychology, Box 2051, Psychology Building, Texas Tech University, Lubbock, TX 79409-2051. jessecae.marsh@ttu.edu. This article was submitted as part of Jessecae K. Marsh's dissertation project. A discussion of Experiments 1 and 2 can be found in a section of the book chapter "Dynamic Interpretations of Covariation Data" in the edited book Causal Learning: Psychology, Philosophy, and Computation (2007). NIH Public Access NIH-PA Author ManuscriptNIH-PA Author Manuscript NIH-PA Author Manuscript covariation data are preclassified into four events: What happens to an effect w...

Decisions about whether a person is in need of mental health care are often made by laypeople with no training in the identification of mental health concerns. For example, the parent of a child displaying problematic behavior has to decide whether this behavior is likely related to mental health concerns and necessitates professional care. The process of identifying mental health concerns is made more complicated by the rich background of real-world environmental factors or contexts in which concerns can present-contexts that might or might not relate to the presence of mental health concerns. We investigated whether laypeople use contextual information to make judgments regarding childhood mental health disorder symptoms. In Experiment 1, we demonstrated that laypeople's judgments of the likelihood of a mental disorder are influenced by non-diagnostic contextual information that surrounds symptoms of the disorder. In Experiment 2, we demonstrated that providing a causal origin for such disorder symptoms accentuates the use of context, regardless of the nature of the causal process (i.e., environmental vs. biological). These findings indicate that contextual influences on judgments about mental health concerns may reflect a more general set of mental reasoning processes than indicated by previous work focused on clinicians' judgments. Consequently, these findings have important implications for how we think about the influence of contextual information on decision making more generally as well as for improving our ability to both reliably identify children in need of mental health care and increase children's access to such care.

We have seen massive global behavioral change as billions of people radically altered their ways of life in response to COVID-19. Here, we review how research on conformity and deviance can inform understanding of and effective responses to the pandemic. Group identities are critical for understanding who is influenced by whom, as well as how partisan divisions can obstruct cohesive and coordinated action. We identify several questions highlighted by the pandemic, including when people will react more harshly to ingroup members who violate health-protective norms (black sheep effect) or to outgroup members violating the same norms (intergroup hypocrisy). As a working hypothesis, we propose a goal-based approach, positing that differentially negative reactions to ingroup and outgroup deviants are likely influenced by the relative salience of goals to protect ingroup image, enforce group norms, maintain intergroup boundaries, and seek justice for potential victims put at risk by norm-violating behavior.