8.1 Introduction

A chapter on a psychiatric category may seem out of place in a book on cognitive ontology. But this chapter argues that the psychiatric category Body Dysmorphic Disorder (BDD) has a basis in perception and cognition. Specifically, we will propose a tentative causal model of the disorder that posits that it has certain key perceptual and cognitive deficits at its core, and argue that this causal profile makes BDD a strong candidate for being a real kind. Based in large part on this causal model, we will contend that BDD has been misclassified in the fifth and most recent edition of the Diagnostic and Statistical Manual of Mental Disorders (2013), DSM-5, which provides the most widely accepted taxonomy of psychiatric disorders. But before proceeding, we will try in this introductory section to respond to some common concerns about psychiatric taxonomy, specifically concerns that psychiatric categories cannot correspond to real kinds.

There has been growing discussion of psychiatric taxonomy in the philosophical literature, and there are a range of views on whether various different psychiatric categories correspond to real kinds (see e.g. Kincaid & Sullivan Reference Kincaid and Sullivan2014; Pober Reference Pober2013; Samuels Reference Samuels, Broome and Bortolotti2009a; Tsou Reference Tsou2013; Weiskopf Reference Weiskopf2017a; Ylikoski & Pöyhönen Reference Ylikoski and Pöyhönen2015). We will not be able to do these discussions justice here and will not attempt to situate our view in relation to them in any detail, but we will outline an account of psychiatric kinds and indicate some ways in which we diverge from some of the prevalent accounts of psychiatric disorders. While we will not try to make the case that the superordinate category psychiatric disorder corresponds to a kind, we will argue that BDD is a good candidate for being a psychiatric kind. As in some of the other cases discussed in this book (e.g. memory and episodic memory; bias and myside bias), we think it is possible for the subordinate category to correspond to a kind but not the superordinate category.

The view of psychiatric disorders that we are operating with considers them to be objective mental states or dispositions and ones that are individuated relative to social contexts. We are in broad agreement with the “harmful dysfunction” view of psychiatric disorders elaborated by Wakefield (Reference Wakefield1992) and other writers on psychiatric or mental disorders, with two important caveats. According to Wakefield (Reference Wakefield1992, 385), “A condition is a mental disorder if and only if (a) the condition causes some harm or deprivation of benefit to the person as judged by the standards of the person’s culture (the value criterion), and (b) the condition results from the inability of some mental mechanism to perform its natural function, wherein a natural function is an effect that is part of the evolutionary explanation of the existence and structure of the mental mechanism (the explanatory criterion).” Although we agree that a mental disorder is a psychological malfunction of a certain type, we do not think the function at issue is exclusively a “proper function” in the evolutionary sense, and malfunctions are hence not necessarily biologically maladaptive (cf. Murphy Reference Murphy2006). Rather, malfunctions can be identified with breakdowns in systems or capacities that might not be associated strictly or directly with a biological adaptation of some kind. For example, some researchers have proposed that schizophrenia represents an evolutionary adaptation (for a review, see Polimeni & Reiss Reference Polimeni and Reiss2003). While the hypothesis may be implausible and has not had wide uptake, it does not seem possible to dismiss it on the grounds that psychiatric disorders are necessarily maladaptive. Functions in this context can be understood in terms of synchronic causal roles, and dysfunctions may involve disruptions in social interactions. The second caveat is that we do not think that identifying the harm involved in a psychiatric disorder involves a value judgment on the part of the clinician, as some writers appear to think. In this context, harm is understood in terms of systematic disruption to the ability of an individual to achieve their goals, avoid suffering, and establish meaningful relationships. This means that the psychiatrist who diagnoses a disorder need not endorse the values prevalent in the relevant social context that lead to the individual’s being harmed. While the individual may suffer harm as a result of social pressures to conform or to fulfill certain duties, due to certain values held by those in the encompassing community, those values need not be shared by a researcher who is interested in categorizing the individual in question. We think that this shows that the researcher who categorizes an individual as having a psychiatric disorder is not thereby making a value judgment. A further point worth emphasizing is that the individual’s abilities (to achieve their goals, avoid suffering, and establish meaningful relationships) are all exercised in a social setting, which means that they are relative to a social context. If an individual has certain psychological malfunctions that systematically and consistently disrupt these abilities and prevent their flourishing, they can be considered to suffer from a psychiatric disorder.

This account may invite the charge that it makes psychiatric conditions socially and culturally relative. But even though there may be some variability in what constitutes a psychiatric disorder on such a view, we do not think that this leads to strong a form of cultural relativism. The kinds of dysfunction that are at issue are ones that lead to disruptions in social relationships, vocational pursuits, and leisure activities, and they will apply across a wide range of human societies. Of course, it may be that some of the disorders listed in the DSM-5 do not qualify under this heading, but we would not regard that as a drawback for this view, since the DSM-5 cannot be considered the final word on what qualifies as a psychiatric disorder. It may also be the case that the current classification system of the DSM casts its net too wide, considering some conditions to be disorders that should not be so regarded. But when it comes to the particular disorder that we are discussing in this chapter, we think that it is clear that it conforms to the picture of psychiatric disorders that we have proposed, as we will try to show in due course.

A closely related concern is that this account of psychiatric disorders might be thought to give rise to what might be called the “drapetomania objection.” The American physician Samuel Cartwright argued in the mid-nineteenth century that American slaves who tried to escape were afflicted with the compulsion to flee, labeling this psychiatric disorder “drapetomania” (see e.g. Wakefield Reference Wakefield1992, 386; Murphy Reference Murphy2006, 27–28). How are we to avoid the conclusion that since this alleged condition arguably prevented its supposed sufferers from flourishing in their society at that historical juncture, it can rightly be considered a psychiatric disorder relative to that society at that time? The response is that genuine psychiatric disorders involve psychological dysfunctions, which can be assessed in terms of the individual’s perceptual, cognitive, and affective abilities. Given that they involve psychological dysfunctions, psychiatric disorders do not just entail that those who suffer from them will violate local norms, but rather that they will issue in behaviors that are harmful to the individual in a broad range of social settings and result in thwarting their ability to achieve their goals. In particular, the harms that ensue cannot just be a result of labeling on the part of the society and the stigma attached to such labeling, since they issue from definite psychological dysfunctions.

If psychiatric disorders are identified with reference to the ability to function in a social setting, does that mean that psychiatric disorder is not a real kind, as characterized in Chapter 1 and subsequent chapters? We will not try to give a definitive answer to this question, but if it is a real kind, it is a relational kind, like many other kinds in both the basic and special sciences, and in this case the crucial relations are social ones. However, we would argue that individual psychiatric disorders may be real kinds, but not the superordinate category psychiatric disorder.Footnote ¹ On this view, having a psychiatric disorder has real causes and effects, though both the causes and effects may be diverse (biological, psychological, and social).Footnote ² This view of psychiatric disorders is generally in accord with the dominant “biopsychosocial model” that is widely adopted in psychiatry and the philosophy of psychiatry. Where we depart from this model is in doubting that each psychiatric disorder has a uniform biological cause (which may not currently be known).Footnote ³ In some cases, the account of a psychiatric disorder will be “psychosocial” rather than “biopsychosocial.” Moreover, we think that the reason that psychiatric disorders are identified as such in the first place is because their effects are primarily of a psychological and social nature, involving social harm and dysfunction, though they may also have biological causes and effects as well.Footnote ⁴ Hence, as in the case of several other cognitive kinds discussed in this book, it is not just that psychiatric disorders have social causes (among other causes), they are partly individuated in terms of social factors. The broader social context enters into their very identification as psychiatric disorders.

A prominent alternative to the view of psychiatric disorders that we have just outlined conceives of them primarily as dysfunctions of biological or neurophysiological systems (see e.g. Tsou Reference Tsou2016). But we would maintain that, even though there will be underlying neurophysiological bases for psychiatric disorders, these may be multiply realized, in accordance with the endorsement of multiple realizability for psychological kinds in earlier chapters. Also, in keeping with a theme sounded repeatedly in this book, the very same neural bases may not be correlated with a psychiatric disorder in different contexts. So the relationship between psychiatric disorders and neural states or dispositions may be many-to-many. For that reason, we would not expect to find that every psychiatric disorder can be identified with a single neurophysiological dysfunction, though this may be the case for some of them.

In this section, we have tried to address some common concerns about the status of psychiatric disorders as real kinds. Our arguments are certainly not decisive and this is not meant to be a full-blown defense of the real kind status of psychiatric disorders. In particular, there are wide variations in the types and levels of harm that ensue from psychiatric disorders, which may render them too heterogeneous to constitute a kind. A specific condition like BDD might be a kind of psychiatric condition with a unified causal profile even though it does not belong to an overarching kind that includes BDD and all (or even a significant number of) the other conditions that are commonly considered psychiatric disorders. In the rest of this chapter, we will begin by describing some of the main features of BDD, drawing on a considerable body of empirical and clinical evidence (Section 8.2). We will then go on to present some preliminary reasons for thinking that BDD has been misclassified in the DSM-5 (Section 8.3). We will then propose a tentative causal model of BDD, which emphasizes the internal states of individuals with BDD rather than simply their outward behaviors (Section 8.4). This will allow us to justify the claim that BDD is a real psychiatric kind. Finally, we will respond to some objections concerning our causal model of BDD and our proposed reclassification (Section 8.5), before coming to a conclusion (Section 8.6).

8.2 Characterization of BDD

According to the DSM-5, BDD involves persistent and intrusive thoughts about a perceived bodily flaw that is not observable or appears slight to others. At some point during the course of the disorder, the individual will have performed repetitive behaviors (e.g. mirror checking, excessive grooming, skin picking, reassurance seeking) or mental acts (e.g. comparing his or her appearance with that of others) in response to concerns about appearance. The preoccupation causes clinically significant distress or impairment in social, occupational, or other areas of functioning. Moreover, the preoccupation with appearance is not better explained by concerns with body fat or weight, and symptoms do not otherwise meet diagnostic criteria for an eating disorder.Footnote ⁵

There has been considerable empirical research, in both clinical and laboratory settings, on the main characteristics of BDD and its distinctive features. This work has revealed a number of important facets of the disorder that set it apart from other psychiatric conditions. In a seminal early monograph on BDD, Phillips (Reference Phillips1996/2005) noted that neuropsychological studies suggest that those with BDD tend to over-focus on minor details when drawing complex figures from memory, compared to those without BDD. This has been corroborated by subsequent research suggesting that people with BDD are more likely to examine details in visual tasks and are less likely to take in the holistic picture. Their tendency to focus on details comes at the expense of processing global or configural aspects of a visual scene, and this is widely thought to be related to their tendency to concentrate on flaws in their own appearance. Feusner, Moller, Altstein, et al. (Reference Bohon, Hembacher, Moller, Moody and Feusner2010) tested this in the laboratory by comparing individuals with BDD with healthy controls on a task involving the identification of inverted faces. In healthy individuals, face recognition relies on both featural information (e.g. skin smoothness, blemishes, lines, hair texture) and configural information (e.g. spatial relationships of features, distances between features, holistic elements). Most experimental participants have been shown to demonstrate a “face inversion effect” whereby they are less accurate and slower at processing inverted faces, which requires greater reliance on featural rather than configural information (since some of those features change with inversion). If people with BDD generally rely more on featural rather than configural information, it was hypothesized that they may be expected to demonstrate less of a face inversion effect than controls, and this hypothesis has been supported by several studies, at least for reaction times though not for accuracy (e.g. Feusner, Moller, Altstein, et al. Reference Feusner, Neziroglu, Wilhelm, Mancusi and Bohon2010). Feusner, Hembacher, Moller, et al. (Reference Feusner, Hembacher, Moller and Moody2011) and Bohon, Hembacher, Moller, et al. (Reference Bohon, Hembacher, Moller, Moody and Feusner2012) confirm that individuals with BDD tend to focus on details at the expense of global aspects of images, especially when it comes to processing human faces and bodies. Moreover, Feusner, Hembacher, Moller, et al. (Reference Feusner, Hembacher, Moller and Moody2011) found that the individuals with BDD “may have general abnormalities [relative to controls] in higher- and lower-order visual processing, beyond that for their own appearance or for faces in general.” Using fMRI data, the same study reported that individuals with BDD showed greater activation in medial prefrontal regions for high spatial frequency images, indicating that they have “abnormal brain activation patterns when viewing objects,” not just faces and bodies (Feusner, Hembacher, Moller, et al. Reference Feusner, Hembacher, Moller and Moody2011, 2385).Footnote ⁶ Some researchers have suggested that these visual processing abnormalities may play a role in the etiology of BDD. By showing that individuals who have not been diagnosed with BDD but are at risk of developing BDD (individuals with high Body Image Concern, BIC) exhibit these abnormalities in visual perception, these researchers hypothesize that the abnormalities may be part of what causes BDD, rather than a consequence of it (Beilharz, Atkins, Duncum, et al. Reference Beilharz, Atkins, Duncum and Mundy2016; see also Mundy & Sadusky Reference Mundy and Sadusky2014). They speculate that “abnormalities in visual perception mechanisms, specifically global and local processing, may be involved in the onset and maintenance of BDD …” (Beilharz, Atkins, Duncum, et al. Reference Beilharz, Atkins, Duncum and Mundy2016; for a review see Beilharz, Castle, Grace, et al. Reference Beilharz, Atkins, Duncum and Mundy2017)

In addition to issues with visual processing, there is evidence that individuals with BDD differ from controls when it comes to their ability to recognize facial expressions and accurately interpret the thoughts of others (see e.g. Buhlmann, Wacker, Dziobek, et al. Reference Buhlmann, Etcoff and Wilhelm2015). These deficits are subtle and do not emerge in all cognitive tasks involving the identification of mental states. Buhlmann, Winter, and Kathmann (Reference Buhlmann, Winter and Kathmann2013) found that individuals with BDD did not differ from a control group with respect to their ability to interpret other people’s emotional states based on images of their eyes, using the standard Reading the Mind in the Eyes Test (RMET), which has been widely used in studying theory of mind and social cognition. However, in a follow-up study, Buhlmann, Wacker, Dziobek, et al. (Reference Buhlmann, Wacker and Dziobek2015) found that individuals with BDD were less accurate than controls in correctly interpreting social situations when asked to evaluate scenarios depicted in video sequences showing interactions among four people. This task is more complex and dynamic than the RMET, since participants are instructed to evaluate the characters’ emotions, thoughts, and intentions based on short videos rather than static images. Moreover, several studies have shown that people with BDD show a deficit in the recognition of emotional expressions in faces, especially when it comes to interpreting ambiguous expressions as negative or threatening and in “self-referent scenarios” (e.g. Buhlmann, Etcoff, & Wilhelm Reference Buhlmann, Etcoff and Wilhelm2006; Grace, Toh, Buchanan, et al. Reference Grace, Toh, Buchanan, Castle and Rossell2019).

Individuals with BDD also differ from controls when it comes to the goals they set for themselves, at least as revealed in questionnaires and interviews. In one study, “internal goals” or the state of “feeling right” was found to be more important for those with BDD than for control participants (Baldock, Anson, & Veale Reference Baldock, Anson and Veale2012). Similarly, Veale and Riley (Reference Veale and Riley2001, 1390) report that “BDD patients are driven by a desire to camouflage their appearance or excessively groom to make themselves look their best or to feel ‘comfortable.’” By contrast, controls are “motivated to use a mirror for more functional reasons such as making themselves look presentable or shaving.” Prior to gazing, BDD patients are driven by the hope that they will look different, the desire to know exactly how they look, a belief that they will feel worse if they resist gazing, and the desire to camouflage themselves. After mirror-gazing, people with BDD often feel worse. No matter how many times they look in the mirror, they do not seem to reach a state of “goal completion” or “feeling right.” Baldock, Anson, and Veale (Reference Baldock, Anson and Veale2012) suggest that those with BDD are more likely to terminate mirror checking for reasons unrelated to their goals. For example, they do so because of frustration or time constraints rather than because of completing their goal and finally “feeling right” about their appearance.

Individuals with BDD are also characterized by delusions about their own appearance, which in some ways resemble delusions seen in psychosis (Toh, Castle, & Mountjoy Reference Toh, Castle, Mountjoy, Buchanan, Farhall and Rossell2017). In the psychiatric literature, delusion and insight are often treated as opposite ends of a spectrum, and delusional beliefs are standardly characterized as “fixed beliefs that are not amenable to change in light of conflicting evidence” (DSM-5). Most accounts of delusions in both scientific and philosophical discussions no longer require that delusions be false, since some delusions may happen to be true, though the patient has no reason to believe in them. Hence, what is distinctive about delusions is thought to be, primarily, the extent to which the thinker resists giving up the belief, despite clear evidence to the contrary. Some researchers have distinguished the types of delusions that occur in BDD from those in schizophrenia and some other types of psychiatric disorders. According to Rosen (Reference Rosen1995, 147), BDD delusions are not “bizarre,” in the sense that (for example) “the patient might be preoccupied with ‘ugly’ vascular marks on the skin, but not with marks of the Devil.” Accordingly, Rosen (Reference Rosen1995, 148) suggests that “overvalued ideas” in BDD fall somewhere between obsessions and delusions in terms of insight, in that “the belief is entrenched and sensible to the patient, but he or she can acknowledge the possibility that it may not be true.” But even though the beliefs associated with BDD may not be on a par with the delusions of schizophrenia, they clearly meet widely accepted criteria for delusions when it comes to their imperviousness to contrary evidence.

In this section, we have surveyed the evidence that shows how BDD is associated with certain significant cognitive deficits, regarding visual processing of faces, interpreting the emotions of others, and harboring certain delusions. In the next section, we will argue that these features and others provide a contrast between BDD and OCD, a disorder that it is classified with in the DSM-5.

8.3 Comparison of BDD with OCD

According to the DSM-5, the category of OCRDs includes five psychiatric disorders: OCD, BDD, trichotillomania, excoriation, and hoarding disorder. We will focus here on the similarities and differences among OCD and BDD, setting aside for these purposes the other three disorders. We take it that demonstrating a wide gap between BDD and OCD and important differences among their central features at least casts doubt on the plausibility of classifying BDD alongside OCD as one of the OCRDs. Others have commented generally on “intracategorical heterogeneity” in psychiatric classification (Held Reference Held2017), and some have even questioned whether BDD ought to be included within the OCRD category (for a review see e.g. Toh, Rossell, & Castle Reference Toh, Rossell and Castle2009), but we think that our reasons for questioning the classification are somewhat novel and shed further light on the nature of BDD. In what follows we will take our cue from the patient’s perspective, particularly their mental states, in driving a wedge between BDD and OCD.

Though BDD and OCD both involve impulse control issues, there are some important differences between them. By trying to understand the patient’s perspective and focusing on the patient’s internal mental states as well as their outward behaviors, one can make a distinction between the emphasis on the content of the disorder (which occurs in BDD) and the emphasis on the form of the disorder (which occurs in OCD). This distinction tells us something important about the nature of these disorders. People with BDD are more focused on their image or particular body part (content), while people with OCD are more concerned with the intrusive preoccupations or thoughts themselves (form) (Rivera & Borda Reference Rivera and Borda2001).

Phillips, Pinto, Hart, et al. (Reference Phillips, Pinto, Hart, Coles, Eisen, Menard and Rasmmussen2012) point out that insight/delusionality is a central construct of psychopathology, and that it is of increasing interest particularly within the area of OCRDs. As mentioned earlier, although BDD delusions are often not considered “bizarre,” they can still be believed with such great conviction that they seriously impact people’s ability to function. Phillips, Pinto, Hart, et al. (Reference Phillips, Pinto, Hart, Coles, Eisen, Menard and Rasmmussen2012) studied individuals primarily diagnosed with OCD and those primarily with BDD, and assessed their levels of insight using the Brown Assessment of Beliefs Scale (BABS). This instrument was developed to assess the degree of insight and delusionality in patients with various psychiatric disorders, and it rates beliefs on various dimensions to emerge with a rating that ranges from good insight to no insight (delusionality).Footnote ⁷ When assessed using this measure, there was a wide range in the levels of insight within both OCD and BDD. But the distribution of scores differed significantly between the two disorders, with the majority of OCD participants showing excellent or good insight and the majority of BDD participants showing poor or absent insight. Another study found significantly reduced overall insight among BDD patients, 39 percent of whom were classified as delusional compared to only 2 percent of OCD patients (Eisen, Phillips, Coles, et al. 2004). Researchers also find that referential thinking is typically associated with BDD, but not OCD (Phillips, Pinto, Hart, et al. Reference Phillips, Pinto, Hart, Coles, Eisen, Menard and Rasmmussen2012; Toh, Castle, Mountjoy, et al. Reference Toh, Rossell and Castle2017). Referential thinking, or the tendency to interpret innocuous, neutral stimuli as having some unique personal meaning, also contributes to one’s overall level of delusionality. While referential thinking is a commonly attested clinical feature of BDD, it is not considered typical of OCD (Phillips, Wilhelm, Koran, et al. Reference Phillips, Pinto, Hart, Coles, Eisen, Menard and Rasmmussen2012, 1294).

In addition to a difference when it comes to levels of insight or delusionality, there is also a difference when it comes to the patients’ attitudes toward their own behaviors. People with BDD engage in compulsions but their obsessions are not usually as intrusive or resisted as staunchly by themselves as those of individuals with OCD, and those with BDD do not typically regard their beliefs as senseless (Oldham, Hollander, & Skol Reference Oldham, Hollander and Skodol1996). This relates to a distinction between “ego-syntonic” beliefs and behaviors, which the individual endorses, and “ego-dystonic” ones, which are not endorsed by the individual himself or herself. While the former are more closely identified with BDD, the latter are typically associated with OCD. There would seem to be a connection between the type and degree of delusion experienced and the syntonicity or dystonicity of the beliefs and behaviors, since those individuals who are unable to acknowledge the senselessness of their own beliefs are likely to be more invested in them and hence to endorse their compulsive behaviors, which is the case with BDD. Meanwhile, individuals with OCD, who tend to have greater insight, are less prone to endorsing their behaviors and associated beliefs.

Another point of contrast between BDD and OCD patients is that the compulsive behavior of BDD patients seems not to relieve their anxiety, unlike the behaviors engaged in by OCD patients. According to Veale and Riley (Reference Veale and Riley2001), mirror checking does not result in a reduction of the anxiety experienced by BDD patients. By contrast, the compulsive behaviors associated with OCD, such as repeatedly checking the stove to make sure that it is not on, or washing hands numerous times to ensure that they are not contaminated, seem to relieve the anxiety of OCD patients at least partially and temporarily. Veale and Riley (Reference Veale and Riley2001) report that prior to mirror gazing, BDD patients are driven by the hope that they will look different, the desire to know exactly how they look, a belief that they will feel worse if they resist gazing, and the desire to camouflage themselves, but after mirror-gazing, participants feel worse than they did before. Further, Phillips, Wilhelm, Koran, et al. (Reference Phillips, Wilhelm, Koran, Didie, Fallon, Feusner and Stein2010, 578) state that “some BDD compulsions (e.g. mirror checking) do not appear to follow a simple model of anxiety reduction, which is more commonly seen in OCD.”

A final contrast between BDD and OCD patients concerns the difference in their respective abilities to understand social situations. As already mentioned, at least in some tasks involving social cognition, BDD patients perform worse than healthy controls, particularly in interpreting the thoughts of others. This difference also holds when they are compared with patients diagnosed with OCD. Those with BDD are found to be less accurate overall in interpreting social situations compared to those with OCD (Buhlmann, Wacker, & Dziobek Reference Buhlmann, Wacker and Dziobek2015). This difference appears related to two attributes of people with BDD already mentioned: their level of insight into their own condition and their tendency to engage in referential thinking. BDD patients suffer from delusions about their appearance or harbor entrenched misguided beliefs that they endorse. These beliefs appear to be partly induced, or at least maintained, by deficits when it comes to interpreting the thoughts of others, since they often think that others are judging them negatively when they are not, and they are not easily persuaded by others when they reassure them about their own appearance. Hence, it seems as though deficits in social cognition are central to BDD and are related to some other aspects of the disorder such as the level of delusionality associated with it.

BDD and OCD both involve behaviors that seemingly cannot be stopped (cf. Abramowitz Reference Abramowitz2018), but they differ when it comes to the individuals’ level of insight or delusionality, the degree to which individuals endorse their behaviors and beliefs, the extent to which individuals’ behaviors relieve their anxiety, and the individuals’ ability to interpret social situations. By emphasizing the internal mental states of BDD patients rather than their external symptoms or behaviors, some important contrasts emerge between BDD and OCD that appear central to understanding BDD. To further justify this claim, we will propose a tentative causal model of BDD in the next section, which further brings out its differences with OCD and suggests reclassifying it.

8.4 Proposal for a Causal Model of BDD

By paying greater attention to the internal mental states of people with BDD rather than their outward symptoms, it becomes apparent that there are important differences between BDD and OCD. We conjecture that these differences point to different causal models of the two disorders. On the surface, it seems as though BDD and OCD patients exhibit similar symptoms of compulsivity and repetitive behaviors, but when one examines the symptoms more closely, it is plausible that these compulsive behaviors involve very different causal processes. The main components of this causal network consist in some of the central characteristics of BDD that set it apart from OCD: focusing on the content of the disorder as opposed to its form; having a lower level of insight; having an ego-syntonic perspective on one’s experiences; and the behaviors’ not having an anxiety reduction function. These differences can be used to sketch a speculative causal model of BDD, as follows. Two of the most basic cognitive features of BDD patients mentioned in Section 8.2 are: (i) deficits in visual processing; (ii) deficits when it comes to evaluating the emotions, thoughts, and intentions of others. As mentioned earlier, the perceptual deficits involve focusing on details at the expense of a global or configural picture. Both clinical observations as well as neurobiological and psychophysical research suggest that individuals with BDD tend to focus on the details of their appearance at the expense of their overall image, and there is some evidence to suggest that abnormalities in visual perception mechanisms are involved in the onset and maintenance of BDD (see Beilharz, Atkins, Grace, et al. Reference Beilharz, Atkins, Duncum and Mundy2016). This implies that individuals are likely to concentrate on specific body parts or features rather than their overall image. Excessive scrutiny of a specific feature may lead in turn to increased dissatisfaction with that feature. Meanwhile, theory of mind deficits involve misinterpreting others’ expressions or statements. While these misinterpretations need not be negative or directed at oneself, it is possible that given that people with BDD are dissatisfied with some aspect of their appearance, these theory of mind deficits may lead to a perception of negative evaluation by others. In addition, there may be an interaction between these two posited causal antecedents of BDD, since dissatisfaction with one’s appearance may be reinforced by a mistaken interpretation of the attitudes of other individuals. This may cause increased dissatisfaction with one’s appearance, which may lead in turn to further negative construals of other people’s attitudes. These negative impressions, of one’s own appearance and of the evaluations of others are likely to reinforce one another. The positive feedback loop may result in beliefs that are difficult to dislodge because of their self-reinforcing nature, and this is a mark of delusional thinking, as defined by the DSM-5 (“fixed beliefs that are not amenable to change in light of conflicting evidence”). However, in order to give rise to full-blown delusions, an additional causal factor may be required in the form of a disposition toward a certain cognitive bias, as we will try to explain.

Recent research has suggested that there is a causal relationship between a bias against disconfirmatory evidence (BADE) and delusions in a number of psychiatric disorders. As argued in Chapter 7, BADE may be one manifestation of a myside heuristic that is attested in a very wide range of human subjects and it occurs when people reject or discount evidence that is contrary to their beliefs. People who are prone to this cognitive bias tend to discount evidence against their beliefs to a greater extent than may be warranted by ideal standards of rationality. As mentioned in Chapter 7, Woodward, Moritz, Cuttler, et al. (Reference Woodward, Moritz, Cuttler and Whitman2006) present evidence that psychiatric patients with delusions differ from controls in this regard, exhibiting an accentuated or extreme version of BADE.Footnote ⁸ Moreover, they suggest that BADE is not just a feature of the delusion itself but that it extends to “delusion-neutral material” in experimental tasks. At least in schizophrenic patients, the tendency toward BADE exists independently from the delusion and may play a role in causing and maintaining the delusion. This causal association between BADE and delusions has also been found in other psychiatric disorders. On the basis of a meta-analysis examining the relationship between two closely related cognitive biases, BADE and Jumping to Conclusions (JTC),Footnote ⁹ McLean, Mattiske, and Balzan (Reference McLean, Mattiske and Balzan2017, 345) conclude: “The association of these biases with delusions in multiple diagnoses would demonstrate this relationship is not limited to schizophrenia, and would support a causal relationship.”Footnote ¹⁰ In addition, BADE has been shown to be associated with delusions in both clinical and nonclinical populations (Bronstein & Cannon Reference Bronstein and Cannon2017). On the basis of these and similar findings, we conjecture that BADE might be involved in the inception and maintenance of delusions present in BDD. In our causal model, we have proposed that perceptual deficits and theory of mind deficits are two of the principal causal factors in BDD, and this additional factor, a disposition toward BADE, may also be causally implicated in causing and maintaining the disorder, specifically in ensuring that the faulty beliefs of BDD patients rise to the level of delusions (see Figure 8.1). People with BDD do not believe others when they try to reassure them that their particular body part is not deformed and they are generally resistant to disconfirming evidence, as we have seen. As already suggested, in some experimental paradigms, most individuals, not just psychiatric patients, are shown to have a bias against disconfirming evidence. However, in individuals with BDD (and in psychiatric patients with delusions more generally), that tendency may exist in a more pronounced and severe form, and when combined with the other causal factors mentioned, may result in the “perfect storm” that is responsible for the emergence of full-fledged BDD. Moreover, since their delusions are rooted in distorted perceptions, BDD delusions are not completely unmoored from reality and hence do not resemble the “bizarre” delusions associated with schizophrenia and other psychiatric disorders. Still, BDD patients maintain certain entrenched beliefs about their appearance that are resistant to change on the basis of contrary evidence.

Figure 8.1. Causal model of body dysmorphic disorder, showing ultimate causes (grey) and proximal causes and effects (white). Dotted arrows indicate causal feedback loops.

Now that we have sketched out a causal model of BDD, we will argue that these components render it a psychiatric kind in its own right and make it categorically distinct from OCD. Our causal model of BDD theorizes that it is at least partly caused by a combination of cognitive and perceptual factors. We have suggested that a perceptual deficit, a theory of mind deficit, and an accentuated cognitive bias against disconfirming evidence are crucial causal antecedents of BDD. In addition to these endogenous causal factors, there may also be various social causes that might contribute to the disorder, either directly or indirectly by influencing these psychological causes. The causal model associates BDD with a distinctive network of effects, since the cognitive deficits interact in such a way as to generate delusional beliefs about one’s physical appearance. These delusional beliefs, in turn, generate a suite of behaviors that aim at fixing perceived flaws in one’s physical appearance, and these behaviors ultimately do not reduce their anxiety. Finally, the beliefs and behaviors involved are such that they constitute a significant departure from rationality and disrupt the functioning of individuals who are characterized by these mental states and behaviors. Departures from rationality, represented primarily by delusional beliefs, lead to disruptive mental states and behaviors, which hamper the ability of individuals with BDD to achieve their goals, avoid suffering, and establish meaningful relationships.Footnote ¹¹ This set of characteristic properties constitutes the causal profile of BDD and distinguishes it as a real kind in the psychiatric domain.

One obvious way that BDD differs from OCD is that BDD involves a perceptual/visual deficit, and OCD does not appear to do so (cf. Kaplan, Rossell, Enticott, et al. Reference Kaplan, Rossell, Enticott and Castle2013). In addition, as already seen, persons with BDD differ from those with OCD when it comes to theory of mind deficits. Moreover, there does not seem to be any evidence that OCD patients have a cognitive bias against disconfirmatory evidence. Indeed, there is evidence that people with OCD have the opposite tendency: They consider alternative possibilities more than controls and may be less discriminating in giving them importance (Dèttore & O’Connor Reference Dèttore and O’Connor2013; for a review of cognitive biases associated with BDD, see Hezel & McNally Reference Hezel and McNally2016). Hence their causal antecedents differ. There are other important ways in which BDD differs from OCD, at least some of which seem to issue from these causal factors. As already mentioned, people with BDD are more focused on their image or particular body part (content), while people with OCD are more concerned with the intrusive preoccupations or thoughts themselves (form) (Rivera & Borda Reference Rivera and Borda2001). Someone with BDD may be concerned about the size or shape of their nose, thinking that it is truly misshaped even though it is not; while someone with OCD is likely to be concerned about the preoccupations of their compulsion, even if they acknowledge that their compulsion is nonsensical. People with BDD and OCD differ not only in their focus, but also in their level of insight. People with OCD tend to have greater levels of insight than those with BDD, which is to say that they are not as delusional. People with OCD can acknowledge that their desire to check the stove multiple times in succession might not make sense, but they still have a desire to do it. People with BDD think that it is perfectly sensible to engage in their “checking” and “fixing” behaviors. When we look at the disorders from the patient’s experiential perspective, it seems that OCD is ego-dystonic, while BDD is ego-syntonic.

Since both the experiences and underlying causal factors associated with BDD differ substantially from those of OCD, we posit that there is no reason to group them in the same category of OCRDs. The experiential differences between the disorders might also mean that it does not make sense to use the same treatment methods for them. The causal model of BDD is important for theoretical and classificatory purposes, but it is also possible that examining etiology could lead to implications for treatment. Treatments and therapies that take causes into account might be more effective than those which do not. We will take this issue up in the following section.

8.5 Objections and Replies

In this section, we will consider three objections to our proposed causal model of BDD. The first objection would question the appropriateness of the causal model that we have proposed. Why single out these features in particular as the crucial causal factors for the emergence and maintenance of BDD? Could an alternative causal model adequately explain the features of BDD? The most explicit causal model of BDD that we have found in the psychiatric literature has been proposed by Veale (Reference Veale2004) (see Figure 8.2.). His model involves seven key components: triggers, negative appraisals of an internal body image, safety behaviors, mood, rumination, processing the self as an aesthetic object, and in a separate section, he mentions selective attention. In the graphic illustrating the model, each of these components is connected by bidirectional arrows, and it is difficult to discern the precise nature of the causal relationship between these components. While it is true that the features mentioned are somehow involved in BDD, the model itself is not clear in explaining how they relate to each other. For example, Veale (Reference Veale2004, 114–115) describes the link between a trigger and a negative appraisal as follows: “It is proposed that the cycle begins when an external representation of the person’s appearance (e.g. looking in a mirror) activates a distorted mental image.” But it is not clear why looking at oneself in the mirror leads to a negative appraisal of one’s internal body image and Veale’s model does not seem to explain this. By contrast, our proposed model highlights the fact that a perceptual deficit, which involves focusing on details at the expense of a configural picture, may lead to exaggerated attention toward one or more bodily features, resulting in an initial negative self-perception, which is then reinforced by a theory of mind deficit. More recently, Feusner, Neziroglu, Wilhelm, et al. (Reference Feusner, Neziroglu, Wilhelm, Mancusi and Bohon2010) also try to enumerate some of the main causal factors for BDD, but although they identify the perceptual processing and theory of mind deficits that we have emphasized, in addition to neurochemical and social factors, they do not mention the cognitive bias against disconfirmatory evidence, nor do they attempt to show how these factors might fit together in a more comprehensive causal model.

Figure 8.2. A cognitive behavioral model of body dysmorphic disorder proposed by Veale (Reference Veale2004)

A second objection questions our understanding of the mental states of BDD patients and sees greater continuity between BDD and OCD. Like us, Abramowitz (Reference Abramowitz2018) suggests that the OCRD category should be reevaluated; however, unlike us, he suggests that of all the OCRDs, BDD is the most similar to OCD. He suggests that OCD compulsions act as a safety behavior, where the behavior allows for one to temporarily escape from distress. He claims that for both OCD and BDD, the patient’s thoughts are intrusive and anxiety provoking, and that the repetitive behaviors have an anxiety reduction function. And, the checking behavior maintains appearance-related preoccupations in BDD in the same way that compulsive rituals maintain obsessive fears in OCD. Our response to Abramowitz’s first point is that, while the intrusive thoughts may be anxiety provoking, the patients experience their anxiety differently. As we mentioned earlier, people with OCD find the thoughts themselves to be concerning, whereas people with BDD appear not to be concerned that they have such thoughts, they are concerned about the body part that is the subject of their thoughts.Footnote ¹² For people with OCD, the thoughts are intrusive and unwanted. They might not endorse those thoughts, but they still have a compulsion to act on them. For example, people with OCD may not strongly believe that their house will in fact burn down unless they check their stove multiple times, but they still feel the need to check. People with BDD, however, endorse their thoughts. For them, it is not that the thoughts or preoccupations are unwanted or intrusive, but the “flawed” body part that is unwanted. Our response to Abramowitz’s second point that both OCD and BDD involve anxiety reduction functions, is that there is a body of evidence already cited (e.g. Veale & Riley Reference Veale and Riley2001; Phillips, Wilhelm, Koran, et al. Reference Phillips, Wilhelm, Koran, Didie, Fallon, Feusner and Stein2010) that strongly suggests that this is not the case. As mentioned earlier, people with BDD do not feel as though they have completed their goal after engaging in their BDD behaviors in the same way that people with OCD have completed their goal once their ritual has been performed and their anxiety has been reduced. To Abramowitz’s third point that the checking behavior maintains appearance-related preoccupations in BDD in the same way that compulsive rituals maintain obsessive fears in OCD, we respond by saying: That the behavior is maintained is different from why the behavior is maintained. We do not deny that both people with BDD and people with OCD engage in checking behaviors, and deal with impulse-control issues. What we have argued is that these issues are experienced differently, and that these behaviors have a different etiology.

Finally, it might be objected that a theoretical causal model should not be the basis for classifying psychiatric disorders, including BDD. Indeed, it might be said that the basis for classification in clinical psychiatry should be treatmentFootnote ¹³ not causal modelling, and that classification ought to be aimed at remedying disorders rather than explaining them. But we doubt that there is such a stark contrast between the aims of treatment and causal explanation. The causal model that we have proposed attempts to identify some of the causal factors that contribute to the emergence of the disorder in people with BDD. Some treatments may be found to be efficacious without understanding the reasons for their efficacy, but these will remain lucky guesses unless the causal process that underlies them is understood. We think that it is unlikely that an informed treatment of psychiatric disorders can refrain from intervening on their causal antecedents in order to attempt to avert the causal process that leads to the emergence of the disorder. While knowledge of the causal process may not always lead to effective treatments, perhaps because it is impracticable to intervene on the relevant causes, such an intervention would seem necessary to an informed treatment (as opposed to a mere guess).

8.6 Conclusion

There is growing interest in psychiatric research and in philosophy of psychiatry in understanding the etiology of mental disorders in terms of specific cognitive dysfunctions or deficits.Footnote ¹⁴ In this chapter, we have tried to synthesize some of the recent research on Body Dysmorphic Disorder to emerge with what we consider to be a plausible causal model of the disorder. The primary causal factors that we posit as part of the etiology of BDD, namely a deficit in visual perception, a theory of mind deficit, and a form of the cognitive bias against disconfirmatory evidence, are certainly not the whole story. But if something like this causal network lies behind BDD and can be used to explain its emergence and persistence, that would support the case for considering it a real kind. Though not exclusively cognitive, this kind has an important cognitive dimension. In previous chapters, the case has been made that many cognitive kinds are likely to resist reduction to neural, neurochemical, or neurophysiological constructs, mainly because they are individuated contextually, etiologically, or both. In this case, one obstacle to a reductive account is the occurrence of a cognitive bias in the causal network. If, as suggested in Chapter 6, the identification of a cognitive heuristic as a bias may require relating it to a broader environmental and social context, then BADE cannot be characterized as such in isolation. In addition, since BADE is thought to be present in an extreme form in individuals with BDD, the severity of this bias is also a contextual matter and cannot be identified without such a context. Moreover, it would seem as though identifying BDD as a disorder is dependent on discerning its effect on the behavior of the individual in a social context. That is, individuals who have this particular combination of psychological traits might not be singled out as a particular human type if it were not for the harms that they endure and the difficulties they face in achieving their goals in a social setting. Still, it might be objected, the causal network that we have identified would seem to constitute a kind in its own right whether or not it has harmful effects on the individuals who possess them. There is something to be said for this view, particularly since we raised doubts in Section 8.1 that the superordinate category of psychiatric disorder might correspond to a kind. Therefore, we would not rule out the possibility that the components of the causal network that we have identified with BDD may each be identified with neural correlates (though they might be multiply realizable neurally), and that their combination would also correspond to a set of neural correlates. Nevertheless, it is the combination of these causal factors that issues in BDD and there is no reason to think that this collection of neural correlates has any claim to unity beyond their issuing in a set of cognitive and behavioral effects. This means that the psychological causal network provides the basis for singling out this collection of features and considering it a unified kind.

Finally, what kind of kind is BDD: Is it a state, event, process, capacity, or what exactly? Though there are events and processes involved in the etiology of BDD as we have described it, as well as (dysfunctional) capacities, the most natural understanding of the category is that it identifies a kind of individual: a person with BDD. The causal process that we have identified takes place in an individual and results in a kind of person who is prone to certain behaviors (e.g. mirror checking) and to having certain states of mind (e.g. delusions about appearance), and hence difficulties in achieving their goals and flourishing in a social setting. It is this kind of person that is primarily a member of the kind BDD, though to say that BDD characterizes a type of person is not to say that persons with BDD are essentially or irreversibly persons of that kind.