Abstract
The available categorical constructs within the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, do not allow forensic psychiatrists to distinguish easily between the varieties of delusion-like belief. This dilemma is especially challenging when seemingly delusional beliefs are shared online. Although the term “extreme overvalued belief” has been proposed to aid with such distinctions, its definition has conceptual problems, including diagnostic overlap with shared delusions, “extremism” that refers to behavior rather than belief, and the potential to be applied with prosecutorial bias to thwart defense strategies attempting to establish connections between criminal behavior and less than optimal mental health. Beliefs and behavior that are not obviously symptomatic of mental illness are best explained by integrating psychiatric expertise with that of other disciplines such as psychology, sociology, and political science. Dimensional quantification of belief conviction and preoccupation as well as established concepts like conspiracy theories and sacred values can help forensic evaluators characterize ideological motives for deviant behavior more accurately to better inform legal decisions about criminal responsibility and therapeutic justice.
… [I]t is important to note that the definition of mental disorder included in DSM-5 [Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition] was developed to meet the needs of clinicians, public health professionals, and research investigators rather than all the technical needs of the courts and legal professions…. When DSM-5 categories, criteria, and textual descriptions are employed for forensic purposes, there is a risk that diagnostic information will be misused or misunderstood. These dangers arise because of the imperfect fit between the questions of ultimate concern to the law and the information contained in clinical diagnosis…. [A]dditional information is usually required beyond that contained in the DSM-5 diagnosis, which might include information about the individual's functional impairments and how these impairments affect the particular abilities in question [Ref. 1, p 25].
A common criticism of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) is that its diagnostic categories lack validity and have failed to properly “carv[e] nature at its joints” (Ref. 2, p 10). In clinical psychiatry, where DSM-5 is used as a rough guide to inform decisions about who needs psychiatric care and what kind of treatment might be appropriate, the fuzzy boundaries of mental illness are tolerable in the name of overall clinical utility3,–,6 and because practicing clinicians are not slaves to diagnostic criteria.7 In forensic psychiatry, where clear boundaries are demanded for life-or-death decisions about criminal responsibility and retributive justice, such ambiguity is less acceptable.8,9 Cases are nonetheless often reduced to debates between forensic psychiatry expert witnesses offering polar opposite diagnostic opinions.
Beyond the competing agendas and biases of prosecution and defense, diagnostic agreement in forensic psychiatry is notably hampered by inadequate DSM-5 categories to account for “delusion-like beliefs” (DLBs), which are beliefs that resemble delusions superficially but fall short on closer dissection. Many such beliefs slip through the cracks of symptom definitions and drift into the gray area between pathological and normal beliefs. In this issue of The Journal, Rahman and colleagues10 further a proposal that such diagnostic dilemmas can be resolved by the addition of a new categorical definition of a DLB variant, the “extreme overvalued belief.” In turn, this commentary critically examines that claim and extends the counterargument that forensic evaluators must look beyond psychiatry to best understand shared beliefs that straddle the boundary between psychopathology and normalcy.9
Delusion-Like Beliefs in DSM-5
Distinguishing delusions from related types of misbelief is critical for guiding proper clinical diagnosis and treatment. Whereas DSM-5 uses criteria to define mental disorders, the glossary definitions of symptoms themselves are brief and incomplete, bypassing the problem of “criteria for criteria”11 but leaving definitional gaps in the process.
Delusions
Delusions are defined in DSM-5 as:
A false belief based on incorrect inference about external reality that is firmly held despite what almost everyone else believes and despite what constitutes incontrovertible and obvious proof or evidence to the contrary. The belief is not ordinarily accepted by other members of the person's culture or subculture (i.e., is not an article of religious faith). When a false belief involves a value judgment, it is regarded as a delusion only when the judgment is so extreme as to defy credibility [Ref. 1, p. 819].
Although the DSM-5 definition has evolved over time, the basic concept of delusions as “fixed, false beliefs” can be traced back to Karl Jaspers, who emphasized their subjective certainty (conviction), incorrigibility (resistance to counterargument), and impossibility of content.11 Jaspers also argued that true delusions, as distinguished from other related phenomena, lie beyond intersubjective understandability.12,13
Delusions have often been described as bizarre (to denote their impossibility), but due to the unfalsifiability of some beliefs and poor interrater reliability for what is possible or impossible,14,15 the diagnostic relevance of bizarre delusions was abandoned in DSM-5. As an extension of Jaspers' un-understandability, unshareability has become the modern proxy for impossibility and falsity. In clinical practice, the idea that delusional beliefs cannot be shared with others relates to their self-referential and grandiose content. For example, it would be easy for an individual to find others who share the belief that God can speak to people or that there will be a second coming of the Messiah, but it would be much harder for an individual to find confederates who agree that God has ordained that he is the Messiah.
Shared Delusions
Although unshareability has been regarded as a defining feature of delusions, it has long been recognized that there are exceptions. The terms folie communiqueé, folie simultaneé, folie imposée, and folie á deux were coined in the late 1800s to describe delusional beliefs transmitted from a primary individual to a usually more passive or subordinate, secondary person.16 Within modern nosology, “shared paranoid disorder” evolved into “induced psychotic disorder” and finally “shared psychotic disorder” from DSM-III in 198017 to DSM-IV18 in 1994. In DSM-IV, shared psychotic disorder was diagnosed when a person developed a delusional belief from a “close relationship with another person, or persons, with an already-established delusion” (Ref. 18, p 306) and did not meet diagnostic criteria for another psychotic disorder. This narrow definition echoed the historical term folie impose, but it presented a contradiction of sorts because effective treatment typically consisted of separation of the person with shared psychotic disorder from the “primary case,” which therefore implied that the secondary individual was not so much psychotic or mentally disordered as impressionable.19,20 Published cases of shared psychotic disorder have revealed that such individuals are not immune to psychiatric comorbidities and that treatment not infrequently includes pharmacotherapy, although whether medication is necessary or even useful is less clear.21 Shared psychotic disorder was eliminated as a distinct entity in DSM-5, leaving clinicians to diagnose individuals with either full-blown delusions or “delusional symptoms in partner of individual with delusional disorder” as an example of “other psychotic disorder” (Ref. 1, p 122).
Although shared psychotic disorder classically occurs within a dyadic relationship, there have been reports in the literature of folie á trois, folie á quatre, and folie á famille.20,22 In addition, the psychodynamic and social underpinnings of shared delusions have been invoked to explain unconventional belief systems maintained within cults, raising the question of whether they could be considered examples of “mass shared psychotic disorder” (Ref. 23, p 515). Alternatively, DSM-5 maintains a section on cultural concepts of stress (formerly called “culture-bound syndromes” in DSM-IV) to remind psychiatrists that seemingly delusional beliefs (e.g., koro, the belief that one's penis is retracting into one's body) can be understood and normalized as culturally sanctioned idioms of distress.1 The resulting lack of clarity surrounding examples of shared delusion leaves unresolved dilemmas about possible overlap between delusions and shared religious, political, and paranormal beliefs; how large an accepting subcultural group must be to normalize unconventional beliefs; and what determines whether group affiliation will have a mitigating or exacerbating effect on delusional thinking.24
Obsessions
Obsessions are hallmark features of obsessive-compulsive disorder (OCD) that are defined in DSM-5 as “recurrent and persistent thoughts, urges, or images that are experienced, at least some time during the disturbance, as intrusive and unwanted and that in most individuals cause marked anxiety or distress” (Ref. 1, p 826).
Because classic obsessions are ego-dystonic, with intact insight as to their irrationality or ridiculousness, they are distinct from delusions and are not, strictly speaking, beliefs at all. But it is well known that erosion of insight can occur in the course of OCD to the extent that obsessions can sometimes become delusional in nature.25 DSM-5 therefore includes a specifier that allows clinicians to diagnose OCD “with absent insight/delusional beliefs” (Ref. 1, p 237).
Overvalued Ideas
In modern psychiatry, overvalued ideas are beliefs held with less than delusional conviction within disorders such as OCD, hypochrondriasis, anorexia, and body dysmorphic disorder.25,–,27 As symptoms of mental disorders, the DSM-5 definition takes care to separate overvalued ideas from shared cultural beliefs:
An [overvalued idea is an] unreasonable and sustained belief that is maintained with less than delusional intensity (i.e., the person is able to acknowledge the possibility that the belief may not be true). The belief is not one that is ordinarily accepted by other members of the person's culture or subculture [Ref. 1, p 826].
The original concept of an overvalued idea has been attributed to Carl Wernicke, who distinguished them from obsessions by virtue of their ego-syntonicity and from delusions based on lessened intensity, but, in contrast to DSM-5, Wernicke regarded them as shareable.26,28 Jaspers likewise opined that overvalued ideas were affect-laden but understandable, not altogether distinct from strong political or religious convictions.25,26 Extending this historical tradition, Rahman and colleagues10 have proposed the term “extreme overvalued belief” to characterize nondelusional but extreme political, religious, and cultural belief, defined as follows:
An extreme overvalued belief is one that is shared by others in a person's cultural, religious, or subcultural group. The belief is often relished, amplified, and defended by the possessor of the belief and should be differentiated from a delusion or obsession. The idea fulminates in the mind of the individual, growing more dominant over time, more refined, and more resistant to change. The individual has an intense emotional commitment to the belief and may carry out violent behavior in its service. It is usually associated with an abnormal personality. [Ref. 28, p 33].
In this issue, Rahman et al.10 extend previous arguments in favor of adopting extreme overvalued beliefs as a term for both forensic psychiatry and DSM-5,28,–,30 with new data reporting near-perfect interrater reliability for distinguishing between extreme overvalued beliefs and DSM-5 definitions of delusions and obsessions.10
Shared Beliefs in the Internet Era
Rahman and colleagues10 have argued that the extreme overvalued belief concept is both valid30 and reliable, but as the critical reading adage warns, “the conclusions giveth but the methods taketh away” (Ref. 31, p 130). In testing for interrater reliability, definitions of delusion, obsessions, and extreme overvalued beliefs were matched against 12 case vignettes that supplied easy contextual clues for proper classification: additional psychotic symptoms such as voice-hearing, disorganization, and negative symptoms in the delusion cases; ego-dystonicity or compulsions in the obsession cases; and good evidence of shared beliefs without other symptoms of mental disorder in the cases of extreme overvalued belief.10 Thus, the near-perfect interrater reliability is most likely a reflection of DLBs portrayed in the presence or absence of other evidence of mental disorder, rather than the reliability of the symptom definitions themselves.
In the same vein, none of the vignettes described cases of monosymptomatic delusions in the setting of delusional disorder where functioning is relatively intact, and none included examples of shared delusion. Indeed, the authors' abbreviated use of the DSM-5 definition of delusion did not allow the possibility that delusions are ever shared. Although the authors note that the “failure to recognize shared versus idiosyncratic beliefs … may lead to inaccurate diagnostic classification” (Ref. 10, p 324), the vignettes exempted evaluators from that challenge by including clear evidence that the extreme overvalued beliefs were shared. The authors crucially belie the complexity of real-life clinical and forensic cases where interrater reliability would be expectedly lower, consistent with research studies predating DSM-5 in which the interrater reliability of delusions averaged κ = 0.69 with a standard deviation of 0.24.14
In promoting the utility of the term extreme overvalued belief, Rahman et al.10 have cited the case of Anders Breivik, a perpetrator of mass murder whose espoused beliefs were variably diagnosed as delusions of schizophrenia and nonpsychotic political beliefs.29,30 Initially, his beliefs were thought to be “bizarre,” but reassessment in the wake of public outrage over an insanity defense determined that Breivik's beliefs conformed to those of right-wing racist groups. Beyond those subculturally sanctioned beliefs, Breivik also claimed during initial evaluations to be the “commander” of the Knights Templar, the “savior of Christianity,” and the future regent of Norway without any evidence that these self-referential beliefs were shared by anyone else.29,32 It is therefore hard to understand how such idiosyncratic and grandiose beliefs, which may have served as the key impetus to commit the terrorist act, would qualify as extreme overvalued beliefs according to the definition proposed by Rahman et al.10 Moreover, rather than providing crisper lines for diagnosis, the conceptualization of Breivik's beliefs as extreme overvalued beliefs would seem to overlook the possibility of having delusional beliefs embedded within a subculturally accepted framework.32,33 Indeed, the original forensic evaluators for Breivik believed that “right-wing extremism [was] not the defendant's primary issue, but a repository of his delusions” (Ref. 34, p 2413).
Excluding shared beliefs from the definition of delusion likewise fails to consider the possibility that “true” delusions can be culturally sanctioned, especially within modern online subcultures. The recent phenomenon of “gang stalking” provides an illustrative case in point. Gang stalking refers to the shared belief that there is ongoing mass surveillance, harassment, and mind-control of self-described “targeted individuals.” Over the past decade, several noteworthy mass shootings have been perpetrated by individuals espousing such beliefs, highlighting their forensic relevance.35 When analyzed individually, such beliefs are best explained as textbook examples of paranoid delusions.36,37 But targeted individuals stereotypically reject such clinical diagnoses and have found validation within an online community of individuals with similar experiences; in this way, their self-referential beliefs have achieved cultural sanctioning. The instant access to individuals all over the world that is now possible via the Internet makes sharing beliefs, even those that are idiosyncratic and self-referential, possible in a way that confounds modern efforts to crisply categorize DLBs. Although the shareability of a belief should detract from its delusionality, the Internet represents a communal space where misinformation and unsubstantiated opinion masquerading as objective evidence can be easily found.9 Rahman and colleagues10 recognize this dilemma,28,30 but they seem shortsighted in their hope that extreme overvalued beliefs will resolve it. In addition, whereas extreme overvalued beliefs are said to “fulminate in the mind of the individual” (Ref. 28, p 33), this definition focuses on misinformation “in the head” to the detriment of considering the larger problem of “misinformation in the world” that “exists across individuals, cultures, and societies” (Ref. 38, p 399).
Nonpsychiatric Models of DLBs
Rahman and colleagues10 acknowledge that it is not the overvalued belief itself that is extreme in their case vignettes (i.e., extreme overvalued beliefs and delusions are undifferentiated by degree of conviction), but rather the criminal behavior. But defining the pathology of a belief based on associated criminal behavior is problematic at best, raising the obvious question of what features might distinguish DLBs that motivate criminal behavior from those that do not. Rahman et al.10 concede that extreme overvalued beliefs do not provide an explanation of why some individuals with such beliefs, but not others, commit socially deviant acts. This explanatory deficit has been criticized as a shortcoming of more general categorical labels that equate extremism with violence and has been used to argue in favor of considering relevant dimensional aspects of religious and political beliefs instead.9,39 With respect to DLBs, extremism more appropriately applies to degree of unwarranted conviction, in contrast to holding beliefs with cognitive flexibility and intellectual humility in a manner more conducive to social functioning and mental health.40 Beyond its political expediency, Anders Breivik's diagnostic revision has been attributed to evidence of the waning conviction of his more self-referential beliefs over time rather than to the discovery that his political beliefs were shared or to a categorical difference between delusions and extreme overvalued beliefs.32
Developing explanatory models for the relationship between beliefs and criminal behavior can be aided not only by modeling misbeliefs along dimensions such as conviction and preoccupation, but by also integrating other categorical constructs that have been elucidated in fields outside of psychiatry.9 For example, Morgellon's syndrome has been characterized as both delusional parasitosis and Internet meme.41 The mass suicides of Jonestown and Heaven's Gate should be considered within the group dynamics of cults and new religious movements.24,42,43 “Sovereign citizen” beliefs can be best conceptualized as a form of political conspiracy theory supported by online misinformation.9,44 Terrorist martyrdom can be best understood as rooted in moral or deontic reasoning based on sacred values, where threatened belief is equated with threatened identity and the need to defend it at all costs.45,46
There is little doubt that, as a medical specialty, psychiatry focuses on mental illness to the extent that it is ill-equipped to account for unusual, odd, or even extreme experiences, beliefs, and behaviors that fall short of clinical psychopathology but are sometimes relevant to forensics. Instead of attempting to rectify this problem by creating a new umbrella term for shared fringe beliefs held with high conviction, forensic evaluators would do well to look beyond psychiatry to explain the nuances of connections between ideology and criminal behavior. Perspectives from psychology, sociology, anthropology, political science, and information science should be integrated to properly characterize the variety and diversity of individual delusion-like beliefs and their role in driving socially deviant acts.
DLBs and Criminal Responsibility
Rahman and colleagues10 advocate that extreme overvalued belief, like overvalued idea, should be included in the DSM-5 glossary. Although overvalued ideas are included as examples of DLBs with less than delusional conviction and as symptoms of nonpsychotic mental disorders like hypochondriasis, it appears that the main intent of the term extreme overvalued belief is to draw a crisp line excluding it as evidence of mental illness. It is therefore not clear why a non-symptom would be included in the DSM-5 glossary. More importantly, it suggests that application of the term would not only be aimed at reducing diagnostic confusion, but at making it easier to prevent DLBs from being used as an insanity defense.
The prosecutorial bias of this goal is again illustrated by the Breivik case. In Norway, “a person is not criminally accountable if psychotic, unconscious, or severely mentally retarded at the time of the crime” (Ref. 33, p 377). For forensic psychiatrists to conclude that Breivik was delusional was therefore to conclude that he was “criminally insane” independent of intent, which the Norwegian public equated with not having to answer for his crimes or that he might be set free.34,47 The resulting public outrage, grounded in the misconception that “the purpose of psychiatry is to get people off” (Ref. 47, p 1564), therefore demanded a psychiatric reevaluation and, in turn, a different diagnostic conclusion that would facilitate retributive justice. Rahman et al.10 likewise seem to suggest that being able to diagnose shared political, religious, and other dogmatic beliefs associated with violence and terrorism as extreme overvalued beliefs would have facilitated the guilty verdict in the case of Breivik as well as that of other terrorists like Ted Kaczynsky.10
In contrast to such crisp line drawing, dimensional perspectives that incorporate perspectives from psychology and other disciplines acknowledge that, although perpetrators of terrorist acts and violent extremism may not have a mental illness per se, they are often far from mentally healthy and may have DLBs that serve as motivations for their acts.48,49 This perspective has particular relevance for defense strategies within the U.S. legal system where not guilty by reason of insanity (NGRI) judgments are variably determined according to state jurisdictions based on the M'Naughten Rule, the Irresistible Impulse Test, and the Model Penal Code. Beyond terrorism and NGRI decisions, DLBs are also relevant to U.S. tax fraud litigation, which, echoing the M'Naughten Rule, depends on a demonstration of willfulness to violate the law. In cases involving sovereign citizens who may not be delusional but demonstrate high levels of conviction for DLBs relating to tax law and have been exposed to online misinformation supplying the evidence for those beliefs, the so-called Cheek defense represents a potentially viable, if rarely successful, defense strategy.9
The fact that some DLBs, such as conspiracy theories, are, unlike delusions, commonplace, shared, and often rooted in exposure to widely available online misinformation50 raises the possibility that purveyors of deliberate disinformation could be held liable for associated harms. For example, Edgar Maddison Welch accepted a plea deal and pled guilty to weapons and assault charges for bringing loaded firearms into a pizzeria to “self-investigate” a conspiracy theory about a Hillary Clinton–affiliated child pornography ring. But could a defense lawyer have argued successfully that sources of misinformation like InfoWars, a website that promoted “Pizzagate,” should have been held accountable as well?51 Indeed, InfoWars' operator Alex Jones is currently being sued for defamation related to promoting conspiracy theories about the mass shooting at Sandy Hook Elementary School being a hoax.52 Such cases illustrate the distributed responsibility and potential liability of spreading disinformation that can inform DLBs and motivate criminal behavior.
To date, defense efforts aimed at obtaining NGRI verdicts in cases of shared psychotic disorder or cult membership involving defendants without mental illness but having shared DLBs have met with mixed, but ultimately limited, success.23,43,53 Such rulings have been rooted in judgments that affiliation with DLBs was voluntary in the first place (similar to the nonviability of the claim that drug use was the cause of criminal behavior) along with the demands of a retributivist American justice system. A commentary for an earlier paper in this journal about extreme overvalued beliefs lauded its authors for helping to avoid conflating overvalued beliefs with “exculpatory mental illness” and to avoid permitting “defendants to exploit untidy areas of our system of classification” (Ref. 8, p 40). By way of contrast, it is argued here that the law should adapt to evolving scientific knowledge about psychopathology, beliefs, and free will, not the other way around, with new concepts about mental illness created to enable outdated legal principles. To that end, a modern U.S. legal system should bypass the slippery slope of equating delusions with reduced culpability by embracing the concept that individuals should always be held accountable for their behavior, but that justice should take the form of consequentialist rather than retributivist sentencing in the service of both correcting individual belief and behavior as well as broader deterrence.54,–,56
Conclusion
Although forensic decisions related to criminal responsibility demand crisp diagnostic boundaries and symptomatic distinctions, those are not inherent features of a psychiatric nosology that has largely failed to “carv[e] nature at its joints” (Ref. 2, p 10). Just so, the distinction between “mad” and “bad” may be morally expedient, but it is more artificial than we would like to imagine.
The contextual utility of DSM-5 for forensic psychiatry is fraught with difficulty,4,6,57 as the passage from the Cautionary Statement for Forensic Use of DSM-5 quoted at the start of this commentary makes clear. The DSM-5 definition of delusion is far from perfect,11 and there are few clear indicators to distinguish delusions from other DLBs in forensic psychiatric assessment.13 In clinical psychiatry, as with the case vignettes presented by Rahman et al.,10 a diagnosis of psychosis often involves best approximations based on associated symptomatic features beyond delusions. When DLBs are the only presenting symptom, the self-referential and grandiose nature of idiosyncratic delusions is an underemphasized feature to disentangle unshareable delusions from sharable DLBs. The fuzzy boundaries of DLBs and the propensity of even self-referential delusional beliefs to be shared in the Internet era nevertheless will likely continue to plague forensic psychiatry and invite debate both in and out of the courtroom for years to come.
Rather than attempting to reify imperfect boundaries to confine psychological phenomena, forensic psychiatry should work toward abolishing the binary conflation of delusion with insanity and nondelusion with criminal culpability. Concerns about NGRI verdicts or undeserved mitigation are rooted in folk intuitions about moral responsibility and our innate desire for retribution that are built into the American justice system. With regard to DLBs, forensic psychiatrists should focus on educating the legal system and the general public about more complex questions related to the relationship between belief conviction and socially deviant behavior while advocating for therapeutic justice,55 not assisting litigators in achieving a particular trial verdict.
Footnotes
Disclosures of financial or other potential conflicts of interest: None.
- © 2020 American Academy of Psychiatry and the Law