Delusional Perception Revisited

The meaning of the concept of schizophrenia has never been set in stone but always evolved. One of the more significant, recent changes of the schizophrenia concept concerns the diagnostic status of the first-rank symptoms. Schneider [1] (p. 134) famously collected a group of abnormal mental phenomena under the label of first-rank symptoms and argued that they had diagnostic importance for schizophrenia: “When any of these modes of experience is undeniably present and no basic somatic illness can be found, we may make the decisive clinical diagnosis of schizophrenia.” The first-rank symptoms were included in the operational diagnostic manuals and their diagnostic significance was greatest in ICD-10 [2], where the presence of one first-rank symptom for a 1-month period was enough to make the schizophrenia diagnosis if the exclusion criteria were not fulfilled. In DSM-5 [3] and ICD-11, the diagnostic weight of first-rank symptoms has been de-emphasized.

This change is not surprising as the diagnostic specificity of first-rank symptoms long has been questioned. For example, some studies found that these symptoms also occur in other mental disorders [4, 5] and even in the general population [6]. Yet, two systematic reviews highlighted conceptual and methodological issues in the empirical studies assessing the diagnostic specificity of first-rank symptoms [7, 8]. Moreover, a lack of conceptual clarity has always enshrouded the first-rank symptoms. Koehler [9], for instance, reported a discrepancy between what symptoms different authors considered as being of first rank. This, too, most likely, contributed to the dilution of their diagnostic specificity [7, 10]. In this vein, several scholars have recently raised concerns regarding the omission of first-rank symptoms (and their conceptual backdrop) from the major diagnostic manuals [11-13].

Delusional perception is one of the first-rank symptoms that has been erased from ICD-11 and which was never included in DSM. Today, knowledge of delusional perception is fading [14] and risks slipping out of clinical awareness. In Germanophone psychopathology, however, delusional perception was for a long time almost considered pathognomonic for schizophrenia. For example, Schneider [1] (p. 106) stated: “Where there is delusional perception we are always dealing with schizophrenic psychosis” (see also [15-17] for similar views). Given this symptom’s importance for the schizophrenia concept, we believe that it is worthwhile to revisit it. Delusional perception can be defined as a sudden, idiosyncratic, and often self-referential delusion triggered by a neutral perceptual content. Historically, especially in Germanophone psychopathology, the psychopathological nature of delusional perception was intensely debated ([18], p. 41; [50], pp. 108-9). In anglophone literature, aspects of this debate have been described by several authors [19-22], but thorough discussion of models – apart from Schneider’s – is sparse. The purpose of this study is to explore the clinical roots of delusional perception, elucidate the phenomenon, present the two predominant conceptualizations, and discuss their relevance for contemporary psychopathology and research on delusion.

Possibly the first conception of delusional perception was provided by Baillarger [23], who argued that it is a false interpretation based on a normal perception (cf. [24]). By stressing that it was triggered by a normal perception, Baillager separated delusional perception from illusion, which designates experiences wherein perceptual objects are mistaken for something else, e.g., “seeing a branch as an arm” ([25], p. 146). Berrios [24] (p. 98) refers to a case of Baillarger, where “a patient […] climbed a wall because he interpreted an (innocent) gesture of the administrator of the asylum as an order to do so.” Crucially, the gesture had been made, implying that the patient’s false interpretation had been triggered by a normal perception.

Jaspers [15], like Séglas [26], differentiated delusional perception from delusional idea, where the latter refers to a sudden delusional realization that is not triggered by a perception. Jaspers offered several examples of delusional perception, e.g., a patient seeing a man in a brown jacket and instantly believing the man to be the dead Archduke, who has resurrected [15]. One of Gruhle’s patients saw three marble tables in a café and was suddenly convinced that the world was about to end [17]. Most examples of delusional perception in the literature portray a perceptual experience in the visual modality, but any sense modality can be involved [1]. For example, one of our patients noticed an odd scent in the interview room (the doctors noticed it too) and instantly knew that the air was toxic, and she would die if she stayed.

The conspicuous gap between the neutral content of the perceptual experience and the delusion triggered by it has traditionally been viewed as an expression of schizophrenic incomprehensibility [15]. Jaspers [15] regarded delusional perception as a primary delusion, because the delusional content is not psychologically reducible, i.e., it cannot be traced back to the content of prior mental states. Similarly, Schneider [27] (p. 136) considered all first-rank symptoms, including delusional perception, as “psychological primaries and irreducible.” Jaspers [15] offered a crucial remark about ordinary perceptual experience, and, by extension, primary delusion, namely, that interpretation usually is not added post hoc to our perceptions, as if we subsequently added meaning to bare sense data. Rather, interpretation is built into our perception. In other words, our perceptions are immediately meaningful to us – we perceive something as something. For example, we do not see some oblong object with a possibly sharp blade in the kitchen drawer; we see a knife. Jaspers [15] (pp. 99–100) stated: “the experiences of primary delusion are analogous to this seeing of meaning,” i.e., delusional perception, an exemplar of primary delusion, is like ordinary perception in the sense that it involves an immediate experience of meaning. In this regard, it is noteworthy that the German term for delusional perception (Wahnwahrnehmung) carries connotations, which the English term does not. The German term for perception (Wahrnehmung) basically means “truth-taking” (Wahr-nehmung), i.e., taking something for true. The prefix (Wahn), a polysemic term, can here be translated into something like ‘mistaken presumption’. Thus, Wahn-wahr-nehmung literally means ‘a mistaken presumption that is taken for true’.

Importantly, Jaspers also offered another clarification, distinguishing delusional perception from ordinary perception, when he argued that in the former “the awareness of meaning undergoes a radical transformation. There is an immediate, intrusive knowledge of the meaning and it is this which is itself the delusional experience” ([15], p. 99). Thus, on Jaspers’ definition, primary delusions are not mediated by inferential errors, gradually solidifying into delusional beliefs, but emerge as an experience of immediate, intrusive knowledge [28-30]. In other words, primary delusions have the experiential character of an epiphany, i.e., the delusional meaning is revealed to the patient and forced upon her, bypassing her ability to distance herself from it [31]. The reality of primary delusion is taken for granted, because, as Blankenburg [32] (p. 99) put it, the delusional “evidence […] is presented before the judgment. It is of a ‘prepredicative’ nature.”

The immediate experience of meaning is indeed, as Jaspers pointed out, like that of ordinary perception in which we also take for granted – or take for true – the reality of perceptual objects. For example, we do not question the reality of the people we see waiting impatiently for a bus to arrive. As Palmer [33] (p. 6) put it, perception seems to be “a perfectly clear window onto reality.” Perhaps it was such considerations that led Berze [17] to suggest that delusional perception amounts to a fusion (Verschmelzung) of idea and perception by which the delusion acquires the immediate certainty that otherwise characterizes perception only – as nicely summarized by Sass [34] (p. 154): “The certainty of the idea stems from the certainty of the perception which was fused with it.”

Yet, the certainty that permeates delusional perception and other primary delusions is not the same as that which characterizes our ordinary, unquestioned certainty or faith in the reality of the perceived world [35, 36]. Although usually taken for granted, we can principally question and correct our perceptions, e.g., through further perceptions, reflection, or evidence provided by others. Thus, ordinary perception is characterized by an immediate, though relative certainty. This does not seem to be the case for delusional perception, which, partly due to its experiential givenness, is immune to doubt [37]. As one of Jaspers’ [15] (p. 100) patients put it: “everything is so dead certain that no amount of seeing to the contrary will make it doubtful.” Consistently, Müller-Suur [38] emphasized the immediate absolute certainty of psychotic experiences in schizophrenia – the correctness of the psychotic experience is directly given as an objective fact and needs no proof. In our view, the absolute certainty of delusional perception is also distinguished from the relative certainty of ordinary perception because the “revealed” delusional reality is not essentially anchored in or pertain to matter of affairs in the perceived world. Several authors have suggested that another, somehow more “real” world, detached from the shared, empirical world, looms up before the patient and that psychotic symptoms in schizophrenia often arise from or concern this other world [15, 31, 39-42]. This was originally encapsulated in Bleuler’s [43] concept of double entry-bookkeeping, emphasizing that patients with schizophrenia often operate with dual-world orientations: a private-solipsistic world and the common world shared with others [44]. In the context of delusional perception, Schneider [1] (p. 104) wrote: “The significance is always of a special kind; it almost always carries great import, is urgent and personal, a sign or message from another world. It is as if some ‘loftier’ reality spoke through the perception.”

Below, we present the two main psychopathological conceptualizations of delusional perception in the literature (for additional views, see [18, 45-49]). We start by outlining the two-link model, primarily espoused by Schneider. Then, we turn to Matussek’s critique of Schneider and his proposal of a Gestalt psychological model, i.e., the one-link model.

In anglophone psychiatry, the two-link model is, if not the sole, then at least the predominant model of delusional perception. It is usually ascribed to Schneider, Gruhle, and Jaspers (see, e.g., [50], pp. 108–110; [19], p. 143; [54], pp. 281–3), but their conceptualizations are in fact not identical. In the following, we focus mainly on Schneider since his model became the predominant one.

According to Schneider [1] (pp. 104, 106), delusional perception occurs “when some abnormal significance, usually with self-reference, is attached to a genuine perception without any comprehensible rational or emotional justification […] Perception itself is not altered but the meaning of it is.” Overall, Schneider’s [1] model appears congruent with the definition laid down by Baillarger, i.e., delusional perception is an abnormal interpretation, often involving self-reference, which is triggered by a “genuine,” “harmless,” or, as Jaspers [15] (p. 100) put it, “normal and unchanged” perception. This allows Schneider [1] to conceptually distinguish between two components or ‘links’ of delusional perception (see also [17]). He specifies the two links as follows:

1. from the perceiver to the perceived object

2. from the perceived object to the abnormal interpretation

Consequently, the disorder in delusional perception lies in the second link, i.e., in the abnormal interpretation of the perceptual object. By contrast, the perception of the object in the first link remains intact. Schneider [1] exemplifies how ordinary perceptual objects such as a furniture van parked in front of a house or an inscription on a gravestone in delusional perception acquires a special, important meaning that is directed towards the patient alone. Although the two-link model appears straightforward, there are aspects of Schneider’s account that are ambiguous, inviting critique and eventually other models to be put forth. We will focus on ambiguities raised by Schneider himself regarding the unchanged nature of perception, which, conveniently, converge with the main point of critique raised by Matussek. Another issue, which we only note in passing, is Schneider’s [1] liberal and, in our view, untenable acceptance of time-gaps, apparently up to years, between the perception and the abnormal interpretation of it.

Schneider [1] (p. 109; [51], pp. 33–5) insisted that delusional perception does not develop from prior emotional states, but often is preceded by a delusional mood or atmosphere, which he also referred to as the “preparatory field.” Delusional mood denotes a global, diffuse, typically anxious but also sometimes elated atmosphere of apprehension, an unspecifiable sense of something impending; an atmosphere that usually becomes increasingly self-referential, i.e., whatever is about to happen, will concern the patient directly [15, 52]. Schneider [1] (p. 112) stressed that “the quality of the whole experience in delusional perception is altogether different, though so elusive to define. It has a peculiar ‘numinous’ atmosphere of its own” and that delusional perceptions “are characteristically embedded in this atmosphere but not derived from it.” These considerations appear to cast some doubt on his central claim that perception itself, i.e., the first link in the two-link model, remains intact. In this regard, the ambiguities Schneider himself introduced to faithfully describe the phenomenon of delusional perception create problems for his own model.

These ambiguities also go to the crux of the difference between Schneider’s and Jaspers’ conceptions of delusional perception, which, despite their differences, often are lumped together as one and the same (see, e.g., [19], p. 143; [22], p. 383; [53], p. 11). Where Schneider focused on the abnormal interpretation, Jaspers [15] (p. 100) insisted that delusion perceptions are “not considered interpretations but direct experiences of meaning,” fundamentally separating his conception of delusional perception from Schneider’s two-link model. Matussek offered an alternative model of delusional perception, which diverges from Jaspers’ claim that perception remains “normal and unchanged” but converges with Jaspers’ ideas about the immediate experience of meaning and that delusional perception takes place within a transformed awareness of meaning – a transformation which Jaspers [15] (p. 99) described as an all-penetrating change (alles durchdringende Veränderung) that discloses “a world of new meanings.”

Central to Matussek’s [54] critique of the two-link model is his claim that it is built upon an outdated psychology. More specifically, he was critical of Wundt’s [55] psychological elementism, which analysed mental processes by breaking them into elements [56], but in doing so, Matussek believed, overlooked or sacrificed aspects of the whole (see similarly [50], p. 122). According to Matussek, it was this methodological point of departure that allowed Schneider to tease apart perception and interpretation and eventually argue that only the latter is disturbed in delusional perception. As Matussek [57] (p. 25, our translation) put it: “Classical psychopathology was of the opinion that no changes to perception undergird delusional perceptions, but that it simply revolves around a distortion of thinking.” He was especially critical of Schneider’s emphasis on the second link of the model, i.e., the abnormal interpretation [54]. Contrary to Schneider, Matussek [54] (p. 310, our translation) argued that the abnormal meaning “is not primarily made manifest, invented or otherwise ‘brought about by thinking,’ but rather is immediately experienced in the object due to a changed perceptual world.” Thus, Matussek [54] proposes a model that offers an explanation of how delusional meaning is met (angetroffen) immediately in the perceptual object.

His proposal draws upon Gestalt psychology. Briefly, Gestalt psychology posits that we do not primarily perceive singular objects but rather organized wholes (Gestalts) within which objects are embedded. The whole is more than the sum of its parts and, contrary to elementism, the whole is therefore not deducible from its parts [58-60]. From this theoretical perspective, Matussek proposed that delusional perception takes place within a changed perceptual world that is characterized by a destabilization of the ordinary meaning structures inherent in perception.

In brief, Matussek’s [54] model posits that, in delusional perception, so-called essential properties (Wesens-eigenschaften) of a perceptual object ‘stand forth’ (hervortreten) in such a way that they achieve a certain ‘weight’, lending them ‘protection’ (Schutz) from other relevant properties that normally would stabilize the meaning of the overall perception. Matussek [54] illustrates his guiding idea by referring to a picture frame. A picture frame encloses a picture and makes it stand out from the background (e.g., the wall and whatever else hangs on it). Similarly, in delusional perception, the destabilization of meaning structures allows certain aspects of the perception to stand out from the perceptual context, becoming ‘framed’ (eingerahmt), and thereby barred from other aspects of the perception decisive to its meaning. Crucially, essential properties also ‘stand forth’ in ordinary perception, e.g., “a peaceful village, the proudly towering castle, […] the tenderness of a young girl, the aged expression of a dystrophic child, the feminine way of speaking” [54] (p. 294, our translation). In delusional perception, however, the ‘standing forth’ of certain properties co-occurs with a blocking-out (Ausschluss) of other relevant properties and a loosening of the natural perceptual context (Auflockerung des natürlichen Wahrnehmungszusammenhanges) [54, 61]. A few of Matussek’s [54] (pp. 298–9, our translation) examples may help illustrate his proposal:

A schizophrenic student reports that one day, as his landlady spoke of cold and warm water, he suddenly realized the true meaning of cold and warm. Warm meant warm-heartedness, sympathy, affection; cold, on the contrary, meant rejection, contempt.

I held my father for the devil, who wants my soul, because my father is limping. One brother looked at me for so long and scrutinizing that he appeared to me like a policeman.

Usually, the meaning of a perception is held in check by an organized perceptual whole, a sort of meaning equilibrium, which prohibits singular aspects of the perception from breaking apart, drawing extraordinary attention to themselves, and changing the perception’s meaning. In delusional perception, the whole’s organizing and structuring effect on the perception’s meaning is weakened, allowing singular, often idiosyncratic, and self-referential aspects of the perception to take over and exert an unimpeded dominance on the perception’s meaning, forcing themselves on the patient. Thus, Matussek’s [54] model consists only of one “link,” i.e., the delusional meaning is contained within the perception itself (cf. [20, 62]).

Conrad [50], also drawing on Gestalt psychology and Matussek’s work, proposed a similar one-link model. In an attempt to refine Matussek’s model, Conrad [50] emphasized what he called the “setting-free of essential properties” (Freisetzung von Wesenseigenschaften). In contrast to Matussek [54, 61], Conrad argued that the essential properties are not intrinsic to the perceptual object but embedded in a “cloud of essential properties” (Wolke von Wesenseigenschaften), which is intersubjectively constituted, modulated, and constrained. These refinements allowed Conrad to clarify exactly what is pathological in the standing forth, blocking-out, and loosening of the natural perceptual context in delusional perception. Paradoxically, the patient is initially surrendered to an extraordinarily broad horizon of meaning (viz. the “setting-free of essential properties”) but becomes captivated by an idiosyncratic meaning, spurred by a single or a few properties’ complete domination of the perception’s meaning. In this regard, Conrad [50] (pp. 86, 93, 142, 270), argued that the patient has lost her ability to transcend the self-centred position (Überstiegsunfähigkeit), i.e., she cannot escape the idiosyncratic, self-referential meaning, leading him to speak of an “imprisonment of I.” In the end, it is this profound dis-location (Ver-rücktheit) from what is intersubjectively and contextually accepted about the perceptual object that makes the delusional perception mad (verrückt).

Delusional perception has long been closely connected to schizophrenia, and despite its absence in DSM and disappearance in ICD-11, it will remain a clinically significant symptom of the disorder. In this article, we presented the two predominant models of delusional perception, emphasizing points of divergence and convergence, in sum, arguing that its psychopathological nature is unclear. In our view, this is not only the case for delusional perception but for many psychopathological concepts, e.g., psychosis [63, 64], delusion [28], and hallucination [65]. When conducting psychopathological or diagnostic assessment in clinical settings or research, we use psychopathological concepts as if they were well-defined – albeit they regularly are not. When psychopathological concepts are ill-defined, it affects their clinical applicability, i.e., unsharp conceptual boundaries stream into the clinical encounter with patients, introducing vagueness and making it difficult to reliably assess if the symptoms, which the concepts are meant to capture, are present or not. Thus, well-defined psychopathological concepts are key to clinical practice and research in psychiatry, but this kind of conceptual research is, unfortunately, continuously underrepresented in psychiatric research [66-68].

To obtain an adequate definition of a psychopathological phenomenon, the first step is to bring it into proper focus [15]. In our investigation of delusional perception, some key questions emerged: is it essentially an abnormal interpretation triggered by a normal perception, or is the delusional meaning rather contained within a changed perception? Is the involved perception intact or not, and, if not, how exactly is it disturbed? The two-link model proposes that perception is intact but acknowledges that delusional perception is ‘embedded’ in a delusional atmosphere, making the quality of the whole experience ‘altogether different.’ The one-link model elucidated the character of the changed perceptual world, proposing that delusional meaning is met directly in the perceptual object and not brought about by an abnormal interpretation. Despite their differences, both models insist that delusional perception is a primary delusion, occurring within a changed experiential framework [15]. These points of convergence are far from trivial. First, primary delusions emerge in a sudden, immediate, revelatory manner that bypasses one’s critical faculties. Consequently, primary delusions are not adequately captured by standard definitions of delusion as erroneous beliefs based on incorrect inferences about worldly matters and held with such certainty that they cannot be corrected [28]. Second, despite the nuances within classical psychopathology concerning delusional perception, most accounts converge on the fact that these pathological phenomena do not emerge ex nihilo. Rather, they emerge within an altered experiential framework, viz., self/world relation. The significance of this insight cannot be overstated – unfortunately, it is all but forgotten in contemporary psychiatry.

To exemplify the latter point, in this manuscript, we have twice quoted Schneider, stating that if delusional perception or a first-rank symptom is present (and no organic illness is found), it suffices to make the diagnosis of schizophrenia. This is precisely the view that is usually ascribed to Schneider, and it was with this diagnostic weight that the first-rank symptoms entered the operational diagnostic manuals. However, our quotes from Schneider regarding the diagnostic significance of the first-rank symptoms were decontextualized, but if we consider the context from which these quotes were extracted, another, more nuanced picture begins to crystalize [69]. Most importantly, prior to describing the first-rank symptoms, Schneider [1] (p. 95) made crucial methodological remarks, stating that a “psychotic phenomenon is not like a defective stone in otherwise perfect mosaic” and that schizophrenia “always involves an over-all change.” In his textbook, he went on to describe this “over-all change” as often involving a “radical qualitative change” of consciousness, epitomized by disturbances of self-experience and failing experiential demarcation of self/other [1] (pp. 100, 120–21, 134). Schneider [1] (p. 98) also argued that if this “over-all change” cannot be found in the patient, then we must be hesitant to assess such apparently abnormal experiences as symptoms of schizophrenia. Essentially, first-rank symptoms were only considered to have diagnostic specificity for schizophrenia if they occurred in a psychopathological context marked by the specific “over-all change.” Classically, this “over-all change” was understood through the prism of Ichstörung [17, 70-72]. In contemporary psychopathological research, Ichstörung has been conceptually elaborated and empirically assessed under the notion of self-disorders, which hyper-aggregate in schizophrenia spectrum disorders [73, 74]. Thus, this psychopathological Gestalt was considered constitutive of the first-rank symptoms’ diagnostic significance, not the symptoms in themselves [7]. Consequently, Schneider [1] insisted on the necessity of a global, contextual assessment of psychopathology, embracing Jaspers’ [15] phenomenological method in which parts (symptoms and signs) and wholes (psychopathological Gestalts) are assessed in a holistic, reciprocal manner. By contrast, when the first-rank symptoms were included as diagnostic criteria in the operational manuals, they were no longer considered as aspects of a specific psychopathological Gestalt but instead viewed as atomistic, isolated symptoms that could be assessed independently of the psychopathological context in which they occur. In our view, the dwindling importance of the first-rank symptoms is related to these conceptual and methodological issues. In brief, we must again start to comprehend and recognize the different psychopathological Gestalts of mental disorders [75, 76] – just like a proper appreciation of delusional perception demands a grasp of the altered experiential framework within which it is embedded.

The phenomenology of delusional perception, e.g., its revelatory givenness, embedment in a profoundly altered experiential framework, and irreducibility to the content of prior mental states, has important implications for current cognitive research on delusion and cognitive psychotherapy of delusion. Generally, the cognitive models of delusion share a blueprint: the patient has an unusual or anomalous experience that she tries to make sense of with a delusional explanation [77, 78]. Maher’s [79, 80] classical one-factor theory encapsulates this common blueprint, i.e., “a delusion is a hypothesis designed to explain unusual perceptual phenomena” [79] (p. 103). In other words, patients are assumed to concoct various theories or explanations in an attempt to understand their anomalous experiences. Maher’s one-factor theory was subsequently found insufficient by some [81, 82], prompting the argument that an additional factor is required for delusions to arise. The so-called two-factor theory thus add aberrant belief evaluation (e.g., disordered use of stored knowledge, deficient Theory of Mind, inability to critically assess experience, etc.) as a second factor to the framework of the one-factor theory.

More presently, the predictive coding model of delusion attempts to subsume the two factors within a Bayesian framework [83] (p. 357). There are multitudinous nuances to this “unified theory” research framework [84], and therefore we refer to the predictive coding model in general terms. Briefly put, the guiding hypothesis of the model is that an imbalance between prior expectations and present sensory data – i.e., so-called prediction errors – give rise to top-down influence and formation of delusion [85-88]. Differently put, due to such prediction errors, a sense of surprise, novelty, or salience [89] occurs that summons the patients to “develop a set of beliefs that must account for a great deal of strange and sometimes contradictory data” [90] (p. 56). Thus, as Corlett et al. [83] (p. 348) writes – citing Maher [79] – “a delusion represents an explanatory mechanism, an attempt to impose order on a disordered perceptual and cognitive world.” Given the complexity of this model, it is not always entirely clear on which explanatory level these explanations unfold. Here, we will not take issue with this model as a theory of sub-personal explanations of delusion (cf. [91], p. 150). However, speaking of delusion as an attempt to make sense of or explain strange experiences due to prediction error dysfunction (see [83], p. 361; [85], p. 636; [90], p. 56) signals that this model not merely pertains to the sub-personal but also to the personal or experiential level.

These cognitive models – one-factor, two-factor, and predictive coding – appear slightly akin to Schneider’s two-link model but, crucially, with opposite operational signs. Contrary to Schneider’s model, the cognitive models emphasize the altered perceptual experience, which then is hypothesized to foster a delusional explanation. Despite this difference between Schneider’s two-link model and the cognitive models, Matussek’s [54] (pp. 306, 310; [57], p. 25) critique of Schneider’s model also applies to the cognitive models, i.e., these models likewise appear overly cognitive or intellectual. Their emphasis is on inferences or top-down beliefs [92] (p. 9), rather than on the patient being confronted with the delusion in an altered experiential world. By striving to explain delusion as such [77, 93], the cognitive models conflate primary and secondary delusions and thus overlook the phenomenological character of primary delusions [91, 94, 95]. Furthermore, especially with predictive coding models, there is a risk that they become excessively internalistic [96] and hereby liable to ignore the profound change of the experiential framework in primary delusions [15, 28], including alterations of intersubjectivity and environmental interaction [97, 98]. In our view, the cognitive models seem more apt for explaining secondary delusions, where there is a gradual, inferential, and psychologically comprehensible solidification of the delusional belief, which is quite different from the revelatory character of primary delusions such as delusional perception. If, by contrast, the cognitive models insist on explaining delusion as a unitary phenomenon, then these models should ideally – instead of keying in on specific anomalous contents of experience – try to integrate the persistent phenomenological findings of primary delusions occurring in a profoundly altered experiential framework and reconsider the influence of this changed framework for delusion formation and maintenance.

This also has implications for cognitive therapies, which we, for the sake of brevity, will divide into “front-door” and “backdoor” approaches [99] (p. 623). Traditional Cognitive-Behavioural Therapy (CBT) constitutes a front-door approach, which regards delusion as a dysfunctional belief and traditionally treats delusion by challenging it on rational grounds, e.g., by exploring alternative explanations, tracing the delusion to prior irrational beliefs, or testing the validity of beliefs (e.g., [100, 101]). Traditional CBT appears relevant for explaining and treating secondary delusions, which occur within a normal experiential framework, but it also appears largely inconsistent with the phenomenology of primary delusions, which occur in an altered experiential framework [28, 102]. Contrastively, Metacognitive Training (MCT) for psychosis constitutes a so-called backdoor approach that seeks to raise awareness of cognitive biases (e.g., attributional bias, ‘jumping to conclusions’, and overconfidence), which are theorized to subserve and attribute to psychosis and delusion [99, 103, 104]. Albeit MCT does not attempt to alter delusional beliefs directly but rather targets what could be denoted as the secondary factor of the two-factor model, i.e., various cognitive biases, the approach is still confined to a cognitive framework. Like traditional CBT, MCT appears best fit to target secondary delusions, as hinted at by Moritz et al. [103] (p. 62): “Raising metacognitive awareness is hoped to intercept the progression from false appraisals of certain subclinical (‘as if’) experiences to fixed false (delusional) beliefs.” In this regard – similar to the cognitive theories of delusions – Matussek’s understanding of delusional perception as an overwhelming delusional confrontation within an altered experiential world raises important questions to contemporary psychotherapeutic approaches for such delusions.

In sum, grasping the nature of psychopathological phenomena is not an outdated enterprise, something that safely can be left to the historians of psychiatry. Psychiatric concepts continue to influence the way we diagnose, treat, and conduct research in psychiatry, regardless of whether we reflect upon these concepts or not. In our view, psychiatry is best served by adopting a curious and critical attitude toward its own concepts – as aptly put by Marková and Berrios [105] (p. 194): “Empirical research in psychiatry will only ever be as good as the delineated concepts.”

The study is of a conceptual nature and only reports already published research. It does not include new data from humans or animals.

The authors have no conflicts of interest to declare.

The authors have not received any funding.

Kasper Møller Nielsen, Julie Nordgaard, and Mads Gram Henriksen identified the object and planned the study. Kasper Møller Nielsen wrote the first draft of the manuscript, which was revised by Julie Nordgaard and Mads Gram Henriksen. Kasper Møller Nielsen, Julie Nordgaard, and Mads Gram Henriksen have contributed to and accepted the final version of the manuscript.