**2.2. The heterogeneous conceptualization and assessment of psychotic-like experiences**

adjusted for co-occurring PLE, negative- and disorganized-like symptoms [51]. Notably, this finding was in line with suggestions that ideas of reference (in contrast with paranoia) must

Odd beliefs as measured by the PAGE-R were prominently represented in the detected negative associations. Importantly, odd beliefs refer not to beliefs in a strict sense, but to experiences characterized by "seeing" patterns in noise (e.g., meaningful linking of separate events, correctly anticipating future events). Importantly, this sets odd beliefs apart from scales assessing e.g., magical thinking that often contain paranormal beliefs rather than experiences, which might be less relevant for the study of subclinical psychosis [54]. Further, in contrast to most studied forms of delusion-like experiences (e.g., suspiciousness), odd beliefs in healthy individuals were particularly enriching and positively-valenced experiences. Nonetheless, odd beliefs are conceptually similar to other delusion-like experiences and may be associated with indicators of psychosis proneness, such as biases in probabilistic reasoning and a tendency to jump to conclusions [55, 56], alterations in attributional styles [57, 58], differences in theory of mind [59], and magical ideation [60]. Importantly, experiences similar to odd beliefs have been suggested to reduce distress in perceptually ambiguous or stressful situations [61, 62] and to facilitate (perceived) control as well as to confer confidence and agency under lack of control [63]. Before this background, it was speculated that odd beliefs in healthy individuals might represent a psychologically stabilizing cognitive response to burdensome experiences [28]. Hence, despite their delusion-like quality, odd beliefs might paradoxically exert a positive effect on psychological well-being. Intriguingly, a new study investigating specificities between PLE and forms of childhood trauma found for the first time that odd beliefs in healthy adults were associated with stronger self-concept of own competences (SC), when adverse childhood experiences were held constant [64]. In contrast, paranoid-like experiences remained negatively associated with SC once adjusted for childhood adversities. SC is the fourth dimension of locus of control according to Rotter's social learning theory [65, 66] and refers to the self-perceived capability to act in new, difficult or ambiguous situations [67]. Notably, addressing SC might also strengthen self-esteem, which has been identified by individuals with schizophrenia to be the most important treatment target [68]. Moreover, strengthening SC in therapy might help to alleviate psychotic symptoms SC [66]. Due to their positive association with SC the question was raised if odd beliefs might contribute to resilience toward mental illness, despite conferring an inaccurate perception of the world [64]. Further, as an individual's inability to give meaning to an adverse experience is important in determining its long-term effect [69], the tendency to have positive delusion-like experiences

The presented findings suggest that despite their tendency to co-occur, PLE may be variously implicated in mental illness and mental health. These results are in line with earlier suggestions that a co-occurrence of characteristics seen in pathological and non-pathological conditions must not necessarily mean that they are indicators of psychopathology [70]. More specifically, some characteristics could simply be by-products of the psychosis dimension but not be clinically relevant *per se*. However, it is cautioned to jump to premature conclusions and these symptom-level insights require further investigation, as there are several limitations to be considered. For example, all studies applied cross-sectional study designs,

not necessarily be burdensome [52, 53].

8 Psychosis - Biopsychosocial and Relational Perspectives

might perhaps be exploited for therapeutic purposes.

Self-report instruments for psychotic-like experiences (PLE) are a central source of information in epidemiological research on subclinical psychosis. However, it is mostly not regarded that these instruments are tied to certain conceptualizations of (subclinical) psychosis and originally served a specific purpose [24, 25, 34]. Notably, many instruments used to assess PLE stem from schizotypy research and are fundamentally influenced by the underlying schizotypy model and the assumed link between schizotypal personality features and schizophrenia. For example, one of the earliest and most frequently used schizotypy scales is the Magical Ideation Scale (MIS, see [41]) [34]. It bases on Meehl's [11] quasi-dimensional schizotypy model and as a screening tool for psychosis proneness (and vulnerability to schizophrenia in particular) its scope is restricted to illness and schizophrenia risk [25]. Accordingly, the MIS conceptualizes "psychotic-like symptoms" as attenuated or milder forms of Schneiderian first-rank symptoms of schizophrenia that manifest in the acceptance of unconventional forms of causality. Hence, the items in the MIS might have a distinct bias toward schizophreniarelated PLE. Furthermore, the selection of items might not be reflective of different forms of PLE in the general population, as items with extremely high and low difficulties were chosen to attain normality of the scale score. In comparison, the popular Schizotypal Personality Questionnaire (SPQ, see [49]) was constructed to screen for schizotypal personality *disorder* according to DSM-III-R criteria and not to assess schizotypal personality organization [25]. Hence, its categorization of PLE into paranormal beliefs/magical thinking, ideas of reference, suspiciousness, and unusual perceptual experiences is entirely derived from a theoretical diagnostic profile. Notably, item-level factor analyses have repeatedly produced incongruent categorizations of the experiences [71]. One of the most widely used self-report instrument to assess PLE not founded on schizotypy research is the Community Assessment of Psychic Experiences Questionnaire (CAPE, see [47]). The CAPE was created against the theoretical background of the extended subclinical psychosis phenotype [15] and is an attractive tool for clinical and research use, as it is comprehensive and measures not only the frequency of PLE but also distress associated with them [48]. In contrast to questionnaires assessing attenuated versions of clinical symptoms, the CAPE inquires symptoms seen in patients with psychotic disorders (albeit toned down by adding "as if" to the questions). Hence, existing instruments assessing PLE may differ regarding the constructs they cover, the qualitative expression of PLE ("symptoms" vs. attenuated versions thereof), as well as their rating format. Importantly, choosing one instrument over the other may profoundly affect the ensuing results [25]. For example, although similar sex differences have been found in non-clinical samples as in schizophrenic patients [72, 73], the non-detection of sex differences in a community sample has led researchers to conclude that they only present themselves in full-blown psychosis but not in sub-threshold states [74]. Interestingly, using a sample of healthy individuals, a study could replicate the detection and non-detection of sex differences in the latter studies, depending on the scales for PLE that were being analyzed [44]. It was suggested that scales including fewer and more severe or difficult items (e.g., someone has power over your thoughts [75]) might not be able to detect sex differences in healthy individuals, whereas scales inquiring more and attenuated experiences might do so (e.g., I have sometimes felt that strangers were reading my mind [41]). Importantly, different populations across the psychosis continuum ranging from non-disordered schizotypes, to prodromal patients, to patients with a schizotypal personality disorder, and to psychotic patients might all experience positive(-like) symptoms such as odd beliefs. However, these groups might differ regarding the relative prevalence of increasingly severe forms of experiences ranging from magical thinking to full-blown delusions [24]. Hence, depending on the sample, the research question, as well as the theoretical model of psychosis, some surveys might be more adequate to be used than others.

symptoms, facilitate more adequate treatment, and counteract stigmatization [29]. Further, it might also help to integrate findings across studies implementing different psychometric

Subtypes of Psychotic-Like Experiences and Their Significance for Mental Health

http://dx.doi.org/10.5772/intechopen.78691

11

More recently, similar categorizations of three basic types of PLE have been proposed, suggesting that: (1) some indicate a specific vulnerability toward psychosis while (2) others might be non-specific and also be implicated in the development of affective disorders, and (3) some might not be associated with any clinical disorder at all [30, 39]. It has been speculated that some PLE such as paranormal beliefs are benign and might explain why they are mostly not associated with mental illness [32]. In contrast, it has been suggested that PLE specifically associated with distress and poor functioning might be more likely to indicate vulnerability to psychotic disorders [30]. However, it yet remains to be clarified to which category certain

Recently, to shed light onto possible categorizations of PLE, a study investigated unique associations of certain PLE with subclinical symptoms relevant for psychosis spectrum disorders [76], i.e. negative-like symptoms, affective symptoms (anxiety, depression), and other psychological difficulties in a sample of healthy adults [28]. Referring to the model introduced above, following categorizations are suggested: Paranoid-like experiences in healthy individuals might specifically indicate vulnerability to psychosis (category 1), as they were the only significant predictor of schizophrenia-like negative symptoms (physical anhedonia, no close friends, and constricted affect) but were not associated with any type of affective symptoms. In contrast, hallucination-like experiences were uniquely associated with experiences from the anxiety spectrum (e.g., phobic anxiety, obsessive-compulsive symptoms) but not with negative-like symptoms. Further, ideas of reference were a positive predictor of anxiety symptoms and depressive symptoms. Therefore, the latter PLE might belong to the category of non-specific PLE, hence, predisposing toward affective and psychosis spectrum disorders (category 2). Lastly, paranormal beliefs and PAGE-R odd beliefs did not positively predict any of the subclinical difficulties, which might reflect that they are not associated with any clinical disorder at all (category 3). The latter categorization was underlined by the observation that paranormal beliefs and odd beliefs were negatively associated with various psychological difficulties. Notably, these findings raise the question if more categories for PLE might be needed that account for associations of PLE with well-being and stronger resilience [51, 64] and lower load of negative-like symptoms. However, it remains to be determined if these findings can be accommodated within a framework encompassing three classes of PLE.

The tentative categorization of PLE presented above requires more data and replications in samples representative of the healthy general population. Ultimately, longitudinal studies are needed to determine if specific PLE predict certain psychosis spectrum disorders more likely than other diagnoses and how they are implicated in the maintenance of mental health. Notably, other symptom factors that are relevant for determining the psychopathological significance of PLE were not regarded. Amongst other factors, these include intrusiveness, distress, and frequency of experiences as well as the associated development of functional impairment [47]. Furthermore, similar analyses are needed including other subclinical difficulties that might be

part of the psychosis phenotype [16], such as disorganized symptoms and mania [51].

instruments and conceptualizations of PLE.

PLE should be assigned [32].

In addition to the mostly non-transparent choice of instruments [34] and their heterogeneous designs, unclear content validity of scales may additionally entail mixed results across studies and contribute to a blurred picture of psychosis [25]. Studies investigating symptom-level associations have applied multiple regression modeling to account for overlapping variance between different PLE scales in order to gain insight into their specific psychopathological significance [29, 30]. Whereas these results are meaningfully interpretable, the reliability of the interpretations ultimately depends on the choice of instruments and the (content) validity of the applied scales. Notably, scales measuring certain PLE may often conflate different constructs impeding a reliable interpretation of results, as exemplified by the MIS [24, 46]. Additionally, the emergence of ever-new concepts and terms as well as the interchangeable use of different terms for PLE with overlapping but not necessarily identical meanings has resulted in a "near Babylonian speech confusion" that hinders clarity in the nomenclature, blurs sources of inconsistencies between findings and constricts their interpretation [24, 32, 36]. Hence, to successfully elucidate the complex structure of psychosis, researchers should have detailed knowledge of existing constructs and be familiar with the limitations of their operationalizations.
