Narrative insight in psychosis: The relationship with spiritual and religious explanatory frameworks

Abstract

Objective

When considering psychosis, the concept of narrative insight has been offered as an alternative to clinical insight in determining individuals’ responses to their difficulties, as it allows for a more holistic and person‐centred framework to be embraced within professional practice. This study aims to explore the validity of the narrative insight construct within a group of people who have experienced psychosis.

Design

Inductive qualitative methods were used to explore how eight participants utilized spiritual or religious explanatory frameworks for their experiences of psychosis and to consider these in relation to the construct of narrative insight.

Methods

Semi‐structured interviews were undertaken with individuals who identified themselves as interested in spiritual or religious ideas and whose self‐reported experiences which were identified as akin to psychosis by experienced academic clinicians. Transcriptions from these interviews were subject to interpretative phenomenological analysis within a broader research question; a selection of themes and data from the resultant phenomenological structure are explored here for their relevance to narrative insight.

Results

Participants discussed spiritual and biological explanations for their experiences and were able to hold alternative potential explanations alongside each other. They were reflective regarding the origins of their explanations and would describe a process of testing and proof in relation to them.

Conclusions

These findings suggest that the narrative insight construct has the potential to be a valid approach to understanding experiences of psychosis, and challenge the dominance of the clinical insight construct within clinical practice.

Practitioner points

• Clinicians should value the explanatory framework for experiences which are provided by individuals experiencing psychosis, and encourage them to develop a framework which is coherent to their own world view rather than predominantly pursuing a biomedical explanation.

• Assessments of psychosis should be adapted to include an understanding of the cohesiveness of the individual's explanatory framework and personal value to them, with a reduced focus on their acceptance of biomedical models of ‘illness’.

• Care and care research for individuals experiencing psychosis should consider the value of narrative insight within future developments.

Background

Qualitative studies regularly demonstrate the complexities in negotiating personal identity that present themselves to people whose experiences are classified as ‘psychosis’.¹ McCarthy‐Jones, Marriott, Knowles, Rowse, and Thompson (2013) produced a meta‐synthesis of inductive qualitative research in the field of psychosis between 2000 and 2010. Identifying four themes related to the psychological journey of psychosis (‘Losing’, ‘Identifying a need for, and seeking, help’, ‘Rebuilding and reforging’, ‘Better than new: gifts from psychosis’), the 97 articles demonstrated vast diversity in reflections amongst these individuals. This demonstrates the importance of the debate regarding appropriate explanatory frameworks for psychosis, including where ownership of explanation belongs (with the individual or with their supporting services).

Narrative insight

Common with most mental health research and treatment, the dominant paradigm regarding psychosis is the biomedical model. Manifestations of distress and difficult psychological processes are taken to be signs of a ‘mental illness’, a state of the psyche in which the individual is significantly – and therefore ‘clinically’ – abnormal. This model is exemplified in the processes of manuals such as the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM–5; American Psychiatric Association [APA], 2013), narrowing the array of human distress to one or another diagnostic category, with an assumption that belonging to one of these categories implies an underlying ‘psychopathology’ shared with others in this category.

For many of the so‐called psychotic disorders, biomedical assumptions result in a circular assumption regarding the individual's explanatory frameworks; specifically, an individual's non‐acceptance of the biomedical paradigm is determined as lacking insight – for example, ‘Unawareness of illness is typically a symptom of schizophrenia itself rather than a coping strategy’ (APA, p.101). Research around insight – often ‘clinical insight’ – is therefore typically framed within this paradigm (e.g., Pijnenborg, van Donkersgoed, David, & Aleman, 2013), and although often described as ‘multidimensional’, the dimensions discussed mostly appear to exist within this overarching framework.

The debate around the appropriateness of psychiatric diagnosis and concepts of ‘illness’ in relation to human distress remains live (e.g., British Psychological Society, 2013; Johnstone & Boyle, 2018); as such, the concept of lacking clinical insight by not accepting psychotic experiences as an illness remains open to question. Lysaker, Yanos, and Roe (2009) present a narrative review of scientific literature, demonstrating the paradoxical findings that follow from studies which adopt the traditional view of insight (such as having both positive and negative effects on measures of recovery). They propose that the construct of insight should move beyond the acceptance or rejection of the one framework; rather, insight is a meaning‐making process in which the awareness of experiences is contextualized within the individual's world view and therefore provides a more robust foundation to move forward in a positive (i.e., ‘recovery’) manner – ‘to possess insight is to have a story that makes personal meaning of what has happened and what can be done in the future to move forward in one's life’ (p.116). The theory suggests that personal meaning‐making can create a narrative that ‘works’ for the individual, potentially regardless of their acceptance of the biomedical paradigm. Some of the authors are also co‐authors in a review of recovery‐oriented service provision (Leonhardt et al., 2017), which proposes that current services need to reconsider a more comprehensive approach to the prioritization of personal meaning‐making within routine care, demonstrating the direct implications of a move away from clinical insight as a dominant idea.

David Roe and colleagues have developed the term narrative insight to expand the insight construct to fit with an individual's personal story (e.g., Lysaker et al., 2009; Roe & Davidson, 2005; Roe, Hasson‐Ohayon, Kravetz, Yanos, & Lysaker, 2008; Vohs, George, Leonhardt, & Lysaker, 2016); similarly, Tranulis and colleagues (e.g., Tranulis, Freudenreich, & Park, 2009) have independently developed the concept, maintaining a broadly similar position. Although this writing mostly maintains ‘illness’ terminology, the paradigm discussed is one in which the capacity and importance of developing personal narratives around psychosis – and of valuing non‐biomedical explanations – are placed in the foreground. For example, Roe et al. (2008) present a mixed methods study; the inductive analysis placed the participants into four distinct groups, which were defined by distinct levels of belonging to each of the four inductive themes: ‘belief in having an illness’, ‘belief in having symptoms’, ‘acceptance of the diagnostic label’, and ‘active involvement in searching for explanations’. Although the four‐dimensional, inductively based framework appeared to describe the sample along four groupings, the traditional measure of insight only distinguished those in a ‘Rejects illness/Searches for name’ group. This suggests that the four groups are defining a unique concept, rather than merely reflecting different points on a continuous spectrum of traditional clinical insight.

The narrative insight construct has gone on to be applied to the analysis of qualitative data in a number of studies. For example, Macnaughton, Sheps, Frankish, and Irwin (2015) reported how their participants explored the illness paradigm as an active decision‐making process and demonstrated that it is possible to develop a coherent narrative in the early stages of psychosis (though a particularly challenging process).

Although theoretically attractive as a valid approach for considering the personal meaning of psychosis, developing narrative insight is not a singularly positive process. For example, the active construction of personal explanations for psychosis can lead to some sense of guilt (Jones et al., 2016), whilst Lysaker et al. (2013) note that insight development can be associated with increased levels of depression. As such, potential negative consequences of developing narrative insight should be acknowledged as factors for management in services. Nevertheless, it is an inevitable feature of human experience to engage in meaning‐making around significant events in our lives (Roe & Davidson, 2005); it is thus a process which will occur in relation to psychosis and should therefore be properly supported for all individuals to create a narrative from which meaningful steps forward can be developed.

Whilst the terminology of narrative insight specifically is relatively novel, the importance of personal meaning‐making in the approach to psychosis has been discussed clinically and in the literature for many years; for example, Geekie and Read (2009) outline in their introductory chapter the long history whereby the individual narrative has been an important part of theories about ‘madness’ in the broadest terms. Narrative insight brings a theoretical addition to this broader discourse as a means to bring personal meaning‐making directly into the mainstream frameworks broadly adopted in psychiatric services, directly addressing clinicians’ concerns about clinical insight or the supposed lack thereof.

Spiritual and religious belief

Narrative insight is proposed as a broad theoretical construct and allows for any number of potential explanatory frameworks by the individual. As an illustration of the way in which the construct can operate with a less mainstream (in psychiatric terms) conceptualization of such experiences, we now focus on the application of spiritual and religious beliefs within the framework.

Spiritual and religious beliefs are important for many people experiencing psychosis, as demonstrated in inductive studies included in the meta‐synthesis (McCarthy, Marriott, et al., 2013). Spirituality has often been explored as a mediating factor in the relationship between psychosis and recovery; for example, two recent reviews (Bonelli & Koenig, 2013; Gearing et al., 2011) found a fairly balanced picture, suggesting that spiritual and religious beliefs can be either positive or negative in people's adjustment to psychosis.

Religion and spirituality are often conceptualized in terms of coping styles (e.g., Mohr et al., 2012), but it also appears that spiritual and religious beliefs affect the experience of psychosis through the influence on explanatory frameworks. Numerous studies demonstrate how people from various cultural or ethnic backgrounds are more likely to use spiritual explanations as a predominant factor in their understanding of psychosis (e.g., Pakistani – Zafar et al., 2008; UK Black African or Black Carribean – Codjoe, Byrne, Lister, McGuire, & Valmaggia, 2013). Spiritual explanatory models of illness were assessed longitudinally by Huguelet, Mohr, Gilliéron, Brandt, and Borras (2010); using five a priori factors, they found that when conceptualized as an explanatory model rather than a coping style, each factor could be either positive or negative, dependent upon how the participant integrated the framework into their experience. They found that the nature of explanatory frameworks changed over time, but found no association between the kind of framework adopted and clinical outcome.

Some qualitative studies provide greater depth into the insights of how spirituality and religion relate to explanatory frameworks for experiences of psychosis. Drinnan and Lavender (2006) found that participants made elaborate use of religious explanations for their experiences, with religious frameworks having both a positive and a negative effect upon their adaptation to psychosis. Nixon, Hagen, and Peters (2010) found that their participants described an active process of coping with the challenges placed upon them by psychosis through engagement in meaning‐making within a spiritual framework. Yang, Narayanasamy, and Chang (2012) explored the experience of long‐term hospitalization with participants in Taiwan and found that they used a spiritual framework as the primary description of their sense of loss through the illness. Given these studies, spiritual and religious beliefs hold clear potential for study as an exemplar of the narrative insight construct.

Research objectives

Given the developing understanding of narrative insight in psychosis, and the theoretical coherence of integrating spiritual or religious beliefs in this framework, the current exploratory empirical study sought to address the open research question: ‘Can inductive analysis of personal accounts of psychosis (in reference to individuals’ spiritual and religious beliefs) support the proposal that narrative insight offers a useful alternative to clinical insight as a theoretical construct in the research and care for people who have these experiences?’ Qualitative methods are most appropriate for this aim, as it is exploratory rather than hypothesis testing.

The current study is taken from a doctoral study exploring the broader interface between spiritual or religious factors and psychosis (Marriott, 2007); for the purposes of this study, we focus specifically on data addressing our objectives regarding narrative insight. The decision to use spiritual and religious beliefs as the exemplar explanatory framework within narrative insight is acknowledged here as an opportunistic sampling decision based on the data available from the original doctoral studies, and the full analysis is available from the reference above.

Method

Design

Inductive qualitative methods were used to develop semi‐structured interviews exploring how eight participants utilized spiritual or religious explanatory frameworks for their experiences of psychosis, before using interpretative phenomenological analysis (IPA; Smith, Flowers, & Larkin, 2009) to explore the data.

Participants and recruitment

Being focussed upon experiential phenomenology, recruitment was based upon self‐reported experiences rather than externally ascribed diagnoses (although most participants also reported diagnostic labels ascribed to them, see Table 1). Participants responded to the question ‘Have you ever experienced hearing voices, having strong beliefs, or any similar experience that has affected your mental wellbeing and general functioning in life?’ Our inclusion criteria for psychosis were psychological experiences that were perceived as real, but were substantially discontinuous from the individual's normal or preferred spectrum of mental experience, and which negatively affected the individual's mental well‐being and level of personal functioning. Whilst this definition is more inclusive than analogous studies requiring ‘objective’ identification of people who meet diagnostic criteria according to trained clinicians – and would potentially not differentiate people who would be diagnosed with ‘schizophrenia’ from people whose experiences could result from ‘other physical explanations’ – it is theoretically coherent within our critical realist approach; we posit that the meaning‐making of the experiences is likely to be sufficiently similar whatever context they arise.

Table 1.

Participant characteristics

Pseudonym	Gender	Age	Ethnicity	Religious identitya	Reported diagnosesa	Reported service accessb
Abdul	M	24	Bangladeshi	Islam	Paranoid Schizophrenia	Early Intervention Team
Barbara	F	54	White British	Roman Catholic	Schizophrenia	Mental Health Team
Chloe	F	41	White British	Spiritualism	Manic Depression	Inpatient and Outpatient Mental Health Team
David	M	59	White British	–	Manic Depression	Inpatient and Outpatient Mental Health Team
Ethel	F	75	White British	I see my truth in them all	None identified	None
Felicity	F	45	White British	Church of England	Depression, Eating Disorders, Tourettes, OCD, Heroin Addiction	Inpatient and Outpatient Mental Health Team
Gordon	M	51	White British	Christian	Toxic Confusional State	Psycho‐legal Assessment
Harry	M	35	White British	Christian (EVJ)	Paranoid Schizophrenia, Borderline Personality Disorder	Secure Inpatient and Outpatient Mental Health Team

^aThe terms reported here are in the participants’ own words, although with regard to the diagnoses, there are clearly based on labels given to them by health professionals. The authors do not wish to imply an acceptance of categorical frameworks with regard to either religious identity or mental health diagnoses.

^bThese terms are paraphrased from the accounts provided by participants.

Consistent with other studies described above (e.g., Drinnan & Lavender, 2006; Huguelet et al., 2010), an open definition of spiritual/religious beliefs was used, and participants responded to the question ‘Have religious or spiritual beliefs ever been important to you?’

Following ethical and governance approval, participants were recruited from a large UK city. An advert was placed in a variety of appropriate locations (e.g., voluntary sector services, community mental health teams, GP surgeries). Upon responding, participants were sent a full information sheet and an opt‐in/screening form which requested demographic details, and asked for a brief description of their experiences, alongside four Likert‐scaled statements about the effect this experience had on their psychological well‐being and the disruption it caused them in social functioning, occupational functioning, and self‐care. Two members of the research team with appropriate clinical training and experience then assessed whether the experience described met our psychosis inclusion criteria and would be recognized as a potential psychosis in mainstream clinical settings. Due to the interviewer's language limitations, no funds to consider an alternative, and the analytic methodology employed, participants had to be fluent in English.

Sixteen individuals responded initially, of whom nine returned their opt‐in, all fulfilling the research criteria. Eight interviews were conducted (see Table 1 – all pseudonyms), with one withdrawal citing ‘life circumstances’. Regarding religious demographics, Gordon recorded Church of England on the form but described ‘having not yet made up [his] mind’ during his interview. David, who recorded ‘–’ on the form, reported a strong belief in God but no identification with any organized religion.

Measures

A semi‐structured schedule was developed through discussion with a collaborator (GC, 3rd Author) who had experiences of psychosis, and also with academic colleagues. Using a selection of open and neutral questions with prompts, the schedule focussed on the following: (1) people's experiences of psychosis and the ways in which these have affected their life, self‐perception, and coping style; (2) their spiritual/religious beliefs, the history of how these beliefs were derived, involvement in spiritual/religious communities, the way spiritual/religious beliefs affect their self‐perception, and any experiences they have had that they would describe as spiritual; and (3) any ways in which the two elements of spiritual/religious beliefs and experiences of psychosis interact (copy available in Marriott, 2007).

Interviews were all conducted by the first author (who had completed two years of training in interview skills as a trainee clinical psychologist in the United Kingdom) and ranged in length from 1 to 2 hr. The schedule was a prompt sheet, and interviews covered all areas but always in a different order. Clarifications, reflections, and further prompts were used as deemed appropriate within the interview. Interviews were recorded and transcribed verbatim with the removal of identifying information.

Analysis

The analysis of the transcribed data followed the principles of interpretative phenomenological analysis. This analytic framework allowed for the adoption of a critical realist perspective, in which the analyst accepts there to be an objective reality to all phenomena – specifically psychological experiences – but also accepts that there is no direct route of access to these phenomena. Having accepted that direct (objective) access to these phenomena is not possible, the analyst seeks instead to derive a partial understanding of psychological experiences through a framework of exploratory meaning‐making by and with the participant(s). Important within this framework is the concept of a double hermeneutic – the participant is constructing an understanding of their own experiences which leads to their account, and the analyst then constructs their own understanding of this account. This acknowledges that both layers of construction are unavoidably affected by the broader psychological process (intra‐ and interpersonal) of each individual, but ensures that the analyst seeks to contextualize as much of their interpretation within their own personal processes and also seeks to validate their ideas through some form of triangulation to ensure that the proposed analysis can be recognized more broadly.

IPA was chosen as the analytic method because our intention was always one in which individual meaning‐making could be brought to the broader academic audience with the application of a systematic methodology situated within a coherent theoretical model; IPA meets these criteria and has been widely used in previous research with this population (see McCarthy‐Jones, Marriott, et al., 2013, for a broad range of inductive studies in the field, including numerous IPA studies).

Holding the principles of IPA and the critical realist perspective in mind throughout, transcripts were analysed primarily by the first author (the project was undertaken as part of their doctoral training, and so the dominance of this author in the analytic process was a pragmatic inevitability). Each transcript was read alongside the original recording and then re‐read. The analyst then noted on one side of the text general points of interest, allowing the development of data codes which held a close representation to the original material. From these, the analyst then undertook a process of abstraction of meaning from the codes and listed on the other side of the transcript initial descriptive themes for each section of data. At this point, these themes were transferred from the originalmaterial, and listed into a word‐processing document along with page or line references.

The lists of themes for each participant were then clustered into groups that appeared relevant to each other as representing a similarity in the way that the participant was making sense of their experiences. The eight sets of clusters were then amalgamated into one master table, where differences in clustering patterns were made apparent. Further global clustering then took place to group the clusters identified in individual transcripts into associated groups, with reliability checked by ensuring that the initial themes of each participant still reflected each new framework and with the opportunity for new clusters to be created.

Quality control

To support the trustworthiness of this analysis (Yardley, 2000), a reflexive log was kept, recording field notes from interviews and the development of any ideas or process of thought throughout the project process. Supervision was used throughout the project to ensure that the analyst monitored the extent to which they brought their own issues and concepts to the research process. This included review of transcripts and analysis at each stage of the process by peers, supervisors, and collaborators with experience of psychosis – in all, subsets of data from 5/8 participants were analysed independently and integrated into the analysis by the first author. This review also included critique of interviewing style (e.g., identification of unintended leading questions, the responses to which should be given less weight within analysis) and consideration of bias within analysis (e.g., some early theme developments were noted as insufficiently rooted in the primary data, apparently shaped more by the analyst's desire to demonstrate their ‘worthiness’ to do such work – corrections were made on the basis of this feedback, better grounding language in the words used by participants).

Results

Participant accounts were organized into three primary themes, each with a selection of secondary themes and further areas of categorization and also with recognition of substantial overlap between some areas and themes; Figure 1 shows how these themes reflect the authors’ understanding of the participants’ accounts.

Figure 1.

Diagrammatic representation of general theme structure.

For the purposes of this study, the primary theme of Explanations, Understandings, and Beliefs is expanded upon further, specifically the two secondary themes relating to Unusual Experiences and Testing and Proof. The analysis and discussion relating to the other themes is available via access to the doctoral thesis (Marriott, 2007).

Within the following analysis, illustrative quotes are provided. Word order has been maintained for authenticity, but both pauses and researcher utterances (e.g., ‘Mm’, ‘Yeah’) are removed for ease of reading.

Explanations, understanding, and beliefs

This theme derived from how the participants’ accounts related to the ways in which they communicated the sense that they make of aspects of the world. Although participants were asked to describe their religious beliefs, the details within the theme represent an analysis that has responded to issues independently identified by participants – participants’ explanations of their unusual experiences were not directly prompted for within the interview schedule (although within individual interviews, further clarification of an explanation was sought when it had been independently volunteered by the participant).

For unusual experiences

Every participant described their unusual experience alongside an explanation for what it was. Everyone had some idea that there could have been a spiritual element to the experience, including direct conversations with God, experience of God in the cosmos, and spirit embodiments. Abdul described how some of his problems were related to djin, a type of spiritual being that was found within his Muslim beliefs.

Abdul, p15: I used to have a djin it used to cause me problems sometimes this, this led, er whenever it come to Christmas like it would stress me out like

However, a number of participants also suggested a biological explanation for the unusual experience, including adoption of medical labels for their experiences – looking again at Abdul, he used medical terminology as well as spiritual explanations, which could be speculated to be the result his close involvement in psychiatric services for a number of years.

Abdul, p2: three years ago I experienced some problems problems but I didn't realise I was getting ill and more I was heading towards that way I just thought I could I started experiencing symptoms like I was getting followed

The potential disagreement between these two types of explanation was recognized by almost all participants in whom the two options were described, which appeared to be grouped within an important category of alternatives. Participants discussed the fact that there were alternative explanations, describing the reasons why they might accept one over the other, discussing the uncertainty about either option, or in Barbara's case, fully accepting both possible explanations as able to mutually coexist:

Barbara, p25: I've been told by my GP actually that, er with me having chickenpox as a child as well the virus can lay dormant which to me, erm, as well as religious beliefs it also, I believed very strongly they were proper visions but also having that virus lay dormant in me that, er, there is a medical side to it as well which to me is now, at long last been recognised … people are actually listening to me understanding me and believing me because it's the medical side and they're switching off to the spiritual side which to me is sad because there is the spiritual part there there is a definite spiritual connection there.

Barbara believed that both a virus she had when younger and a recent stroke meant that her brain was more susceptible to having unusual experiences; however, she also believed that these were genuine spiritual experiences, representing protection from her grandmother and the Virgin Mary. She appears to be disappointed that although she was able to see how the two factors could both be true, the majority of people she spoke to about it focussed solely on the biological side rather than the spiritual; this may represent her experience of either professionals’ or societies’ preference for reductionist biomedical explanations for psychosis. Similarly, Chloe explained her difficulties in terms which adopted both spiritual and biomedical aspects.

Chloe, p.16: A large part of me, because I have, still have got my own questions about chemicals in your body getting out of hand and what I believe in, erm, but yeah, a large of me does think that it was an evil spirit that, that's why I, that's why my body broke down

Chloe demonstrates an understanding that there are explanations for her experiences which encompass chemical concepts; however, her framework appears to suggest that these biomedical elements are the process by which difficulties occur, but the actual causation of this in the first place can be driven by evil spirits.

Each participant was also able to explain the origins of their explanations, showing what had led them to hold these explanations; sometimes, this was through read material, contact with others, or a result of the framework for life and the world that they already held. At other times, the reality of the experience seemed enough to explain the origin spiritually. Felicity felt that God's voice, which she heard most times that she prayed, was completely different from her internal thoughts; this was enough justification for her to be sure that it was genuinely God's voice.

Felicity, p27: I do have a little God voice that I know is God [R: Can you tell me a bit more about, erm, the God voice then, what, what's God's voice like?] It's gentle, it's loving, it's, erm it's deeper than my own thoughts erm it's authoritative, erm it's truthful, erm it's warm and welcoming erm and away from my own thought processes.

The adjectives chosen by Felicity give a sense of the comfort she experiences when hearing this voice; perhaps, the comfort experienced allows the alignment with an explanation that includes a benevolent God figure. However, in proffering explanations for their unusual experiences, the explanations were often the same regardless of the negative or positive appraisals, and participants rarely made a distinction between them. One notable exception was David, whose unusual experiences were primarily related to conversations he had with God; these had started when he was in a psychiatric institution following a suicide attempt, but had continued virtually on demand, helping him to develop a world view with which he felt more comfortable. David differentiated between a positive and negative feeling to decide whether a spiritual explanation was correct for that particular experience; however, he remained convinced throughout that the positive experiences were divine in origin.

David, p.37: when you begin to distrust God erm, you know that what you're talking to isn't God (laughs) and there are times when I could only talk to this thing I distrusted which was me I knew it was but luckily, since then I've been talking to, I've talked to God quite frequently.

Testing and proof

This secondary theme within Explanations, Understandings, and Beliefs represents part of the rationale‐giving process that was engaged with for each of the other two secondary themes. Participants would describe how their broader religious frameworks were something which they had engaged with testing, with alternatives being explored.

Gordon, p.25: I think, erm, and, erm, you know, my understanding of, of, religion, erm, and Christianity maybe, I don't know, I mean I don't know about other forms but that is quite a big element, of you know, of the teachings about, you know, doing wrong and evil, and being cast down to Hell and all that sort of stuff and that, that erm, there's part of me that thinks ‘well God, if that's right, you know,that's scary’ and then I think ‘blimey, hope it's not right, that it's all manufactured by man and it's got nothing to do with that’.

For Gordon, his religious beliefs pose challenges for him in what might lie before him with regard to punishment. In those moments, he recognizes the alternative perspectives of religion being a manufactured construct, but ultimately maintains his position that the religious framework is an appropriate understanding of the world.

Frequently, participants would (unprompted) use examples of things that had happened which illustrated why their account was true; sometimes, these examples would be circumstances which had arisen as a direct result of the participant challenging or testing the beliefs themselves – this could mean being told something by voices (identified as God or spirits) which they could not possibly have known and was later proven correct.

Ethel, p.41: I was meditating one day and, er, I was told my brother had got a beautiful baby boy and, er, I thought why did they give my brother's name and not his wife's and, er, I'm waiting for her to tell me she's pregnant so that I could tell her she's going to have a baby boy, nothing happened, three years later she got on the phone, did I know that my brother had got this child, er, baby, er, a boy with another woman. Are you with me? The, the things has, this is why I, I, these voices I've heard have always told me the truth.

Noting her surprise at the time of the voice telling her about her brother's child, Ethel is providing details in her narrative which demonstrate that she had a genuine experience of premonition, allowing her to accept the potential truth in spiritual explanations for her voice hearing. For other participants, the proof of spiritual explanations lay in prayers being answered.

Harry, p.23: [R: What (.) can you tell me a bit about the (.) protection from spiritual harm?] Right, this t‐, right okay, erm, just, just simple, simple things like, erm you'll get, you'll get on a bus and someone's arguing with someone and, er, and they might turn on you, you know, and you'll just say, ‘Lord, help me’, you know, ‘make this person go away from me or, or, erm, make him get off the bus’, you know, and you'll say a prayer in your head and somehow, someone will speak up or you might say something or that person might go away from the situation and then you'll say, ‘Lord, thank you thank you for that protection, thank you for that guidance’

In these circumstances, Harry holds a clear belief that God is acting directly within his life, influencing situations. He frames this as a protective action, but it is interesting to note that Harry also holds responsibility for the outcome – it is through Harry choosing to pray that God acts. Harry described in his interview a history of assaulting others, and it can be seen that his belief in external (spiritual) protection here still requires him to take a proactive decision to avoid conflict, with apparently prosocial results.

With testing and proof as a theme, we see how participants were often describing a process of engagement with the debate around their beliefs; they developed their beliefs in a way which in part relied upon a process of evidence‐gathering and testing, even if the evidence itself may not be universally accepted. Reflexive notes from the interviews and analysis suggest that, when approached with an open mind, the explanations provided by participants often felt coherent and convincing to the researcher, despite the mutual exclusivity of some of the explanations offered between different participants.

Discussion

Taken from a broader study exploring individuals’ reflections on the interface between their psychosis and their spiritual or religious beliefs, this inductive analysis supports the idea that the construct of narrative insight might be a more appropriate approach to support and research for people who have experienced psychosis, compared with the narrower construct of clinical insight currently dominant in the discourse of research and service provision.

Flexibility and reasoning in explanations

In line with the narrative insight theory, our participants engaged in a complex and individualized process of seeking to understand and explain their experiences. Given the sampling, it was unsurprising that participants used spiritual/religious references in their explanations, but the nature of these explanations was notable. Whilst some literature suggests that religion can be seen as providing simple heuristics that help the individual make rapid, rather than complex, decisions (e.g., Carone & Barone, 2001), our participants considered both their dominant explanatory frameworks and some alternatives. They evaluated reasons for their preferences and often engaged in a process of testing their explanations. The majority of the participants were aware of biological explanations, and some accepted the validity of these explanations alongside a spiritual explanation. This demonstrates a subtle process of insight – in line with the narrative insight grouping process of Roe et al. (2008) – than is typically understood in the clinical insight model.

Spiritual vs. psychotic experiences and explanations

There is a discussion about the extent to which experiences labelled as psychotic and those labelled as spiritual can be distinguished (e.g., Jackson, 2001; Jackson & Fulford, 1997; McCarthy‐Jones, Waegeli, & Watkins, 2013). The suggestion is that the underlying psychological correlates of either experience would be similar, with the only difference being in the manner in which they affect the individual's life. For most of our individuals, there was only one time that a distinction in the type of attribution (Spiritual vs. Illness) made between the positive and negative experiences; rather, negative experiences were often be related to a different aspect of the spiritual framework (e.g., good spirits vs. bad spirits), despite the fact that all of the negatively appraised experiences would have suited the application of a label of psychosis.

Limitations

There were limits in the sample, such as only one non‐white participant; whilst this does reflect the demographics of the study's catchment area, explanatory frameworks will be culturally bound, so some of our findings might be skewed in this respect. The majority of participants’ experienced voice hearing, experienced as an independent entity – explaining the origin of something experienced as separate may be more conducive to coherent frameworks than other experiences deemed as psychosis, such as ‘delusions’.

Although the majority of the participants had psychosis within their very recent history, as a self‐selected sample able to make the decision to participate in this research, all might be considered as relatively well‐functioning in the spectrum of people who have experienced psychosis; it can further be argued that by adopting a phenomenological approach to psychosis, we are privileging the accounts of people able to offer coherent reflections on their experiences, excluding a full communication of the suffering that can characterize the lived world of these experiences at their height.

Implications for clinical practice

These findings support the suggestion that the construct of narrative insight is appropriate and that clinical practice could prioritize individualized explorations of people's explanatory frameworks for their psychosis (Macnaughton et al., 2015). Narrative enhancement and cognitive therapy (NECT; Yanos, Roe, & Lysaker, 2011) is one such example, being a manualized group therapy which is based on the concept that narrative coherence is valuable in recovery and explicitly addresses self‐stigma, and with positive findings in a first evaluation (Roe et al., 2014). A wider impact would be for assessments of insight across clinical care moving from clinical insight to narrative insight as a matter of routine for all individuals with experiences of psychosis; this would fit with the re‐visioned model of recovery‐oriented care promoted by Leonhardt et al. (2017). This might also include a broader involvement of the social context in which all individuals are constantly defining their meaning‐making processes, recognizing the importance of this as clearly as the professional meaning‐making is currently enacted.

With regard specifically to our exploration of spiritual and religious beliefs in this sample, our work suggests that services should explicitly ask how individuals’ personal understanding of spirituality and/or religion may be part of their narrative development (McCarthy‐Jones, Marriott, et al., 2013; Heffernan Neil, Thomas, & Weatherhead, 2016). Services need to develop an acceptance that individuals who have psychotic experiences may have a good rationale behind their own frameworks, which can validly include religious aspects. However, whilst there is literature exploring how to approach spirituality in such settings (e.g., Weisman de Mamani, Tuchman, & Duarte, 2010), and evidence that this approach is acceptable and useful to patients (Huguelet et al., 2011), evidence also suggests that clinicians can be reluctant to engage in conversations about spiritual or religious beliefs (Crossley & Salter, 2005; Huguelet et al., 2011). Such barriers need to be explicitly addressed at a service level.

Further research

Participants often described how their previous conversations with others contributed to the development of their narrative. The focus of the overarching doctoral project – and therefore interview schedule, interview conduct, and the material derived – was on the participants’ understanding of their experiences; hence, phenomenology was appropriate, and the data would not be appropriate for narrative analysis. Further projects could approach narrative insight by exploring the process of narrative development using a narrative analysis approach, which would develop an understanding of how shared conceptualization between service users and clinicians might affect the process of narrative insight.

As noted above, this study focusses on spiritual and religious beliefs as an exemplar for the construct of narrative insight, opportunistically using data available from a project with different aims at inception. Further research might therefore explore whether other non‐diagnostic frameworks (such as the concept of post‐traumatic growth in psychosis) also demonstrate a goodness of fit within the theoretical construct of narrative insight.