Angela Woods*,1, Nev Jones2, Marco Bernini3, Felicity Callard4, Ben Alderson-Day5, Johanna C. Badcock6, Vaughan Bell7, Chris C. H. Cook8, Thomas Csordas9, Clara Humpston7, Joel Krueger10, Frank Larøi11, Simon McCarthy-Jones12, Peter Moseley5, Hilary Powell13, Andrea Raballo14, David Smailes5, and Charles Fernyhough5

Woods, A. and Jones, N. and Bernini, M. and Callard, Felicity and AldersonDay, B. and Badcock, J. and Bell, V. and Cook, C. and Csordas, T. and Humpston, C. and Krueger, J. and Larøi, F. and McCarthy-Jones, S. and Moseley, P. and Powell, H. and Raballo, A. and Smailes, D. and Fernyhough, C. (2014) Interdisciplinary approaches to the phenomenology of auditory verbal hallucinations. Schizophrenia Bulletin 40 (S), S246-S254. ISSN 05867614.


Introduction
The term "voice-hearing," or auditory verbal hallucination (AVH), typically refers to hearing a voice or other sound in the absence of an external stimulus. The apparent simplicity of this mainstream definition belies the diversity of the experiences it names. Writing at the turn of the last century in what would become one of psychiatry's most important textbooks, Emil Kraepelin 1 described in detail the kinds of auditory hallucinations reported by patients suffering from what he called "dementia praecox." Eschewing the relatively dry, flat language of the clinic, the patients spoke of: "resonant voices," "organ voices," "voices of conscience," "voices which do not speak with words," "false voices," "abortive voices," an "inner feeling in the soul," an "inward voice in the thoughts," something "between hearing and foreboding," "the brain talk[ing]," "voices in the whole body," "murmurings and natural spirit-voices," "underground voices from the air," "telephone gossip," "good voices," and "whispering voices from the whole of mankind." How are we to make sense of these descriptions, which are echoed in the narratives of voice hearers today? 2 Can conventional perception-centered definitions of AVHs, and even the mainstream metaphor of "hearing voices," do justice to their complexity? How should we understand hallucinatory experiences within the arc of an individual's life, and across different clinical and nonclinical populations? S247 Interdisciplinary Approaches to the Phenomenology of AVH With the etiology and underlying mechanisms of AVHs a matter of ongoing research and debate, experience itself remains an important area of investigation. 3,4 Given that AVHs are privy and particular to the individual, no research can avoid making decisions about the kinds of data and analytical tools that are to be considered sufficiently robust when it comes to investigating these most complex of scientific objects. In cognitive science, psychology, and psychiatry, empirical studies of AVH phenomenology typically utilize standardized scales in order to assess particular aspects of experience within a given population. 3,[5][6][7][8] However, while the validity, reproducibility, reliability, and foci of particular scales and measures are central to their scientific and clinical evaluation, 9 hallucinations research can arguably benefit from more sustained engagement with the philosophical, epistemological, and theoretical issues underpinning such investigations.
One of the greatest contributions that might be made by the humanities and social sciences to the study of AVHs is in offering methods through which to conceptualize, delimit, identify, elicit, and analyze the so-called "subjective" data that form such a central component of this research. Whether or not it is addressed explicitly, all research inevitably has to negotiate what voice-hearing experiences denote, how they can be extracted from or identified in the fabric of an individual's life, how they change over time, and if and how an AVH or voice is differentiated from other "normal" and "anomalous" forms of inner experience. Other important questions include: to what extent does representing and reporting AVHsparticularly in light of different cultural frameworks and available terms and constructs-impact upon the nature of that experience? And how might the relationship between the person experiencing the AVH and the person researching it be best conceptualized (not least in situations in which that person is the one and the same)? There are rich bodies of literature located in and across phenomenological psychiatry, anthropology, sociology, theology and religious studies, literary studies, history, medical humanities, and "mad studies"/service user led research that have much to offer vis-à-vis these questions. This literature can open up a variety of frameworks, methods, and analytical tools through which psychiatric researchers might reconsider what it is that they study when they study AVHs, and how they might go about acquiring and analyzing that data.
Recognizing that psychiatry has a long history of engagement with other disciplinary perspectives, the first sections of this article outline the purpose and value of an interdisciplinary approach to the phenomenology of AVH. Following Aboelela et al, 10 we understand interdisciplinary research to involve collaboration between researchers from different disciplines "that links or integrates theoretical frameworks from those disciplines, uses study design and methodology that is not limited to any one field, and requires the use of perspectives and skills of the involved disciplines throughout multiple phases of the research process." Durham University's "Hearing the Voice" project is an example of an interdisciplinary approach to the study of AVH. With respect to phenomenology, such an approach would utilize rigorous and context-appropriate methodologies to analyze a wide range of first-person accounts of AVH at 3 contextual levels: (1) cultural, social, and historical; (2) experiential (in relationship to changes in the structure of experience); and (3) biographical (in relationship to the arc of an individual's life). In the final sections we suggest 3 potential benefits for voice hearers, clinicians, and researchers: (1) informing the development and refinement of subtypes of hallucinations within and across diagnostic categories; (2) "front-loading" research in cognitive neuroscience; and (3) suggesting new possibilities for therapeutic intervention. Our focus is not on the wider contribution of the humanities and social sciences to the study of mental (ill) health and clinical practice; more narrowly, we seek to show how work in the humanities and social sciences can help us better understand the experience of hearing voices.

AVHs in Cultural, Social, and Historical Context
Has hearing voices always been part of human experience? The fact that "hallucination" itself is a recent 11 and contested 12 term raises the question of whether experiences so named have been the same across time, place, and culture. Anthropological studies of psychosis and schizophrenia have challenged the view that culture is of minor relevance to understanding AVH, [13][14][15][16] showing that "local theory of mind-the features of perception, intention, and inference that the community treats as important-and local practices of mental cultivation will affect both the kinds of unusual sensory experiences that individuals report and the frequency of those experiences." 17 If the complexity of the relationship between culture and hallucinatory experience is to be adequately theorized and empirically investigated, researchers must make use of tools and disciplinary approaches which do not simply reduce "culture" to a one-dimensional variable (for which country of residence frequently functions as a proxy). Instead, as cultural psychiatrists and others 18,19 have argued, ethnographic and qualitative approaches have a vital role to play in investigating the ways in which communities interpret, legitimize, support, and even produce different voice-hearing experiences.
Recognizing that mainstream biological psychiatry constitutes one such (albeit powerful) community can bring medical understandings of hallucination into dialog with other explanatory frameworks. Religious and spiritual accounts of voice hearing-which take seriously the meaning of voices and their origin in a realm beyond empirical studycall into question distinctions between "pathological" and S248 A. Woods et al other kinds of AVH. For example, recent anthropological studies of Pentecostal and charismatic Christian groups have found that members of these communities not infrequently report hearing the voice of God out loud and in external space, in ways they find spiritually encouraging and which are not associated with evidence of mental illness. 20,21 Data collected through narrative and semistructured interviews, quantitative hallucination and absorption scales, ethnographic observation, and the study of theological and religious traditions, are here used to build an in-depth picture of voice-hearing experiences regarded as acceptable and even desirable within these Christian communities. As well as their intrinsic value (in contributing to our collective knowledge about the complexities of human experience), studies such as these also have instrumental value in clinical and research contexts (eg, in informing future taxonomies of AVH, as discussed below).
Of course, complex, multidimensional accounts of voice hearing are not confined to the present, so an interdisciplinary approach to the phenomenology of AVH must recognize its own historical specificity by taking into account the dis/continuities between contemporary accounts of voice hearing and those reported in earlier historical periods. Hearing the voice of God in the Judeo-Christian tradition again provides a useful case study. For the theologian studying scripture, as for the historian working with medieval Miracula, investigating the voices and visions of the past requires an understanding of the languages, practices, and beliefs which are reflected in and constituted by particular social, cultural, and political environments, as well as embedded in particular linguistic conventions. The voices heard by religious figures such as Ezekiel and Joan of Arc have been attributed to mental disorder by commentators, often in ways which reflect ignorance of the historical and textual worlds from which they speak to us. 22 Given that scholarly engagement with the past is mediated primarily through textual forms, the act and status of writing becomes inextricable from the question of experience. Medieval mystic Margery Kempe's experience of hearing voices was recorded by an amanuensis, which means that her "first-person account" might say as much about the process of transcription as about "experience" itself. Reading historical texts involves not only translation but interpretation: asking questions about why a text was written, how it was produced, the context of dissemination, its "social logic," and even the reasons for its survival. 23 The expert analysis of these texts provides insight into a wider realm of hallucinatory experiences as well as the sophisticated schemas for differentiating voice hearing that existed prior to its medicalization. From 16th century Spanish Carmelites such as St John of the Cross 24 to contemporary charismatic Chicagoans, 21 theologians, anthropologists, and historians have shown that the need to exercise discrimination in relation to voices-both literal and metaphoricalcontinues to be important to Christians, as to others, in the present day. Multiplying the frameworks through which voice-hearing experiences can be identified and interpreted, interdisciplinary approaches can offer new perspectives on the shared and nonshared features of these heterogeneous experiences, and so avoid circular arguments about what are to be counted as "pathological" and "non-pathological" voices.

AVHs and the Structure of Experience
Phenomenology (literally, "the science of appearances") is a philosophical movement which seeks to reflect upon the basic structures of experience-ie, experience from the first-person perspective of the subject-and to understand how these basic structures give our experience of the world and ourselves its formal coherence. 25 For phenomenologists, this first-person emphasis entails a consideration of topics such as intentionality, selfawareness, temporality, embodiment, spatiality, agency, and intersubjectivity. Phenomenology has enjoyed a fruitful interaction with psychiatry and psychopathology for over a century, with phenomenological psychiatrists and psychologists (among them Jaspers, Minkowski, Blankenburg, Fuchs, Sass, Parnas, and Stanghellini) as well as nonclinical philosophers (including Gallagher, Zahavi, and Ratcliffe) making pioneering contributions to the analysis and theorization of psychosis. [26][27][28][29][30][31][32] Following Husserl, there are generally thought to be 3 phases at the heart of phenomenological methodology: an initial bracketing of taken-for-granted assumptions or judgments about the cause, normality, or reality of what is experienced (including diagnostic or etiological considerations), which enables the investigator to focus on the character of the experience itself; discerning the prototypical features of the experience, its essence, or eidos, and generating descriptions that account for these features; finally, assessing the adequacy and fit of these descriptions and, when necessary, subjecting them to further elaboration and refinement. In a clinical context, a central part of this process involves focusing on the character and meaning of the person's experience from their perspective, and continually submitting resultant descriptions to intersubjective scrutiny. Phenomenologists thus seek to analyze the specific structure of subtle qualitative changes in perceptual, proprioceptive, and intersubjective experience that might underlie psychopathological phenomena. 32 Despite empirical findings corroborating a continuity in the severity of AVHs among clinical and nonclinical subjects 33 studies have not yet addressed whether there is a corresponding continuum of phenomenology. 34,35 Philosophical phenomenology (PP) is a methodology well suited to the investigation of subtle but potentially highly significant differences between the voice-hearing experiences of clinical and nonclinical individuals, 7 and between different diagnostic groups including severe anxiety, military S249 Interdisciplinary Approaches to the Phenomenology of AVH posttraumatic stress disorder, depression, and schizophrenia spectrum disorders. 36 The auditory quality of voices is a case in point. Although descriptive psychopathologists have long recognized the nonauditory thought-like quality of many otherwise "psychotic" voices, 37 contemporary measures and clinical interviews generally assume that self-reported voices are both literally auditory and identifiable in terms of auditory characteristics. Bracketing this assumption and analyzing the variety of ways in which voices are "heard" (as reported so clearly by Kraepelin's patients quoted above) could illuminate differences which in turn have the potential to revise existing empirical measures. PP is also especially well placed to address hallucinatory complexities which are less frequently analyzed, such as the boundaries between externalized or externally located "thoughts" and literally "auditory" hallucinations, and the relationship between auditory and/or verbal aspects of hallucinations and their tactile, visual, affective, nonverbal, and/or somatic aspects. 38 Should voice hearing be approached as a unitary, static, crystallized phenomenon, or as dynamic and subject to change over time? 39 Do AVHs sometimes or always occur against the backdrop of a more encompassing transformation in one's sense of reality? Through the rigorous analysis of individuals' selfreports PP can identify fine-tuned structural differences and variances in how AVHs are experienced in relation to a person's being in the world. [40][41][42]

AVHs in Biographical Context
The narrative and biographical context(s) in which AVHs develop and are sustained have served as a key point of departure for many clinical and psychotherapeutic approaches to voices and psychosis, including psychoanalytic/psychodynamic therapies, existential and narrative therapies, trauma-informed care, and techniques developed within the international Hearing Voices Movement. [43][44][45][46][47][48] Across these approaches, life events and internal struggles are viewed not only as significant causal contributors to the onset of AVHs, but also as major influences on their content, phenomenological form and structure, and degree of associated distress or disability. 44,49,50 At the same time, and often in close proximity to these therapeutic approaches to voices, social scientists and humanities scholars have developed diverse models through which to understand and analyze the overarching concepts of "narrative" and "biography" and to consider what exactly constitutes the "context" of a life. While longitudinal psychiatric research typically makes some attempt to contextualize changes in symptomatology or functional disability, 51,52 narrative biographical interviewing techniques might instead attempt to explore the subjective impact of factors identified as important by the participant in light of their own personal history. 53 Narratological theories, finally, can highlight the importance of everyday storytelling in the processing and communication of life events and experiences, and the limitations imposed when opportunities to tell certain kinds of stories are constricted by cultural norms or sociopolitical factors. 54 In the context of psychosis, a breakdown in the ability to tell integrated stories about the self has been implicated in the genesis of psychopathology 55 and cultivated storytelling therefore underpins innovative work in narrative therapy. 56, 57 We emphasize that such methods extend beyond more familiar qualitative mental health research techniques such as thematic analysis 58 or brief Interpretative Phenomenological Analysis. 59 More critical and in-depth analytic methodologies seek to move past cross-sectional themes and categories and devise new ways to understand how conceptual frameworks, available cultural scripts, and biographical and embodied experiences might help structure and constrain both the subjective experience and communicated phenomenological form of AVHs. 60 Much available qualitative mental health research remains committed to-and arguably constrained bythe categories and empirical methods currently dominant in mainstream schizophrenia research; we want to emphasize here, instead, the benefits of analytic frameworks that attempt to understand how language, narrative and embodied experience can both structure experience over time and provide potential tools for healing.
Complementing approaches from phenomenological psychiatry, sophisticated narrative and biographical approaches focus on the ways in which individuals negotiate their experiences of perceptual, affective, cognitive, and/or interpersonal change within specific social and cultural contexts. In addition, they help draw attention to the divergent and heterogeneous longitudinal trajectories of individuals with AVHs, including full recovery, periods of remission, deterioration, or more complex and varied changes in the nature, content, and valence of symptoms over time. Notably, much creative research and writing on narrative has been conducted by service users/those with lived experience, 61 who have offered different accounts of what recovery might mean 62,63 and whose work thereby demands that we attend more carefully to some of the standard categories (eg, "chronicity") that are commonly used, often as shorthands, in psychiatric research on AVHs. Working collaboratively to refine these categories is an important task for future researchers. Narrative and biographical approaches may also help to link work in socioenvironmental epidemiology, clinical phenomenology and outcomes, for instance by suggesting or ruling out potential confounds or third variables, and connecting neurodevelopmental and epigenetic changes with subjective experience.

Informing Development and Refinement of Subtypes
One potential benefit of the approach advocated here is the development of a more accurate and nuanced S250 A. Woods et al phenomenological portrait of AVHs and the foregrounding of the radical phenomenological heterogeneity of the experience. Delineating new facets of AVHs challenges contemporary neurocognitive models of AVHs to account for the phenomena as they are, and not merely idealized, simplified, or partial versions of the experience. 64 As we have argued, historical texts can constitute a valuable alternative "lens" through which to identify overlooked phenomenological aspects of voice hearing. 65 For example, in his detailed taxonomy of hallucinations, 24 St John of the Cross stressed the density of meaning of some voice-hearing experiences (instances in which voices may be experienced as communicating much more than simply what the words say). 65 This resonates with and could inform the development of contemporary models of inner speech 66 (especially those proposing a specific condensed variety which loses most of the acoustic and structural qualities of external speech and approaches the state of "thinking in pure meanings" described by Vygotsky 67 ). Contemporary concretization of the term "hearing voices" has also led to the neglect of experience of "soundless voices" reported by St John of the Cross and across the centuries, including in the work of pioneering psychiatrists Bleuler, 68 Kraepelin, 1 and Jaspers. 26 Such overlooked experiences are likely to offer important clues into the types of mentation that may form the raw material of AVHs, and remind us that the data of experience must be prioritized over existing theoretical accounts. 69 Finally, the rich and detailed phenomenological descriptions we have argued for here should also aid the extension of the contemporary approach of using cluster analytic techniques on the phenomenological properties of AVHs to identify subtypes. 8,70 This is particularly important, as identification of AVH subtypes offers the potential for tailored clinical interventions for specific subtypes of voice-hearing experiences. 35,36

"Front-loading" Cognitive-Neuroscientific Research
Given the National Institute of Mental Health's shift to neuropsychiatric domains in place of categorical diagnoses or diagnostically specific symptoms, the exigency of addressing translational continuities and discontinuities between phenomenology and underlying biological change seems clear. 71 Most experimental and correlational research continues to depend on self-report measures and tentative assumptions concerning the reliability of self-report. Neuroimaging studies, for instance, depend on self-report measures in order to categorize subjects experiencing or not experiencing AVHs, as well as on accurate "real-time" self-report within the scanner. More finely tuned phenomenological distinctions, including spatial location, loudness, and subjective reality, have proven important in understanding differential functional and structural alterations. [72][73][74] Conversely, unquestioned assumptions about AVH phenomenology can potentially lead to the false aggregation or summation of potentially neuropathologically distinct phenomena and risk compromising the validity of the constructs used to study AVHs.
Research on the relationship between "auditory false perceptions" and AVH is a case in point. One commonly used cognitive paradigm used to study AVHs is the "signal detection" task, in which participants are asked to indicate with a button press whether they heard a voice in a burst of white noise. Findings suggest that people who experience AVHs are more likely to report that they heard a voice when no voice was present. It has therefore been suggested that these "auditory false perceptions" are, to some extent, analogous to "true" voice-hearing experiences 75 (a finding supported by neuroimaging evidence suggesting similar brain activations during false alarm responses and AVHs). 76 However, the extent of this analogy remains unclear: what do people "experience" when they make a false alarm response? Although data suggest that voice hearers show a decisional bias toward responding "yes," it is also unclear what drives this biaseg, are false alarm responses accompanied by a feeling of non-self-generation and alien-ness, as some AVHs are? 77 Would different kinds of AVHs, such as those described as "hypervigilance" hallucinations, 78 be more analogous to false alarm responses?
Researchers must decide how best to make the link between the "phenomenon" of voice hearing and the specific aspects of brain, body, or behavior under investigation, and how the relevant aspects of experience should be identified and used within experimental contexts. Advocating an interdisciplinary approach, philosopher Shaun Gallagher has urged a "front-loaded phenomenology," whose guiding idea is to incorporate insights from phenomenological philosophy into experimental design. 79 Philosophically informed phenomenological characterizations of AVH are notable for their richness and complexity; while selection from and even simplification of these accounts will be required in order to enable sufficient control over the many variables being studied, phenomenologically inspired studies of self-disturbances 80 suggest this might be a fruitful direction for cognitive neuroscientific research into AVH.

Enhancing Therapeutic Practice
Just as an interdisciplinary approach to the phenomenology of AVH can open up new ways of approaching taxonomies of voice hearing and the design of empirical studies, so too can the humanities, arts, and social sciences-in foregrounding the complexity, multidimensionality, and affective qualities of hallucinatory experiences-offer practical resources for individuals and clinicians seeking to understand and alleviate the distress which can accompany them. In this section we show how the narrative arts (cinema and literature), the tools of narrative and linguistic analysis, and approaches S251 Interdisciplinary Approaches to the Phenomenology of AVH which take seriously the spiritual significance of voices, can offer valuable resources in the therapeutic encounter.
Phenomenological psychiatrist Karl Jaspers 26 claimed that the process of developing a psychotic symptom could be subject to objective or causal explanation (erklären) but not an ability to empathically grasp the subjective experience and know what it is like (verstehen). While art has explored madness over millennia, the fact that the formation of psychotic symptoms is an extended process suggests that narrative arts are some of the best placed to help communicate the subjective experiences which Jaspers famously claimed were "ununderstandable." As a medium including both sensory and narrative aspects, cinema can comprehensively characterize complex psychological disturbances and help make them empathically available to the spectator. 81 Masterpieces of cinema d'auteur (as directed by Bergman, Buñuel, and Cronenberg, eg) offer a "vision from within," capturing aspects of psychosis in all its existential and interpersonal impact. Enigmatic and sometimes impenetrable subjective experiences can be depicted on screen in several ways: either by observationwhere the viewer sees the behavior of someone experiencing AVHs; by insightful representation-where the viewer sees the world as the voice hearer does, in full knowledge that the experiences are not shared by others; by firstperson representation-where the viewer is not aware, at least at the time, that the experiences are not shared; or by analogy-where psychosis-like experiences are accurately represented but not identified as such in the story. 81 Literature provides another invaluable repertoire for studying the distinguishing features of AVHs and problematizing the alleged commonality of the experience. The techniques of narrative fiction enable the reader to experience a "transparency" in accessing multiple fictional minds. 82 For example, through the fiction of Virginia Woolf, 83 Samuel Beckett, 84 and Hilary Mantel 85 we can gain access to shifting perspectives on the content of hallucinatory experience; its evolution across time; and its embodied, interpersonal, and social dimensions. Immersion in a fictional world can also trigger in the reader a complex empathetic experience, allowing the reader to enact a simulative experience of the event. 86 The value of fiction here is underscored by the relative paucity of longitudinal studies of AVH and the difficulties faced by empirical researchers in accessing and measuring subtle changes within individuals' experiences.
Voices have the distinction of being symptoms which speak: ie, symptoms which often appear as language. Literary theory and in particular narratology can provide powerful, if underrecognized, tools with which to investigate self-reports of AVHs. First-person accounts are often shaped as narratives in which the voice is described as a speaking character (ie, personalized with name, intentions, temper). Complementing psychological studies of persons' relationship with their voices, 87 a narratological analysis of such talkative acts can highlight structural commonalities or specific features such as the voice's narrative distance and point of view (speaking in first person, second person, third person, or a plural "we"); spatial and temporal indexing (here, there, this, that time, and so on); and the consonant or dissonant narration (in terms of content, timeline, ideology, information) the voice produces in relationship to the voice hearer. Furthermore, if the voice hearer as interpreter of her voices can also be considered as a reader, contemporary cognitive narratology can shed light on how-in this interpretive activity -she is able to construct a "continuing-consciousness frame" 88 for this voice or fill the gaps in the voice's narration. Analyses of linguistic style and word use 89 likewise can provide simple but powerful tools to assist voice hearers to track for themselves how they are thinking and communicating in their daily lives (eg, via emails and diaries) both with voices and with others. While these are important goals in their own right, the potential for literary and linguistic methods to fine-tune taxonomies of voice hearing and so feed into translational research also warrants further exploration.
Within the experience of those who are diagnosed as suffering from mental disorder, spiritual and religious themes arise frequently. Among a group of people diagnosed as suffering from schizophrenia, 60% reported that they found religion helpful in coping with their experience of hearing voices. 90 Psychiatrists are less likely to be religious than their patients, 91 and so the potential for pathological interpretation of religious experience-thus denying its importance as a coping resource-is significant. In fact, first-hand accounts of voice hearing, from those of medieval mystics 92 to contemporary users of mental health services 93,94 suggest that people can and do distinguish between "spiritual" voices and voices which they themselves consider to be a manifestation of illness. Similarly, some professionals have suggested systematized criteria by which distinctions between spiritual and pathological voices can be made 95 whereas others have suggested that a range of unusual experiences-including divine or mystical voices and paranormal phenomena-might all in fact be a part of a realm of the "borderline" or "transliminal." 96 Interdisciplinary approaches to phenomenology as outlined here can have practical benefits in helping to develop better patient-therapist relationships. It has been argued that, without a good patient-therapist relationship, the efficacy of cognitive behavioral therapy (CBT) is reduced, and that therapist empathy is important in fostering this relationship. 97 This claim is supported by the reports of patients with a diagnosis of schizophrenia 98 and by data showing that patient ratings of therapist empathy are a predictor of therapeutic alliance in CBT for psychosis. 99 If the core feature of empathy is "the ability to understand the patient's situation, perspective and feelings" 100 it seems likely that interdisciplinary approaches to phenomenology would be helpful in communicating these features of a patient's world to clinicians who might otherwise struggle to empathize with hallucinatory experiences. 101

Conclusion
This article has argued for the value of an interdisciplinary approach to the phenomenology of AVH to complement, challenge, enrich, and extend mainstream hallucination research. AVHs do not exist except in context: the context of human consciousness, of an individual's life, and as made meaningful within the explanatory frameworks available in particular places and historical periods. While it can be extremely fruitful to take AVHs "aside" and examine them in isolation and from different disciplinary perspectives, we also must not forget that AVHs are part of the gestalt of the person. Empirical psychological and psychiatric accounts of AVH phenomenology can be enriched, we have argued, by an interdisciplinary approach which utilizes the robust methodologies of the humanities and social sciences to fully realize this contextual complexity, inextricable as it is from experience itself. This has significant potential benefits for the refinement of AVH subtypes, informing empirical work, and enhancing therapeutic practice.
Interdisciplinary approaches from the humanities and the social sciences demand that we all-no matter what our own expertise or topic of research-need to examine which sources, frameworks, and models we explicitly or implicitly foreground in the production, analysis, and dissemination of data relating to the experience of voice hearing. As well as offering new ways of working with and thinking through what the "subjective" might mean in research on AVH, these literatures raise difficult questions concerning which disciplines, which theoretical perspectives, and which kinds of expertise have most authority in determining how concepts are defined, how the category of "voice-hearing" is delineated from other experiences, and which questions are most pressing for researchers to answer. The challenges of interdisciplinary research are practical as well as philosophical. Work that endeavors to cross epistemic, disciplinary, discursive, and professional-cultural divides can be difficult and time consuming to produce, and harder still to disseminate. Yet multiperspectival approaches can nourish the ethical core of scientific enquiry by challenging prevailing interpretive paradigms and showing that these are, as with every paradigm, culturally and historically grounded. 102 They also offer rich possibilities for voice hearers in making sense of their experiences outside the relatively narrow frameworks of conventional psychiatric frameworks. As indicated in the supplementary table, the real challenge now is to work toward ways in which an interdisciplinary approach that foregrounds and values multiple forms of expertise-professional and experiential-can be fully integrated into mainstream psychiatric and hallucination research.