The voice and you: development and ... - Wiley Online Library

Clinical Psychology and Psychotherapy Clin. Psychol. Psychother. 15, 45–52 (2008) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/cpp.561

Assessment

The Voice and You: Development and Psychometric Evaluation of a Measure of Relationships with Voices Mark Hayward,1* Joanna Denney,2 Sam Vaughan3 and David Fowler3 1

University of Surrey University of Southampton 3 University of East Anglia 2

The experience of hearing voices has recently been conceptualized within a relational framework. Birtchnell’s Relating Theory offers a framework capable of exploring the power and intimacy within the relationship between the hearer and the voice. However, measures of relationships with voices derived from the theory, such as the Hearer to Voice (HTV) and Voice to Hearer (VTH) by Vaughan and Fowler, have lacked robust psychometric properties. Data were available from 71 participants who completed the HTV and VTH, and analysis of these data generated a new 29-item measure, the Voice and You (VAY), capable of assessing the ‘interrelating’ between the hearer and the voice. The VAY was completed by a further 30 participants and was found to be internally consistent, stable over time and associated with other measures of the voice-hearing experience. The VAY offers a psychometrically stable measure of the relationship between the hearer and the voice. It may be used as an adjunct to the clinical interview and/or a measure of outcome. Copyright © 2008 John Wiley & Sons, Ltd.

INTRODUCTION Over the past two decades the experience of hearing voices (‘auditory hallucinations’) has been considered within a relational framework. The inter-

* Correspondence to: Mark Hayward, Psychology Department, University of Surrey, Guildford GU2 7XH, UK. E-mail: [email protected]

Copyright © 2008 John Wiley & Sons, Ltd.

personal nature of the voice-hearing experience was first explored by Benjamin (1989), who, using the Structural Analysis of Social Behaviour, found hearers to have ‘integrated, personally coherent relationships with their voice’ (p. 308). Conceiving of the voice(s) as an interpersonal ‘other’, subsequent research began to explore the characteristics of the relationships that individuals may develop with the voices they hear and how such characteristics may be associated with distress (Vaughan

46 & Fowler, 2004) and social relating (Birchwood et al., 2004; Birchwood, Meaden, Trower, Gilbert, & Plaistow, 2000; Hayward, 2003). Two theories of interpersonal relating have recently been utilized to explore relationships with voices: Gilbert’s Social Rank Theory (Gilbert & Allan, 1998) and Birtchnell’s (1996, 2002) Relating Theory. Social Rank Theory focuses primarily on issues of power and has been utilized therapeutically to modify the relationships that hearers have with voices that issue commands (Trower et al., 2004). Relating Theory focuses additionally on issues of proximity and intimacy, themes that have been reported as prevalent within the experience of hearing voices (Birchwood & Chadwick, 1997; Garrett & Silva, 2003; Nayani & David, 1996). Birtchnell’s (1996, 2002) Relating Theory describes relating along the intersecting axes of power and proximity, represented at their two poles by ‘upperlower’ and ‘distant-close’, respectively. A person can relate positively or negatively in each of these four positions. An ability to relate positively from all four positions and move confidently between them is termed versatility, and is likely to be contextualized by a history of satisfactory social relationships. A negative relater, having experienced relatively impoverished social relationships, will occupy only a limited number of positions, possibly at odds with the relational needs of others. Birtchnell (1996, 2002) further defines ‘interrelating’ as the process by which two parties experience and are influenced by the relating of each other. The additional focus upon proximity afforded by Relating Theory has gleaned insights into styles of relating that the hearer may adopt in response to the dominance of a voice (e.g., ‘distance’—attempting to withdraw and escape) (Vaughan & Fowler, 2004), and the similarities between relating to voices and relating to others socially (Hayward, 2003). Prior to the therapeutic application of Relating Theory to voice-hearing, a measure of relating was required. Birtchnell, Voortman, De Jong and Gordon (2006) developed a reliable and valid set of questionnaires to assess the relating of couples, but these measures required modification in order to assess relating to voices. Vaughan and Fowler (2004) made the necessary modifications to produce two 40-item questionnaires: the Hearer To Voice (HTV), which assesses the relating of the hearer over four subscales—upperness, lowerness, closeness and distance; and the Voice To Hearer (VTH), which assesses the hearer’s perceptions of the relating of the predominant voice over the same four subscales (see Vaughan & Fowler, Copyright © 2008 John Wiley & Sons, Ltd.

M. Hayward et al. 2004 for a detailed description of the HTV and VTH). Examination of the psychometric properties of the HTV and VTH revealed some subscales that were not sufficiently reliable. Specifically, the upperness scale of the HTV (Chronbach’s alpha of 0.44) (Hayward, 2003) and the lowerness and distance scales of the VTH (Cronbach’s alpha of 0.50 and 0.47, respectively) (Vaughan & Fowler, 2004) were found to lack internal consistency. Hayward (2003) suggested that these inconsistencies may have been attributable to ‘. . . a paradigm shift too far . . . . To construe the voice-hearing experience as one that involved interrelating (a combination of relating and being related to), as opposed to a unidirectional experience of being related to by the voice, was one paradigm shift. To then conceive of themselves as in any way relating from a position of upperness [or the voice relating from positions of lowerness or distance] within this relationship may have seemed too far-fetched for many participants’ (Hayward, 2003, p. 379). As a consequence of the poor reliability of some of their scales, Vaughan and Fowler (2004) recommended that the HTV and VTH be subjected to further psychometric evaluation and possible revision. An evaluation was conducted by the current authors (reported in the methods section) and necessitated modification of the HTV and VTH. The modifications resulted in the two measures being combined to develop a single questionnaire assessing the interrelating between the hearer and the voice—the Voice and You (VAY). This study aimed to explore the reliability and validity of the VAY. Reliability was assessed in relation to the internal consistency of scales and their stability over time. The assessment of validity was facilitated through the exploration of associations with established measures of emotional responses to voices—the Psychotic Symptoms Rating Scale (PSYRATS) by Haddock, McCarron, Tarrier and Faragher (1999) and cognitive–behavioural conceptualizations of the perceived power and intent of the voice and the responses of the hearer—Beliefs about Voices Questionnaire–Revised (BAVQ-R) by Chadwick, Lees and Birchwood (2000). Specifically, it was hypothesized that 1. The scales of the VAY will be found to be internally consistent, demonstrated by a Cronbach’s alpha of ≥0.7. 2. The scales of the VAY will be found to be stable over time, demonstrated by test–retest correlations of ≥0.7. Clin. Psychol. Psychother. 15, 45–52 (2008) DOI: 10.1002/cpp

Assessment of Voice-Hearing Experiences within a Relational Framework 3. The scales of the VAY will demonstrate concurrent validity with scales from the BAVQ-R by Chadwick et al. (2000) and the PSYRATS by Haddock et al. (1999): (a) Voice dominance and intrusiveness will be significantly associated with beliefs about the voice’s ‘malevolence’ and ‘omnipotence’ (as measured by the BAVQ-R), and ‘intensity of distress’ (as measured by the PSYRATS). (b) Hearer distance will be significantly associated with ‘resistance’ (BAVQ-R) and intensity of distress (PSYRATS). (c) Hearer dependence will be significantly associated with beliefs about the voice’s ‘benevolence’ and ‘engagement’ (BAVQR) and significantly associated with intensity of distress in a negative direction (PSYRATS).

METHOD Development of the VAY Due to the poor reliability of some of their scales the HTV and VTH required further psychometric evaluation. Data were available from 71 participants: 14 participants had completed only the VTH; 27 participants had completed only the HTV. Thirty participants completed both the HTV and the VTH (total n = 57 for HTV; total n = 44 for VTH). Participants were served by rehabilitation and community mental health teams across three National Health Service (NHS) Trusts. Inclusion criteria required participants to have heard voices for at least 6 months, irrespective of diagnosis. Prospective participants were excluded if they had been given primary diagnoses related to organic disorders or substance misuse. Thirty nine participants were men (55%) and 32 were women (45%). Ages ranged from 21 to 63 years, with a mean (M) age of 39.42 (Standard Deviation [SD] = 10.49). The participants had been given the following diagnoses: 61 with schizophrenia; two with schizoaffective disorder; one with bipolar affective disorder; two with personality disorder; two with depression; and one with post-natal psychosis. Two people had no diagnosis. Duration of voice-hearing ranged from 1 to 50 years (M = 13.84; SD = 10.49). Consistent with the reliability analysis of Vaughan and Fowler (2004) and Hayward (2003), the upperness scale of the HTV and the lowerness and distance scales of the VTH were removed as they were Copyright © 2008 John Wiley & Sons, Ltd.

47

found to lack internal consistency. The remaining 50 items across five subscales (distance, lowerness and closeness of the HTV, upperness and closeness of the VTH) were analysed with respect to keyed response and item–total correlations. Kline (1986) recommends that an affirmative response to an item should account for between 20% and 80% of responses. For the HTV, seven items did not meet this criterion and were removed: three from lowerness (e.g., ‘I let my voice take responsibility for me’) and four from closeness (e.g., ‘When my voice is absent, I feel anxious until it returns’). For the VTH, one item from the closeness scale did not meet the criteria and was removed (‘My voice complains that I do not pay him/her enough attention’). Each remaining item was correlated with the total score for its subscale. Following correction for removal from the total of the item being assessed, each item was required to correlate at 0.3 or above with its subscale and to correlate less strongly with any other subscale for that measure (Kline, 1986). For the HTV, eight items did not meet the criteria and were removed: three items from the distance subscale (e.g., ‘I tend to escape from my voice into a world of my own’); two from lowerness (e.g., ‘I ask my voice to help me solve my problems’); and two from closeness (e.g., ‘I can be very demanding of my voice’s attention’). For the VTH, six items did not meet the criteria and were removed: three items from the upperness subscale (e.g., My voice puts me off the things I like’) and three from closeness (e.g., ‘Whenever my voice wants to talk I have to be ready to talk’). Following analysis for keyed response and item– total correlation, 29 items remained across five subscales. Further modification concerned criticism reported by Hayward (2001) with respect to the interaction between response scale and some items producing double negatives, and the strong association between items on the HTV scales of lowerness and closeness (r = 0.73, p < 0.01). Consequently, the response scale was changed to ‘nearly always true’ to ‘rarely true’, three items were phrased less negatively, and the lowerness and closeness scales were collapsed to form a single scale of ‘dependence’ (Birtchnell, Falkowski, & Steffert, 1992). The further psychometric evaluation of the HTV and VTH revealed four scales that were reliable and valid. These subscales were renamed ‘hearer distance’ (seven items), ‘hearer dependence’ (nine items), ‘voice dominance’ (seven items) and ‘voice intrusiveness’ (six items), and were combined to form a new measure capable of assessing the interClin. Psychol. Psychother. 15, 45–52 (2008) DOI: 10.1002/cpp

48 relating between the hearer and the voice: the VAY. This measure required piloting on a new sample.

Participants A new sample of 30 participants was recruited from community and rehabilitation teams across two NHS Trusts. Inclusion criteria required participants to have heard voices for at least 6 months, irrespective of diagnosis. Prospective participants were excluded if they had been given primary diagnoses related to organic disorders or substance misuse. Twelve of the participants were female (40%) and 18 were male (60%). Ages ranged from 22 to 65 years, with a mean age of 39.52 (SD = 10.54). The participants had been given the following diagnoses: 19 with schizophrenia; four with schizoaffective disorder; three with depression with psychotic features; and three with personality disorder. One person had not received a diagnosis. Duration of voice-hearing ranged from 1 to 49 years (M = 15.40, SD = 13.27).

Measures Semi-Structured Interview The introductory questions from the cognitive assessment of voices (Chadwick & Birchwood, 1994) were adapted to form the basis of a short semi-structured interview aimed at providing some background information about the participant’s voice-hearing experience. The interview also facilitated the identification of the predominant voice that would form the focus of subsequent measurement.

Voice and You The VAY is a 29-item measure of interrelating between the hearer and their predominant voice (see above for a description of the development of this measure). Relating is measured across four scales; two concerning the hearer’s perception of the relating of the voice—voice dominance and voice intrusiveness; and two concerning the relating of the hearer—hearer distance and hearer dependence. Each item is measured on a fourpoint scale (0–3).

M. Hayward et al. sity of distress and controllability. Each item is measured by the rater on a five-point scale ranging from 0 to 4. The PSYRATS was used in this study to provide a detailed profile of the voice-hearing experience of participants. Data from the ‘intensity of distress’ item was used to test hypotheses exploring the concurrent validity of the VAY.

The Revised Beliefs about Voices Questionnaire (BAVQ-R) The BAVQ-R is a 35-item questionnaire that generates five subscales: three concerning beliefs about the dominant voice (malevolence, benevolence and omnipotence) and two concerning emotional and behavioural reactions (resistance and engagement) (Chadwick et al., 2000). Each item is rated on a four-point scale (0–3). Data from the BAVQ-R was used to test hypotheses exploring the concurrent validity of the VAY.

Procedure All participants were assessed by the second author, who had received formal training in the administration of the measures. Prior to assessment participants gave written consent. The measures were administered in the following order: semistructured interview, PSYRATS, VAY, BAVQ-R. Assessment took place over two sessions at the venue the participant chose.

Statistical Analysis The scales of the VAY were explored descriptively to provide information on the distribution of scores. Reliability was assessed in two ways: Cronbach’s alpha was used to assess internal consistency and test–retest reliability was assessed using Pearson correlations. Validity was measured in relation to existing measures of the voice-hearing experience (BAVQ-R and PSYRATS) using Pearson correlations. Analysis of data was performed using the Statistical Package for Social Sciences (SPSS Inc., Chicago) for Windows (version 13). All reported p values are two-tailed.

PSYRATS: Auditory Hallucinations Scale (AHRS)

RESULTS

The AHRS is an 11-item rating scale designed to measure the severity of different dimensions of the voice-hearing experience (Haddock et al., 1999). Items include frequency, duration, loudness, inten-

A Description of the Voice-Hearing Experience


Participants reported the gender of their predominant voices as follows: male (n = 18), female (n = Clin. Psychol. Psychother. 15, 45–52 (2008) DOI: 10.1002/cpp

Assessment of Voice-Hearing Experiences within a Relational Framework 6) and uncertain (n = 6). Ten participants heard voices continuously and the majority heard them at least once a day. Twenty-five of the participants reported the majority of voice content to be negative, primarily in the form of personal abuse and threats to the self. Two participants reported no negative content. The vast majority of participants experienced their voices as very (n = 12) and extremely (n = 12) distressing. An inability to control the voices was also apparent with 22 of the participants stating they had little or no control.

49

The Cronbach’s alpha for the scale of voice intrusiveness was also acceptable (a = 0.77). However, item 3 (‘My voice needs to have me around a lot’) exhibited a very low item–total correlation (0.07) and was consequently removed as it made no unique contribution to the scale. The revised Cronbach’s alpha for the five-item scale was a = 0.82. It should also be noted that missing scores affected this scale. Consequently, missing values for voice intrusiveness were replaced by participants’ median scores for the scale. This replacement may have artificially inflated the alpha.

Descriptive Analysis of the VAY The distribution of scores was skewed for each of the scales of the VAY: for voice dominance, voice intrusiveness and hearer distance this was in a negative direction with skewness statistics of −1.39, −0.76 and −1.54, respectively; for hearer dependence this was in a positive direction with a skewness statistic of 1.76. The mean scores for the VAY suggested that participants related to their predominant voices primarily from a position of distance (M = 15.73, SD = 5.42), in response to a voice perceived as dominant (M = 16.10, SD = 6.77) and intrusive (M = 10.03, SD = 4.80). The mean score for hearer dependence was lower (M = 6.67, SD = 5.86), suggesting that participants did not make extensive use of this style of relating to the voice.

Hypothesis 1: Internal Consistency of the VAY The internal consistency of the VAY was analysed using Cronbach’s alpha and was found to be acceptable for the scales of voice dominance (a = 0.92), hearer dependence (a = 0.78) and hearer distance (a = 0.81).

Table 1.

Hypothesis 2: Test–Retest Reliability of VAY Seventeen participants (57% of the sample) completed the VAY again within 3 weeks of the initial assessment. Acceptable test–retest correlations were found for each of the scales: voice dominance (r = 0.91, p < 0.01), voice intrusiveness (r = 0.88, p < 0.01), hearer distance (r = 0.91, p < 0.01) and hearer dependence (r = 0.72, p < 0.01). Table 1 shows the summary of descriptive and reliability statistics.

Hypothesis 3: Concurrent Validity The total scores from the scales of the VAY were correlated with those from the relevant scales of the BAVQ-R and PSYRATS: 1. A perception of the voice relating dominantly and intrusively was significantly associated with malevolence (r = 0.87, p < 0.01; r = 0.76, p < 0.01, respectively) and omnipotence (r = 0.59, p < 0.01; r = 0.64, p < 0.05, respectively). In addition, voice dominance and voice intrusiveness were each associated with distress (r = 0.70, p < 0.01; r = 0.48, p < 0.01, respectively).

Descriptive and reliability statistics for the Voice and You

No. of items Range of scores Mean (SD) Range of corrected item–total correlation Cronbach’s alpha Test–retest reliability

Voice dominance

Voice intrusiveness

Hearer distance

Hearer dependence

7 0–21 16.10 (6.77) 0.69–0.89

5* 1–15 10.03 (4.80) 0.52–0.79

7 0–21 15.73 (5.42) 0.32–0.80

9 0–25 6.67 (5.86) 0.30–0.81

0.92 0.91

0.82 0.88

0.81 0.91

0.78 0.72

* Following removal of item 3. SD = Standard Deviation.


Clin. Psychol. Psychother. 15, 45–52 (2008) DOI: 10.1002/cpp

50 2. The hearer relating to the voice from a position of distance was associated with resistance (r = 0.77; p < 0.01) and distress (r = 0.61, p < 0.01). 3. The hearer relating to the voice from a position of dependence was associated with engagement (r = 0.63, p < 0.01) with a voice perceived as benevolent (r = 0.55, p < 0.01). In addition, hearer dependence was negatively associated with distress (r = −0.37, p < 0.05).

Additional Analysis: Correlations between Scales of the VAY The correlations of the different scales of the VAY were examined. The following significant associations were found: voice dominance and voice intrusiveness (r = 0.66, p < 0.01); voice dominance and hearer distance (r = 0.79, p < 0.01); voice intrusiveness and hearer distance (r = 0.58, p < 0.01); and a negative correlation between hearer distance and hearer dependence (r = −0.56, p < 0.01). Possible interpretations of these associations will be discussed below.

DISCUSSION An evolving literature suggests that the conceptualization of the voice-hearing experience within a relational framework may be of therapeutic value (Trower et al., 2004). More specifically, Birtchnell’s (1996, 2002) Relating Theory offers a framework capable of exploring issues of both power and proximity within the relationship with the voice (Hayward, 2003; Vaughan & Fowler, 2004). Prior to the therapeutic application of Relating Theory, it was necessary to develop a measure that could assess the appropriate relating variables within the voice-hearing experience. This study has developed and explored the psychometric properties of the VAY and has shown it to be a measure of the interrelating between hearer and voice that is both reliable and valid (see Appendix for a list of VAY items). Reliability has been demonstrated through the internal consistency of the four scales and the stability of their measurement over time. The validity of the VAY has been demonstrated through the strong associations found between its scales and those of existing measures of the voice-hearing experience. The VAY measures the hearer’s perceptions of the negative relating of the voice with respect to dominance and intrusiveness, and the negative relating of the hearer with respect to distance and dependence. Consistent with previous studies, the Copyright © 2008 John Wiley & Sons, Ltd.

M. Hayward et al. participants within this study primarily reported a desire to distance themselves from a voice perceived as dominant (Hayward, 2003; Vaughan & Fowler, 2004). Distant relating by the hearer was significantly associated with distress, offering support to the suggestion of Vaughan and Fowler (2004) that attempts to distance and distract oneself from such complex experiences, possibly rooted in an individual’s interpersonal history, may be of limited utility.

LIMITATIONS The psychometric evaluation of the VAY has identified a number of issues that require conceptual clarification. The significant association between the relational variables and beliefs about voices raises the issue of their distinctiveness. This issue was explored by Vaughan and Fowler (2004), who found a tendency for the voice to relate dominantly and the hearer to respond through distancing to be uniquely associated with distress. Their interpretation of this uniqueness concerned the possibility that ‘it is not the perceived powerfulness of the voice per se that is problematic, but perhaps the way in which the voice is perceived to use its power’ (Vaughan & Fowler, 2004, p. 150). The robustness of the unique contribution of relating variables relative to beliefs about voices is currently being explored in a replication of the Vaughan and Fowler (2004) study and will be reported in a separate paper. A further need for clarity concerned the significant associations found between several of the scales of the VAY. Most notably this concerned the scales of voice dominance and voice intrusiveness; a correlation of r = 0.66 suggesting that the two scales may have been measuring the same underlying construct of negative relating by the voice. Despite this possibility, a decision was made to maintain the separateness of the two scales for clinical reasons. First, the retention of the intrusiveness scale maintains an emphasis upon issues of proximity and intimacy, characteristics of the voice-hearing experience prevalent within phenomenological studies (Garrett & Silva, 2003; Nayani & David, 1996). Second, the intrusiveness scale uniquely highlights the perceived relational needs of the voice, an important prerequisite to a consideration of the motives of the voice and a possible process of negotiation between parties with conflicting agendas, referred to by Birchwood and Chadwick (1997) as a modus vivendi. Finally, a therapeutic focus upon acceptance of the Clin. Psychol. Psychother. 15, 45–52 (2008) DOI: 10.1002/cpp

Assessment of Voice-Hearing Experiences within a Relational Framework voice-hearing experience, such as through the use of mindfulness approaches (Chadwick, NewmanTaylor, & Abba, 2005), may benefit conceptually from the possibility and measurement of proximal change. Future research should attempt to clarify the distinctiveness of the VAY scales by recruiting a sample of a sufficient size for the conducting of a factor analysis. More general limitations of this study concerned the small sample size and a focus upon a clinical sample. The former limitation increases the risk of Type II errors, and findings should be interpreted with caution. The latter issue negates the possibility of learning from people who do not use mental health services and who report less distress (Honig et al., 1998).

Hearer Views and Therapeutic Applications From the perspective of the hearers the distinctiveness and value of assessing and working therapeutically within a relational framework is becoming apparent. Many of the participants interviewed by Denney (2004) ‘commented that completing the questionnaire [the VAY] had prompted them to consider the detail of the inter-relationship for the first time’ (p. 61). The intuitive nature of this conceptualization to some hearers has also been demonstrated through in-depth qualitative analysis by Chin, Hayward and Drinnan (2006), who found that 7 out of 10 participants experienced their voice as a relational other (albeit reluctantly in some cases) with whom attachments and a sense of ‘we-ness’ could be formed. Therapeutically, engagement with a relational framework will offer treatment opportunities to ‘people [who] do not wish to alter their views that voices reflect real interpersonal experience’ (Vaughan & Fowler, 2004, p. 152). Such an approach has the potential to normalize voice-hearing by locating it within a known realm of experience and can create possibilities for relational change within the social world of the hearer (Birchwood, Meaden, Trower, & Gilbert, 2002). Analysis of the experiences of participants within a recent case series of individual relating therapy for voices has also revealed the acceptability of this approach to the hearers (Hayward & May, 2007).

Clinical Implications The clinical uses of the VAY may be varied. Used formally as a self-report measure, it may act as an Copyright © 2008 John Wiley & Sons, Ltd.

51

adjunct to the clinical interview and provide valuable information for the case formulation about a variable that is increasingly being recognized as a mediator of distress within the voice-hearing experience. The VAY can also be used to assess change at the conclusion of therapy that attempts to modify the interrelating between the hearer and voice. More informally, the VAY may be used as a ‘catalyst for conversation’ between the therapist and the hearer, possibly triggering curiosity about the relational nature of the voicehearing experience and its connections with past and present interpersonal relating, which has been referred to by Romme and Escher (2000) as a ‘construct’.

ACKNOWLEDGEMENTS The authors would like to thank John Birtchnell, Eleanor Sorrell and Arlene Vetere for their comments on earlier drafts of this paper.

REFERENCES Benjamin, L.S. (1989). Is chronicity a function of the relationship between the person and the auditory hallucination? Schizophrenia Bulletin, 15, 291–310. Birchwood, M., & Chadwick, P. (1997). The omnipotence of voices: Testing the validity of a cognitive model. Psychological Medicine, 27, 1345–1353. Birchwood, M., Gilbert, P., Gilbert, J., Trower, P., Meaden, A., Hay, J. Murray, E., & Miles, J.N.V. (2004). Interpersonal and role related schema influence the relationship with the dominant ‘voice’ in schizophrenia: A comparison of three models. Psychological Medicine, 34, 1571–1580. Birchwood, M., Meaden, A., Trower, P., & Gilbert, P. (2002). Shame, humiliation and entrapment in psychosis. A social rank theory approach to cognitive intervention with voices and delusions. In A.P. Morrison (Ed.), A casebook of cognitive therapy for psychosis (pp. 108–131). Hove: Brunner-Routledge. Birchwood, M., Meaden, A., Trower, P., Gilbert, P., & Plaistow, J. (2000). The power and omnipotence of voices: Subordination and entrapment by voices and significant others. Psychological Medicine, 30, 337– 344. Birtchnell, J. (1996). How humans relate: A new interpersonal theory. Hove: Psychology Press. Birtchnell, J. (2002). Relating in sychotherapy: The application of a new theory. Hove: Brunner-Routledge. Birtchnell, J., Falkowski, J., & Steffert, B. (1992). The negative relating of depressed patients: A new approach. Journal of Affective Disorders, 24, 165–176. Birtchnell, J., Voortman, S., De Jong, C., & Gordon, D. (2006). Measuring interrelating in couples: The Couple’s Relating to Each Other Questionnaires (CREOQ).


52 Psychology and Psychotherapy: Theory, Research & Practice, 79, 339–364. Chadwick, P.D.J. & Birchwood, M.J. (1994). Challenging the omnipotence of voices: A cognitive approach to auditory hallucinations. British Journal of Psychiatry, 164, 190–201. Chadwick, P., Lees, S., & Birchwood, M. (2000). The revised Beliefs about Voices Questionnaire (BAVQ-R). British Journal of Psychiatry, 177, 229–232. Chadwick, P., Newman-Taylor, K., & Abba, N. (2005). Mindfulness groups for people with psychosis. Behavioural and Cognitive Psychotherapy, 33, 351–359. Chin, J., Hayward, M., & Drinnan, A. (2006). Relating to voices: Exploring the relevance of this concept to people who hear voices. Manuscript submitted for publication. Denney, J. (2004). Refinement and confirmation of the psychometric properties of the ‘Voice and You’ questionnaire. University of Southampton unpublished Master’s thesis. Garrett, M., & Silva, R. (2003). Auditory hallucinations, source monitoring, and the belief that ‘voices’ are real. Schizophrenia Bulletin, 29, 445–457. Gilbert, P., & Allan, S. (1998). The role of defeat and entrapment (arrested flight) in depression: An exploration of an evolutionary view. Psychological Medicine, 28, 585–598. Haddock, G., McCarron, J., Tarrier, N., & Faragher, E.B. (1999). Scales to measure dimensions of hallucinations and delusions: The Psychotic Symptoms Rating Scale (PSYRATS). Psychological Medicine, 29, 879–889. Hayward, M. (2001). An exploration of the ways in which people with auditory hallucinations relate to their voices. University of Leicester unpublished thesis. Hayward, M. (2003). Interpersonal relating and voice hearing: To what extent does relating to the voice reflect social relating? Psychology and Psychotherapy: Theory Research and Practice, 76, 369–383. Hayward, M., & May, R. (2007). Daring to talk back. Mental Health Practice, 10, 12–15. Honig, A., Romme, M., Ensink, B., Escher, S., Pennings, M., & deVries, M. (1998). Auditory hallucination: A comparison between patients and non-patients. Journal of Nervous and Mental Disease, 186, 646–651. Kline, P. (1986). A handbook of test construction: Introduction to psychometric design. London, UK: Methuen. Nayani, T.H., & David, A.S. (1996). The auditory hallucination: A phenomenological survey. Psychological Medicine, 26, 177–189. Romme, M., & Escher, S. (2000). Making sense of voices. London: MIND Publications. Trower, P., Birchwood, M., Meadon, A., Byrne, S. Nelson, A., & Ross, K. (2004). Cognitive therapy for command hallucinations: Randomised controlled trial. British Journal of Psychiatry, 184, 312–320. Vaughan, S., & Fowler, D. (2004). The distress experienced by voice hearers is associated with the perceived relationship between the voice hearer and the voice. British Journal of Clinical Psychology, 43, 143–153.


M. Hayward et al.

APPENDIX: VOICE AND YOU SCALE ITEMS Voice Dominance 1) My voice wants things done his/her way. 4) My voice makes hurtful remarks to me. 8) My voice constantly reminds me of my failings. 13) My voice does not give me credit for the good things I do. 17) My voice tries to get the better of me. 21) My voice makes me feel useless. 25) My voice tries to make me out to be stupid. Voice Intrusiveness 5) My voice does not let me have time to myself. 9) My voice dislikes it when I exclude him/her by showing an interest in other people. 14) My voice tries to accompany me when I go out. 18) My voice dislikes spending time on his/her own. 26) My voice finds it hard to allow me to have time away from him/her. Hearer Dependence 2) 6) 10) 12) 15) 19) 22)

My voice helps me make up my mind. I have a tendency to look up to my voice. I allow my voice to take control of me. It is easy for my voice to change my mind. I feel deserted when my voice is not around. My voice’s judgement is better than mine. I need to have my voice around me a great deal. 24) I have difficulty letting go of my voice. 27) I have a great need to talk to my voice. Hearer Distance 3) I prefer to keep my voice at a safe distance. 7) When my voice gets too close to me, it makes me feel uneasy. 11) I feel I have little to offer my voice. 16) I try to hide my feelings from my voice. 20) I do not like to get too involved with my voice. 23) I don’t like my voice to know what I am thinking. 28) I don’t wish to spend much time listening to my voice.
