This article was published in Psychology and Psychotherapy Theory Research

and Practice

Reference: Varese, F., Mansell, W. and Tai, S. J. (2017), What is distressing about auditory verbal hallucinations? The contribution of goal interference and goal facilitation. Psychol Psychother Theory Res Pract. doi:10.1111/papt.12135

Abstract

Objectives: Distressing as well as pleasant/positive voices (auditory verbal

hallucinations) are common in both clinical and non-clinical voice-hearers.

Identifying factors that contribute to emotional reactions to voices is essential for

developing effective psychological interventions. Several theories propose that

facilitation and interference with personal goals are important predictors of distress

and well-being. This study examined whether voice-related distress is related to the

degree to which voices interfere with personal goals, and whether pleasantness of

voices is influenced by the extent to which they facilitate goals.

Design: Cross-sectional with clinical and non-clinical voice-hears

Method: 22 clinical and 18 non-clinical voice-hearers completed interviews and

self-report measures assessing (i) personal goals, (ii) content, characteristics and

affective reactions to voices, and (iii) ratings of the extent to which voices facilitated

and/or interfered with achievement of important personal goals.

Results: Affective reactions were strongly correlated with measures of goal

interference and goal facilitation. Regression analyses revealed these associations

remained significant when controlling for important covariates (e.g. participant

1

grouping; content, frequency and duration of voices). Goal interference was

specifically associated with distress, whereas goal facilitation specifically predicted

perceived pleasantness of voices.

Conclusions: This study provides a novel perspective on the factors that might

contribute to distress in people who hear voices. The findings suggest that perceived

impact of voices on valued personal goals may be an important contributor of voice-

related distress. We propose that clinical assessments, formulations and interventions

could benefit from the careful analysis of the perceived impact of voices on goals.

Practitioner points:

These findings suggest that variability in voice-related distress is closely

linked to the perceived impact of voices on personal goals.

These strong effects observed highlight the importance of considering the

role of personal goals in future research on the psychological mechanisms

leading to distress associated with voice-hearing.

Psychological assessments may benefit from the careful exploration of the

impact of voice hearing on valued goals, and interventions promoting control

over personal goals may be explored as treatment options for clients with

distressing voices.

2

Hearing voices is frequently regarded as characteristic symptoms of

schizophrenia-spectrum disorders, but are also relatively common amongst

individuals with other diagnoses such as bipolar disorder, post-traumatic stress

disorder, depression, and personality disorders (for reviews of prevalence studies,

see Aleman & Larøi, 2008; McCarthy-Jones, 2012). Furthermore, a sizable minority

of individuals in the general population without mental health difficulties also

experience voices (e.g. Beavan, Read, & Cartwright, 2011; Johns, Nazroo,

Bebbington, & Kuipers, 2002). Comparisons of clinical and non-clinical voice-

hearers suggest that non-clinical voices are characterised by lower levels of negative

content and distress (e.g. Daalman et al., 2010; Daalman & Diederen, 2013; Hill &

Linden, 2012; Hill, Varese, Jackson, & Linden, 2012). However, voices described by

psychiatric patients are not always experienced as distressing (Oorschot et al., 2012).

Previous studies indicate that positive/pleasant voices are relatively common

amongst clinical voice-hearers and often associated with the person’s desire to

“preserve” and engage with voices (e.g. Jenner, Rutten, Beuckens, Boonstra, &

Sytema, 2008; Miller, O'Connor, & DiPasquale, 1993; Sanjuan, Gonzalez, Aguilar,

Leal, & Van Os, 2004). The reasons for such variability in emotional responses are

unclear, with the majority of previous research focusing primarily on the ways in

which voices are idiosyncratically appraised individuals (Chadwick, Lees, &

Birchwood, 2000; Morrison, Nothard, Bowe, & Wells, 2004; Peters, Williams,

Cooke, & Kuipers, 2012)

Personal goals are a psychological construct that may provide an integrative

perspective on the apparent variability in the affective reactions to voices. Goals are

defined as ‘internal representations of desired states, where states are broadly

construed as outcomes, events or processes’ (Austin & Vancouver, 1996, p. 338).

3

Goals can convey both information facilitating an individual in progressing toward

or maintaining desirable outcomes (e.g. to be successful), and also inhibiting

undesirable ones (e.g. to avoid failure; Dickson & MacLeod, 2004; Elliot & Sheldon,

1998; Elliot, Sheldon, & Church, 1997). Several theories propose that goals are

internally organised within hierarchical structures of interrelated internal standards,

ranging from innate or gene-driven needs and predispositions (e.g. the range of

optimal blood sugar levels, hunger, need for affiliation) to more complex desired

states formed from past perceptual experiences and sociocultural influences (e.g. a

desired body shape, the person’s self-concept; Austin & Vancouver, 1996). Implicit

to this hierarchical organisation is the notion that lower-order, concrete goals (e.g. to

call a sick relative) can be “traced” to more abstract and general higher order goals

(to be a loving person).

Within several control theory accounts of human behaviour and cognition

(e.g. Perceptual Control Theory, PCT; Powers, 1973, 2005; Powers, Clark, &

McFarland, 1960a, 1960b) the processes through which goals are achieved and

maintained has been often defined in terms of negative feedback control (e.g. Ashby,

1952; Wiener, 1948). In this context, control is defined as a process which reduces

the discrepancy between a desired state (a goal) and the person’s current experience,

leading to a process of continuous dynamic corrections to achieve and maintain

certain desired states. Previous research has demonstrated that successful pursuit of

personally valued goals (in other words, successful control over one’s goals) is

related to positive affect and subjective well-being (Diener, 1984; Emmons, 1986;

Oishi, Diener, Suh, & Lucas, 1999). In contrast, when important personal goals are

threatened and remain unachieved, negative affect and distress may be experienced,

especially when goals that are particularly valued or fundamental to the individual

4

remain in a state of error (Emmons, 1986; Martin, Tesser, & McIntosh, 1993;

McIntosh, Harlow, & Martin, 1995; Srull & Wyer, 1986).

These goals- and control-based theories provide a number of hypotheses

regarding the mechanisms leading to different affective reactions to voices. As

distress is the manifestation of reduced or lost perceived control over desired states,

and in line with findings suggesting that positive and negative affect are experienced

when individuals are successful or unsuccessful in attaining/progressing towards

valued goals (e.g. Emmons, 1986), it is predicted that different affective reactions to

voices depends on the extent to which voices either facilitate or interfere with

important personal goals. In the current study, we tested several hypotheses

consistent with this prediction. First, we examined whether voice-related distress is

associated with the degree to which voices interfere with goals attainment, and

whether perceived pleasantness of voices is related by the extent to which voices

facilitate goal attainment. Second, using regression methods, we tested whether these

hypothesised associations remained significant after accounting for other

determinants of affective reactions to voices (including the amount of negative voice

content, frequency frequency and duration of voices, and the amount of “disruption”

to daily functioning caused by the voices; Haddock, McCarron, Tarrier, & Faragher,

1999; Morrison et al., 2004; Varese et al., under review). Third, we examined the

related hypothesis that clinical voice-hearers (i.e. individuals generally characterised

by heightened levels of voice-related distress, despite considerable individual

variability) would present higher levels of goal interference and lower levels of goal

facilitation when compared to non-clinical voice-hearers.

5

Method

Participants

Forty voice-hearers were recruited for this study. In line with symptom-

specific approaches to the study of psychotic experiences (e.g. Bentall, 2003), no

restriction was placed on eligible psychiatric diagnoses. All participants met the

following eligibility criteria: (i) aged above 16-years-old, (ii) had experienced voices

in the two weeks prior to participating in the study, (iii) history of voice-hearing had

a minimum of six months duration, (iv) the experience of voices was not due to

organic illness (e.g. traumatic brain injury), hypnagogic/ hypnopompic states, sleep

disorders (e.g. sleep paralysis), or alcohol/ drugs intoxication. Demographic

characteristics for the sample are reported in Table 1.

[Insert Table 1 approximately here]

For the purpose of some of the analyses included in the present paper, two

participant subgroups were created: a non-clinical voice-hearers group (n = 18),

comprising individuals with no current or past mental health difficulties requiring

contact with mental health services and/or formal psychiatric diagnoses, and a

clinical voice-hearers group (n = 22), comprising individuals who disclosed having

received a psychiatric diagnosis and/or present or past use of inpatient and/or

outpatient mental health services. Group allocation was informed by self-reported

information provided as part of a short structured interview adapted from the

“Demographic Data”, “Education and Work History” and “Treatment and

Hospitalization History” sections of the Overview module of the Structural Clinical

Interview for DSM-IV-TR; Axis I Disorder (First, Spitzer, M., & Williams, 2002;

see Appendix A). Demographic and clinical characteristics for the sample are

6

reported in Table 1. The two groups were comparable in terms of age, gender and

ethnicity (White Caucasian vs other ethnicity).

Measures

Content and characteristics of voices. A semi-structured interview was

used to assess voice characteristics. Firstly, the interview schedule included the

auditory hallucinations subscale of the Psychotic Symptoms Rating Scale

(PSYRATS-AH; Haddock et al., 1999), a multi-dimensional semi-structured

interview assessing cognitive, affective and “physical” (e.g. loudness, location)

dimensions of voices on 5-point scales (0 to 4). The PSYRATS-AH has

demonstrated good validity and reliability in previous studies, including samples

comprising clinical and non-clinical voice-hearers (e.g. Sommer et al., 2010). For

the present analyses, three separate PSYRATS-AH scores were computed based on

previous factor analytic findings (e.g. Steel et al., 2007; Varese et al., 2016): a total

distress score, defined as the combination of the PSYRATS-AH amount and

intensity of distress items (Spearman-Brown reliability = 0.96 in this sample); a total

negative content score, defined as the amount and degree of negative content items

(Spearman-Brown reliability = 0.94), and a total score comprising the PSYRATS-

AH frequency and duration items (Spearman-Brown reliability = 0.76). We also used

the PSTRATS-AH single-item disruption scale, assessing the impact of voices on the

hearer’s daily living functioning. Furthermore, as the PSYRATS-AH only provides

measures of voice-related distress, but not other affective responses to voices,

participants were asked to provide self-reported ratings of voice pleasantness (“On a

scale from 0 to 10, how pleasant are the voices?”; 0 = “Not at all”, 10 = “Very much

so”). A similar item assessing voice-related distress (“On a scale from 0 to 10, how

7

distressing are the voices?”) was also included to ensure that similar measures of

distress and pleasantness were available for statistical analysis.

Voice facilitation and interference with goals. Participants completed a

series of procedures to generate personally salient “higher-order” (i.e. cross-

situational and/or self-descriptive) goals, followed by a set of questions exploring

further the extent to which their voices influenced their perceived ability to achieve

these goals. An adapted version of the Goal Task (Dickson & MacLeod, 2004) was

used to assess personal goals. This task was modified following piloting work with

two clinical and three non-clinical pilot participants, with the aim of adapting the

procedure to match current study aims and participant population. The task involved

two separate stages. Firstly, participants were asked to describe as many goals that

came to mind within a 3-minute period (i.e. goals were described to participants as

experiences that individuals typically try to accomplish or avoid), and rate the

perceived importance of each goal on a 11-point scale (i.e. “How important is for

you to accomplish this goal?”; 0 = “Not at all”, 10 = “Very much so”). Secondly, as

pilot work indicated that the goals generated using the above procedure were often

“concrete” (context- or activity-specific e.g. “Finish my nursing degree by

September”), an additional component was included to encourage participants to

generate more abstract and general goals. This involved the participants being asked

a series of “why questions” (a questioning style used in several goal constructs

approaches to demonstrate the hierarchical nature of goals; Carey, 2006; Mansell,

Carey, & Tai, 2012) inquiring why they considered important achieving a certain

goal. This was repeated until the participant identified a general cross-situational or

self-definitional goal (e.g. “Being a helpful person”) that underpinned the more

8

concrete goal previously generated (e.g. “Finish my nursing degree by September”).

This procedure was limited to the three goals rated as the most important by the

participant, and follow-up questions were used to ensure that the information

provided during this procedure were indeed regarded as salient goals by the

participant (“Is being an helpful person one of your goals?, “How important is for

you to accomplish this goal on a scale from 0 to 10”). Finally, participants rated how

much voices facilitated and/or hindered their achievement of goals. Separate

facilitation and interference ratings were provided for the three abstract/general goals

identified in the modified Goal Task (e.g. “How much do the voices help with

achieving this goal?” and “How much do the voices interfere with achieving this

goal?”; 0 = not at all, 10 = very much so). Ratings were averaged to provide two

separate summary scores for goal facilitation and goal interference. Both scores

presented good internal consistency in the present study (Cronbach’s αs = 0.89 and

0.91, respectively).

Emotional distress. The short Depression, Anxiety and Stress Scales

(DASS-21; Lovibond & Lovibond, 1995)is a questionnaire assessing emotional

distress. This measure was included to increase comparability of our findings with

those of previous studies which have included measures of emotional symptoms in

addition to or instead of voice-related distress, (e.g. Peters et al., 2012). The DASS-

21 requires participants to rate the extent to which they experienced 21 different

symptoms of anxiety, depression and stress in the previous week on a 4-point scale

(0 to 3). In this study, the DASS-21 presented excellent internal consistency

(Chronbach’s α = 0.96).

9

Procedure

After receiving NHS Ethics and R&D approval, prospective participants were

recruited in the Greater Manchester region between June 2013 and April 2014.

Participants were approached through (i) liaison with mental health services, mental

health charities and Hearing Voices Network groups, (ii) display of poster adverts on

University of Manchester premises, GP surgeries and public venues (e.g. libraries,

community notice boards), (iii) liaison with local Spiritualist churches (given

previous findings suggesting high prevalence of non-clinical voice-hearers in these

communities; e.g. Hill & Linden, 2012; Hill et al., 2012), (iv) adverts on social

network groups, specialist websites and through local patient and public

involvements initiatives. Testing took place in locations convenient to participants,

including University premises, hospitals, community mental health centres, Spiritual

churches and the participants’ homes. After providing informed consent, participants

were asked to complete the demographic and clinical history interview and the

battery of research measures. Participants were then fully debriefed, and received £7

reimbursement.

Data Analysis

Group differences were examined using bootstrap for independent sample

tests (with 1000 bootstrap samples) to test for differences between clinical and non-

clinical voice-hearers on voice characteristics and test for the hypothesized

differences on measures of goal interference and facilitation between clinical and

non-clinical voice-hearers. Non-parametric correlational analyses (Spearman’s rho)

were carried out to test whether goal interference scores were positively associated

10

with measures of distress (PSYRATS-AH distress, self-reported voice-related

distress, and DASS-21 scores), and negatively related to voice pleasantness; and

whether goal facilitation scores positively associated with voice-pleasantness but

negatively related to distress measures. Hierarchical regression analyses were

employed to examine whether goal interference/facilitation ratings significantly

predicted distress and voice pleasantness when controlling for participant group and

other characteristics of voices that might influence affective reactions to voices (i.e.

PSYRATS-AH measures of frequency/duration of voices, measures of negative

content and the amount of disruption to life functioning). All correlational analyses

were carried out both within the entire participant sample and separately within the

clinical and non-clinical groups in order to examine consistent and divergent patterns

of findings between groups. Furthermore, all regression analyses controlled for

participant grouping (by including the dichotomous predictor “group”; 0 = clinical, 1

= non-clinical) to ensure that the findings reported were unbiased by possible group-

differences (Fox, 2008). All statistical analyses were carried out using SPSS version

20.

Results

Differences Between Clinical and Non-clinical Participants

The results of group comparisons between clinical and non-clinical voice-

hearers are reported in Table 1. Clinical voice-hearers scored significantly higher on

the PSYRATS-AH measures of distress, voice frequency/duration, negative voice

content, and disruption to daily life caused by the voices. They also presented

significantly higher DASS-21 and self-reported voice-related distress scores. There

11

was a trend towards significance for the analysis of voice pleasantness, suggestive of

lower pleasantness scores amongst clinical voice-hearers. Consistent with our

hypotheses, clinical participants presented significantly higher interference scores,

and lower facilitation scores compared to non-clinical voice-hearers.

Goal Interference/Facilitation Scores Are Associated with Voice-Related

Distress and Pleasantness

Non-parametric correlational analyses (Spearman’s rho) for associations

between between goal interference/facilitation scores, distress measures (PSYRATS

distress, self-reported voice-related distress, and DASS-21 scores) and the self-

reported measures of voice pleasantness are reported in Table 2. To examine whether

these associations were consistent in both clinical and the non-clinical participants,

separate analyses were carried out within the two participant groups and the Fisher’s

procedure for comparing independent correlation coefficients (Fisher, 1921) was

used to identify any significantly difference in findings obtained from these two

subgroup analyses.

[Insert Table 2 approximately here]

Goal interference and facilitation ratings were significantly associated for the

whole sample (rs = -.72, p = .001) and within subgroup analyses (rs = -.56, p = .021

in clinical and rs = -.73, p = .001 in non-clinical participants). For the total sample,

the associations between interference scores and the distress measures (DASS-21

12

scores, PSYRATS distress, and self-reported voice-related distress ratings) were all

robust and significant. Conversely, significant inverse relationships were found

between goal facilitation and all distress measures. Analyses also revealed that

perceived pleasantness of voices was inversely related to interference ratings,

whereas a large positive association was observed between voice pleasantness and

goal facilitation scores. Subgroup analyses for the clinical and the non-clinical

groups are also displayed in Table 2. Fisher’s z test indicated that the association

between goal facilitation and voice pleasantness was significantly larger in the non-

clinical group compared to the clinical group (z = 3.60, p = .001). All remaining

analyses were non-significant (all zs < 1.96, all ps > .05), indicating that the effects

observed in the two groups were comparable in magnitude.

Two separate hierarchical multiple regression analyses were then used to test

whether goal interference and facilitation scores were significant predictors of: (i)

self-reported voice-related distress ratings when controlling for the effect of other

voice characteristics (frequency/duration, negative content and amount of disruption

to life) and the dichotomous predictor “group” (0 = clinical, 1 = non-clinical), and of

(ii) voice pleasantness when controlling for the same covariates. Interference and

facilitation ratings were included as separate predictors as their bivariate association,

although large, was not of a magnitude that may create spurious findings due to

perfect or quasi-perfect collinearity (r ≥ .80 or .90; Field 2009). For both models

there was no evidence of violations of the assumption of independence of errors

(Durbin-Watson test = 1.80 for both analyses), linearity, homoscedasticity and

normally distributed errors (examined through visual inspection of standardised

residuals). Inspection of collinearity diagnostics found no evidence of

multicollinearity amongst the predictors in the regression models (i.e. all Variance

13

Inflation Factor statistics were < 10; all Tollerance statistics were > .10; Field,

2009).

Results of regression analyses are displayed in Table 3. For voice-related

distress, with participant group and other voice-characteristic (frequency/duration,

negative content and disruption) included in the first step of the analysis, the adjusted

R2 was estimated at .85, indicating that the predictors explained 85% of the variance

in distress. In the first step of the analysis, negative voice content and disruption to

life caused by the voices significantly predicted distress. At step 2, the inclusion of

goal interference and facilitation scores significantly improved the prediction of

distress scores, explaining an additional 4% of the observed variance (R2 = .89,

significance of R2change p = .017). In the final model, goal interference, but not goal

facilitation, significantly predicted distress when controlling for the other variables

included in the model. The effects of negative content and voice disruption were also

significant, whereas frequency/duration of voices and group were not associated with

distress.

[Insert Table 3 approximately here]

A similar hierarchical regression model was estimated with voice

pleasantness as the dependent variable. After the inclusion group and other voice-

characteristic at step 1, the adjusted R2 was estimates at .39, indicating that the model

accounted for 39% of the observed variance in voice pleasantness. In the first step of

the analysis only negative voice content significantly predicted pleasantness scores.

The inclusion of goal facilitation and in reference ratings at step 2 largely improved

the prediction of pleasantness scores, and accounted for an additional 38% of the

14

observed variance (R2 = .77, significance of R2change p < .001). In the final step of

the analysis, goal facilitation was the only significant predictor of pleasantness.

Discussion

The current study examined whether affective reactions to voices (voice-

related distress and pleasantness) are linked to the extent to which hearing voices

interfere or facilitate the persons’ ability to achieve important personal goals. Our

results revealed that goal interference was significantly associated with higher

distress and lower perceived voice pleasantness, whereas goal facilitation was

significantly associated with higher pleasantness and lower voice-related distress.

Subsequent regression analyses revealed that goal interference specifically predicted

distress, whereas goal facilitation was specifically associated with perceived

pleasantness of voices. These analyses found that the extent to which voices

facilitate or hider personal goals are strongly predictive of affective reactions to

voices. Furthermore, these effects remained statistically significant event after

accounting for several covariates that might influence affective reactions to voices

(i.e. voice frequency/duration, amount and degree of negative content of voices,

amount of disruption to daily life caused by the voices, and the participants’ clinical

or non-clinical status).

The current study suggests that goals, and the extent to which individuals

experience voices as facilitating or interfering with personally meaningful goals,

may play a central role in determining whether voices are perceived as distressing or

as pleasant experiences. In contrast with previous investigations which often

15

assessed the impact of hearing voices using arbitrarily defined measures of

psychosocial and occupational functioning (e.g. Hall, 1995) or disruption to daily

life activities (Haddock et al., 1999), the present study employed a more

“idiographic” approach where the impact of voices was subjectively measured in

relation to self-generated and personally meaningful goals. This can be regarded as a

relative strength, in accordance with recommendations of employing

subjective/idiographic methods to understand the idiosyncratic nature of psychotic

experiences (e.g. Haddock et al., 2011; Pitt, Kilbride, Nothard, Welford, &

Morrison, 2007).

Our findings provide a novel perspective on processes that might contribute

to distress in people who hear voices. Within PCT, psychological distress is

experienced whenever control over personally meaningful and salient goals is

disrupted (e.g. Carey, 2006; Mansell et al., 2012; Powers, 1973, 2005). Findings of a

significant relationship between goal interference and distress are therefore in line

with the principles of PCT. Of notable interest, our results compare favourably with

previous investigations of other possible psychological determinants of distress in

voice hearing. For example, the effects on distress observed in the present

investigation were larger in terms of effect sizes than reports from previous studies

focusing on negative beliefs and appraisals of voices (e.g. Chadwick et al., 2000;

Morris, Garety, & Peters, 2014; Morrison et al., 2004; Peters et al., 2012),

maladaptive metacognitive beliefs (e.g. Brett, Johns, Peters, & McGuire, 2009; Hill

et al., 2012), insecure attachment styles (e.g. Berry, Wearden, Barrowclough,

Oakland, & Bradley, 2012; Robson & Mason, 2014) and, to a lesser extent, measures

of the perceived relationship between voice-hearers and their voices (e.g.

Birchwood, Meaden, Trower, Gilbert, & Plaistow, 2000; Vaughan & Fowler, 2004)..

16

Future studies could help to clarify the current findings by examining the possible

interrelationships between other predictors of distress (e.g. negative beliefs and

appraisals of voices) and the goal interference measures considered in the present

study, and the extent to which the constructs considered in this investigation are

independent from other predictors of voice-related distress. It is possible, for

example, that goal interference may represent a more “proximal” predictor of

distress that could mediate the observed association between the way specific voices

are appraised and subsequent distress. Conversely, a reverse association might be

also plausible, with specific positive and negative beliefs/appraisals developing as a

consequence of the perceived impact of voice hearing on personal goals.

The relationship between goal facilitation and voice pleasantness is also of

interest given the paucity of studies investigating the specific determinants of

pleasant hallucinatory experiences, and more generally, possible positive aspects of

psychotic experiences (e.g. Haddock et al., 2011). In both clinical and non-clinical

voice hearers a robust association was observed between the degree to which voices

were perceived as facilitating the ability to achieve personal goals and the extent to

which the voices were experienced as pleasant. This is consistent with previous

surveys suggesting that voices are experienced as useful and constructive events by a

substantial proportion of hearers (e.g. Jenner et al., 2008; Sanjuan et al., 2004).

Furthermore, our comparisons between clinical and non-clinical voice-hearers (in

addition to replicating well-documented differences on several voice-related

variables, e.g. distress, frequency/duration and negative content of voices; Daalman

et al., 2010; Daalman & Diederen, 2013) suggest marked differences in the extent to

which voices are perceived in relation to the ability to progress towards valued goals,

with clinical voice-hearers reporting significantly higher interference, and lower

17

levels of facilitation. The findings of our regression analyses could be interpreted as

evidence that the relative contribution of goal facilitation to pleasantness of voices is

stronger than that of goal interference on distress (the inclusion of step 2 predictors

increases the proportion of explained variance by 38% in models predicting goal

pleasantness, whereas only 4% additional variance is explained in the voice-related

distress analysis). However, it should be noted that these analyses controlled for

covariates that are particularly relevant to the prediction of distress (e.g. amount of

negative voice content assessed using the PSYRATS-AH) rather than pleasantness of

voices. The assessment of “positive” aspects of voices is neglected by the vast

majority of measures focusing on specific dimensions of voice-hearing experiences

like the PSYRATS. Future studies may consider replicating our findings by

controlling for positive voice dimensions that could complement well features of

voices assessed by the PSYRATS (for example “amount of positive voice content”).

A number of caveats should be considered when interpreting the current

findings. As with most psychological investigations relying on voluntary

participation and informed consent, the present study might be vulnerable to

selection bias. The sample consisted primarily of white, British volunteers, which

may limit the generalisability of our findings. The cross-sectional nature of our data

and the modest sample size are also caveats to consider when appraising these

findings. Future investigations considering larger samples of clinical and non-clinical

voice-hearers and/or employing alternative research designs might help to clarify the

nature of the relationships considered in the present study (e.g. experience sampling

methods to examine temporal dynamics amongst the variables considered, and/or

intervention studies considering treatment approaches specifically targeting the

extent to which voices interfere with valued goals). Although it can be argued that

18

self-report measures are unavoidable when investigating phenomena that are

exclusively accessible through introspection, it is recognised that these methods are

vulnerable to several biases (e.g. retrospective reporting, demand characteristics,

social desirability), and may lead to overinflated relationships between the constructs

considered. Several of our research measures were specifically developed to address

the research questions of this study (i.e. goal interference and facilitation ratings) or

adapted following service users consultation to increase their acceptability to the

target population (i.e. the modified Goal Task) which might benefit from further

psychometric evaluation in future studies. The results of this study should be

interpreted in the light of these limitations. Finally, the study did not include a

validated diagnostic tool to assign participants to the clinical and non-clinical groups

and only relied on participant’s self-reported diagnosis and service history. Although

this is not problematic for the symptom-specific approach employed in this instance,

the authors recognise that the lack of such instruments might have introduced bias in

the subgroup analyses reported. Future studies aiming to consider potential effects of

specific diagnoses, or investigate more precisely differences between clinical and

non-clinical voice-hearers may benefit from the use of more detailed diagnostic

assessments.

The study bears a number of clinical implications. The finding that voice-

related distress is closely associated to interference with important goals stresses the

importance of placing individual’s valued goals at the core of any psychological

intervention for distressing voices. The careful exploration of client’s goals, and the

extent to which experiences of voice hearing influence these goals, may allow for

more effective idiographic psychological formulations of clients presenting

difficulty, and the delivery of interventions that may be more acceptable to

19

individual clients. Rather than focusing on symptom reduction, interventions

focusing on promoting the client’s ability to progress towards valued life goals might

represent a meaningful treatment option for individuals with distressing voices. An

example of such approaches is Acceptance and Commitment Therapy interventions

for distressing voices (Shawyer, Thomas, Morris, & Farhall, 2013; Thomas, Morris,

Shawyer, & Farhall, 2013), which have increasingly been used with individuals with

distressing psychotic symptoms, with encouraging preliminary outcomes (Valmaggia

& Morris, 2010). Within PCT and other control theories, psychological distress is

assumed to arise whenever a person experiences loss of control over valued goals,

most commonly resulting from chronic conflict between important strivings and

goals (Carey, Mansell, Tai, & Turkington, 2014), and that long-term distress

resolution can only be achieved when information pertaining to higher-order goals is

accessed. Method of Levels therapy (MoL), a transdiagnostic cognitive therapy

based on PCT (Carey, 2006; Mansell, Carey & Tai, 2012), is a type of therapy,

which specifically aims to shift awareness to higher-level goals and perceptions to

promote long-term resolution of conflict. The application of MoL may therefore

prove beneficial to clients with distressing voices in terms of resolving the possible

distress associated with these experiences (Tai, 2009; Tai & Turkington, 2009).

20

References

Aleman, A., & Larøi, F. (2008). Hallucinations: The science of idiosyncratic perception. Washington, DC: American Psychological Association.

Ashby, R. (1952). Design for a brain. New York: Whiley.Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure, process,

and content. Psychological Bulletin, 120(3), 338-375. doi: 10.1037/0033-2909.120.3.338

Beavan, V., Read, J., & Cartwright, C. (2011). The prevalence of voice-hearers in the general population: A literature review. Journal of Mental Health, 20(3), 281-292. doi: 10.3109/09638237.2011.562262

Bentall, R. P. (2003). Madness explained: Psychosis and the human nature. London: Penguin Books Ltd.

Berry, K., Wearden, A., Barrowclough, C., Oakland, L., & Bradley, J. (2012). An investigation of adult attachment and the nature of relationships with voices. British Journal of Clinical Psychology, 51(3), 280-291. doi: 10.1111/j.2044-8260.2011.02027.x

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(2), 337-344. doi: 10.1017/S0033291799001828

Brett, C. M. C., Johns, L. C., Peters, E. P., & McGuire, P. K. (2009). The role of metacognitive beliefs in determining the impact of anomalous experiences: A comparison of help-seeking and non-help-seeking groups of people experiencing psychotic-like anomalies. Psychological Medicine, 39, 939-950.

Carey, T. A. (2006). Method of Levels: How to Do Psychotherapy without Getting in the Way. Hayward, CA: Living Control Systems Publishing.

Carey, T. A., Mansell, W., Tai, S. J., & Turkington, D. (2014). Conflicted control systems: the neural architecture of trauma. The Lancet Psychiatry, 1(4), 316-318. doi: 10.1016/s2215-0366(14)70306-2

Chadwick, P., Lees, S., & Birchwood, M. (2000). The revised Beliefs About Voices Questionnaire (BAVQ-R). The British Journal of Psychiatry, 177(3), 229-232. doi: 10.1192/bjp.177.3.229

Daalman, K., Boks, M. P. M., Diederen, K. M., de Weijer, A., D., Blom, D. J., Kahn, R. S., & Sommer, I. (2010). The same or different? A phenomenological comparison of auditory verbal hallucinations in healthy and psychotic individuals. Journal of Clinical Psychiatry, 72(3), 320-325. doi: 10.4088/JCP.09m05797yel

Daalman, K., & Diederen, K. M. (2013). A final common pathway to hearing voices: examining differences and similarities in clinical and non-clinical individuals. Psychosis, 5(3), 236-246. doi: 10.1080/17522439.2013.796402

Dickson, J., & MacLeod, A. (2004). Approach and avoidance goals and plans: Their relationship to anxiety and depression. Cognitive Therapy and Research, 28(3), 415-432. doi: 10.1023/B:COTR.0000031809.20488.ee

Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95(3), 542-575. doi: 10.1037/0033-2909.95.3.542

Elliot, A. J., & Sheldon, K. M. (1998). Avoidance personal goals and the personality–illness relationship. Journal of Personality and Social Psychology, 75(5), 1282-1299. doi: 10.1037/0022-3514.75.5.1282

Elliot, A. J., Sheldon, K. M., & Church, M. A. (1997). Avoidance personal goals and subjective well-being. Personality and Social Psychology Bulletin, 23(9), 915-927. doi: 10.1177/0146167297239001

21

Emmons, R. A. (1986). Personal strivings: An approach to personality and subjective well-being. Journal of Personality and Social Psychology, 51(5), 1058-1068. doi: 10.1037/0022-3514.51.5.1058

Field, A. (2009). Discovering statistics using SPSS (3rd ed.). London: Sage.First, M. B., Spitzer, R. L., M., G., & Williams, J. B. W. (2002). Structured Clinical Interview for

DSM-IV-TR Axis I Disorders, Research Version, Patient Edition. (SCID-I/P). New York: Biometrics Research.

Fisher, R. A. (1921). On the `probable error' of a coefficient of correlation deduced from a small sample. Metron, 1, 3-32.

Fox, J. (2008). Applied regression analysis and generalised linear modes (2nd ed.). London: Sage.

Haddock, G., McCarron, J., Tarrier, N., & Faragher, E. B. (1999). Scales to measure dimensions of hallucinations and delusions: The Psychotic Symptom Rating Scales (PSYRATS). Psychological Medicine, 29(4), 879-889. doi: 10.1017/s0033291799008661

Haddock, G., Wood, L., Watts, R., Dunn, G., Morrison, A. P., & Price, J. (2011). The Subjective Experiences of Psychosis Scale (SEPS): Psychometric evaluation of a scale to assess outcome in psychosis. Schizophrenia Research, 133(1–3), 244-249. doi: 10.1016/j.schres.2011.09.023

Hall, R. C. W. (1995). Global Assessment of Functioning: A modified scale. Psychosomatics, 36(3), 267-275. doi: 10.1016/S0033-3182(95)71666-8

Hill, K., & Linden, D. (2012). Hallucinations in non-clinical populations. In R. Jardri, A. Cachia, P. Thomas & D. Pins (Eds.), The Neuroscience of Hallucinations (pp. 21-42). London: Springer.

Hill, K., Varese, F., Jackson, M., & Linden, D. E. J. (2012). The relationship between metacognitive beliefs, auditory hallucinations, and hallucination-related distress in clinical and non-clinical voice-hearers. British Journal of Clinical Psychology, 51(4), 434-447. doi: 10.1111/j.2044-8260.2012.02039.x

Jenner, J. A., Rutten, S., Beuckens, J., Boonstra, N., & Sytema, S. (2008). Positive and useful auditory vocal hallucinations: Prevalence, characteristics, attributions, and implications for treatment. Acta Psychiatrica Scandinavica, 118(3), 238-245. doi: 10.1111/j.1600-0447.2008.01226.x

Johns, L. C., Nazroo, J. Y., Bebbington, P., & Kuipers, E. (2002). Occurrence of hallucinatory experiences in a community sample and ethnic variations. British Journal of Psychiatry, 180(1), 174-178. doi: 10.1192/bjp.180.2.174

Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour Research and Therapy, 33(3), 335-343.

Mansell, W., Carey, T. A., & Tai, S. (2012). A transdiagnostic approach to CBT using Method of Levels therapy. Hove, UK: Routledge.

Martin, L. L., Tesser, A., & McIntosh, W. D. (1993). Wanting but not having: The effects of unattained goals on thoughts and feelings. In D. M. Wegner & J. W. Pennebaker (Eds.), Handbook of mental control (pp. 552-572). Englewood Cliffs, NJ, US: Prentice-Hall, Inc.

McCarthy-Jones, S. (2012). Hearing voices: The histories, causes and meanings of auditory verbal hallucinations. Cambridge, UK: Cambridge University Press.

McIntosh, W. D., Harlow, T. F., & Martin, L. L. (1995). Linkers and nonlinkers: Goal beliefs as a moderator of the effects of everyday hassles on rumination, depression, and physical complaints. Journal of Applied Social Psychology, 25(14), 1231-1244. doi: 10.1111/j.1559-1816.1995.tb02616.x

22

Miller, L. J., O'Connor, E., & DiPasquale, T. (1993). Patients' attitudes toward hallucinations. The American Journal of Psychiatry, 150(4), 584-588.

Morris, E. M., Garety, P., & Peters, E. (2014). Psychological flexibility and nonjudgemental acceptance in voice hearers: Relationships with omnipotence and distress. Australian and New Zealand Journal of Psychiatry. doi: 10.1177/0004867414535671

Morrison, A. P., Nothard, S., Bowe, S. E., & Wells, A. (2004). Interpretations of voices in patients with hallucinations and non-patient controls: A comparison and predictors of distress in patients. Behaviour Research and Therapy, 42(11), 1315-1323. doi: 10.1016/j.brat.2003.08.009

Oishi, S., Diener, E., Suh, E., & Lucas, R. E. (1999). Value as a moderator in subjective well-being. Journal of Personality, 67(1), 157-184. doi: 10.1111/1467-6494.00051

Oorschot, M., Lataster, T., Thewissen, V., Bentall, R., Delespaul, P., & Myin-Germeys, I. (2012). Temporal dynamics of visual and auditory hallucinations in psychosis. Schizophrenia Research, 140(1-3), 77-82. doi: 10.1016/j.schres.2012.06.010

Peters, E. R., Williams, S. L., Cooke, M. A., & Kuipers, E. (2012). It's not what you hear, it's the way you think about it: Appraisals as determinants of affect and behaviour in voice hearers. Psychological Medicine, 42(7), 1507-1514. doi: 10.1017/S0033291711002650

Pitt, L., Kilbride, M., Nothard, S., Welford, M., & Morrison, A. P. (2007). Researching recovery from psychosis: A user-led project. Psychiatric Bulletin, 31(2), 55-60. doi: 10.1192/pb.bp.105.008532

Powers, W. T. (1973). Behaviour: The control of perception. New Canaan, CT: Benchmark Publications.

Powers, W. T. (2005). Behaviour: The control of perception (2nd ed.). New Canaan, CT: Benchmark Publications.

Powers, W. T., Clark, R. K., & McFarland, R. L. (1960a). A general feedback theory of human behaviour. Part I. Perceptual and Motor Skills, 11(1), 71-88. doi: 10.2466/pms.1960.11.1.71

Powers, W. T., Clark, R. K., & McFarland, R. L. (1960b). A general feedback theory of human behaviour. Part II. Perceptual and Motor Skills, 11(3), 309-323. doi: 10.2466/pms.1960.11.3.309

Robson, G., & Mason, O. (2014). Interpersonal processes and attachment in voice-hearers. Behavioural and Cognitive Psychotherapy. doi: doi:10.1017/S1352465814000125

Sanjuan, J., Gonzalez, J. C., Aguilar, E. J., Leal, C., & Van Os, J. (2004). Pleasurable auditory hallucinations. Acta Psychiatrica Scandinavica, 110(4), 273-278. doi: 10.1111/j.1600-0447.2004.00336.x

Shawyer, F., Thomas, N., Morris, E. M. J., & Farhall, J. (2013). Theory on voices. In E. M. J. Morris, L. C. Johns & J. E. Oliver (Eds.), Acceptance and Commitment Therapy and Mindfilness for Psychosis (pp. 12-32). Chichester, UK: Wiley-Blackwell.

Sommer, I. E. C., Daalman, K., Rietkerk, T., Diederen, K. M., Bakker, S., Wijkstra, J., & Boks, M. P. M. (2010). Healthy individuals with auditory verbal hallucinations; Who are they? Psychiatric assessment of a selected sample of 103 subjects. Schizophrenia Bulletin, 36(3), 633-641. doi: 10.1093/schbul/sbn130

Srull, T. K., & Wyer, R. S. (1986). The role of chronic and temporary goals in social information processing. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (pp. 503-549). New York: Guilford Press.

Steel, C., Garety, P. A., Freeman, D., Craig, E., Kuipers, E., Bebbington, P., . . . Dunn, G. (2007). The multidimensional measurement of the positive symptoms of psychosis.

23

International Journal of Methods in Psychiatric Research, 16(2), 88-96. doi: 10.1002/mpr.203

Tai, S. J. (2009). Using Perceptual Control Theory and the Method of Levels to work with people who experience psychosis. The Cognitive Behaviour Therapist, 2(03), 227-242. doi: 10.1017/S1754470X09990134

Tai, S. J., & Turkington, D. (2009). The evolution of cognitive behaviour therapy for scizophrenia: Current practice and recent developments. Schizophrenia Bulletin, 35(5), 865-873.

Thomas, N., Morris, E. M. J., Shawyer, F., & Farhall, J. (2013). Acceptance and Commitment Therapy for voices. In E. M. J. Morris, L. C. Johns & J. E. Oliver (Eds.), Acceptance and Commitment Therapy and Mindfulness for Psychosis (pp. 95-111). London: John Wiley & Sons.

Valmaggia, L. R., & Morris, M. (2010). Attention Training Technique and Acceptance and Commitment Therapy for distressing auditory hallucinations. In F. Larøi & A. Aleman (Eds.), Hallucinations: A guide to treatment and management (pp. 123-142). Oxford: Oxford University Press.

Varese, F., Morrison, A. P., Beck, R., Hefferman, S., Law, H., & Bentall, R. P. (under review). Experiential avoidance and appraisals of voices are specifically related to distress but not to frequency and duration of voices. Manuscript submitted for publication.

Varese, F., Morrison, A. P., Beck, R., Heffernan, S., Law, H., & Bentall, R. P. (2016). Experiential avoidance and appraisals of voices as predictors of voice-related distress. British Journal of Clinical Psychology, n/a-n/a. doi: 10.1111/bjc.12102

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(2), 143-153. doi: 10.1348/014466504323088024

Wiener, N. (1948). Cybernetics: Control and communication in the animal and the machine. Cambridge, MA: MIT Press.

24

Table 1: Demographic and clinical characteristics

Whole sample Clinical Non-clinical Group differencep

values(N = 40) (n = 22) (n = 18)

Demographic measures

Age 38.8 (14.5) 37.8 (12.3) 39.9 (16.9) 95% CI [-12.86, 7.52] .70

Gender 17 Males 10 Males 7 Males Fisher's exact test .80

Ethnicity 35 White 19 White 16 White Fisher's exact test .94

Clinical/voice-related measures

PSYRATS-AH frequency/duration 4.5 (2.2) 5.7 (2.1) 3.0 (1.3) 95% CI [0.86, 3.52] .007

PSYRATS-AH negative content 3.7 (2.7) 5.6 (2.3) 2.1 (2.4) 95% CI [1.98, 5.31] .001

PSYRATS-AH disruption 1.0 (1.2) 1.7 (1.2) 0.2 (0.4) 95% CI [1.02, 2.16] .002

PSYRATS-AH distress 3.3 (2.7) 4.7 (2.3) 1.9 (2.4) 95% CI [1.12, 4.46] .004

self-reported distress 4.1 (3.3) 5.9 (2.6) 2.2 (2.8) 95% CI [1.79, 5.75] .001

self-reported pleasantness 4.8 (3.9) 3.6 (3.7) 6.1 (3.7) 95% CI [-5.12, 0.17] .08

DASS-21 24.6 (16.9) 31.1 (17.1) 17.7 (13.7) 95% CI [3.46, 23.97] .02

Goals measures

Facilitation scores 4.2 (3.7) 2.5 (3.1) 5.9 (3.6) 95% CI [-5.70, -0.91] .01

Interference scores 3.8 (3.3) 5.7 (2.7) 1.8 (2.7) 95% CI [1.82, 5.77] .002

25

Table 2: Spearman’s rho between goal interference/facilitation scores and the distress and voice pleasantness measures

Whole sample (N = 40) Clinical (n = 22) Non-clinical (n = 18)

Facilitation Interference Facilitation Interferenc

e Facilitation Interference

DASS-21 total -.66 *** .81 *** -.60 ** .84 *** -.77 *** .71 **

PSYRATS distress -.70 *** .89 *** -.65 ** .91 *** -.47 * .79 **

Self-report distress -.71 *** .93 *** -.55 * .93 *** -.72 ** .80 ***

Self-report pleasantness .86 *** -.69 *** .69 ** -.60 ** .94 *** -.72 **

Note: * p < .05; ** p < .01; *** p < .001

26

Table 3: Summary of the hierarchical regression analyses

voice-related distress analysis voice pleasantness analysis

Predictor B SE β Predictor B SE β

Step 1 Step 1

Group -1.11 .68 -.17 Group 0.75 1.65 .01

Negative content .81 .11 .72 ** Negative content -1.02 .28 -.77 ***

Disruption 1.01 .30 .39 ** Disruption .13 .72 .04

Frequency/duration .10 .12 .07 Frequency/duration .11 .29 .06

Step 2 Step 2

Group -.99 .61 -.15 Group 1.51 1.03 .20

Negative content .50 .14 .45 ** Negative content -.34 .24 -.26

Disruption .75 .29 .27 * Disruption .51 .49 .16

Frequency/duration -.06 .11 -.04 Frequency/duration .06 .19 .03

Goal facilitation -.07 .08 -.08 Goal facilitation .78 .13 .76 ***

Goal interference .34 .14 .34 * Goal interference -.19 .15 -.15

Note. * p < .05, ** p < .01, *** p < .001

27

28