This article was downloaded by: [Laurentian University] On: 09 December 2014, At: 00:13 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Cognitive Neuropsychiatry Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/pcnp20

Auditory verbal hallucinations reflect stable auditory attention deficits: a prospective study ab

c

a

Else-Marie Løberg , Hugo A. Jørgensen , Rune A. Kroken & Erik Johnsen

ac

a

Division of Psychiatry, Haukeland University Hospital, Bergen, Norway b

Department of Clinical Psychology, University of Bergen, Bergen, Norway c

Department of Clinical Medicine, University of Bergen, Bergen, Norway Published online: 11 Nov 2014.

To cite this article: Else-Marie Løberg, Hugo A. Jørgensen, Rune A. Kroken & Erik Johnsen (2015) Auditory verbal hallucinations reflect stable auditory attention deficits: a prospective study, Cognitive Neuropsychiatry, 20:1, 81-94, DOI: 10.1080/13546805.2014.977857 To link to this article: http://dx.doi.org/10.1080/13546805.2014.977857

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Downloaded by [Laurentian University] at 00:13 09 December 2014

Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

Cognitive Neuropsychiatry, 2015 Vol. 20, No. 1, 81–94, http://dx.doi.org/10.1080/13546805.2014.977857

Auditory verbal hallucinations reflect stable auditory attention deficits: a prospective study Else-Marie Løberga,b*, Hugo A. Jørgensenc, Rune A. Krokena and Erik Johnsena,c

Downloaded by [Laurentian University] at 00:13 09 December 2014

a Division of Psychiatry, Haukeland University Hospital, Bergen, Norway; bDepartment of Clinical Psychology, University of Bergen, Bergen, Norway; cDepartment of Clinical Medicine, University of Bergen, Bergen, Norway

(Received 12 August 2013; accepted 10 October 2014) Introduction. Previous studies have shown that auditory verbal hallucinations (AVHs) in psychosis are associated with reduced verbal auditory attention. Whether this is an effect of ongoing AVH or reflects a more stable cognitive vulnerability also present after treating the AVH is unknown. The aim of this study was to follow patients with acute psychosis with and without AVH, and to test their auditory attention in a more stabilised clinical phase. Methods. Fifty patients (35 males and 15 females) were examined when admitted to an acute psychiatry ward and tested three months later with a dichotic listening test with attention instructions. The patients were divided into a frequent (n = 33) and nonfrequent (n = 17) AVH group based on their score on the Positive and Negative Syndrome Scale item hallucinatory behaviour (≥4 and ≤3, respectively) at baseline. Results. A significant interaction emerged between AVH group and attention instruction condition; the frequent AVH group failed to control their auditory attention as opposed to the non-frequent AVH group. Conclusions. Patients with frequent AVH in an acute psychotic state showed impaired auditory attention three months after their AVH had been treated, indicating a stable cognitive vulnerability factor for experiencing AVH. Keywords: psychoses; schizophrenia; positive symptoms; dichotic listening; attention

Auditory verbal hallucinations (AVHs) are frequent and distressing symptoms in psychosis. AVHs are perceptual experiences of voices or sounds as real and salient despite the absence of acoustic signals; current therapeutic strategies often aim to decrease the significance of and attentional focus on these experiences, suggesting the influence of cognitive abnormalities such as deviant auditory attention. In line with this, a relationship between AVH and abnormal auditory processing and attentional control has been reported (Hugdahl et al., 2008; Løberg et al., 2006). We do not know, however, whether these cognitive difficulties are an effect of ongoing AVH driving the attentional focus or whether auditory attention is still compromised after AVH has been treated, suggesting a state-like vulnerability for deviant processing and AVH. In this prospective study, we aimed to test the presence of stable auditory attention deficits in AVH prone individuals.

*Corresponding author. Email: [email protected] © 2014 Taylor & Francis

Downloaded by [Laurentian University] at 00:13 09 December 2014

82

E.-M. Løberg et al.

Behavioural paradigms for understanding AVH have often focused on the combination of auditory processing and attentional control, as seen in the dichotic listening paradigm (Hugdahl et al., 2008; Løberg et al., 2006). This simple and non-invasive task is based on an ear advantage (EA) for dichotically presented (two different sounds at the same time) and competing stimuli. The EA is generated by hemispheric dominance for certain auditory stimuli in normal healthy subjects. For verbal material, there is a right EA (REA) attributed to a bias for verbal processing in the left temporal lobe language areas in combination with stronger contralateral pathways and a suppression of ipsilateral pathways (Bryden, 1988; Hugdahl, 1995; Kimura, 1961). Thus, the right ear perceives verbal stimuli faster than the left ear. One of the most widely used dichotic listening paradigms is the consonant-vowels (CVs) syllables paradigm, used in more than 2000 published articles and with an EA effect that has repeatedly been shown to be both valid and reliable (Hugdahl, 2011). Early work with verbal dichotic listening tests found a reduced REA in schizophrenia (Bruder, 1995; Hugdahl, 2011; Kaufman & Trachenko, 1981). These effects were not consistent across different patients with schizophrenia groups, however, as more chronified or acute psychotic patients showed more deviant laterality; a relationship with positive symptoms was also suggested (Løberg et al., 2006; Løberg, Jørgensen, & Hugdahl, 2002, 2004). Later studies have shown that patients with ongoing AVH in particular have reduced brain asymmetry for verbal auditory material on dichotic listening tests (Bruder et al., 1995; Green, Hugdahl, & Mitchell, 1994; Levitan, Ward, & Catts, 1999; Løberg, Hugdahl, & Green, 1999; Løberg et al., 2006). Furthermore, studies comparing this effect of AVH with the effect of negative symptoms on dichotic listening laterality did not find a relationship with negative symptoms, signifying a specific effect of AVH and not symptom severity in general (Hugdahl et al., 2012, 2013). Ocklenburg and colleagues performed two meta-analyses on reduced lateralisation in schizophrenia supporting this conclusion (Ocklenburg, Westerhausen, Hirnstein, & Hugdahl, 2013). In the first (n = 1407), support for a general reduced laterality in the dichotic listening task for patients with schizophrenia as compared to healthy controls emerged, but due to a small effect size and substantial inter-study variability, the experience of auditory hallucinations was suggested as a possible mediating factor. The second meta-analysis (n = 407) found that patients with schizophrenia with AVH showed a significantly reduced REA in the dichotic listening task when compared to non-hallucinating controls, and the effect size of this comparison was substantially larger than the one observed for the comparison of all patients with schizophrenia with healthy controls (Ocklenburg et al., 2013). As suggested, verbal content and competing sounds in the verbal dichotic paradigm may be particularly suitable for modelling the interfering auditory experience of AVH (Hugdahl et al., 2013), and most early studies utilised the language processing systems of the brain to explain their findings, e.g. attributing the anatomical basis for the REA effect to the left peri-Sylvian region including the superior temporal gyrus and planum temporale (Hugdahl et al., 2013). However, abnormal performance and/or laterality in relation to other dichotically presented stimulus in schizophrenia indicated the involvement of other processing systems. Dichotic listening tests with emotional content typically gave rise to a left EA (LEA) in healthy controls (Gadea, Espert, Salvador, & Marti-Bonmati, 2011), and emotional prosody (tone of voice) have also yielded aberrant laterality in schizophrenia and AVH (Alba-Ferrara, de Erausquin, Hirnstein, Weis, & Hausmann, 2013); no LEA for emotional prosody in patients with AVH as compared to

Downloaded by [Laurentian University] at 00:13 09 December 2014

Cognitive Neuropsychiatry

83

non-AVH and healthy controls emerged suggesting deviant lateralisation for emotional stimuli (Alba-Ferrara et al., 2013). In addition, types of stimuli, e.g. words versus syllables, may actually influence the magnitude of the EA and group differences in relation to schizophrenia and controls (Smucny, Wylie, & Tregellas, 2012). Thus, to understand the EA effect in dichotic listening, it is necessary to expand the basic neuroanatomical model beyond the language processing systems of the brain. Together with an increased understanding of the influence of dynamic processes like verbal priming, training and context on the EA (Hiscock & Kinsbourne, 2011), attentional models for the dichotic listening effect have emerged. Hugdahl and colleagues have been developing and testing such a model in several studies by manipulating attentional demand on the dichotic task (Hugdahl, 2011; Hugdahl & Andersson, 1986; Løberg et al., 1999). In this model, the stimuli-based laterality effect in the original noattention demanding condition (no-instruction condition) was mainly attributed to bottom-up cognitive possessing. By adding attentional instruction and/or distractor conditions to the dichotic listening paradigm, it was possible to increase the attention load and the involvement of executive control. For instance, instructions about which ear the subjects should listen to tested the ability to override the stimulus-driven bottom-up processing by top-down attentional efforts, influencing the EA accordingly (see also Løberg et al., 1999, 2002). Thus, the ability to change the EA in the attention instruction conditions indicated the ability for controlled attention. Furthermore, it has been suggested that overriding the typical EA by attentional control, e.g. creating a LEA for typical REA stimuli, may have a particular loading on attention focus and cognitive control interference related to the tendency to experience and being interrupted by AVH (Hugdahl et al., 2013). Also for emotional stimuli, inefficient top-down attentional control has been shown in patients with schizophrenia and AVH by showing more sensitivity to (and less laterality for) interference effects from emotional prosody when attending to dichotically presented voices (Alba-Ferrara et al., 2013). Possibly ongoing AVH and AVH vulnerability influenced these two conditions of the dichotic listening test differently. A previous study examined whether the different aspects of dichotic listening abnormalities were state or trait markers for AVH, by comparing the effect of ongoing AVH versus remitted AVH with healthy controls (Løberg et al., 2004). Decreased laterality in the no-instruction condition was only observed in the ongoing AVH patients (Løberg et al., 2004), suggesting a state phenomenon. In contrast, problems with changing the EA in the forced attention conditions by means of attentional control was seen in both patients with ongoing AVH and only a history of AVH (Løberg et al., 2004), suggesting a trait phenomenon. This previous study was, however, weakened by a cross-sectional design and a small sample of only 26 patients (Løberg et al., 2004). Thus a conclusion on whether AVH reflected stable auditory attention deficits, and not only state markers related to the experience of AVH as such, could not be reached. A larger sample and a prospective design are needed to better examine this issue. Therefore, the present study tested for trait or vulnerability markers that were present after the symptoms had been stabilised. Furthermore, to capture the full variability of AVH and to ensure valid AVH grouping, the acute psychotic phase was chosen as a baseline. More stabilised phases of psychosis can easily camouflage variability in AVH vulnerability. Thus, the aim of the present study was to follow patients admitted during a state of acute psychosis with and without frequent AVH, and to test their dichotic listening performance during a more stabilised clinical phase three months later. It was expected that the functional asymmetry in the no-instruction condition would be related

84

E.-M. Løberg et al.

to the state of AVH, and not present at follow-up in either of the two groups. Furthermore, it was expected that the ability to dynamically change the EA as a result of instruction in the attention conditions, reflecting auditory attention abilities, would be impaired in the frequent AVH group as compared with the non-frequent AVH group at follow-up.

Downloaded by [Laurentian University] at 00:13 09 December 2014

Materials and methods Participants Fifty patients (35 males and 15 females) in an active psychosis phase were included. Active psychosis was defined as a score of at least 4 on one or more of the items: Delusions, Hallucinatory Behaviour, Grandiosity, Suspiciousness/Persecution or Unusual Thought Content from the Positive and Negative Syndrome Scale (PANSS, Structured Clinical Interview for the PANSS [SCI-PANSS]; Kay, Opler, & Lindenmayer, 1989). The patient group was split into two AVH subgroups based on their score on the PANSS item hallucinatory behaviour (auditory verbal content for all patients) using the cut-off scores for symptom severity proposed by Morrison and colleagues (2004): a frequent AVH group with a score of 4 or higher (n = 33; 23 males, 10 females) and non-frequent AVH group with a score of 3 or lower (n = 17; 12 males, 5 females). For the frequent AVH group, four (12.12%) patients were left-handed and for the non-frequent AVH group, two (11.76%) patients were left-handed. The patients’ symptom profiles were examined at admission in the acute psychiatric ward and were tested with a dichotic listening test three months later. See Table 1 for means, standard deviations (SDs) and group comparisons for the demographic, clinical and cognitive data. Diagnoses following the ICD-10 (WHO, 1992) were determined by the hospital’s psychiatrists or specialists in clinical psychology at baseline. For the frequent and nonfrequent AVH groups, the distribution of the diagnoses was as follows: acute and transient psychotic disorders (F23): 27.27%, 35.29%; schizophrenia (F20): 24.24%, 11.76%; drug-induced psychosis (F12, F19): 18.18%, 17.65%; persistent delusional disorders (F22): 12.12%, 11.76%; affective psychosis (F31, F33): 6.06%, 17.65%; and non-organic psychotic disorders (F28, F29): 6.06%, 5.88%, respectively. Two patients in the frequent AVH group had missing data. Patients with drug-induced psychoses were included only when the condition did not resolve within a few days and when antipsychotic drug therapy was indicated. The present study was part of a larger naturalistic, prospective and randomised study on the effectiveness of olanzapine, quetiapine, risperidone and ziprasidone (the Bergen Psychosis Project; Johnsen, Kroken, Wentzel-Larsen, & Jørgensen, 2010). The study was approved by the Regional Committee for Medical Research Ethics and the Norwegian Social Science Data Services. Patients were excluded from the study due to the following: a clinically significant neurological disease, a history of head injury, the presence of manic psychosis or for other behavioural or mental reasons related to the state of illness were unable to cooperate with assessments, did not understand spoken Norwegian, had hearing impairments or were unable to use oral medication. Hearing impairments were defined as failing to detect 30 dB at 500, 1000, 2000 or 4000 Hz, or having a right-left ear difference threshold of more than 10 dB. Patients medicated with clozapine on admittance were also excluded as clozapine is predominantly used in treatment resistant schizophrenia when other antipsychotics have failed. Accordingly these patients were not eligible for randomisation to the antipsychotic medication under investigation.

Cognitive Neuropsychiatry

85

Table 1. Mean (SD) demographic, clinical and cognitive data by group.

Downloaded by [Laurentian University] at 00:13 09 December 2014

Item Age (years) Education (years) Baseline PANSS delusions PANSS hallucinatory behaviour PANSS positive subscale PANSS negative subscale PANSS gen. psychopath. subscale PANSS total scores GAF function scores Verbal abilities (RBANS) Visuospatial abilities (RBANS) Learning (RBANS) Memory (RBANS) Attention (RBANS) 3 months Medication (DDD) PANSS delusions PANSS hallucinatory behaviour PANSS positive subscale PANSS negative subscale PANSS gen. psychopath. subscale PANSS total scores GAF function scores Verbal abilities (composite score) Visuospatial abilities (composite score) Learning (composite score) Memory (composite score) Attention (composite score) DL non-forced condition right ear DL non-forced condition left ear DL non-forced condition total DL non-forced condition laterality index DL forced-right condition right ear DL forced-right condition left ear DL forced-right condition total DL forced-right condition laterality index DL forced-left condition right ear DL forced-left condition left ear DL forced-left condition total DL forced-left condition laterality index

Frequent AVH

Nonfrequent AVH

t

p

−0.59 1.49

.56 .14

33.61 (14.47) 12.39 (3.13)

31.25 (10.75) 13.71 (2.57)

4.61 4.69 20.52 17.52 36.61 74.65 30.66 38.28 42.14 36.93 38.50 30.50

(0.88) (0.64) (4.63) (8.01) (7.57) (16.64) (7.24) (7.71) (13.00) (10.10) (10.72) (8.33)

4.12 1.71 18.18 17.94 33.24 69.88 30.18 41.06 49.56 37.38 39.50 32.50

(1.11) (0.77) (4.11) (6.88) (6.54) (12.99) (5.49) (8.80) (12.65) (9.82) (14.18) (10.19)

196.50* 0.00* −1.74 0.18 −1.55 −1.02 −0.24 1.10 1.84 0.14 0.26 0.71

1.07 2.19 2.00 10.47 13.59 24.25 48.31 51.31 44.05 43.72 43.71 43.56 41.36 14.30 12.45 26.76 6.79 16.21 11.15 27.36 18.59 13.21 13.03 26.24 0.48

(0.51) (1.47) (1.46) (3.71) (6.43) (7.38) (14.61) (18.34) (7.92) (13.49) (5.90) (6.79) (8.35) (5.17) (4.80) (3.54) (33.40) (5.01) (5.05) (2.25) (36.15) (4.72) (4.57) (2.48) (34.28)

1.03 2.29 1.18 10.41 11.59 24.06 46.06 46.12 47.34 48.73 46.58 47.66 43.66 14.29 12.12 26.41 6.86 18.71 8.59 27.29 36.44 12.00 15.71 27.71 −14.33

(0.67) (1.31) (0.73) (3.28) (4.74) (5.86) (11.19) (15.17) (7.90) (8.93) (7.01) (7.22) (6.36) (5.64) (4.73) (2.48) (38.78) (5.81) (5.23) (2.02) (39.75) (7.47) (7.02) (1.83) (52.08)

0.22 .83 253.00* .70 187.00* .08 −0.05 .96 −1.13 .26 −0.09 .93 −0.55 .58 −1.00 .32 1.39 .17 1.37 .18 1.52 .14 1.97 .06 0.99 .33 −0.01 1.00 −0.24 .81 −0.36 .72 0.01 .99 1.58 .12 −1.68 .10 −0.11 .92 1.60 .12 −0.70 .49 1.63 .11 2.15 .04** −1.21 .23

.15 .00** .09 .85 .13 .31 .81 .28 .07 .89 .80 .48

Note: PANSS = The Positive and Negative Syndrome Scale for Schizophrenia. GAF = Global Assessment of Functioning Scale. RBANS = The Repeatable Battery for the Assessment of Neuropsychological Status. DDD = the assumed average maintenance dose per day for a drug used for its main indication in adults (WHO). Verbal abilities composite score = WAIS-III Similarities, Vocabulary; COWAT letters/animals. Visuospatial abilities composite score = WAIS-III Digit Symbol coding, Block design; ROCF Copy. Attention composite score = WAIS-III Digit Span, Digit Vigilance test, CalCAP Choice Reaction Time/Sequential Reaction Time, Trail Making test B, Stroop colour/word conflict. Learning composite score = CVLT-II Immediate Recall, WMS-R Logical Memory, WAIS-III Digit Span. Memory composite score = CVLT-II Delayed Recall, ROCF Delayed recall. DL = dichotic listening. Total = right ear score + left ear score. Laterality index = [(right ear score  left ear score)/(right ear score + left ear score)] × 100. *For the ordinal and non-normally distributed data from single PANSS items, Mann–Whitney U tests were used. **p < .05.

86

E.-M. Løberg et al.

Downloaded by [Laurentian University] at 00:13 09 December 2014

In addition, data on a control group from a previous study (Løberg et al., 2004) were included. This group included 21 healthy control subjects (11 males and 10 females), recruited through a government employment agency to match the patient group for education (mean year = 13.57, SD = 2.68) and age (mean year = 27.33, SD = 10.06). Exclusion criteria were a history of substance abuse, neurological disease, head injury or hearing impairments. The control subjects received about 24 EUR (200 NOK January 2014) for participating. The dichotic listening test The dichotic listening test (Hugdahl, 1995) was administered under three instruction conditions: a no-instruction condition without attention instructions, a forced-right instruction condition in which subjects were instructed to attend to stimuli presented in the right ear and a forced-left instruction condition in which subjects were instructed to attend to stimuli in the left ear. The no-instruction condition was given first as a baseline condition, and the order of the two forced attention instruction conditions was counterbalanced across subjects to avoid systematic practice effects. The dichotic stimuli consisted of the six stop-consonants (/b/, /d/, /g/, /t/ and /k/) with vowel /a/ to form the following CV syllables: /ba/, /da/, /ga/, /ta/ and /ka/. A male (actor) voice read the recorded syllables. The syllables were paired with each other for all possible combinations, yielding 36 dichotic pairs and six trials in which the same CV were presented to both ears (homonymic condition). The stimuli were prepared on a computer using a standard sound editing programme designed so that the CVs were presented to each ear simultaneously. Each CV syllable lasted approximately 350–450 ms and the inter-trial interval was approximately 4 s. All subjects were given five practice trials to ensure that they understood the nature of the task, and had a sheet in front of them that listed all six possible responses vertically. They were instructed to report the syllable they heard best on each trial, emphasising that only one answer should be given for each trial. During the forced-instruction conditions, an arrow was placed in front of the subjects to remind them which ear to attend to. Correct responses were scored separately for the right and left ear for all three instruction conditions, in addition to a total performance score created by adding right ear scores and left ear scores for each condition. A laterality index score was calculated according to the following formula: [(Right ear score) − (Left ear score)]/(Right ear score + Left ear score) × 100. A changing laterality index as a result of attention instructions signified the ability to focus auditory attention.

Results Demographic, clinical and cognitive group differences The data confirmed that the majority of patients in the frequent AVH group were in a stable clinical phase at follow-up three months later. Only one patient in the non-frequent and seven patients in the frequent AVH group had clinically significant AVH, defined as a score of 4 or more on the PANSS hallucinatory behaviour item, at this time point. This group difference was not significant, as examined by Fisher’s exact test ( p = .24). In addition, to further test for a potential group effect as a result of still present AVH at follow-up (comparing AVH and no AVH groups at follow-up for differences in relation to dichotic listening performance), one-way analysis of variances were performed for all the

Downloaded by [Laurentian University] at 00:13 09 December 2014

Cognitive Neuropsychiatry

87

dichotic listening performance variables and indices in all instruction conditions; for right and left ear correct reports, total correct report (right ear + left ear), and laterality indices (Jäncke & Shah, 2002, p. 113), in the two forced-instruction condition and the nonforced-instruction condition emerged (p = ns). Differences between the two patient groups for the cognitive variables, the demographic and clinical data at baseline and at three months from Table 1 were examined using t-tests, except the group differences for the ordinal and non-normally distributed data from single PANSS items were examined using the Mann–Whitney U test two-sample rank-sum test. No significant group differences emerged, except for the forced-left total performance, t(48) = 2.15, p = .04, and as expected the baseline PANSS auditory hallucinations item, U = 0.00, Z = −5.70, p < .001. See Table 1 for details. To further test for potential differences between the groups, the change in the same clinical and cognitive variables from baseline to follow-up was examined using repeated measures two-way 2 (Time: baseline, three months) × 2 (Group: frequent AVH, nonfrequent AVH) ANOVAs. Again, no significant effects of group emerged (p = ns for all comparisons). Differences between the two patient groups for gender, handedness and the distribution of the diagnostic categories were examined using Fisher’s exact test, and no significant group differences emerged (p = ns for all comparisons). Differences between the two patients groups and the control group for age and education were examined using one-way ANOVAs, followed by post-hoc Unequal N HSD (honest significant difference) tests, and for gender using Fisher’s exact test for all combinations of the means. Again, no significant group differences emerged (p = ns for all comparisons). Thus, there were no group differences in relation to any other clinical or cognitive variables other than the PANSS hallucination item. Furthermore, there were no significant correlations (Spearman rank order) between any of the clinical variables (excluding PANSS hallucinatory behaviour at baseline) at both time points and the dichotic listening performance variables and indices, except for a significant relationship between PANSS negative symptoms at follow-up and the forced right laterality index and right and left ear performance (rs(47) > 0.39, p < .01), as well as age and the forced-left performance (rs(48) > 0.29, p < .05). Negative symptoms were associated with less ability to focus as instructed in the forced-right condition, and age was associated with less ability to focus as instructed in the forced-left condition. Effect of AVH group at baseline on dichotic listening at follow-up The distribution of the PANSS hallucination item scores at baseline was bimodal (n = 8 for Score 1, n = 6 for Score 2, n = 3 for Score 3, n = 14 for Score 4, n = 16 for Score 5, n = 3 for Score 6), and did not reach normality using the Shapiro–Wilk W test (W = 0.86, p < .01). The distribution showed that data were best represented as two groups, with a split at PANSS hallucination item score = 4. In line with the hypothesis and the distribution of the data, group-based analyses were executed. A two-way 2 (Group: frequent AVH, non-frequent AVH) × 3 (Attention condition: no-instruction, forced-right, forced-left) repeated measures ANOVA was performed to test for effects on the dichotic listening laterality index, with Group as a between-subjects factor and Attention condition as within-subjects factor. There was a significant two-way interaction between Group and the Attention condition, F(2, 96) = 4.59, p < .01. The observed power of this interaction effect was .77, indicating a satisfactory level of power. Post-hoc Unequal N HSD tests were used to compare the means for the two patient

E.-M. Løberg et al.

subgroups for each of the attention instruction conditions. The non-frequent AVH group changed their EA (laterality index) as a result of instructions (p < .01 for all comparisons), as opposed to the frequent AVH group (p = ns for all comparisons). Only the non-frequent AVH group managed to significantly change their REA from the no-instruction condition to the forced-right condition and from the forced-right condition to the forced-left condition, indicating a group difference in relation to the pattern of performance; i.e. an interaction effect. The frequent AVH patients did not manage to change their attention as a result of instructions to the same degree. See Figure 1 for corresponding means and SD for the two patient groups. In addition, the means for the control group are included in the figure to facilitate understanding of the figure. Including the control group in the two-way repeated measures ANOVA did not change the results, (F(4, 136) = 5.98, p < .01), or post-hoc tests (p < .05). To control for the potential effect of ongoing AVH in the eight patients still having clinically significant AVH at three months, the two-way repeated measures ANOVA was repeated excluding these patients. The results remained significant, (F(2, 78) = 3.18, p < .05), and also for the post-hoc tests (p < .03). To control for the potential effect of lefthanded hand preference, the same two-way repeated measures ANOVA was repeated excluding the six left-handers. Again, the results remained the same, (F(2, 86) = 3.16, p < .05), and for the post-hoc tests (p < .05). These analyses must be interpreted with caution, however, due to decreased power. To further control for the effects of potential confounders, PANSS negative symptoms at follow-up were included as a covariate, since this variable influenced the dichotic listening results. Again, the results remained significant, (F(2, 90) = 4.35, p < .02), and also for the post-hoc tests (p < .02).

60

Laterality index score (RE – LE)/(RE + LE) 100

Downloaded by [Laurentian University] at 00:13 09 December 2014

88

40

20

0 Non-frequent AVH group Frequent AVH group Control group

–20

–40

No-instruction cond.

Forced-right cond.

Forced-left cond.

Figure 1. Laterality index scores and 95% confidence intervals in the no-instruction, forced-right and forced-left condition by group.

Downloaded by [Laurentian University] at 00:13 09 December 2014

Cognitive Neuropsychiatry

89

Discussion Frequent AVH in an active psychotic state is related to auditory attention deficits three months later. This was seen as a failure in the frequent AVH group to change their EA as a result of instructions, as compared to the non-frequent AVH group, reflecting the inability to dynamically control and modify their auditory attention. As expected, the two AVH groups showed the same level of functional asymmetry, however, seen as a REA in the no-instruction condition. The deficit auditory attention after three months could not be explained by ongoing AVH, since the results remained the same after excluding the few patients who still had AVH. Therefore, the present findings support a stable deficit in patients with psychosis who are prone to AVH. Furthermore, the relationship between AVH and the dichotic listening test emerged in a relatively heterogeneous group of patient with psychosis, supporting the notion of AVH being a more dynamic crossdiagnosis phenomenon. The stable deficit seems more consistent with a trait, and not state, marker replicating a previous cross-sectional finding in a smaller sample (Løberg et al., 2004). In addition, findings of anatomical brain changes related to AVH have usually been interpreted as trait markers (Barta, Pearlson, Powers, Richards, & Tune, 1990; Flaum et al., 1995; Levitan et al., 1999). Two recent meta-analyses concluded the following structural correlates to the severity of AVH: left insula and right superior temporal gyrus (Palaniyappan, Balain, Radua, & Liddle, 2012) and bilateral superior temporal gyri (Modinos et al., 2013), respectively, but it was not possible to differentiate between state and trait effects in these studies. Also functional brain studies have focused on ongoing AVH, which may be related to more basic abnormal asymmetry, and it has been proposed that ongoing AVH occupies the left-sided language processing system (Løberg et al., 2004), in line with most studies on the experience of AVH (Kompus, Westerhausen, & Hugdahl, 2011). The picture is not clear-cut, however, as one study for instance focused on right-sided language areas of the brain during AVH (Sommer et al., 2008). Kompus and colleagues performed two meta-analyses of functional neuroimaging studies in patients with schizophrenia experiencing AVH focusing on both activation during and in the absence of auditory stimulation. Increased activation in the left primary auditory cortex, but also in the right rostral prefrontal cortex during the experience of hallucinations without external auditory stimulation, emerged for this patient group (Kompus et al., 2011). Another meta-analysis concluded that the experience of AVH was associated with increased activity in fronto-temporal areas and within structures in the medial temporal lobe, suggesting the involvement of distributed network involvement in speech production, generation and verbal memory for example (Jardri, Pouchet, Pins, & Thomas, 2011). The experience of actual AVH may activate different brain regions than those implicated in the vulnerability for AVH however, and to our knowledge, only one metaanalysis focused on the brain correlates with AVH from a “trait” point of view defined in this study as comparing periods of presence and absence of AVHs within a subject. The authors suggested that the AVH state was more related to speech production areas, and trait was related more related to auditory and speech perception areas of the brain (Kuhn & Gallinat, 2012). Corresponding to the “dual deficit” (see Løberg et al., 1999) understanding of the dichotic listening test, Hugdahl and colleagues put forward a two-step perceptual model to explain the psychotic experience of AVH (Hugdahl, Løberg, & Nygard, 2009). The tendency to actually hear the AVH originated from abnormal activation of the lateralised

Downloaded by [Laurentian University] at 00:13 09 December 2014

90

E.-M. Løberg et al.

speech perception areas in the left temporal lobe, and this abnormal activation influenced the lateralisation of verbal processing, whereas the psychotic interpretation and misattribution of the AVH as actual real life verbal stimuli would be dependent on top-down cognitive control. The model would predict both laterality and attention control deficits during the psychotic, ongoing AVH phase, consistent with previous dichotic listening studies (Bruder et al., 1995; Collinson, Mackay, O, James, & Crow, 2009; Green et al., 1994; Levitan et al., 1999; Løberg et al., 2004). In the absence of AVH in AVH prone individuals, the vulnerability for psychotic interpretations may still be present as a trait marker, but the laterality deficit was no longer present, as seen in the present study. In addition, in the Hugdahl model, the deficit top-down cognitive control may be conceptualised as executive functioning, as seen especially in the forced-left condition, due to the conflict between the bottom-up processing (stimulus-driven REA) and the topdown attentional control instruction (attend left). In line with this, group differences in the present study emerged in the forced-left total performance, suggesting that this was an especially challenging and attention demanding condition for the AVH prone patients. We did not test the patients with dichotic listening in the acute phase, but inferred from previous studies and the present results that the observed deficits were also present in this phase. It seems unlikely that the auditory attention difficulties would be present in the stable phase, and not the acute psychotic phase, but it cannot be completely ruled out by the present study as a period of frequent AVH may have influenced the results by changing auditory attention capacity via processes related to neuroplasticity for example, or this particular patient group had non-deficit dichotic listening performance in their acute phase in contrast to other patients with schizophrenia and active psychosis. Furthermore, the finding of a symptom-related vulnerability seems specific to AVH, since there were no relationships between positive symptoms or negative symptoms at baseline and the dichotic listening performance. There was a relationship between negative symptoms at follow-up and the dichotic listening performance however, but since there were no group differences in relation to negative symptoms, and negative symptoms did not influence the effect of AVH on the dichotic listening performance, the effect of negative symptoms on dichotic listening performance was interpreted as an independent effect. The relationship between age and forced-left performance was interpreted the same way. The lack of a relationship between cognition and most clinical symptoms was consistent with the typical finding in psychosis and schizophrenia (see for instance a systematic review by Dominguez Mde, Viechtbauer, Simons, van Os, & Krabbendam, 2009). The lack of an overall group difference in relation to performance supported that the results were not due to the frequent AVH group not voluntarily following the instructions or understanding the test, but that they specifically did not manage to control their attention in line with the instructions in the forced attention conditions. The main finding was an interaction effect revealing a different pattern of attentional change as a result of instructions between the two groups. Hence, except for the forcedleft total performance, the differences between the groups were not with respect to performance at each time point, but reflected different dynamics in how the two groups adjusted their attention to the instructions. The non-frequent AVH group managed to change their attentional focus and override the stimulus-based bottom-up processing by top-down executive control from each condition to another, whereas the frequent AVH group did not show a significant change in their dichotic listening performance in the three conditions. The forced-left total performance group difference further validated the

Downloaded by [Laurentian University] at 00:13 09 December 2014

Cognitive Neuropsychiatry

91

focus on attentional control to explain the findings, as this condition was suggested to be especially sensitive to attention control deficits (Hugdahl et al., 2009). The present dichotic listening results were found for nonsense syllables, and a previous study suggested that group differences in schizophrenia may be particularly large in relation to this type of stimuli (Smucny et al., 2012), and in addition, other tests combining attention and verbal material have been shown to be related to AVH (Gisselgard et al., 2014). The present study did not test the influence of type of stimuli, however, and it is not known at this point whether the current effect may be attributed to verbal stimuli, or whether other types of auditory stimuli would have yielded the same effects, as suggested by decreased laterality for emotional prosody as shown by AlbaFerrara and colleagues (Alba-Ferrara et al., 2013). Contrasting different types of auditory stimuli would be needed to test this, and could have implications for understanding the brain-based underpinnings behind the dichotic listening effect in AVH. Authors focusing on the verbal properties of their dichotic listening tasks have implicated the integrity of the left temporal lobe language areas in this test, further validated by brain imaging studies. During a dichotic listening test with verbal stimuli, activation predominantly in the superior temporal gyrus in the left temporal lobe was reported (Jäncke & Shah, 2002). In addition, reduced REA on this task was associated with structural changes in the temporal lobes, especially the speech-related areas like the left planum temporale and left peri-Sylvian areas (Hugdahl et al., 2009). Alternatively, the present effect could be attributed to deviant functioning of prefrontally mediated cognitive control (Hugdahl, 2009) and the breakdown of fronto-temporal functional connectivity (Løberg et al., 2004), suggesting the influence of more widespread cortical circuitry and attentional networks. Involvement of the prefrontal and parietal circuitries involved in cognitive control may be particularly relevant for the attentional aspects of the dichotic listening findings (Hugdahl, 2009). Interestingly, a recent functional brain imaging study found that a fronto-parietal network, including the pre-supplementary motor area, anterior cingulate cortex, inferior frontal junction, insula and inferior parietal lobe, interpreted by the authors as a top-down cognitive control network, was engaged in especially attention demanding conditions of the dichotic listening test (Falkenberg, Specht, & Westerhausen, 2011). Future studies with longer follow-ups are needed to fully understand the relationship between auditory attention control and AVH. In addition, studies on patients in prepsychotic phases and how this cognitive deficit relates to changes in the frequency and interpretation of AVH would be of value. A neurocognitive understanding of AVH is of clinical importance and has psycho-educational implications. Attributing AVH to auditory attention vulnerability can make these alternative perceptual experiences less anxiety provoking, and thus reduce the need for delusional and psychotic explanations of these often frightening experiences. Acknowledgements The authors wish to thank psychiatric research nurses Ingvild Helle and Marianne Langeland at the Research Department, Division of Psychiatry, Haukeland University Hospital who assisted with data collection. We also wish to thank the patients and clinical staff at Division of Psychiatry, Haukeland University Hospital for their participation in this study and their support.

Funding This work was supported by the Research Council of Norway and Haukeland University Hospital.

92

E.-M. Løberg et al.

Downloaded by [Laurentian University] at 00:13 09 December 2014

References Alba-Ferrara, L., de Erausquin, G. A., Hirnstein, M., Weis, S., & Hausmann, M. (2013). Emotional prosody modulates attention in schizophrenia patients with hallucinations. Frontiers in Human Neuroscience, 7, 59. doi:10.3389/fnhum.2013.00059 Barta, P. E., Pearlson, G. D., Powers, R. E., Richards, S. S., & Tune, L. E. (1990). Auditory hallucinations and smaller superior temporal gyral volume in schizophrenia. American Journal of Psychiatry, 147, 1457–1462. Bruder, G. E. (1995). Cerebral laterality and psychopathology: Perceptual and event-related potential asymmetries in affective and schizophrenic disorders. In R. J. Davidson & K. Hugdahl (Eds.), Brain asymmetry (pp. 123–156). Cambridge: MIT Press. Bruder, G. E., Rabinoviz, E., Towey, J. P., Brown, A., Kaufman, C. A., Amador, W., … Gorman, J. M. (1995). Smaller right (left hemisphere) advantage for dichotic fused words in patients with schizophrenia. American Journal of Psychiatry, 152, 932–935. Bryden, M. P. (1988). An overview of the dichotic listening procedures and its relation to cerebral organisation. In K. Hugdahl (Ed.), Handbook of dichotic listening: Theory, methods and research (pp. 1–45). Chichester: Wiley. Collinson, S. L., Mackay, C. E., O, J., James, A. C., & Crow, T. J. (2009). Dichotic listening impairments in early onset schizophrenia are associated with reduced left temporal lobe volume. Schizophrenia Research, 112, 24–31. doi:10.1016/j.schres.2009.03.034 Dominguez Mde, G., Viechtbauer, W., Simons, C. J., van Os, J., & Krabbendam, L. (2009). Are psychotic psychopathology and neurocognition orthogonal? A systematic review of their associations. Psychological Bulletin, 135, 157–171. doi:10.1037/a0014415 Falkenberg, L. E., Specht, K., & Westerhausen, R. (2011). Attention and cognitive control networks assessed in a dichotic listening fMRI study. Brain and Cognition, 76, 276–285. doi:10.1016/j. bandc.2011.02.006 Flaum, M., O'Leary, D. S., Swayze, V. W., Miller, D. D., Arndt, S., & Andreasen, N. C. (1995). Symptom dimensions and brain morphology in schizophrenia and related psychotic disorders. Journal of Psychiatric Research, 29, 261–276. doi:10.1016/0022-3956(94)00046-T Gadea, M., Espert, R., Salvador, A., & Marti-Bonmati, L. (2011). The sad, the angry, and the asymmetrical brain: Dichotic listening studies of negative affect and depression. Brain and Cognition, 76, 294–299. doi:10.1016/j.bandc.2011.03.003 Gisselgard, J., Anda, L. G., Bronnick, K., Langeveld, J., Ten Velden Hegelstad, W., Joa, I., … Larsen, T. K. (2014). Verbal working memory deficits predict levels of auditory hallucination in first-episode psychosis. Schizophrenia Research, 153(1–3), 38–41. doi:10.1016/j. schres.2013.12.018 Green, M. F., Hugdahl, K., & Mitchell, S. (1994). Dichotic listening during auditory hallucinations in schizophrenia. American Journal of Psychiatry, 151, 357–362. Hiscock, M., & Kinsbourne, M. (2011). Attention and the right-ear advantage: What is the connection? Brain and Cognition, 76, 263–275. doi:10.1016/j.bandc.2011.03.016 Hugdahl, K. (1995). Dichotic listening: Probing temporal lobe functional integrity. In K. Davison & K. Hugdahl (Eds.), Brain asymmetry (pp. 123–156). Cambridge: American Psychiatric Press. Hugdahl, K. (2009). “Hearing voices”: Auditory hallucinations as failure of top-down control of bottom-up perceptual processes. Scandinavian Journal of Psychology, 50, 553–560. doi:10.1111/ j.1467-9450.2009.00775.x Hugdahl, K. (2011). Fifty years of dichotic listening research – Still going and going and …. Brain and Cognition, 76, 211–213. doi:10.1016/j.bandc.2011.03.006 Hugdahl, K., & Andersson, L. (1986). The “forced-attention paradigm” in dichotic listening to CVsyllables: A comparison between adults and children. Cortex, 22, 417–432. doi:10.1016/S00109452(86)80005-3 Hugdahl, K., Løberg, E. M., Falkenberg, L. E., Johnsen, E., Kompus, K., Kroken, R. A., … Ozgoren, M. (2012). Auditory verbal hallucinations in schizophrenia as aberrant lateralized speech perception: Evidence from dichotic listening. Schizophrenia Research, 140, 59–64. doi:10.1016/j.schres.2012.06.019 Hugdahl, K., Løberg, E. M., Jørgensen, H. A., Lundervold, A., Lund, A., Green, M. F., & Rund, B. (2008). Left hemisphere lateralisation of auditory hallucinations in schizophrenia: A dichotic listening study. Cognitive Neuropsychiatry, 13, 166–179. doi:10.1080/13546800801906808

Downloaded by [Laurentian University] at 00:13 09 December 2014

Cognitive Neuropsychiatry

93

Hugdahl, K., Løberg, E. M., & Nygard, M. (2009). Left temporal lobe structural and functional abnormality underlying auditory hallucinations in schizophrenia. Frontiers in Neuroscience, 3, 34–45. doi:10.3389/neuro.01.001.2009 Hugdahl, K., Nygard, M., Falkenberg, L. E., Kompus, K., Westerhausen, R., Kroken, R., … Løberg, E. M. (2013). Failure of attention focus and cognitive control in schizophrenia patients with auditory verbal hallucinations: Evidence from dichotic listening. Schizophrenia Research, 147, 301–309. doi:10.1016/j.schres.2013.04.005 Jäncke, L., & Shah, N. J. (2002). Does dichotic listening probe temporal lobe functions? Neurology, 58, 736–743. Jardri, R., Pouchet, A., Pins, D., & Thomas, P. (2011). Cortical activations during auditory verbal hallucinations in schizophrenia: A coordinate-based meta-analysis. The American Journal of Psychiatry, 168, 73–81. doi:10.1176/appi.ajp.2010.09101522 Johnsen, E., Kroken, R. A., Wentzel-Larsen, T., & Jørgensen, H. A. (2010). Effectiveness of second-generation antipsychotics: a naturalistic, randomized comparison of olanzapine, quetiapine, risperidone, and ziprasidone. BMC Psychiatry, 10, 26. doi:10.1186/1471-244X-10-26 Kaufman, D. A., & Trachenko, O. P. (1981). A study of interhemispheric asymmetry in healthy and schizophrenic individuals by a dichotic testing method. Human Physiology, 7, 449–455. Kay, S. R., Opler, L. A., & Lindenmayer, J. P. (1989). The Positive and Negative Syndrome Scale (PANSS): Rationale and standardisation. The British Journal of Psychiatry, 7(Suppl.), 59–67. Kimura, D. (1961). Cerebral dominance and the perception of verbal stimuli. Canadian Journal of Psychology, 15, 166–171. doi:10.1037/h0083219 Kompus, K., Westerhausen, R., & Hugdahl, K. (2011). The “paradoxical” engagement of the primary auditory cortex in patients with auditory verbal hallucinations: A meta-analysis of functional neuroimaging studies. Neuropsychologia, 49, 3361–3369. doi:10.1016/j. neuropsychologia.2011.08.010 Kuhn, S., & Gallinat, J. (2012). Quantitative meta-analysis on state and trait aspects of auditory verbal hallucinations in schizophrenia. Schizophrenia Bulletin, 38, 779–786. doi:10.1093/schbul/ sbq152 Levitan, C., Ward, P. B., & Catts, S. V. (1999). Superior temporal gyral volumes and laterality correlates of auditory hallucinations in schizophrenia. Biological Psychiatry, 46, 955–962. doi:10.1016/S0006-3223(98)00373-4 Løberg, E. M., Hugdahl, K., & Green, M. F. (1999). Hemispheric asymmetry in schizophrenia: A “dual deficits” model. Biological Psychiatry, 45, 76–81. Løberg, E. M., Jørgensen, H., & Hugdahl, K. (2002). Functional brain asymmetry and attention modulation in young, psychiatric stable schizophrenic patients: A dichotic listening study. Psychiatry Research, 109, 281–287. Løberg, E. M., Jørgensen, H. A., Green, M. F., Rund, B. R., Lund, A., Diseth, A., … Hugdahl, K. (2006). Positive symptoms and duration of illness predict functional laterality and attention modulation in schizophrenia. Acta Psychiatrica Scandinavica, 113, 322–331. doi:10.1111/j.16000447.2005.00627.x Løberg, E. M., Jørgensen, H. A., & Hugdahl, K. (2004). Dichotic listening in schizophrenic patients: Effects of previous vs. ongoing auditory hallucinations. Psychiatry Research, 128, 167– 174. doi:10.1016/j.psychres.2004.01.011 Modinos, G., Costafreda, S. G., van Tol, M. J., McGuire, P. K., Aleman, A., & Allen, P. (2013). Neuroanatomy of auditory verbal hallucinations in schizophrenia: A quantitative meta-analysis of voxel-based morphometry studies. Cortex, 49, 1046–1055. doi:10.1016/j.cortex.2012.01.009 Morrison, A. P., French, P., Walford, L., Lewis, S. W., Kilcommons, A., Green, J., … Bentall, R. P. (2004). Cognitive therapy for the prevention of psychosis in people at ultra-high risk: Randomised controlled trial. The British Journal of Psychiatry, 185, 291–297. doi:10.1192/ bjp.185.4.291 Ocklenburg, S., Westerhausen, R., Hirnstein, M., & Hugdahl, K. (2013). Auditory hallucinations and reduced language lateralization in schizophrenia: A meta-analysis of dichotic listening studies. Journal of the International Neuropsychological Society, 19, 410–418. doi:10.1017/ S1355617712001476 Palaniyappan, L., Balain, V., Radua, J., & Liddle, P. F. (2012). Structural correlates of auditory hallucinations in schizophrenia: A meta-analysis. Schizophrenia Research, 137, 169–173. doi:10.1016/j.schres.2012.01.038

94

E.-M. Løberg et al.

Downloaded by [Laurentian University] at 00:13 09 December 2014

Smucny, J., Wylie, K., & Tregellas, J. (2012). Stimulus-dependent effects on right ear advantage in schizophrenia. Journal of Neuropsychiatric Disease and Treatment, 8, 423–427. doi:10.2147/ NDT.S36277 Sommer, I. E., Diederen, K. M., Blom, J. D., Willems, A., Kushan, L., Slotema, K., … Kahn, R. S. (2008). Auditory verbal hallucinations predominantly activate the right inferior frontal area. Brain: A Journal of Neurology, 131, 3169–3177. doi:10.1093/brain/awn251 WHO. (1992). The ICD-10 classification of mental and behavioral disorders. Clinical descriptions and diagnostic guidelines. Geneva: Author.