Available online at www.sciencedirect.com

ScienceDirect Comprehensive Psychiatry 55 (2014) 928 – 937 www.elsevier.com/locate/comppsych

Comprehension of metaphors in patients with schizophrenia-spectrum disorders Nilufar Mossaheb a, c,⁎, Harald N. Aschauer c , Susanne Stoettner b , Michaela Schmoeger b , Nicole Pils b, d , Monika Raab b, e , Ulrike Willinger b a

Department of Psychiatry and Psychotherapy, Clinical Division of Social Psychiatry, Medical University Vienna, Austria b Department of Neurology, Medical University Vienna, Austria c Department of Psychiatry and Psychotherapy, Division of Biological Psychiatry, Medical University Vienna, Austria d Landesklinikum Thermenregion Baden, Department of Psychiatry and Psychotherapeutic medicine, Baden bei Wien, Austria e SMZ Baumgartner Hoehe, Otto Wagner Hospital, Department of Substance abuse, Vienna, Austria

Abstract Background: Metaphors, mainly proverbs and idiomatic expressions of ordinary life are commonly used as a model for concretism. Previous studies have shown impaired metaphor comprehension in patients with schizophrenia-spectrum disorders compared to either psychiatric or non-psychiatric control subject. The aim of this study was to detect possible quantitative differences in figurative processing between patients with schizophrenia-spectrum disorders and healthy controls. Methods: In order to analyse possible dissociations of different aspects of figurative speech, a range of metaphor tasks was used to distinguish between recognition of familiar metaphors, paraphrasing the meaning of the latter and generating novel metaphors: we used a standard proverb test for conventional metaphors consisting of a multiple-choice and a paraphrasing task, and the Metaphoric Triads Test for the assessment of novel metaphors. We included 40 patients with schizophrenia-spectrum disorders and 43 healthy control subjects. Results: Our results showed that patients had impaired figurative speech processing regarding novel and conventional metaphors. Associations with cognitive functions were detected. Performance on the paraphrasing task was associated with the severity of negative symptoms. Conclusion: We conclude that patients with schizophrenia-spectrum disorders do exhibit impairments in the recognition and paraphrasing of conventional and the generation of novel metaphors and that some cognitive domains as well the extent of negative symptoms might be associated with these deficits. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Patients with schizophrenia have long been shown to often exhibit language impairments, such as concretism [1–7], a characteristical deficit in non-literal language comprehension, where metaphoric meaning is not understood. However, the manifestation of abnormal language in schizophrenia disorders varies [8]. The basic structure of language, in particular the understanding of literal language, often remains intact [9,10]. Deficits are mainly related to the processing of high-level language, i.e. the pragmatic aspect of language, ⁎ Corresponding author at: Medical University of Vienna, Department of Psychiatry and Psychotherapy, Clinical Division of Social Psychiatry, Waehringer Guertel 18–20, 1090 Vienna, Austria. Tel.: +43 1 40400 3603; fax: +43 1 40400 3605. E-mail address: [email protected] (N. Mossaheb). 0010-440X/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.comppsych.2013.12.021

usually impacting the processing and comprehension of figurative language in particular metaphors, proverbs, idiomatic expressions and irony [2,3,11]. Figurative language requires the ability to process more than just the literal meaning of a statement in order to comprehend the speaker’s intention [9,11,12]. Metaphors, mainly proverbs and idiomatic expressions of ordinary life are commonly used as a model for concretism. The neuroanatomical basis of metaphorical comprehension somewhat parallels that of schizophrenia-spectrum disorders [13–17] with functional MRI studies showing that the following regions seem related to figurative language in both controls and patients with schizophrenia: left inferior frontal gyrus [17,18], right lateral temporal cortex [17]. Furthermore, some groups showed a lateralisation, yet data seem inconclusive [19], as some findings point towards an importance of the left hemisphere [18,20,21]; whereas

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937

others, more often in the context of innovative metaphors and possibly depending on familiarity and difficulty of the metaphor, enforce the importance of an intact right hemisphere [22–24]. Furthermore, from a neurobiological point of view the ability to adequately interpret metaphors has exhibited some association with dopaminergic activity in the frontotemporal region [25,26]. Studies in neurological patients with frontal lobe damage have demonstrated massive deficits in metaphor comprehension and a tendency to interpret these on a literal level [27–29]. Here too, overlaps with brain regions affected in schizophrenia are evident [30,31]. Furthermore, the frontal lobe is an important region for cognitive functions such as semantic processing [32] and is involved in higher executive processes [33], such as working memory, cognitive flexibility, reasoning, attention, as well as verbal fluency, inhibition and information processing speed [34]. As comprehension of figurative language is an essential part of social interaction, further understanding of concretism in patients with schizophrenia might lead to more insight into specificities of social and cognitive impairments in schizophrenia. Those studies that have analysed concretism, have mainly used figurative proverbs to assess metaphoric comprehension and have shown reduced capacity of abstraction in patients with schizophrenia compared to healthy controls [2,35,36] or to patients with depressive disorders [37]. Conventional metaphors include previously novel expressions that have lost their metaphorical novelty through time and have been completely integrated into everyday language. Metaphoric proverbs are therefore defined as being conventional metaphors, because they do not need to be newly invented, but are present in popular language [38]. According to the Contemporary Theory of Metaphors, conventional metaphors are described as belonging to an underlying conceptual mapping system of a language [39] (in [40]), whereas the Career of Metaphor Model assumes an evolution from novel metaphors that have a comparative nature to the adoption of a metaphoric categorization in the process of being conventionalized [41]. On the other hand, non-conventional, innovative or novel metaphors refer to the creation of novel “images”. Note, that the differentiation between conventional and novel metaphors can be seen as being on a gradual continuum, rather than being dichotomous [42]. Also, conventional metaphors are described as being more salient than novel metaphors due to their familiarity (Gradient Salience model, [43]). Indeed, the sole use of assessments of proverb comprehension in studies has sometimes shown insufficient reliability [44] due to the influence of familiarity. Therefore, in this study, additionally to metaphoric proverbs, we assessed figurative language comprehension in an operationalised manner with the Metaphoric Triads Test [45] — a test in which innovative metaphors have to be developed by the individual. The aim of this study was to detect possible quantitative differences in figurative processing between patients with schizophrenia or schizoaffective disorders and healthy con-

929

trols. In order to analyze possible dissociations of different aspects of figurative speech, we used a range of metaphor tasks to distinguish between recognition of familiar metaphors, paraphrasing the meaning of the latter and generating novel metaphors. We hypothesized that patients would show reduced comprehension of figurative language in both conventional and novel metaphor tasks and that this impairment would be associated to frontally mediated cognitive dysfunctions and to psychopathology.

2. Materials and Methods 2.1. Subjects Patients were recruited from in- and outpatient services of the Psychiatric Hospital Social–Medical Centre Otto-Wagner, Vienna, the Department of Psychiatry Social–Medical Centre East, Vienna, a therapeutic service in Vienna and the Sociopsychiatric Service Moedling, all of the above being in Austria. Of the initial 46 patients recruited for the study two were excluded due to discrepancies between clinician- and SCIDinterview diagnosis, further three dropped out during the assessments without specific reason, one was excluded for more adequate comparison with the control group. Inclusion diagnoses were assessed by the treating psychiatrist and reassessed through the German version of the Structured Clinical Interview for DSM-IV (SCID Axis I and II, [46,47]). Consensus diagnoses were made by two experienced, independent and blinded clinicians using information from the SCID I/II and from blinded case reports and hospital records. Prior to testing, a PANSS (Positive and Negative Syndrome Scale [48]) rating was performed. In order to avoid potential biases due to acute conditions, only patients with a PANSS total score of up to 95 (with a score on the positive symptoms scale under 38 and on the negative symptoms score under 36) were included in the study. Healthy controls without any psychiatric disorder were recruited through concentric circles and matched with the patient group for sociodemographic variables. To ensure lack of psychiatric diagnoses, all healthy individuals also underwent SCID-interviews. All participants were native speakers. The study was approved by the local ethics committee and was conducted according to the Declaration of Helsinki and Tokyo (1975) and further amendments. All participants were thoroughly informed about the study procedure and testing was started only after subjects had signed the informed consent form. The duration of the entire assessment was approximately 3–10 h, performed in several sessions where necessary. 2.2. Assessment of psychopathology Current psychopathology of patients was assessed by a trained rater using the PANSS [48], a widely used and wellknown semi-structured interview, which characterises

930

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937

positive, negative and general aspects of psychopathology. The scale comprises 30 items rated from 1 to 7. The item “abstract thinking” (N5) was omitted in the rating because this dimension was thoroughly evaluated with other assessment instruments described below. 2.3. Metaphor tests 2.3.1. Metaphor Triads Test (MTT) [45] For the assessment of generation and comprehension of novel metaphors, the German version of the MTT was used. It comprises 24 items consisting of triads, each triad offering three pairing possibilities, only one of which has a metaphorical context, the others being pairs due to functional, locational or categorical similarities. Metaphor categories are based on three modalities according to Kogan et al. [49]: physiognomic metaphors (P), which are based on an affective association (e.g. angry man–thunderstorm); configural metaphors (CF) which are based on perceptual similarities between two relevant attributes (e.g. winding river–snake) and conceptual metaphors (CC) that have category-overlapping similarities (e.g. moldy cheese, broken-down house). Subjects were asked to make as many pairings as possible and to explain their decisions. Scoring was made according to Schmöger [50], via the sum of metaphorically correct answers, the highest total score being 48, consisting of the following scoring system: 2 points for metaphorically correct answers with precise explanations, 1 point for metaphorically correct answers with imprecise explanations, 0 point for metaphorically incorrect answers. See Table 2. 2.3.2. Metaphoric proverbs [38] Furthermore, 12 conventional German metaphoric proverbs were presented to the participants, whereof 6 offered multiple-choice answers as a measure of recognition of conventional metaphors. The others were to be paraphrased freely. Scoring for paraphrasing was made according to Strasser [51] with 1 point for metaphorically correct explanations. Multiple-choice answers were scored according to Schaunig et al. [52] and categorized into metaphorically correct (1 point), metaphorically incorrect, literally correct and literally incorrect. For each method a maximum of 12 points is obtainable, adding up to a total score of 24. 2.4. Neuropsychological assessments 2.4.1. Verbal Intelligence The Wechsler Adult Intelligence Scale (HAWIE-R, [53]) was used to assess verbal IQ. It consists of six subtests: information, comprehension, digit span, vocabulary, arithmetic, and similarities. The numbers of correct items was recorded. 2.4.2. Nonverbal Intelligence The Raven’s Standard Progressive Matrices (SPM) were used as a measurement of the nonverbal components of general intelligence, especially of analogical thinking. They consist of five subtests with 12 items each [54,55]. The number of correct items was recorded.

2.4.3. Information processing speed and cognitive flexibility The Trail Making Test (TMT, [56]) measures information processing speed and cognitive flexibility. In this test, the participants had to draw lines to connect numbers (part A) or numbers and letters (part B) in ascending order as fast as possible. The time needed to complete part A and part B was recorded. The Wisconsin card sorting test (WCST, [57]) is a widelyused test for the assessment of cognitive flexibility in terms of executive function. The task consists in matching a set of cards on the background of changing reinforcement schedules. The total number of correct items was recorded. 2.4.4. Inhibition Inhibition of a habitual response was measured by the Stroop test [58]. The subject is presented a set of words that are the names of colours, written in incongruent colours. The task is to respond with the colour that is seen and not the one that is read. The tasks consist in reading the colour words as fast as possible (“reading”), naming the colour of coloured lines (“naming”) and naming the colour that is seen and not the one that is read (“interference”). The time taken for each task was recorded.

3. Data analysis All calculations were performed with IBM SPSS© version 20. Differences in means were calculated using t-test or repeated-measures ANOVA, where applicable. Chi-square was used for comparative analysis between groups regarding categorical or dichotomic variables. Additionally, analyses of covariances (ANCOVA) were conducted to compare patients and healthy controls on measures of metaphor comprehension using cognitive functions as covariates. Correlation analyses based on Pearson´s product–moment correlation were conducted to examine the relationship between metaphor tests and cognitive functions as well as between clinical status and metaphor tests respectively cognitive functions. Bonferroni correction was applied after multiple testing [59]. Post hoc power analysis conducted with G*Power 3.1 provided a power of 0.491 (α = 0.05, ρ = 0.3) in the given sample of 40 patients. Finally, the association between cognitive functions and metaphor comprehension in patients with schizophrenia was analysed by three separate backward multiple linear regression analyses, with the number of correct items on the metaphor tests as dependent variables. All statistical tests were based on a level of significance of .05.

4. Results 40 patients with a DSM-IV-TR diagnosis of schizophrenia and schizoaffective disorder (15 males, 25 females) were included in the study.

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937 Table 1 Sociodemographic and clinical characteristics of patients and healthy controls. Patients

Age (SD) Gender, male: female Education University studies High school diploma Apprenticeship Compulsory education PANSS positive (SD) PANSS negative (SD) PANSS global (SD)

Healthy controls

Difference

n = 40

n = 43

p-value

39.6 (9.938) 15: 25

42.9 (13.2) 19: 24

.201 .536 .350

1 13 15 11 18.54 (1.14) 14.83 (0.73) 25.1 (6.49)

0 21 14 8 – – –

– – –

PANSS = Positive and Negative Syndrome Scale.

27 of 40 patients had a diagnosis of paranoid type schizophrenia, 1 was diagnosed with catatonic type schizophrenia, 2 with residual type schizophrenia and 10 patients had a diagnosis of schizoaffective disorder. Mean age at first episode of illness was 21.8 years (SD = 6.715). Within the patient group there was no significant difference regarding age of first manifestation of illness and sex (χ 2(1) = 20.94, p = 0.340). 43 healthy controls (19 males, 24 females) were included in the study. There were no significant differences between the groups on sociodemoraphic variables (see Table 1). For further details on the sociodemographic and clinical, see Table 1.

931

4.1. Metaphor comprehension 4.1.1. Novel metaphors There were significant differences between patients and controls with respect to the comprehension of novel metaphors with more errors in the patient group [t (81) = 2.65, p b .05, r = .28] (Table 3). 4.1.2. Conventional metaphors Regarding multiple-choice as well as paraphrasic answers in conventional metaphor comprehension significant group differences were found [multiple choice answers: t (81) = 4.62, p ≤ .0001, r = .21; paraphrasic answers: t (81) = 3.55, p ≤ .0001, r = .37] with lower comprehension of metaphors in patients with schizophrenia and schizoaffective disorder.

4.1.3. Error distribution More detailed analysis of error distribution revealed that patients made more errors in the comprehension of conventional metaphors, in both the paraphrasing (46% errors in patients with schizophrenia/schizoaffective disorder versus 31% in healthy controls, p b 0.0001) and the multiple-choice tasks (13% errors in patients versus 2% in controls, p = 0.0425). More specifically, they tended to make more errors on a literal level: In the paraphrasing task patients gave 15% literal answers, whereas controls only 2%; in the multiple-choice task patients’ errors on a literal level were up to 73%, whereas controls’ were 11% (p = 0.0425).

Table 2 Metaphoric Triads Test (Kogan & Chadrow, 1986). Modality Coiled snake * Man in the rain Grandfather * Dancing ballerina * Rat House with shades pulled down * Ocean Watering can Foggy street corner * Waves running into sand castle * Watering can Sunflower * Violin * Wilted plant * Rifle Snorting bull * Old man * City lit up at night * Car Rooster crowing * Willow tree * Vase on table Ambulance Fly in spider´s web*

Winding river * Thunderstorm * Rocking chair Girl playing Moldy Swiss cheese * Bed Plane on fire * Baby in garden * Veiled woman * Melting snowman * Woman with long hair * Greenhouse Singing canary * Hot, tired runner * Marching soldiers * Man boxing * Smoking candle nearly burnt out * Classy city street Spoked wheel * Farmer showing muscles * Woman crying * Old woman sick in bed * Explosion * Old boat

Metaphorical pairs are marked with an asterisk. P = perceptual metaphors; CF = configural metaphors; CC = conceptual metaphors.

Fish Angry man * Ancient tree * Spinning top * Broken-down house * Woman with closed eyes * Fish on hook * Rose bud * Person driving car Girl standing in sun Hanging plant * Tall thin woman * Tree Glass of water Flock of birds * Leather gloves Lit pipe Woman with sparkling jewellery * Traffic circle * Barnyard with animals Park bench Wilted flowers * Man in rage * Fish caught in net*

CF P CF CC CC P CC CC P CC CF CF CF P CF CC CC P CC CC P CC CF CF

932

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937

Table 3 Performance on metaphor tasks and cognitive functioning tests in patients and healthy controls (means, standard deviations, and p-value). Patients

MTT score MC score Paraphrasing score HAWIE-R information HAWIE-R digit span HAWIE-R vocabulary HAWIE-R arithmetic HAWIE-R comprehension HAWIE-R similarities Raven’s SPM Trail Making Test-A Trail Making Test-B Stroop Test Reading Naming Interference WCST Total number of correct Perseverative responses Number of categories completed

Healthy controls

p-value

(n = 40)

(n = 43)

15.55 (7) 10.42 (1.8) 6.4 (2.7) 14.45 (4.8) 12.73 (3.4) 15.08 (5) 9.35 (3.8) 14.80 (5.4) 20.18 (5.9) 38.98 (10.2) 42.90 (15.9) 121.85 (55)

20.46 (9.6) 11.76 (0.7) 8.2 (1.9) 15.72 (3.9) 13.6 (3.3) 20.56 (4.9) 12.71 (3.4) 19.26 (3.4) 26.02 (3.4) 48.19 (8.5) 30.7 (14) 75.9 (31.8)

b .05 ≤ .0001 ≤ .0001 n.s. n.s. ≤ .0001 ≤ .0001 ≤ .0001 ≤.0001 ≤ .0001 ≤ .0001 ≤ .0001

37.25 (6.8) 60.38 (18.8) 103.65 (38)

30.3 (5.7) 45.5 (7.9) 74.14 (14.2)

≤ .0001 ≤ .0001 ≤ .0001

45.20 (9.2) 10.08 (5.7) 2.75 (1.4)

46.67 (9.8) 9.23 (6.1) 3.44 (0.6)

n.s. n.s. ≤ .0001

ns = Statistically not significant. MTT = Metaphors Triad Test; MC = multiple choice task; TMT = Trail Making Test; WCST = Wisconsin Card Sorting Test.

4.2. Cognitive functions Patients scored significantly lower than controls in the SPM [T (81) = 4.48, p ≤ .0001, r = .45], in the subscale “number of completed categories” of the WCST [T (81) = 2.124, p ≤ .0001, r = .23] and on several subtests of the HAWIE-R, such as “similarities” [T (81) = 5.54, p ≤ .0001, r = .52), “arithmetics” [T (81) = 4.26, p ≤ .0001, r = .43], “vocabulary” [T (81) = 5.06, p ≤ .0001, r = .49] and “comprehension” [T (81) = 4.53, p ≤ .0001, r = .45]. A repeated-measure ANOVA of the TMT part A and part B yielded a significant effect of the conditions [F (1, 81) = 225.72, p ≤ .0001] and a significant interaction between the two groups and the condition [F (1, 81) = 16.64, p ≤ .0001]. The difference between TMT part B and part A was significantly longer for patients compared to the healthy controls. Separate t-tests of TMT part A and part B yielded significant differences between patients and the healthy control group for the time taken to complete TMT part A [T (81) = −3.72, p ≤ .0001, r = .38] and part B [T (81) = −4.70, p ≤ .0001, r = .46]. This was due to significantly longer reaction times for the schizophrenic patients compared to the healthy controls in any of the two conditions. Repeated-measures ANOVA yielded a significant effect for condition [F(2,81) =198.39, p ≤ .0001] and a significant group and condition interaction [F(2, 81) = 8.261, p ≤ .0001] in the Stroop Test. This was due to a significantly larger difference between the conditions “interference” and

“naming the color of colored lines” of the patient group in comparison with the healthy controls (both p ≤ .0001). Ttests of “reading color words” and “naming colored lines,” showed significant effects [T(81) = − 5.057; p ≤ .0001, r = .49;T (81) = −4.751, p ≤ .0001, r = .47; and T (81) = −4.76, p ≤ .0001, r = .47]. These effects were due to significantly longer reaction times in the patient group in comparison to the healthy control group. No other significant effects were observed (all p N .05). 4.3. Association of metaphor comprehension, cognitive functions and psychopathology The results of the covariance analyses showed that the differences between patients and healthy controls in the comprehension of conventional metaphors were significantly related to scores on the HAWIE-R Digit span subtest [F(1, 67) = 5.36, p = .024, r = .07], the WCST number of correct items [F (1, 67) = 7.268, p = .009, r = .09] and number of perseverative responses subtests [F (1, 67) = 4.828, p = .031, r = .06]. The differences in metaphor tasks remained significant even after adjustment for covariates (all p ≤ .0001). Correlations between metaphor comprehension tasks and measures of cognitive functions are detailed in Table 4. Multiple regression analyses were conducted to examine the relative contribution of cognitive functions to the comprehension of novel and conventional metaphors in patients (see Table 5). For MTT scores significant effects were found for the HAWIE-R vocabulary subtest. TMT part B showed significant effects on both task of conventional metaphor comprehension. With respect to correlations with clinical status, there were no correlations between the PANSS positive symptoms score and novel (p = 0.48) or conventional metaphor (paraphrasing p = 0.99, multiple-choice p = 0.61) comprehension, however there was a weak to moderate correlation between the PANSS negative symptoms score (excluding the concretism-item of the scale) and scores on paraphrasic answers (r = −0.346, p = 0.029). Also, PANSS negative symptoms scores were weakly to moderately associated with the HAWIE-R Verbal part subtests information (r = −0.39, p = 0.012), arithmetic (r = −0.41, p = 0.009), comprehension (r = −0.357, p = 0.024), and similarities (r = −0.324, p = 0.041), as well as with part B of the Trail Making Test (r = 0.412, p = 0.008), Stroop test reading part (r = 0.399, p = 0.011) and several subtests of the WCST (total number correct r = −0.416, p = 0.008, total number errors r = 0.416, p = 0.008, perseverative responses r = 0.336, p = 0.034). Although the reported correlations are no longer significant after multiple testing correction was applied (Bonferroni-corrected α-level of 0.0038, [59]) the coefficients of determination (R 2), ranging between 0.105 and 0.173, indicate that 10.50% to 17.30% of the variation is explained by the particular variables.

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937

933

Table 4 Correlations of neurocognitive battery and metaphor comprehension tasks. MTT

HAWIE-R information HAWIE-R digit span HAWIE-R vocabulary HAWIE-R arithmetic HAWIE-R comprehension HAWIE-R similarities Raven’s SPM TMT-A TMT-B Stroop test reading Stroop test naming Stroop test interference WCST total number correct WCST perseverative responses WCST number of categories completed

MC

Paraphrasing

Patients

Controls

Patients

Controls

Patients

Controls

.32 * n.s. .46 ** n.s. n.s. n.s. n.s. n.s. –.32 * n.s. n.s. n.s. n.s. n.s. n.s.

.35* .38 ** .36** .35** .47 ** n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

.51 ** .36 * .43 ** .47 ** .42 ** .39 ** .60 ** −.50 ** –.61** –.47** n.s. n.s. 0.43 ** n.s. 0.38 *

n.s. n.s. .36 * .34 * n.s. n.s. n.s. –.32 * n.s. n.s. –.33 * n.s. n.s. n.s. n.s.

n.s. n.s. n.s. n.s. n.s. n.s. n.s. –.34 * –.39 ** –.39 ** n.s. n.s. n.s. n.s. n.s.

n.s. n.s. .45 ** n.s. .41 ** n.s. .41 ** n.s. n.s. n.s. n.s. n.s. .37 * –.42 ** .36 *

* p-value b .05, ** p-value b .01, n.s. = not significant. MTT = Metaphors Triad Test; MC = multiple choice task, TMT = Trail Making Test, WCST = Wisconsin Card Sorting Test.

5. Discussion The aim of our study was to analyse potential differences between individuals with schizophrenia-spectrum disorders and healthy control subjects in their ability to recognise and paraphrase conventional metaphors and to generate novel metaphors. Our results clearly indicate impairments in patients’ performance in all metaphor appreciation tasks used in this study with medium to large effect sizes. These findings confirm our hypotheses and are in line with previous studies that have shown deficits in figurative language comprehension in patients with schizophrenia compared to either psychiatric or non-psychiatric controls [2,5,36,37,60–63]. On a descriptive level, patients tended to interpret idiomatic expressions more often on a literal level, whereas controls made more errors on a metaphorical level. The differences observed between patients and controls remained even after cognitive functions – that were impaired Table 5 Backward regression analysis predicting novel and conventional metaphor comprehension. Variable

B SEB

Novel metaphor comprehension task (MTT) Vocabulary .65 .20

ß

p-value .47

.003

Conventional metaphor comprehension task (paraphrasing) TMT part B –.02 –.40 .01

.012

Conventional metaphor comprehension task (MC) TMT part B –.02 –.61 .01

.0001

For novel metaphor comprehension: R2 = .21, p = .003. For conventional metaphors (paraphrasing): R2 = .16, p = .012. For conventional metaphor comprehension (MC): R2 = .37, p b 0001.

in patients – were controlled for. The findings do indeed indicate that comprehension of metaphors is related to cognitive abilities and that these associations are stronger in patients than in controls. Specifically, in controls, innovative metaphor comprehension was related to verbal IQ, whereas understanding of conventional metaphors was linked to verbal vocabulary and cognitive flexibility (both tasks), as well as with analogical thinking (paraphrasing task) and information processing speed and inhibition (multiple-choice task). In patients, recognition of conventional metaphors was also associated with non-verbal analogical thinking, inhibition, information processing speed, cognitive flexibility and additionally with a range of verbal IQ subtests. The latter three functions were also related to generation of novel metaphors. This raises the question of whether literal language is easier to understand in the context of impaired cognitive functioning and psychopathology. In the existing linguistic literature there are two theories on the issue: one posits a multi-stage process of metaphor comprehension that first involves rejection of literal meaning [64–66], i.e. demands a certain cognitive ability to do so, the other postulates a salience-mediated understanding of figurative expressions due to contextual interpretation [43,67,68]. However, it seems that the latter imposes a certain familiarity with either certain idiomatic expressions [69] or a capacity of abstractive thinking in the context of informal so-called “social contracts” [36]. An important factor in the assessment of understanding of abstractive language is the choice of the testing instrument. In schizophrenia, the most common test to assess concretism is the proverb test [37,70]. Our findings reveal that the paraphrasing task of this test is affected by the intensity of negative symptoms. Thus, it can be argued that this task not only assesses comprehension of common idiomatic expressions but also demands a certain semantic understanding and

934

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937

more importantly a level of verbal fluency and ability of executive and expressive language; this again has been shown to be affected by negative symptoms in schizophrenia-spectrum disorders [71,72]. We have assessed metaphorical comprehension not only with a commonly used proverb test, but also by means of a test for novel metaphors; with this testing instrument, the issue of familiarity in the context of common proverbs can be excluded and speech fluency is a less important variable. The next question that arises when looking at our results and the existing literature is whether we are dealing with a specific, pathognomonic feature of schizophrenia-spectrum disorders, or whether this impairment is a sign of a general cognitive deficit, shown in these patients [73,74]. Interestingly, reduced comprehension of figurative language has also been shown in patients with Alzheimer’s disease [75], which might underline the theory of a general cognitive deficit. Compared to healthy controls, our patients showed significant deficits within verbal IQ and non-verbal analogical thinking, but also a variety of executive function tasks. Also, a range of these functions was associated with performance on the metaphor tasks, with highest correlations depicted for recognition of figurative speech (multiple-choice task). Specifically, in the present study, cognitive flexibility, measured with the TMT part B, influenced both recognition and paraphrasing of conventional metaphors. Also, the ability to comprehend vocabulary was a significant predictor for performance on novel metaphor comprehension. Indeed, executive functions are thought to require competencies in planning, reasoning, problem-solving, mental flexibility, long-term knowledge searching and long-term goal pursuits, but also abstraction [76]. Thus, our findings might point towards abstraction deficits being part of a more generalized deficit in executive functions known in schizophrenia-spectrum disorders. From a neurobiological perspective, there is a regional overlap between brain areas involved in these cognitive functions and those affected in schizophrenia with those crucial to metaphor comprehension [17,18,77]. Thus, it could be speculated that impairments in understanding figurative language might be linked to brain alterations associated with frontally mediated cognitive deficits or with the illness itself. Metaphorical comprehension implies that one is able to understand the analogy behind the obvious semantic meaning of words; it implies the capacity to abstract, to put into the right context. From a neurodevelopmental perspective, studies have shown that adequate proverb interpretation, i.e. comprehension of existing and common metaphors, usually improves between the ages of 7 and 11 years [52]. The developmental pathway of metaphor comprehension also appears in the error distribution: younger children tend to make more literally correct, but not-metaphoric interpretations, whereas older children seem to understand that they are dealing with a metaphor, thus make more metaphorically incorrect errors [52]. Similarly, our patients tended to make more errors on a literal level than

the non-psychiatric controls. Recently, Reichenberg and colleagues presented results of a representative longitudinal cohort study of individuals that were followed-up from age 3 to 32 years [78]. Their cohort shows early emerging, stable and static cognitive impairments on verbal and visual knowledge acquisition as well as reasoning and conceptualization tasks in children aged 7 to 13 years that later develop schizophrenia. These findings certainly fortify the neurodevelopmental model of schizophrenia. We did not conduct a longitudinal study from childhood to transition to psychosis assessing patients in acute and remitted states; the lack of state-dependent assessments and long-term followups can indeed be seen as a limitation of this study. However, we aimed to depict and to better understand the nature of the impairment of differentiated aspects of figurative processing in a cross-sectional sample of patients with schizophreniaspectrum disorders. Another question that arises is whether metaphorical comprehension can be seen as a form of social cognition. Social cognition refers to the abilities necessary to perceive and process cues relevant to social systems and to adapt one’s behaviour accordingly [79]. The results of our study point towards a relationship between metaphorical comprehension and specific cognitive abilities and psychopathology. Specifically, we found correlations of verbal IQ, non-verbal analogical thinking and executive functions with generation of novel metaphors and recognition of conventional metaphors as well as associations of negative symptoms with standard proverb comprehension. However, the study did not assess selective attention deficits, which might have impeded metaphor comprehension at the very beginning of the information processing chain. Since metaphors are part of everyday language, the inability to understand these adequately can lead to misinterpretation, which in turn can induce a reduction of social contacts, social withdrawal and possibly increasing negative symptoms. Similarly, in patients with Asperger’s syndrome, a condition known to be associated with social interaction deficits, metaphoric comprehension is reduced [80]. From a different point of view, misinterpretation and misunderstanding of speech may further enhance increased suspiciousness in the context of salience and enforce a continuum of compensation leading to paranoid ideations [81]. Adequate use and comprehension of common language are crucial to normal social interaction and social interaction has been shown to be important for rehabilitation [82]; furthermore, cognitive impairments and concretism are associated with reduced outcomes of rehabilitation [83]. Thus, more insight into the relationship between metaphor comprehension, psychopathology, cognitive and social deficits might help to develop early trainings and preventative strategies. 5.1. Limitations Further to the mentioned limitations regarding longitudinal observations, the study bears some other limitations. Although

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937

a non-psychiatric control group offers the benefits of a comparison with healthy subjects, it may impede the differentiation between schizophrenia-specific pathology and non-characteristic, pathological features of psychiatric illnesses or effects of medication and hospitalisation. As a matter of fact, we have not assessed our data regarding medication effects nor have we included another psychiatric control group. Thus, the interpretation of our data is limited to the confines of this study. Given the heterogeneity of schizophrenia-spectrum disorders it might have been useful to look at figurative speech processing deficits within different disorder subtypes, however the size and distribution of diagnoses of our sample would not have permitted a useful analysis. 6. Conclusion In this study, patients with schizophrenia-spectrum disorders exhibited impairments in novel and conventional metaphor comprehension compared to healthy controls and these deficits showed some association with cognitive domains as well as with the extent of negative symptoms. However, there is still a plethora of questions that need to be answered through further research combining clinical, neurobiological and linguistic methods. Declaration of interests The authors declare no conflicting interests. Appendix 1. Conventional German metaphoric proverbs (adapted from Röhrich 1999) 1.1. Paraphrasing task (examples) 1. Wir sitzen alle im gleichen Boot (We are all in the same boat) 2. Paul hat eine lange Leitung. (Paul is a slow coach) 3. Du bringst mich auf die Palme (You’re setting my teeth on edge) 4. Ich lasse Sabine ein bisschen zappeln. (I am keeping Sabine on tenterhooks) 5. Sie beißt nun in den sauren Apfel. (She will be biting the bullet) 6. Barbara und Lisa geraten sich ständig in die Haare. (Barbara and Lisa are constantly at loggerheads) 7. Johanna hat die Katze aus dem Sack gelassen. (Johanna let the cat out of the bag) 8. Er streut Salz in ihre Wunden. (He rubs salt into her wounds) 9. Sie hat viel Staub aufgewirbelt. (She raises a lot of dust) 10. Jetzt brennt ihm der Hut. (Now the roof is on fire) 11. Ich musste in letzter Zeit vieles schlucken. (Recently, I had to swallow a lot)

935

12. Ich rede anscheinend gegen eine Mauer. (Apparently, I am talking to a brick wall) 1.2. Multiple-Choice task (examples) 1. Du bringst mich auf die Palme (You are setting my teeth on edge (figurative)/You are getting me up the palm tree (literal)) A. Du machst mich wütend. (You are making me angry) B. Mir gefallen Palmen nicht so gut wie dir. (I don’t like palm trees as much as you do) C. Ich helfe dir gerne. (I will be helping you gladly) D. Für dich klettere ich auf die Palme. (I will go up a palm tree for you) 2. Barbara und Lisa geraten sich ständig in die Haare. (Barbara and Lisa were constantly at loggerheads (figurative)/Barbara and Lisa are constantly getting into each other's hair (literal)) A. Barbara und Lisa reißen sich oft gegenseitig an den Haaren. (Barbara and Lisa pull each others’ hair a lot) B. Barbara und Lisa streiten sich sehr häufig. (Barbara and Lisa argue a lot) C. Barbara und Lisa kämmen sich oft gegenseitig die Haare. (Barbara and Lisa comb each others’ hair a lot) D. Barbara und Lisa sind immer füreinander da. (Barbara and Lisa are always there for each other) 3. Ich musste in letzter Zeit vieles schlucken. (Recently I had to swallow a lot (both figurative and literal)) A. Ich musste in letzter Zeit viele Tabletten schlucken. (Recently, I had to swallow a lot of pills) B. Ich musste in letzter Zeit viel essen und trinken. (Recently, I had to eat and drink a lot) C. Ich musste in letzter Zeit viel ertragen. (Recently, I had to bear a lot) D. Ich musste in letzter Zeit viel lernen. (Recently, I had to learn a lot) 4. Wir sitzen alle im gleichen Boot. (We are all in the same boat) A. Wir befinden uns im selben Boot. (We are all situated in the same boat (both figurative and literal)) B. Wir befinden uns in gleicher Lage und sind aufeinander angewiesen. (We are all in the same situation and are dependent on each other) C. Wir müssen die Sache ausdiskutieren. (We have to discuss the matter) D. Wir kaufen uns zusammen ein Boot. (We are buying a boat together) References [1] Bleuler E. Dementia praecox or the group of schizophrenias. Leipzig-Wien: International Universities Press; 1911.

936

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937

[2] Brune M, Bodenstein L. Proverb comprehension reconsidered— ‘theory of mind' and the pragmatic use of language in schizophrenia. Schizophr Res 2005;75:233-9. [3] Chapman LJ. Confusion of figurative and literal usages of words by schizophrenics and brain damaged patients. J Abnorm Soc Psychol 1960;60:412-6. [4] Gorham DR. Use of the proverbs test for differentiating schizophrenics from normals. J Consult Psychol 1956;20:435-40. [5] Iakimova G, Passerieux C, Laurent JP, Hardy-Bayle MC. ERPs of metaphoric, literal, and incongruous semantic processing in schizophrenia. Psychophysiology 2005;42:380-90. [6] Kraepelin E. Dementia praecox and paraphrenia. Edinburgh, Scotland: Ligingburg; 1919. [7] Strobl R, Resch F. Schizophrenic concretism. Nervenarzt 1988;59:99-102. [8] Kuperberg GR. Language in schizophrenia. Part 1: an introduction. Lang Linguist Compass 2010;4:576-89. [9] Champagne-Lavau M, Stip E. Pragmatic and executive dysfunction in schizophrenia. J Neurolinguistics 2010;23:286-96. [10] Martin I, McDonald S. Weak coherence, no theory of mind, or executive dysfunction? Solving the puzzle of pragmatic language disorders. Brain Lang 2003;85:451-66. [11] Gavilàn JM, Garcia-Albea JE. Theory of mind and language comprehension in schizophrenia: poor mindreading affects figurative language comprehenion beyond language deficits. J Neurolinguistics 2011;14:54-69. [12] Champagne-Lavau M, Joanette Y. Pragmatics, theory of mind and executive functions after a right hemisphere lesion: different patterns of deficits. J Neurolinguistics 2009;22:413-26. [13] Andreasen NC. A unitary model of schizophrenia: Bleuler's “fragmented phrene" as schizencephaly. Arch Gen Psychiatry 1999;56:781-7. [14] Crow TJ. Schizophrenia as failure of hemispheric dominance for language. Trends Neurosci 1997;20:339-43. [15] Flor-Henry P. Psychosis and temporal lobe epilepsy. A controlled investigation. Epilepsia 1969;10:363-95. [16] Friston KJ. The disconnection hypothesis. Schizophr Res 1998;30:115-25. [17] Kircher TT, Leube DT, Erb M, Grodd W, Rapp AM. Neural correlates of metaphor processing in schizophrenia. Neuroimage 2007;34:281-9. [18] Stringaris AK, Medford NC, Giampietro V, Brammer MJ, David AS. Deriving meaning: distinct neural mechanisms for metaphoric, literal, and non-meaningful sentences. Brain Lang 2007;100:150-62. [19] Schmidt GL, Kranjec A, Cardillo ER, Chatterjee A. Beyond laterality: a critical assessment of research on the neural basis of metaphor. J Int Neuropsychol Soc 2010;16:1-5. [20] Giora R, Zaidel E, Soroker N, Batori G, Kascher A. Differential effects of right- and left-hemisphere damage on understanding of sarcasm and metaphor. Metaphor Symb 2000;15:63-83. [21] Rapp AM, Leube DT, Erb M, Grodd W, Kircher TT. Neural correlates of metaphor processing. Brain Res Cogn Brain Res 2004;20:395-402. [22] Mashal N, Faust M. Conventionalisation of novel metaphors: a shift in hemispheric asymmetry. Laterality 2009;14:573-89. [23] Pobric G, Mashal N, Faust M, Lavidor M. The role of the right cerebral hemisphere in processing novel metaphoric expressions: a transcranial magnetic stimulation study. J Cogn Neurosci 2008;20:170-81. [24] Schmidt GL, Seger CA. Neural correlates of metaphor processing: the roles of figurativeness, familiarity and difficulty. Brain Cogn 2009;71:375-86. [25] Chenery HJ, Angwin AJ, Copland DA. The basal ganglia circuits, dopamine, and ambiguous word processing: a neurobiological account of priming studies in Parkinson's disease. J Int Neuropsychol Soc 2008;14:351-64. [26] Flaherty AW. Frontotemporal and dopaminergic control of idea generation and creative drive. J Comp Neurol 2005;493:147-53. [27] Chobor KL, Schweiger A. Processing of lexical ambiguity in patients with traumatic brain injury. J Neurolinguistics 1998;11:119-36. [28] Jakobson R, Halle M. Fundamentals of language. Paris: Mouton; 1972. [29] Stuss DT, Benson DF, Kaplan EF, Weir WS, Naeser MA, Lieberman I, et al. The involvement of orbitofrontal cerebrum in cognitive tasks. Neuropsychologia 1983;21:235-48.

[30] Bora E, Fornito A, Radua J, Walterfang M, Seal M, Wood SJ, et al. Neuroanatomical abnormalities in schizophrenia: a multimodal voxelwise meta-analysis and meta-regression analysis. Schizophr Res 2011;127:46-57. [31] Brown GG, Thompson WK. Functional brain imaging in schizophrenia: selected results and methods. Curr Top Behav Neurosci 2010;4:181-214. [32] Bookheimer S. Functional MRI of language: new approaches to understanding the cortical organization of semantic processing. Annu Rev Neurosci 2002;25:151-88. [33] Samson AC, Hempelmann CF, Huber O, Zysset S. Neural substrates of incongruity-resolution and nonsense humor. Neuropsychologia 2009;47:1023-33. [34] Antonova E, Sharma T, Morris R, Kumari V. The relationship between brain structure and neurocognition in schizophrenia: a selective review. Schizophr Res 2004;70:117-45. [35] Holm-Hadulla RM, Haug F. Interpretation of proverbs as a clinical method for the detection of schizophrenic thinking, speech and symbolism disorders. Nervenarzt 1984;55:496-503. [36] Mo S, Su Y, Chan RC, Liu J. Comprehension of metaphor and irony in schizophrenia during remission: the role of theory of mind and IQ. Psychiatry Res 2008;157:21-9. [37] Barth A, Kufferle B. Development of a proverb test for assessment of concrete thinking problems in schizophrenic patients. Nervenarzt 2001;72:853-8. [38] Röhrich L. Lexikon der sprichwörtlichen Redensarten. Wien: Herder, Freiburg im Breisgau; 1994. [39] Lakoff G, Johnson M. Metaphors we live by. Chicago, IL: University of Chicago Press; 1980. [40] Lai VT, Curran T, Menn L. Comprehending conventional and novel metaphors: an ERP study. Brain Res 2009;1284:145-55. [41] Bowdle B, Gentner D. The career of metaphor. Psychol Rev 2005;112:193-216. [42] Frieling G. Untersuchungen zur Theorie der Metapher: das Metapherverständnis als sprachlich kognitiver Verarbeitungsprozess. Osnabrück: Universitätsverlag Rasch; 1996. [43] Giora R. On our mind: salience, context and figurative language. New York: Oxford University Press; 2003. [44] Andreasen NC. Reliability and validity of proverb interpretation to assess mental status. Compr Psychiatry 1977;18:465-72. [45] Kogan N, Chadrow M. Children´s comprehension of metaphor in the pictorial and verbal modality. J Behav Dev 1986;9:285-95. [46] First MB, Spitzer RL, M. G, Williams JBW. Structured clinical interview for DSM-IV-TR axis I disorders, research version, patient edition (SCIDI/P). New York: New York State Psychiatric Institute; 2002. [47] Wittchen HU, Zaudig M, Fydrich T. Strukturiertes Klinisches Interview für DSM-IV-TR, Achse I und II (SKID-I, SKID-II). Göttingen: Hogrefe; 1997. [48] Kay SR, Fiszbein A, Opler LA. The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophr Bull 1987;13:261-76. [49] Kogan N, Connor K, Gross A, Fava D. Understanding visual metaphor: developmental and individual differences. Monogr Soc Res Child Dev 1980;45:1-78. [50] Schmöger M. Metaphernverständnis und nonverbale Fähigkeiten (Metaphor comprehension and non-verbal abilities). Vienna, Austria: Faculty of Psychology. University of Vienna; 2004. [51] Strasser E. Einflussfaktoren auf das Verstehen von Metaphern (Factors influencing metaphor comprehension). Vienna, Austria: Faculty of Psychology. University of Vienna; 2005. [52] Schaunig I, Willinger U, Formann AK. Das Verständnis metaphorischer Sprache bei Grundschulkindern. Ger J Educ Psychol 2004;18:53-61. [53] Tewes U. Hamburg Wechsler Intelligenztest für Erwachsene-R (HAWIE-R)-Verbalteil. Bern: Huber-Verlag; 1991. [54] Raven JC. Standard progressive matrices (SPM). Göttingen: Beltz; 1998. [55] Raven J, Raven JC, Court HH. Raven manual: section 3: standard progressive matrices. Oxford: Oxford Psychologists Press Ltd.; 2000. [56] Reitan RM. Trail Making Test. Göttingen: Beltz; 1979. [57] Kongs SK, Thompson LL, Iverson GL. Wisconsin Card Sorting Test (WCST-64). Odessa: PAR; 2000.

N. Mossaheb et al. / Comprehensive Psychiatry 55 (2014) 928–937 [58] Bäumler G. Farb-Wort-Interferenztest (Stroop test). Göttingen: Hogrefe; 1985. [59] Bonferroni CE. Il calcolo delle assicurazioni su gruppi di teste. Studi in onore del Professore Salvatore Ortu Carboni; 1935. p. 13-60. [Rome, Italy]. [60] Cutting J, Murphy D. Preference for denotative as opposed to connotative meanings in schizophrenics. Brain Lang 1990;39:459-68. [61] de Bonis M, Epelbaum C, Deffez V, Feline A. The comprehension of metaphors in schizophrenia. Psychopathology 1997;30:149-54. [62] Drury VM, Robinson EJ, Birchwood M. ‘Theory of mind' skills during an acute episode of psychosis and following recovery. Psychol Med 1998;28:1101-12. [63] Tavano A, Sponda S, Fabbro F, Perlini C, Rambaldelli G, Ferro A, et al. Specific linguistic and pragmatic deficits in Italian patients with schizophrenia. Schizophr Res 2008;102:53-62. [64] Langdon R, Coltheart M, Ward PB, Catts SV. Disturbed communication in schizophrenia: the role of poor pragmatics and poor mindreading. Psychol Med 2002;32:1273-84. [65] Spitzer M, Lukas M, Maier S, Hermle L. Comprehension of metaphoric speech by healthy probands and schizophrenic patients. An experimental psychopathologic contribution to concretism. Nervenarzt 1994;65:282-92. [66] Titone D, Holzman PS, Levy DL. Idiom processing in schizophrenia: literal implausibility saves the day for idiom priming. J Abnorm Psychol 2002;111:313-20. [67] Giora R. On the priority of salient meanings: studies of literal and figurative language. J Pragmat 1999:919-29. [68] Glucksberg S. The psycholinguistics of metaphor. Trends Cogn Sci 2003;7:92-6. [69] Blasko DG, Connine CM. Effects of familiarity and aptness on metaphor processing. J Exp Psychol Learn Mem Cogn 1993;19:295-308. [70] Benjamin J. A method for distinguishing and evaluating formal thinking disorders in schizophrenia. reprint 1964. In: & Kasanin J, editor. Language and thought in schizophrenia. New York: W. Norton; 1944. p. 65-90.

937

[71] Howanitz E, Cicalese C, Harvey PD. Verbal fluency and psychiatric symptoms in geriatric schizophrenia. Schizophr Res 2000;42:167-9. [72] Kerns JG, Berenbaum H, Barch DM, Banich MT, Stolar N. Word production in schizophrenia and its relationship to positive symptoms. Psychiatry Res 1999;87:29-37. [73] Green MF, Kern RS, Heaton RK. Longitudinal studies of cognition and functional outcome in schizophrenia: implications for MATRICS. Schizophr Res 2004;72:41-51. [74] Harvey PD, Green MF, Bowie C, Loebel A. The dimensions of clinical and cognitive change in schizophrenia: evidence for independence of improvements. Psychopharmacology (Berl) 2006;187:356-63. [75] Papagno C. Comprehension of metaphors and idioms in patients with Alzheimer's disease: a longitudinal study. Brain 2001;124:1450-60. [76] Reichenberg A. The assessment of neuropsychological functioning in schizophrenia. Dialogues Clin Neurosci 2010;12:383-92. [77] Nenadic I, Sauer H, Gaser C. Distinct pattern of brain structural deficits in subsyndromes of schizophrenia delineated by psychopathology. Neuroimage 2010;49:1153-60. [78] Reichenberg A, Caspi A, Harrington H, Houts R, Keefe RS, Murray RM, et al. Static and dynamic cognitive deficits in childhood preceding adult schizophrenia: a 30-year study. Am J Psychiatry 2010;167:160-9. [79] Adolphs R. The neurobiology of social cognition. Curr Opin Neurobiol 2001;11:231-9. [80] Dennis M, Lazenby AL, Lockyer L. Inferential language in highfunction children with autism. J Autism Dev Disord 2001;31:47-54. [81] Seeman MV. Symptoms of schizophrenia: normal adaptations to inability. Med Hypotheses 2007;69:253-7. [82] Brekke JS, Hoe M, Long J, Green MF. How neurocognition and social cognition influence functional change during community-based psychosocial rehabilitation for individuals with schizophrenia. Schizophr Bull 2007;33:1247-56. [83] Lysaker PH, Bell MD, Bioty SM. Cognitive deficits in schizophrenia. Prediction of symptom change for participators in work rehabilitation. J Nerv Ment Dis 1995;183:332-6.