You are on page 1of 6

Neuropsychologia 49 (2011) 203208

Contents lists available at ScienceDirect

Neuropsychologia
journal homepage: www.elsevier.com/locate/neuropsychologia

The executive control of attention differentiates patients with schizophrenia, their rst-degree relatives and healthy controls
Florence Breton a , Aurlie Plant b , Cindy Legauffre a , Nastassja Morel b , Jean Ads a,c , Philip Gorwood b , Nicolas Ramoz b , Caroline Dubertret a,b,c,
a b c

AP-HP, Department of Psychiatry, Louis Mourier Hospital, Colombes, France INSERM U675-U894, Center of Psychiatry and Neurosciences, Paris, France University Paris 7 Denis Diderot, Faculty of Medicine, Paris, France

a r t i c l e

i n f o

a b s t r a c t
Attentional and executive impairments have been reported in patients with schizophrenia and in their healthy rst-degree relatives. However, its nature remains unclear and discrepancies between studies have been observed. These might be due to differences in the clinical severity of the illness or in sociodemographic factors. The objective of the present work was to explore the efciency of three attention networks: alerting, orienting and executive control (conict inhibition) dened anatomically, using patients, their relatives and controls, assessing the possibility to use them as endophenotypes. We used three tests, the Attention Network Test (ANT), the Wisconsin Card Sorting Test (WCST) and the Stroop Test, and compared 52 patients with schizophrenia, 55 of their rst-degree relatives and 53 unrelated healthy controls, taking into account demographic variables (age, sex and years of education) and clinical symptoms of schizophrenia. Patients had a longer overall mean reaction-time (p < 0.001), and took longer to resolve the ANT conict (ANTc) (p = 0.04) than the control group. In the schizophrenia group, the SSPI disorganization score was signicantly correlated to the ANTc performance. Additionally, rst-degree relatives of patients with schizophrenia also performed signicantly worse than controls in attention performance test. Our ndings support a specic decit in executive control of attention in patients with schizophrenia. This decit was shown to be correlated with the intensity of the disorganization score in patients. Relative presented an intermediate phenotype between patients and controls; the ANT reaction time (but not the ANTc) may thus be considered as possible endophenotype marker for schizophrenia. 2010 Elsevier Ltd. All rights reserved.

Article history: Received 21 June 2010 Received in revised form 14 September 2010 Accepted 16 November 2010 Available online 21 November 2010 Keywords: Schizophrenia Attention Attentional network Executive dysfunction Endophenotype

1. Introduction Cognitive decits in executive performance, working memory and attention are considered to be core features in patients with schizophrenia, because these decits are present from the rst psychotic episode (Bilder, Goldman, Robinson, & Reiter, 2000; Ma et al., 2007). As they are also observed in relatives of patients, they are attractive as potential endophenotypes for studying the genetic background of schizophrenia (Gottesman & Shields, 1973; Leboyer, 2003). Most of the studies of neuropsychological decits in patients with schizophrenia (see meta-analyses by Heinrichs & Zakzanis, 1998; Johnson-Selfridge & Zalewski, 2001) and their rst-degree relatives (see meta-analyses by Snitz, MacDonald, & Carter, 2006; Szke et al., 2005) lead to the con-

Corresponding author at: Service de psychiatrie adulte, Hpital Louis Mourier, 178, rue des Renouillers, 92701 Colombes Cedex, France. Tel.: +33 1 47 60 64 16; fax: +33 1 47 60 67 40. E-mail address: caroline.dubertret@lmr.aphp.fr (C. Dubertret). 0028-3932/$ see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropsychologia.2010.11.019

vergent conclusion that these two populations display cognitive decits affecting the same cognitive functions (attention, executive functions and memory), and that less profound impairments of the same type are also observed in rst-degree relatives. The Stroop test and the Wisconsin Card Sorting Test (WCST) have both been shown to be particularly sensitive to these decits. However, there are discrepancies between studies investigating attentional and executive functions, which might be in part explained by differences in the inclusion criteria for patients (age at onset, duration and severity of illness, type of treatment received), for rst-degree relatives (age at inclusion, which might be related to differences in the risk of developing schizophrenia) and for controls (not identical compared to relatives, and with unknown psychiatric family history). In this study, we used homogenous strict inclusion criteria for patients, relatives and controls to compare the neuropsychological impairment of attention and of executive functions avoiding such inclusion criteria confonding factors. Attention is dened as reecting the efciency of three anatomically-distinct, organized networks (alerting, orienting and

204

F. Breton et al. / Neuropsychologia 49 (2011) 203208

executive control) that are widely distributed across frontal, parietal and thalamic sites (Posner & Peterson, 1990). The alerting component involves the capacity to sustain an alert cognitive state and can be assessed with continuous performance and vigilance task, whereas the orienting network involves the focused identication and selection of sensory information and can be tested with cue task. The executive control of attention governs the capacity to decide among conicting responses and to give a response to one aspect of a stimuli by ignoring a dominant aspect. Several previous studies have found a decit in one of the alerting, orienting and executive control components of attention in schizophrenia using different employed experimental manipulations. An approach of attention impairment has been developed to assess the three attention network within one test. The attention network test (ANT) is a recent paradigm designed to investigate attention efciencies of alerting, orienting, and executive control of attention (Fan, McCandliss, Sommer, Raz, & Posner, 2002), on the basis of a concept of an integrative selective attention system (Posner & Peterson, 1990). It is a combination of a cued reaction time task (Posner, 1980) and a anker task (Eriksen & Eriksen, 1974). Validation study in a healthy population showed sufcient independence of the three networks and gave reliable single subject estimates of alerting, orienting and executive function. In particular, the assessment of executive functions seems to reveal robust conict effects reected in the reaction time (RT) (Fan et al., 2002). In healthy subjects, orienting is not inuenced by age, executive function remained stable after the age of seven and alerting is improved with age (Rueda et al., 2004). Given prior ndings of impaired performances by schizophrenia patients on the measure of executive functioning and attention with the Wisconsin Card Sorting Test (WCST) (Grant & Berg, 1948; Heaton, 1981) and the Stroop Word-Color Test (Stroop WC) (Stroop, 1935), we selected these two well established tests as external validated tests in our samples. Wang et al. (2005) rstly administrated ANT to a large sample of inpatients with schizophrenia compared to controls and found that schizophrenia patients showed impairments of the executive and orienting networks. Gooding, Braun, and Studer (2006) also found an increased ANT conict (ANTc) effect in schizophrenia outpatients, with no signicant difference for the alertness and orientation metrics compared to controls, suggesting a specic decit of the executive control of attention in schizophrenia. Other studies have produced controversial results and have reported signicantly lower conict effect scores in individuals with schizophrenia (Neuhaus et al., 2007; Opgen-Rhein et al., 2008), or similar levels of executive controls (Nestor et al., 2007; Urbanek et al., 2009), suggesting discrepancies with the ANT studies in schizophrenia. Schizophrenia is a heterogeneous disorder that is likely the result of diversity in the underlying pathophysiological processes in different forms of schizophrenia. Factorial analytic studies lead to a tri-syndromic model of schizophrenia, including a negative, a positive and a disorganization factor (Liddle, 1987). Each syndrome has been associated with distinct patterns of neuropsychological decits and a distinct pattern of aberrant cerebral activity. For example, thought disorders, a cardinal feature of disorganization, can inuence performance in attention-dependent functions (Braff & Light, 2004). Some studies report an association between the disorganized syndrome and defective performance in tasks involving executive and attentional functions (Basso, Nasrallah, Olson, & Bornstein, 1998; Daban et al., 2003; Ngan & Liddle, 2000). These associations have been consistently demonstrated in patients with persistent symptoms, but may differ in patients with remitting illness. It seems possible that the inconsistencies between previous studies may be attributable, at least in part, to differences in socio-demographic

factors or in the clinical severity of the illness between patient samples. Furthermore, the efciency of the executive control of ANT was previously found to be highly heritable (Fan, Wu, Fossella, & Posner, 2001), which may make it an attractive potential endophenotype in schizophrenia. The present study thus evaluated for the rst time whether the executive control of attention decit measured using ANT is characteristic of a specic liability to schizophrenia or may represent an endophenotype. In this study, we tested the hypothesis that patients with schizophrenia and their rst-degree relatives would perform less well in the conict task and/or display longer reaction times (RTs) than healthy controls. To do this, we rst compared executive control of attention using the Attention Network test in patients affected with schizophrenia and in a group of healthy volunteers to nd out whether these impairments constitute a feature of schizophrenia. We then compared the performance of their healthy rst-degree relatives to that of healthy controls to nd out whether executive control of attention, like that of other executive functions or attention, has a familial component. We also evaluated whether the decits are related to variations on clinical characteristics of patients, such as symptoms scores, age, duration of the illness, age at onset and other decits known to involve executive function.
2. Methods 2.1. Participants We recruited 52 outpatients with schizophrenia, 55 healthy rst-degree relatives, and 53 healthy controls from a psychiatric department of a French teaching hospital in Paris suburb. A total of 160 subjects were evaluated by a trained psychiatrist using the Diagnostic Interview for Genetic Studies (DIGS) (Nurnberger et al., 1994), a semi-structured interview leading to a lifetime diagnosis according to the DSM-IV criteria for schizophrenia and other psychiatric diseases. Past family history was obtained from each proband and from at least one rst-degree relative using a semi-structured interview, the Family History-Research Diagnostic Criteria (FHRDC) (Andreasen, Endicott, Spitzer, & Winokur, 1977), which assesses psychiatric morbidity for all rst-degree relatives. Inclusion for patients with schizophrenia required a combination of criteria including (1) having a stable state with no change in medication or symptoms for at least 4 weeks before the cognitive evaluation, (2) being treated exclusively with atypical antipsychotics, with no anticholinergic agents or mood stabilizers, (3) having had a relatively short course of illness (less than 10 years). Clinical symptoms were assessed using the positive and negative syndrome scale (PANSS) (Kay, Fiszbein, & Opler, 1987) and Signs and Symptoms of Psychotic Illness (SSPI) (Liddle, Ngan, Dufeld, Kho, & Warren, 2002). A disorganization dimension score was based on four SSPI criteria (attention impairment, affective attering, poverty of speech and formal thought disorder) (Liddle et al., 2002). All patients are clinically stable as assessed with the PANSS and fully able to cooperate with testing. Controls and relatives were included only if they were between 28 and 65 years in age, to minimize the risk that they would go on to develop schizophrenia. The control sample was recruited through partners of patients admitted in the psychiatric department. Control subjects had no rst-degree family history of bipolar or schizophrenia disorder. All subjects (patients, relatives and controls) were without any history of current substance abuse, brain injury or neurological disease, medical condition or medication known to be associated with neuropsychological impairment. All participants had normal or corrected-to-normal vision. All recruited subjects were euthymic, as evaluated by the Montgomery and Asberg Depression Rating Scale (MADRS) (Montgomery & Asberg, 1979) and the Young Mania Rating Scale (Young, Biggs, Ziegler, & Meyer, 1978), and they had no mental retardation as evaluated by the NART (National Adult Reading Test) (Bright, Jaldow, & Kopelman, 2002). Educational level distinguished 3 groups: (1) less than high school, (2) high school graduate and (3) some college or higher education. Patients, relatives and healthy volunteers all gave their written informed consent after hearing a complete description of the study. The study protocol was approved by the French Ethics and Data Protection and Freedom of Information Commissions (Comit National dEthique and the Commission Nationale Informatique et Libert). 2.2. Procedure Patients, relatives and controls administered three neurological tests (ANT, WCST, Stroop test) that included an assessment of attention and executive functions.

F. Breton et al. / Neuropsychologia 49 (2011) 203208 The Attention Network was originally developed to assay the three independent attention networks, namely the alerting (ANTa), orienting (ANTo), and executive control (ANTc) networks, within a single experiment. This is a computerized test, which is easy to use and assesses ve metrics: alertness, orientation, conict, median reaction time (ANT-RT), and percentage of correct answers (ANT%). The stimuli consist of a row of ve visually-presented, white lines, with arrowheads pointing leftward or rightward, against a black background; the target is a leftward or rightward arrowhead at the center. This target is anked on either side by two arrows pointing in the same direction (congruent condition), by two arrows in the opposite direction (incongruent condition), or by 2 lines (neutral condition). The participants task is to indicate the direction of the central target by right- or leftclicking the mouse as quickly as possible. Cues console consists of a 100 ms asterisk presented 400 ms before the target. There were four cue conditions: no cue, central cue which appear at the central xation point, double-cue in which two warming cues correspond to two possible target positions above and below the central xation point, and spatial cue which was presented on the exact target localization. The effect of alerting cue was calculated by subtracting the mean reaction times (RT) of the condition with double-cue from the mean RT of the condition with no cue. The orienting effect was calculated by subtracting the mean reaction times (RT) of the condition with spatial-cue from the mean RT of the condition with center-cue. The executive effect was calculated by subtracting the mean reaction times (RT) of the condition with congruent ankers from the mean RT of the condition with incongruent ankers. For all participants, neuropsychological test scores were available on the Wisconsin Card sorting test (WSCT; Heaton, 1981) that was a well establish measure of executive function, and the card version of the Stroop test employed by Stroop (1935) that involved conict network of attention. The Stroop Word/Color interference test measures selective attention (Stroop, 1935). This task tests the ability to attend only to the color in which a word is written while ignoring the distractor, which is what the word means. Participants are presented with a series of cards containing 100 stimuli from one condition and are asked to respond to each stimulus without stopping. The total time per card is the measure of performance in a given condition. The subject is asked rst to read the names of colors written in black ink (rst condition), then is asked to name the color of noncolor words written in colored ink (second condition). The third condition elicits what is called the Stroop interference effect, where the subject must name the color of the ink in which the names of various colors are written while ignoring the word. We determined the Color-word interference score (Stroop WC), which is the latency for naming all of the colors correctly in this third condition minus the time need of the second condition. The Wisconsin Card Sorting Test (WCST) was used to assess executive processes. This task includes four stimulus cards and 128 response cards that contain various gures. The cards vary for color (red, yellow, blue, green), type of gure (crosses, circles, triangles or stars), and number of gures (one to four). The test was discontinued after the completion of six categories or when no more response card remained. The chosen variables were the numbers of perseverative errors (WCST-PE), which reects the tendency towards perseveration, the number of nonperseverative errors (WCST-NPE) and the number of categories completed (WCST-NC) given the number of times that 10 consecutive correct responses were made, reecting overall success.

205

3. Results 3.1. Demographic characteristics of the sample We recruited 52 outpatients with DSM-IV criteria for schizophrenia who had neuropsychological exam results out of 210 patients that did not met inclusion (duration of the illness longer than 10 years or treatment exclusion criteria). Thirty-three patients (63%) had prominent positive symptoms (paranoid type), 9 (17%) showed mainly negative symptoms (disorganized type), 10 (20%) exhibited both positive and negative symptoms. Patients received atypical antipsychotic medication consisting of Risperidone (48%), Olanzapine (19%), Aripiprazole (14%) and Clozapine (19%). The mean age at onset was 23.1 years (6.5), the mean duration of the illness was 7.2 years (3.4), the mean chlorpromazine equivalent daily dose was 323.46 mg (120.26) and the mean number of hospitalizations was three (between 1 and 10 hospitalizations). The mean PANSS score was 76.3 (18.9), and a mean total SSPI score was 19.9 (6.6). We recruited 55 rst-degree relatives (30 parents and 25 siblings), and 53 healthy controls, with symmetric and identical inclusion criteria; they were free of any psychiatric disease (DSM-IV axis I), and any condition known to affect cognitive function. The demographic characteristics of the three samples are shown in Table 1. All patients except one in each group were right handedness. Patients were younger than their relatives and than the healthy control subjects (p = 0.009 and p < 0.0001, respectively). There was a higher proportion of women in the control and relative groups than in the patient group (p = 0.001 and p < 0.0001, respectively). Controls had a higher level of education (college or higher education) than the patient and relative groups (respectively, p = 0.0003 and p = 0.02). 3.2. Cognitive differences between groups The results of the neuropsychological tests for the patients, healthy relatives and controls are shown in Table 2 and Fig. 1. A oneway ANOVA revealed signicant differences between these three groups for the ANTc, ANT-RT, WCST-NPE, WCST-PE and Stroop WC tests. To verify conict effects independent of main RT, we calculated attention network ratios as respective ANTc effect/mean RT and the comparison mean of the ration remained signicantly different between groups (data not shown). Analyses controlling for the inuence of sex, level of education and age at interview were thus performed between groups for data displaying signicant global p values. Patients had longer overall mean reaction times (p = 0 <0.001), and took longer to resolve the ANTc conict (signicantly less efcient executive attention) (p = 0.044) than the control group, whereas effects on the alerting and orienting networks did not differ between the groups. As expected, patients performed worse in tests of selective attention (Stroop WC, p = 0.002), and made signicant more category and perseverative errors than controls in the WCST-NC (p = 0.042), WCST-PE (p = 0.014) and WCST-NPE (p = 0.023). Signicant differences were found between patients and their healthy rst-degree relatives only for ANTc, ANT-RT, ANT% and Stroop WC (Table 2). First-degree relatives also differed from the control groups in their overall mean ANT reaction time (p = 0.021) and their selective attention performance (Stroop WC, p < 0.0001), however no such difference was observed in any part of the WCST test except WCST-NPE. Correlation analyses were conducted for the rst time in this study to explore the linear relationship (for the whole group of subjects and patients) between ANT performance and other

2.3. Statistical analyses Patients, relatives and healthy controls were compared using chi-square ( 2 ) analyses for sociodemographic distributions. Continuous variables were compared between the three groups using a univariate analysis of variance (one way ANOVA). For all the cognitive tasks showing signicant global between group differences, the difference in executive and attentional performances were determined by bilateral comparisons of interest between the groups of patients with schizophrenia and healthy controls, between the groups of patients and their respective relatives, and between the groups of relatives and controls. We rst compared the Attention Network test performance in patients affected with schizophrenia and in healthy volunteers to nd out whether the impairments constitute a feature of schizophrenia. We then compared the performance of their healthy rst-degree relatives to that of healthy controls to nd out whether attention network, like that of other executive functions or attention, has a familial component. Given that groups differed signicantly on demographic variables, an adjustment was done on gender; age at interview and level of education for all the inter-group comparisons using analyses of covariance (ANCOVA). Adjusted p values are presented in the tables and text. Pearson correlations were performed to analyze the relationship between ANT performance (as a continuous variable) and WCST or Stroop WC performance, the severity of schizophrenia symptoms (PANSS scores and SSPI disorganization scores), or clinical characteristic that might inuence cognitive performance (duration of illness, age at onset, medication level). Difference were considered to be signicant if p < 0.05. Statistical analyses were carried out using the SPSS for Windows (version 17.0). We initially computed the sample size required for ANT-RT. Opgen-Rhein et al. (2008) found a higher ANT-RT in schizophrenia (644 130) compared to controls (561 85). This indicated that the required sample size for each group would be 22 (with an alpha risk of 5% and a power of 80%).

206

F. Breton et al. / Neuropsychologia 49 (2011) 203208

Table 1 Demographic and clinical characteristics of the three samples. Controls N = 53 Age, years (mean, SD) Age at onset (mean, SD) PANSS pos (mean, SD) PANSS neg (mean, SD) PANSS gen (mean, SD) PANSS tot (mean, SD) SSPI tot (mean, SD) Antipsychotic dose (mg/CPZeq/day) 43.77 (12.18) Controls N = 53 Sex (% men) Educational level (%) Less than high school High school graduate Some college or higher 36 8 28 64 Relatives N = 55 46.15 (13.43) Relatives N = 55 25 42 16 42 Patients with schizophrenia N = 52 32.65 (9.68) 23.07 (6.45) 16.02 (6.47) 26.02 (7.86) 36.70 (8.31) 76.27 (18.88) 19.90 (7.79) 323.46 (120.76) Patients with schizophrenia N = 52 69 11 60 29 F (df = 2) 19.37 p <0.001

(df = 2)

p <0.001 <0.001 <0.001 0.001

22.59 23.25 23.46 13.56

PANSS pos: PANSS positive symptom subscale; PANSS neg: PANSS negative symptom subscale; PANSS gen: PANSS general symptom subscale; PANSS tot: PANSS total score; SSPI: signs and symptoms of psychotic illness total score; CPZeq: chlorpromazine equivalent. Table 2 Mean score values (SD) in the neurocognitive tests used for patients, relatives and controls. Mean (SD) Controls (N = 53) Relatives (N = 55) Patients with schizophrenia (N = 52) ANOVA between-groups ANCOVA between two-group differences

F (df = 2)

Schizophrenia versus controls pa 0.045 <0.001 <0.001 0.042 0.014 0.023 0.002

Schizophrenia versus relatives pa 0.031 0.005 0.001 0.663 0.444 0.061 0.054

Relatives versus controls pa 0.590 0.021 0.057 0.128 0.187 0.020 <0.001

ANTa (ms) ANTo (ms) ANTc (ms) ANTRT (ms) ANT (%) WCST-NC WCST-PE WCST-NPE Stroop WC

34.73 (26.63) 44.08 (27.01) 152.02 (36.96) 704.96 (118.35) 97.36 (4.36) 5.91 (0.45) 8.66 (5.71) 7.72 (5.33) 14.79 (4.92)

37.82 (30.45) 43.80 (35.09) 166.93 (59.80) 776.87 (129.07) 94.44 (8.01) 5.58 (0.97) 12.57 (10.07) 9.82 (7.25) 29.07 (14.86)

30.58 (39.02) 38.94 (42.49) 203.13 (110.31) 807.46 (187.21) 87.06 (16.00) 5.35 (1.11) 16.71 (12.36) 11.59 (7.36) 33.48 (30.73)

0.64 0.35 6.36 6.62 13.11 5.11 8.87 4.35 12.61

0.53 0.706 0.002 0.002 <0.001 0.007 <0.001 0.015 <0.001

WCST: Wisconsin card sorting test; WCST-PE: number of perseverative errors; WCST-NPE: number of nonpersevarative errors; WCST-NC: number of categories completed; Stroop WC: Color-word interference score. The attention network test (ANT) assesses alertness (ANTa), orientation (ANTo), conict (ANTc), median reaction time (ANT-RT), and percentage of correct answers (ANT%). a Adjusted for sociodemographic variables (sex, educational level, age at interview).

test that measured the same executive and attention function (WCST, Stroop WC). Pearson correlations revealed strong correlations (all p < 0.001) between ANTc conict performance or ANT-RT and WCST-NC, WCST-PE, WCSTNPE and Stroop WC.

3.3. Clinical correlation We examined the relation of the attention network tests with clinical symptom scores, age at onset, duration of the illness, treat-

Fig. 1. ANT conict (ANTc), median reaction time (ANT-RT) and Stroop WC performances in patients, relatives and controls.

F. Breton et al. / Neuropsychologia 49 (2011) 203208

207

ment daily dose in schizophrenia patients. We found a positive correlation between the poorer performance in the ANTc and the disorganization score (from the SSPI scale) (r = 0.47, p = 0.01). No correlation was found between cognitive impairments and PANSS scores (positive, p = 0.20; negative, p = 0.23; total, p = 0.59, respectively, for ANTc particularly). There was also no evidence of correlation among ANTc and illness duration, age at onset or antipsychotic treatment level (p = 0.141, p = 0.30; p = 0.67, respectively).

4. Discussion The main ndings of our study were (1) the identication of a specic decit in the executive control of attention (ANTc) in schizophrenia, which seem to be directly attributable to the illness itself, and particularly to the severity of the disorganization, and (2) validation of the Stroop WC test and ANT overall reaction times as probable endophenotype markers for genetic vulnerability to schizophrenia. In our sample, the patients with schizophrenia had a relatively short illness follow-up time and were all being treated with atypical antipsychotics which are supposed to have benecial effect on cognition (Harvey, Rabinowitz, Eerdekens, & Davidson, 2005; Volavka et al., 2002) with no other drugs that affect cognition. They constituted a relatively homogenous group of outpatients with schizophrenia, and displayed an overall decit compared to controls in executive functions measured using the ANT, WCST and Stroop tests. The ANTc performance was positively correlated with the WCST and Stroop WC performance, which further validate the fact that it is executive functions which were assessed in our sample. As a result of the inclusion criteria, the patients with schizophrenia were younger when interviewed than either the control or relatives groups. However, ANTc performance was not inuenced by age after seven (Rueda et al., 2004). The sex ratio was different in the different groups, and the control group had received more years of education than either the patient or relative groups. To take these differences into account, our results were reported before and after controlling for these three confounding demographic variables. This suggested that the decits observed in patients with schizophrenia could be attributed directly to the illness, rather than to these variables that might affect cognition performance. We conrmed that patients had a specic attention decit, and performed signicantly less well in their executive control attention network tests (ANTc and ANT-RT), whereas their alerting (ANTa) and orienting (ANTo) network performances did not differ from those of the healthy control subjects. Our ndings are consistent with several previous studies using conventional executive tests (Heinrichs & Zakzanis, 1998; Johnson-Selfridge & Zalewski, 2001), and with previous studies using ANT in particular (Gooding et al., 2006; Wang et al., 2005). However, a study focusing on male patients only (Nestor et al., 2007; Opgen-Rhein et al., 2008) or stratied for gender (Urbanek et al., 2009), and another focusing on less symptomatic patients measured by the PANSS (Neuhaus et al., 2007) did not report these positive results. Sex differences and residual psychotic productivity might potentially explain these conicting results for the attention network. Interestingly, our results remained signicant after controlling for the effect of sex on performance in the test. Another explanation for the discrepancy has been proposed by Nestor et al. (2007) and Urbanek et al. (2009). The monitoring failures in their schizophrenia patients could have masked executive control impairments in patients with older age and longer duration of illness compared to our patients. Moreover, we found a signicant correlation between the ANTc score and disorganization score but not PANSS scores. In our sample, schizophrenia patients decit in executive function

attention network is related to abnormal performance of other task involving executive functioning processes (Stroop WC and WCST) and some of the characteristic symptoms of the disorder, such as positive formal thought disorder and loose association. Disorganization is one of the cardinal features of schizophrenia (Andreasen & Flaum, 1991; Liddle, 1987), and differing intensity of these symptoms in the schizophrenia patients may explain the inconsistent ndings in studies that failed to take the inuence of disorganization symptoms into account. This study demonstrated that rst-degree relatives of patients with schizophrenia share some of the attention decits observed in schizophrenia. First-degree relatives performed signicantly less well than healthy controls for the Stroop WC test and ANT reaction time, therefore they appear as potential schizophrenia-associated intermediate phenotype. In contrast, the effect of conict on attention appeared to be illness decit. We thus demonstrated for the rst time a familial resemblance for two tests assessing the executive function of attention, but not for the WCST test in families of patients with schizophrenia. The WCST results for executive dysfunction are partially inconsistent with those of two previous meta-analyses (Snitz et al., 2006; Szke et al., 2005). Nevertheless, here we used strict and identical inclusion criteria for the healthy relatives and control groups, and excluded all moderator variables known to affect cognitive function, such as psychiatric disease, somatic disease or medical treatment known to severely affect cognition. Moreover, the healthy relatives had all passed the peak age for the risk of schizophrenia. One other plausible reason that could more probably explain the lack of any signicant difference in our study is its statistical power, which may account for the detection of familial resemblance in schizophrenia families. Nevertheless, signicant results were found for the Stroop WC test and ANT-RT, which all involved selective attention, even though the same sample size was used. Further studies are required to replicate these results in larger samples. In conclusion, the ANT is particularly interesting for the study of executive and attentional processes in patients with schizophrenia. Firstly, it reveals that schizophrenia patients seem to have a specic executive attention decit, in accordance with other studies (Gooding et al., 2006; Wang et al., 2005). The executive control network orchestrates the capacity to decide between conicting or competing responses based upon a principle or goal. Interestingly, functional imaging studies of healthy controls conducted during the performance of ANT conrmed the anatomical and functional independence of the alerting, orientating and conict resolution (executive) functions of attention, which activate distinct cortical and subcortical anatomical networks (Fan et al., 2002; Fan, McCandliss, Fossella, Flombaum, & Posner, 2005). Particularly, these studies suggest that the midline and lateral frontal areas of the cortex are activated, especially the anterior cingulate cortex, and that dopamine may be implicated in modulating these areas during executive control tasks (Ingwar & Franzen, 1974). Moreover, the conict condition of the ANT could be used in genetic studies. Fan et al. (2001) demonstrated that the executive control measured by the ANT is heritable (89%), while lower levels of heritability were observed for the alerting and median reaction times, 18% and 16%, respectively. The ANT is an attractive endophenotypes for schizophrenia because it assesses easily distinct components of attention, each of which activate specic cortical and subcortical brain regions and neurotransmitter systems (Fan et al., 2002, 2005). Interestingly, the implication of prefrontal areas in the performance of the executive attention network could be used to identify a candidate gene that affects dopaminergic signaling. For example, one single nucleotide polymorphism (rs1800497/TaqIA) of the ANKK1 gene (located downstream from the DRD2 gene) has been associated both with the executive control of ANT (Fossella, Green, & Fan, 2006) and

208

F. Breton et al. / Neuropsychologia 49 (2011) 203208 Heinrichs, R. W., & Zakzanis, K. (1998). Neurocognitive decit in schizophrenia: A quantitative review of the evidence. Neuropsychology, 3, 426445. Ingwar, D. H., & Franzen, G. (1974). Abnormalities of cerebral blood ow distribution in patients with chronic schizophrenia. Acta Psychiatrica Scandinavia, 50, 425462. Johnson-Selfridge, M., & Zalewski, C. (2001). Moderator variables of executive functioning in schizophrenia: Meta-analytic ndings. Schizophrenia Bulletin, 27(2), 305316. Kay, S. R., Fiszbein, A., & Opler, L. A. (1987). The Positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia Bulletin, 13(2), 261276. Leboyer, M. (2003). Searching for alternative phenotypes in psychiatric genetics. Methods in Molecular Medicine, 77, 145161. Le Strat, Y., Ramoz, N., & Gorwood, P. (2009). The role of genes involved in neuroplasticity and neurogenesis in the observation of a gene-environment interaction (GxE) in schizophrenia. Current Molecular Medicine, 9(4), 506518. Liddle, P. F., Ngan, E. T. C., Dufeld, G., Kho, K., & Warren, A. J. (2002). Signs and symptoms of psychotic illness (SSPI): A rating scale. British Journal of Psychiatry, 180, 4550. Liddle, P. F. (1987). The symptoms of chronic schizophrenia. A re-examination of the positivenegative dichotomy. British Journal of Psychiatry, 151, 145151. Ma, X., Wang, Q., Sham, P. C., Liu, X., Rabe-Hesketh, S., Sun, X., et al. (2007). Neurocognitive decits in rst-episode schizophrenic patients and their rst-degree relatives. American Journal of Medical Genetic B: Neuropsychiatry Genetic, 144B(4), 407416. Montgomery, S. A., & Asberg, M. (1979). A new depression scale designed to be sensitive to change. British Journal of Psychiatry, 134, 382389. Nestor, P. G., Kubicki, M., Spencer, K. M., Niznikiewicz, M., McCarley, R. W., & Shenton, M. E. (2007). Attentional networks and cingulum bundle in chronic schizophrenia. Schizophrenia Research, 90(13), 308315. Neuhaus, A. H., Koehler, S., Opgen-Rhein, C., Urbanek, C., Hahn, E., & Dettling, M. (2007). Selective anterior cingulate cortex decit during conict solution in schizophrenia: An event-related potential study. Journal of Psychiatric Research, 41(8), 635644. Ngan, E. T. C., & Liddle, P. F. (2000). Reaction time, symptom prole and course of illness in schizophrenia. Schizophrenia Research, 46, 195201. Nurnberger, J. I, Blehar, M. C., Kaufmann, C. A., York-Cooler, C., Simpson, S. G., Harkavy-Friedman, J., et al. (1994). Diagnostic interview for genetic studies. Rationale, unique features, and training. NIMH genetics initiative. Archives of General Psychiatry, 51(11), 849859. Opgen-Rhein, C., Neuhaus, A. H., Urbanek, C., Hahn, E., Sander, T., & Dettling, M. (2008). Executive attention in schizophrenic males and the impact of COMT Val108/158Met genotype on performance on the attention network test. Schizophrenia Bulletin, 34(6), 12311239. Posner, M. I. (1980). Orienting of attention. The Quarterly journal of experimental psychology, 32(1), 325. Posner, M. I., & Peterson, S. E. (1990). The attention system of the human brain. Annual Review of Neurosciences, 13, 2542. Rueda, M. R., Fan, J., McCandliss, B. D., Halparin, J. D., Gruber, D. B., Lercari, L. P., et al. (2004). Development of attentional networks in childhood. Neuropsychologia, 42, 10291040. Snitz, B. E., MacDonald, A. W., III, & Carter, C. S. (2006). Cognitive decits in unaffected rst-degree relatives of schizophrenia patients: A meta-analytic review of putative endophenotypes. Schizophrenia Bulletin, 32, 179194. Stroop, J. R. (1935). Studies of interference in series verbal reactions. Schizophrenia Bulletin, 18, 643662. Szke, A., Schrhoff, F., Mathieu, F., Meary, A., Ionescu, S., & Leboyer, M. (2005). Tests of executive functions in rst-degree relatives of schizophrenic patients: A meta-analysis. Psychology Medicine, 35, 771782. Urbanek, C., Neuhaus, A. H., Opgen-Rhein, C., Strathmann, S., Wieseke, N., Schaub, R., et al. (2009). Attention network test (ANT) reveals gender-specic alterations of executive function in schizophrenia. Psychiatry Research, 168, 102109. Volavka, J., Czobor, P., Sheitman, B., Lindenmayer, J. P., Citrome, L., McEvoy, J. P., et al. (2002). Clozapine, olanzapine, risperidone, and haloperidol in the treatment of patients with chronic schizophrenia and schizoaffective disorder. American Journal of Psychiatry, 159, 255262. Wang, K., Fan, J., Dong, Y., Wang, C. Q., Lee, T. M., & Posner, M. I. (2005). Selective impairment of attentional networks of orienting and executive control in schizophrenia. Schizophrenia Research, 78, 235241. Young, R. C., Biggs, J. T., Ziegler, V. E., & Meyer, D. A. (1978). A rating scale for mania: Reliability, validity and sensitivity. British Journal of Psychiatry, 133, 429435.

with schizophrenia (Dubertret et al., 2004, 2010; Le Strat, Ramoz, & Gorwood, 2009). This constitutes the next challenge, and studies following on from this neurocognitive study in schizophrenia should include both cognitive tests and genetic investigations. Acknowledgments INSERM (RBM99003) has supported this work. We thank Monika Ghosh for the English editing of the manuscript. References
Andreasen, N. C., Endicott, J., Spitzer, R. L., & Winokur, G. (1977). The family history method using diagnostic criteria. Reliability and validity. Archives of General Psychiatry, 34(10), 12291235. Andreasen, N. C., & Flaum, M. (1991). Schizophrenia: The characteristic symptoms. Schizophrenia Bulletin, 17, 2749. Basso, M. R., Nasrallah, H. A., Olson, S. C., & Bornstein, R. A. (1998). Neuropsychological correlates of negative, disorganized and psychotic symptoms in schizophrenia. Schizophrenia Research, 31, 99111. Bilder, R. M., Goldman, R. S., Robinson, D., & Reiter, G. (2000). Neuropsychology of rst-episode schizophrenia: Initial characterization and clinical correlates. American Journal of Psychiatry, 157, 549559. Braff, D. L., & Light, G. A. (2004). Preattentional and attentional cognitive decits as targets for treating schizophrenia. Psychopharmacology, 174, 7585. Bright, P., Jaldow, E., & Kopelman, M. D. (2002). The national adult reading test as a measure of premorbid intelligence: A comparison with estimates derived from demographic variables. Journal of International Neuropsychological Society, 8, 847854. Daban, C., Amado, I., Gut, A., Leonardi, S., Willard, D., Bourdel, M. C., et al. (2003). Disorganisation syndrome is correlated to working memory decits in unmedicated patients with recent onset schizophrenia. Schizophrenia Research, 61, 323324. Dubertret, C., Gouya, L., Hanoun, N., Deybach, J., Ades, J., Hamon, M., et al. (2004). The 3 region of the DRD2 gene is involved in genetic susceptibility to schizophrenia. Schizophrenia Research, 67, 7585. Dubertret, C., Bardel, C., Ramoz, N., Martin, P. M., Deybach, J. C., Ads, J., et al. (2010). A genetic schizophrenia-susceptibility region located between the ANKK1 and DRD2 genes. Progress in Neuro-psychopharmacology & Biological Psychiatry, 34(3), 492499. Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon the identication of a target letter in a nonsearch task. Perception and Psychophysics, 16, 143149. Fan, J., McCandliss, B. D., Fossella, J., Flombaum, J. I., & Posner, M. I. (2005). The activation of attentional networks. Neuroimage, 26, 471479. Fan, J., McCandliss, B. D., Sommer, T., Raz, A., & Posner, M. I. (2002). Testing the efciency and independence of attentional networks. Journal of Cognitive Neuroscience, 14, 340347. Fan, J., Wu, Y., Fossella, J., & Posner, M. I. (2001). Assessing the heritability of attention networks. BMC Neuroscience, 2, 14. Fossella, J., Green, A. E., & Fan, J. (2006). Evaluation of a structural polymorphism in the ankyrin repeat and kinase containing 1 (ANKK1) gene and the activation of executive attention networks. Cognitive Affective and Behavioral Neuroscience, 6, 7178. Gooding, D. C., Braun, J. G., & Studer, J. A. (2006). Attentional network task performance in patients with schizophrenia-spectrum disorders: Evidence of a specic decit. Schizophrenia Research, 88(13), 169178. Gottesman, I. I., & Shields, J. (1973). Genetic theorizing and schizophrenia. British Journal of Psychiatry, 122, 1530. Grant, D. A., & Berg, E. A. (1948). A behavioral analysis of degree of reinforcement and ease of shifting to new responses on a Weigl type card sorting problem. Journal of Experimental Psychology, 38, 404411. Harvey, P. D., Rabinowitz, J., Eerdekens, M., & Davidson, M. (2005). Treatment of cognitive impairment in early psychosis: A comparison of risperidone and haloperidol in a large long-term trial. American Journal of Psychiatry, 162, 18881895. Heaton, R. K. (1981). Wisconsin card sorting test. Odessa, FL: Psychological Assesment Resources.