PropertyValue
is nif:broaderContext of
nif:broaderContext
is schema:hasPart of
schema:isPartOf
nif:isString
  • The study was approved by the Grampian Local Research Ethics Committee. Potential patients and controls were given the Ethics approved Information Sheet and encouraged to discuss the study with others and take several days to decide on any questions. After a few days subjects were invited to meet with one of the researchers (JDS) and a discussion determined whether they understood the nature of the study and if they had further questions which were then answered. If they understood the task, wished to participate and met the inclusion criteria they were recruited. Written informed consent was obtained from all participants. The consent form was signed by the participants themselves. Data was acquired from two groups of subjects: a group of 15 patients with DSM IV schizophrenia and a group of 20 healthy controls. Exclusion criteria were any neurological disorder, claustrophobia, or other DSM IV Axis I or II diagnosis. Patients were recruited via NHS Consultant Psychiatrists from their Community Mental Health Teams (CMHTs). All were outpatients in long term follow up at the time of scanning with stable chronic symptomatic illness despite on-going antipsychotic treatment. No advertisements were used for recruitment. All patients had been diagnosed with schizophrenia by Consultant Psychiatrists at least 2 years prior to recruitment and in many cases had been diagnosed decades earlier. For patients with a primary diagnosis of schizophrenia in long term NHS follow up, there is often significant comorbidity, particularly mood disorder and substance misuse. No patients satisfied criteria for a depressive illness at the time of scanning and no patients had a significant problem with substance misuse. With the exception of comorbidity, patients were typical of those seen in NHS outpatients. Four control and two patient data sets were excluded because of structural brain abnormalities, failure to understand the task or scanner hardware failure. Sixteen controls and thirteen schizophrenia patients were finally included in the analysis. The two groups did not differ on a between group t-test with respect to age and National Adult Reading Test estimated pre-morbid IQ (Nelson and Wilson, 1991). Given the smaller proportion of females in the schizophrenia group than in the control group, gender was used as a covariate for the behavioural and image analyses. Details of subjects included in the analysis are presented in Table 1. Table S1 describes patient's antipsychotic medication at the time of the study. Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0042608.t001 Patient and control details. Values are mean ± DS; NART, National Adult Reading Test; BDI, Beck depression inventory; SP, Spielberg anxiety scale; RSES, Rosenberg Self-Esteem Scale; PANSS, Positive and Negative Syndrome Scale; (*) significant difference between groups. For blood oxygen level dependent (BOLD) response imaging, T2* weighted gradient echo planar images were obtained using a GE Medical Systems Signa 1.5 T MRI scanner. A total of 30 axially orientated 5 mm thick contiguous sequential slices were obtained for each volume, 244 volumes being obtained with a TR of 2.5 s, TE 30 ms, flip 90°, FOV 240 mm and matrix 64×64. The first four volumes were discarded to allow for transient effects. A T1 weighted image was obtained to exclude gross structural brain abnormality. Subjects performed a version of the ‘Cyberball’ social exclusion task whilst being scanned [19]. In this task, subjects play a ball passing game with two cartoon animated figures on a screen, with the subject being represented by an animated hand (Fig. S1). Subjects were instructed to press either of two buttons to pass the ball to one of the cartoon figures. In turn, each cartoon figure either passed the ball to the subject or passed it to the other cartoon figure. Throughout the task, the extent to which the subject was excluded in the game (ball not being passed to the participant) was systematically varied from 0% (ball equally shared between all three players) to 100% (ball only passed between the two animated figures). Specifically, the task was divided into 17 blocks with the following percentage levels of exclusion: 0, 25, 50, 75, 100, 75, 50, 25, 0, 25, 50, 75, 100, 75, 50, 25 and 0. As in previous studies, the behaviour of the two figures was driven by a computer program and the catching actions were performed automatically. Participants had a short training session with Cyberball before playing the task in the scanner. They were instructed ‘when you receive the ball, just pass it back’. Subjects were not told that the object of the game was to study the effects of varying social exclusion. Participants were not told they were going to play with real people but were encouraged to ‘imagine the game as being with real people’. This was because a fully believable story about playing with others was impractical given the limited (controlled) behaviour of the cartoon figures. Also, previous research has shown that subjects experience similar level of distress when playing Cyberball against a computer as when they think they are playing against real people [20]. A very similar approach has recently been used in another Cyberball study [13]. To enhance the sense that the two cartoons represented real people taking decisions, the time that the cartoon figures took to pass the ball was randomly varied between 800 and 3000 milliseconds, simulating ‘decision making’. The task lasted for ∼10 minutes and was completed on a single run. All blocks lasted for the same length of time (∼35 sec) but varied slightly depending on the reaction time of the participant and the ‘decision making’ variation. There were no rest blocks. Clinical, behavioural and social ratings: Immediately before scanning, all subjects completed the Beck Depression Inventory (BDI) [21] Spielberger State Anxiety scale [22] and the Rosenberg Self-Esteem Scale [23]. Patients were additionally assessed using the Schizophrenia Positive and Negative Syndrome Scale (PANSS) [24]. Recruitment, clinical and rating scale assessment of all subjects, was by JDS, a Consultant Psychiatrist with considerable experience in the NHS. After scanning, each subject was assessed using a self-report ‘social distress’ rating questionnaire [10] used in previous Cyberball studies [13]. This measure is predicated on the idea that ostracism threatens four primary social ‘needs’: belonging, self-esteem, control and meaningful existence [10]. Each need was assessed by a 0 to 10 point question, ranging from 0 (not at all) to 10 (very much). Belonging was assessed by the question “How much do you feel you belonged to the group? ", self-esteem by the question “To what extent do you think the other participants value you as a person? ", meaningful existence by the question “How true is the statement: ‘Life is meaningless’?" and control by the question “How true is the statement: ‘I am in control of my life’?". The questions were scored so that higher scores indicate a greater challenge to the social need. Additionally, participants completed a manipulation check similar to Williams and colleagues [10] that assessed mood, perceived intensity of ostracism and perception of group cohesiveness during the game. Social distress and emotional impact measures were analysed using multiple linear regression with group as a fixed factor and gender as a covariate. SPM8 (http://www.fil.ion.ucl.ac.uk/spm) was used for analysis. For pre-processing, global effects were removed from the fMRI time series using a voxel-level linear model of the global signal [25] (http://code.google.com/p/lmgs4spm). Images were slice-time corrected and realigned to the first image in each time series. The average realigned image was used to derive parameters for spatial normalization to the SPM8 Montreal Neurological Institute (MNI) template with the parameters applied to each image of the time-series. The resultant time-series realigned and spatially normalised images were finally smoothed with an 8 mm Gaussian kernel. For first level analysis, a blocked design was implemented as a parametric modulation of percentage of social exclusion. This aimed to identify brain regions which activated as the degree of social exclusion systematically increased or decreased. Each subject's motor response times were included as a regressor to control for motor and associated cognitive processes during the task. The six head motion realignment terms where also included as further covariates of no interest, to allow for residual movement artefacts not removed by pre-processing realignment. The social exclusion and motor response regressors were convolved with the SPM8 haemodynamic response function without time or dispersion derivatives. For each subject, the covariate image used for second level analyses was the SPM8 ‘beta’ image, which comprised the estimated linear regression coefficient between the percentage of social exclusion and observed BOLD signal. Two second level random effects analyses were conducted. The first consisted of testing the null hypothesis of no significant relationship between systematically increasing (or decreasing) exclusion and the observed brain response within each group (controls and schizophrenia). This was done by entering the covariate images of interest into two one-group t-tests. The second (second level) analyses consisted of testing the null hypothesis of no difference between control and patient groups in the imaging parameter estimates corresponding to the parametric regressor of social exclusion. The between groups comparison was performed using multiple linear regression with group as a covariate of interest and gender as a covariate of no interest. Both for the within and between group analyses, regions are reported that are significant at a cluster threshold of p<0.05 with whole brain correction. Monte Carlo simulations [26] indicated this was achieved by the simultaneous requirement for a voxel level threshold of p<0.005 and at least 106 continuous resampled voxels. Next we investigated whether abnormal neural responses to social exclusion correlated with illness severity measures in the schizophrenia group. First, the positive and negative symptom scales of the PANSS were used in separate analyses as explanatory variables for a random effects whole brain regression analysis, of the parameter estimates for increasing social exclusion. For these regression maps we applied a cluster extent threshold of 141 voxels to ensure a p<0.05 threshold corrected for multiple comparisons across the whole brain with an individual voxel threshold of p = 0.05. Second, we examined whether significant activations in the regression maps overlapped with the mPFC region where patients differed from controls in their neural responses to social exclusion. In an analogous way, we tested for correlations with the self-report scores from the social distress and manipulation check questionnaires both for the control and schizophrenia group (results from this last analysis are reported in the supplementary material. To examine whether neural response abnormalities in schizophrenia were secondary to antipsychotic medication we tested for correlations between brain activations in response to increasing degrees of exclusion and medication dose as chlorpromazine dose equivalents [27], [28] at a less stringent threshold of p<0.05 uncorrected.
rdf:type