دیدگاه مراحل اولیه در اختلال اسکیزوفرنی و شخصیت اسکیزوتایپال

Abstract Previous studies of visual perception have reported deficits in contrast sensitivity and dot motion discrimination in schizophrenia. We tested whether these deficits also appear in schizotypal personality disorder (SPD). SPD appears to be genetically and symptomatically related to schizophrenia, but without the marked psychosocial impairment associated with psychotic disorders. The present study investigated contrast sensitivity for moving and static gratings, form discrimination and dot motion discrimination in 24 patients with schizophrenia or schizoaffective disorder (SZ), 16 individuals with SPD, and 40 control subjects. SZ, but not SPD subjects, showed impairments on tests of contrast sensitivity for static and moving gratings, form discrimination in noise, and dot motion discrimination. Visual performance did not differ between medicated SZ patients and patients withdrawn from medication. These results confirm early stage visual deficits in schizophrenia regardless of medication status. SPD subjects, in contrast, show intact early stage visual processing despite the presence of marked schizotypal symptoms.

Introduction Schizophrenia is associated with disturbances of visual perception. Both interview and self-report scales indicate that patients with schizophrenia frequently suffer from visual distortions, which appear in the earliest stages of the illness (Bunney et al., 1999, Cutting and Dunne, 1986 and Phillipson and Harris, 1985). Subjective disturbances are accompanied by visual processing deficits on psychophysical tests (Butler and Javitt, 2005). These psychophysical deficits have often been interpreted in terms of deficits in specific visual pathways or channels. Psychophysical tests measure visual performance thresholds as a function of such factors as contrast, noise, stimulus duration, or stimulus similarity. In primates and humans, two neural pathways for visual processing have been characterized by the differing response properties of the magnocellular (M) and parvocellular (P) neurons of the lateral geniculate nucleus (Livingstone and Hubel, 1988 and Wandell, 1995). The M pathway is characterized by high contrast sensitivity, high temporal resolution, low spatial resolution, and insensitivity to color. The P pathway has low contrast sensitivity, low temporal resolution, high spatial resolution, and strong color opponency responses. Human psychophysical performance also suggests the existence of a transient or broad-band visual channel, whose response properties are similar to those of the M pathway, and the sustained channel, which resembles the P pathway (Legge, 1978, Livingstone and Hubel, 1988 and Merigan and Maunsell, 1993). Functional differentiation continues into the cortex, with a ventral cortical pathway from the occipital to the inferior temporal lobe for the analysis of color and object properties, and a dorsal pathway from the occipital to the parietal lobe for motion and spatial relationships (Ungerleider, 1985, Van Essen and Gallant, 1994 and Merigan et al., 1997). The M pathway primarily projects to the dorsal stream of the cortex, while the P pathway primarily projects to the ventral cortical stream. A number of investigators have proposed that schizophrenia is associated with a more severe disturbance of M or transient channel relative to P or sustained channel processing (Butler and Javitt, 2005, O'Donnell et al., 1996, Green et al., 1994 and Kéri et al., 2004). In the following section, findings are reviewed from studies of perception of coherent dot motion and contrast sensitivity for static and moving grating stimuli in terms of visual pathway function. One of the most consistent findings in schizophrenia has been impaired discrimination of the trajectory of moving dots in both medicated (Brenner et al., 2003, Chen et al., 2003b, Chen et al., 2005, Hooker and Park, 2000, Li, 2002, O'Donnell et al., 1996 and Stuve et al., 1997) and unmedicated patients (Richardson et al., 1996). In a widely used paradigm, the motion coherence or dot kinetogram test, the percentage of dots moving in a uniform direction is varied (e.g. Brenner et al., 2003 and Chen et al., 2003b). This percentage is usually referred to as motion coherence and constitutes the signal in the image. The percentage of motion coherence required for a specific level of discrimination performance, or threshold, is usually higher in schizophrenia than in control subjects. Li (2002) showed that this deficit was due to reduced sensitivity, rather than altered response bias. With respect to neural mechanisms, animal and human studies indicate that the dorsal visual pathway, particularly cortical region MT, is involved in the perception of coherent motion from an array of moving dots (Newsome and Pare, 1988). Chen et al. (2003b) found that patients with schizophrenia were impaired at discrimination of coherent motion of moving dots, which requires global motion processing, but not at discrimination of moving gratings, which can be accomplished by local visual cues. Chen and colleagues argued that these findings suggest a late stage motion processing deficit in schizophrenia, probably in cortical regions specialized for motion perception. Contrast sensitivity for sinusoidal grating stimuli has also been studied in schizophrenia. Contrast sensitivity is the inverse of contrast threshold (the minimum physical contrast needed to reliably detect a stimulus). Some investigators (Butler et al., 2005, Kéri et al., 2002, Schwartz et al., 1987, Slaghuis, 1998 and Slaghuis, 2004) but not all (Chen et al., 1999a and Chen et al., 1999b) have found deficient contrast sensitivity in schizophrenia. In monkeys, magnocellular lesions have their greatest impact on contrast sensitivity for low spatial frequencies (< 4 cycles/degree) at temporal frequencies above 8 Hz, while parvocellular lesions have their greatest impact at temporal frequencies below 4 Hz (Merigan and Maunsell, 1993). Human psychophysical studies indicate that transient channels are insensitive at spatial frequencies above 4 cycles/degree (Legge, 1978). Consequently, a differential contrast sensitivity deficit at low spatial and high temporal frequencies would be supportive of an M deficit. This pattern has not been consistently found. Schwartz et al. (1987) reported that contrast sensitivity deficits were most reliably observed for temporally modulated gratings, rather than static gratings, suggestive of a transient channel deficit. Butler et al. (2005) reported a greater schizophrenia deficit at low compared to high spatial frequencies. Slaghuis, 1998 and Slaghuis, 2004, on the other hand, found that negative symptom schizophrenic patients showed a deficit for both stationary and moving patterns for both low and high spatial frequencies. Positive symptom patients showed deficits only at medium to high spatial frequencies (Slaghuis, 1998) or no impairment at any spatial frequency (Slaghuis, 2004). Chen et al. (2003a) hypothesized that the differences among studies of contrast sensitivity may have been related to medication levels. Chen and colleagues reported that patients receiving typical anti-psychotic medications showed elevated contrast thresholds and patients receiving novel anti-psychotic medications showed unimpaired contrast sensitivity levels for a moving grating stimulus. Moreover, unmedicated patients demonstrated decreased contrast threshold levels, indicative of performance which was better than that of control subjects. Chen et al. suggested that these medication effects might be mediated by dopaminergic cells in the retina which influence contrast sensitivity. Increased dopaminergic activity may increase contrast sensitivity (Tagliati et al., 1994), while reduced dopaminergic activity may reduce contrast sensitivity (Bodis-Wollner and Tagliati, 1993). An unresolved issue is whether these visual disturbances might be a liability marker, or endophenotype, for schizophrenia. Two groups have been evaluated to investigate this possibility: relatives of patients with schizophrenia, and individuals with schizotypal personality disorder (SPD). SPD is related to schizophrenia in terms of symptoms and genetic risk (Kety et al., 1994 and Siever and Davis, 2004), but is usually not associated with severe psychosocial impairment or treatment with anti-psychotic medication. SPD thereby provides a vehicle to identify which neural or cognitive processes are common to schizophrenia spectrum disorders, and which are only associated with psychosis. SPD has been associated with subjective perceptual distortions (Camisa et al., 2005 and Raine, 1991), and with deficits on tests of visual backward masking, cognition, and working memory (Cadenhead et al., 1999 and Siever and Davis, 2004). With respect to early stage vision, a previous study by Farmer et al. (2000) found intact thresholds for discrimination of form and motion in noise in SPD, suggestive of spared early stage visual processing. Studies of first-degree relatives of patients with schizophrenia have yielded inconsistent findings. Chen et al. (1999b) reported that first-degree relatives of patients with schizophrenia showed a deficit in velocity discrimination for moving grating stimuli. Subsequently, Chen et al. (2005) found that first-degree relatives were unimpaired on a test of coherent dot motion discrimination (Chen et al., 2005). The aim of the present study was to clarify whether early stage visual processing is differentially affected in schizophrenia and SPD. Tests of contrast sensitivity were designed to probe the M and P pathways. Motion and form coherence tests were used to assess the dorsal and ventral cortical pathways. By measuring performance thresholds, these paradigms allow for matching of task difficulty and internal reliability across conditions (Chapman and Chapman, 1973 and Brenner et al., 2003). Because medication status may affect performance, patients receiving medication were compared with patients who had recently discontinued medication. Because visual task performance may be affected by the general intellectual deficit observed in a broad range of cognitive measures (Brenner et al., 2002, Chapman and Chapman, 1973 and Mohamed et al., 1999), we also evaluated aspects of current intellectual function.

Results 3.1. Form and dot discrimination For both tests, control and SPD subjects had comparable thresholds, while SZ patients were impaired as indicated by higher coherence thresholds (Table 2). ANOVA for the motion coherence threshold revealed an effect of Group (F(2,77) = 8.79, p < .001). T-tests indicated that SZ subjects had higher motion coherence thresholds than control (t(62) = 3.85, p < .001) or SPD groups (t(38) = 2.17, p < .04), while SPD and control subjects did not differ. An ANOVA on form coherence also revealed an effect of Group (F(2,75) = 9.41, p < .001). Again, t-tests indicated that SZ subjects had higher coherence thresholds compared to control (t(60) = 3.91, p < .001) and SPD subjects (t(38) = 2.61, p = .01), while SPD and control subjects did not differ. Table 2. Visual test performance by control, unmedicated, and medicated patients Control SPD Medicated SZ Unmedicated SZ Coherence thresholds Motion 12.2 (9.1) 14.6 (10.9) 24.6 (22.8) 35.7 (28.6) Form 27.4 (10.2) 28.9 (7.7) 42.0 (22.2) 40.0 (8.7) Contrast sensitivity Static 1.74 (.45) 1.76 (.46) 1.31 (.53) 1.45 (.42) Moving (2.1 Hz) 2.13 (.13) 2.13 (.11) 2.00 (.28) 1.88 (.56) Moving (9.3 Hz) 2.08 (.23) 2.13 (.18) 1.96 (.40) 1.87 (.71) Moving (18.8 Hz) 1.57 (.19) 1.56 (.13) 1.49 (.24) 1.29 (.78) Mean values are provided with S.D. in parentheses. SPD: Schizotypal Personality Disorder. SZ: Schizophrenia or schizophrenia or schizoaffective disorder. Coherence thresholds indicate signal percentage required to discriminate a visual feature. Lower coherence thresholds indicate better performance. Contrast sensitivity indicates the log10 contrast sensitivity. Higher contrast sensitivity values indicate better performance. Table options 3.2. Contrast sensitivity for gratings The control and SPD groups had comparable performance on both measures, while the SZ group showed impaired contrast sensitivity (Table 2). ANOVA on log10 contrast sensitivity with the factors of Stimulus (4: static, 2.1 Hz, 9.3 Hz, 18.8 Hz) and Group (3) revealed a main effect of Group (F(2,77) = 6.52, p = .002) and a main effect of Test (F(3,231) = 66.1, p < .001). Follow-up ANOVAs confirmed that the control and SPD groups did not differ (F(1,54) = .14, p = .71), while the SZ group had poorer thresholds compared to the control group (F(1,62) = 9.67, p = .003) as well as the SPD group (F(1,38) = 5.58, p = .02). 3.3. Neuropsychological tests Table 1 provides group means for the three neuropsychological tests and estimated Full Scale IQ. One-way ANOVAs revealed group differences on all measures: Picture Completion (F(2,74) = 12.35, p < .001), Similarities (F(2,74) = 4.91, p = .01), Digit Symbol (F(2,74) = 19.24, p < .001), and IQ (F(2,74) = 21.34, p < .001). For all tests, t-tests indicated that control (t(59) > 3.02, p < .004) and SPD (t(35) > 2.23, p < .04) subjects had higher scores than SZ subjects, while the control and SPD groups did not differ. 3.4. Medication and performance in schizophrenia In order to determine whether medication withdrawal affected visual test performance, control subjects were compared with SZ patients receiving medication (N = 14) and patients who had been withdrawn from medication (N = 10) ( Table 2). No SPD subjects were included in this analysis. For motion coherence, ANOVA revealed a main effect of Group (F(2,61) = 8.87, p < .001). Follow-up t-tests indicated that both medicated SZ (t(52) = 2.87, p = .006) and unmedicated subjects (t(48) = 4.47, p < .001) required higher coherence levels to discriminate motion trajectory. For form coherence, ANOVA showed a main effect of Group (F(2,59) = 7.59, p = .001). Both unmedicated SZ (t(46) = 3.59, p = .001) and medicated SZ subjects (t(50) = 3.25, p = .002) were impaired compared to control subjects. Medicated and unmedicated patients did not differ on either test. For contrast sensitivity, ANOVA with the factors Group (3) and Stimulus (4) revealed a main effect of Group (F(2,61) = 4.97, p = .01) and a main effect of Stimulus (F(3,183) = 43.27, p < .001). Follow-up ANOVAs comparing each pair of groups showed that both the unmedicated patients (F(1,48) = 7.16, p = .01) and the medicated patients (F(1,52) = 8.55, p = .005) were impaired compared to the control group. The medicated and unmedicated patients did not differ. 3.5. SPQ T-tests were used to compare SPQ scale scores between control and SPD subjects ( Table 1). As expected, SPD subjects showed higher scores on all three scales (t(53) > 4.2, p < .001). There was only one significant correlation between SPQ scores and thresholds on the visual tests: a correlation between the SPQ positive scale score and contrast sensitivity for the 9.3 Hz modulation rate, r = .30, p = .03. 3.6. Correlations between clinical measures and visual performance in SZ Table 3 lists correlation coefficients between each visual test and clinical measures in SZ patients. Picture Completion was positive correlated with three contrast sensitivity measures, indicating that contrast sensitivity improved with better Picture Completion performance. Full Scale IQ correlated with one of the contrast sensitivity measures. More severe negative symptoms were associated with poorer contrast sensitivity at a 2.1 Hz temporal modulation rate. Positive symptom severity and illness duration were not correlated with visual test performance. Table 3. Correlation of visual tests and clinical measures in SZ patients Duration Positive Negative Pic Comp Simil Dig Sym IQ (N) (24) (16) (16) (21) (21) (21) (21) Coherence thresholds Motion .16 − .04 − .17 − .25 − .34 − .32 − .40 Form − .06 − .09 .21 − .14 − .22 − .34 − .09 Contrast sensitivity Static − .24 − .27 .00 .51⁎ .05 .05 .29 Moving (2.1 Hz) − .23 − .36 −.64⁎ .53⁎ .32 .17 .46⁎ Moving (9.3 Hz) .00 − .04 .02 .14 .15 − .25 .02 Moving (18.8 Hz) − .16 .12 − .14 .52⁎ .06 − .01 .26 Duration = illness duration in years; Positive = PANSS Positive Symptom Score; Negative = PANSS Negative Symptom Score; Pic Comp = WAIS-III Picture Completion; Simil = WAIS-III Similarities; Dig Sym = WAIS-III Digit Symbol; IQ = Full Scale Intelligence Quotient. ⁎ p < .05.