تراکم، گروه بندی، و افزایش کنترل در اسکیزوفرنی

Visual paradigms are versatile tools to investigate the pathophysiology of schizophrenia. Contextual modulation refers to a class of paradigms where a target is flanked by neighbouring elements, which either deteriorate or facilitate target perception. It is often proposed that contextual modulation is weakened in schizophrenia compared to controls, with facilitating contexts being less facilitating and deteriorating contexts being less deteriorating. However, results are mixed. In addition, facilitating and deteriorating effects are usually determined in different paradigms, making comparisons difficult. Here, we used a crowding paradigm in which both facilitation and deterioration effects can be determined all together. We found a main effect of group, i.e., patients performed worse in all conditions compared to controls. However, when we discounted for this main effect, facilitation and deterioration were well comparable to controls. Our results indicate that contextual modulation can be intact in schizophrenia patients.

The obvious symptoms of schizophrenia are hallucinations, delusions, and cognitive dysfunctions. However, schizophrenia patients have many other abnormalities, including visual impairments (Kraepelin, 1893, Chapman, 1966, Silverstein and Keane, 2011 and Chen, 2011). Visual paradigms are versatile tools of schizophrenia research because patients׳ deficits are often very pronounced (Espeseth et al., 2007, Chkonia et al., 2010, Silverstein and Keane, 2011 and Bakanidze et al., 2013). One class of interesting visual paradigms relates to contextual modulation, where perception of a target is strongly influenced by surrounding elements. For example, all spatial illusions are versions of contextual modulation. Other examples are surround suppression, contour integration, and crowding, which all are abnormal in schizophrenia patients. Dakin et al. (2005) presented a medium contrast patch together within a high-contrast surround. Controls perceived the patch as of much lower than the true contrast. In schizophrenia patients this effect was strongly diminished, i.e., patients reported a value closer to the true contrast. Interestingly, contrast discrimination itself is strongly deteriorated in schizophrenia patients ( Slaghuis, 1998 and Keri et al., 2002). Another example of contextual modulation is crowding, where perception of a target deteriorates when flanked by neighbouring elements (see Fig. 1). As in the previous paradigm, patients show less interference by the neighbouring elements ( Robol et al., 2013). Full-size image (42 K) Fig. 1. (A) Mean thresholds for the unflanked (vernier is presented alone) condition and for three crowding conditions (vernier is flanked by 2×8 equal, short or long flankers). When the vernier was flanked by 2×8 same-length flankers, thresholds increased dramatically compared to the other conditions. Performance in patients is worse in all conditions compared to controls. (B) We discounted for the main effect of Group by dividing the thresholds of each observer in the crowding conditions by the threshold in the unflanked condition of this observer (threshold elevation). After discounting, performance is very similar in patients and controls. Error bars indicate the standard error of the mean for 16 schizophrenia patients and 15 healthy controls. Figure options Contextual modulation is usually explained by interactions between neighbouring neurons that mutually influence each other, for example via gain control or long range excitation, which are proposed to be weaker than in controls (e.g., Butler et al., 2008 and Phillips and Silverstein, 2013). Accordingly, it seems that contextual modulation is in general weaker in patients, i.e., patients benefit less from helpful contexts but are also less affected by deleterious ones (Robol et al., 2013). These diminished neural interactions are also in agreement with the broader claim that contextual processing is deteriorated in general, including in non-visual examples such as verbal and cognitive context memory (e.g., Phillips and Silverstein, 2003, Uhlhaas et al., 2006 and Phillips and Silverstein, 2013). It might be that all sorts of abnormal contextual modulation are due to higher levels of cognitive disorganisation, as already proposed by Bleuler (1911). However, the situation is more complex. For example, Tibber et al. (2013) used a similar paradigm to Dakin et al. (2005) mentioned above, together with a paradigm where the orientation of target lines was modulated by surrounding lines. As in Dakin et al. (2005), contextual modulation for the contrast was weaker in the patients compared to controls but there were no differences in the orientation paradigm. Hence, not all types of contextual modulation are affected in schizophrenia patients (see also Yang et al., 2013). Moreover, Yoon et al. (2009) reported that patients show weaker modulation when target and surround have the same orientation but no differences when they are orthogonal. In addition to mixed findings, results are often hard to interpret because facilitating (release of deterioration) and deteriorating effects are not determined in one, but in different paradigms. Related to the neural causes, the neural mechanisms of contextual modulation are under debate. For example, we have shown that in healthy participants contextual modulation cannot be explained by simple local interactions between neighbouring neurons. To the contrary, complex Gestalt aspects determine processing (e.g., crowding: Malania et al., 2007, Pelli and Tillman, 2008, Whitney and Levi, 2011 and Manassi et al., 2012; surround modulation: Saarela and Herzog, 2008; visual masking: Herzog and Fahle, 2002). Particularly interesting is crowding, where we developed a task which allows one to test contextual facilitation and deterioration within one paradigm. We presented a vernier, which comprises two vertical bars that are offset slightly either to the left or right. Observers indicated the offset direction. Performance strongly deteriorated when the vernier was flanked by arrays of lines of the same length as the vernier (Fig. 1). This is a classic crowding effect. However when we presented arrays of shorter or longer lines, performance improved, challenging most models of crowding and of contextual modulation in general. Particularly, models cannot explain why longer lines, with more stimulus energy, improve performance compared to the equal length condition (e.g., Malania et al., 2007 and Manassi et al., 2012). We proposed that grouping, rather than low level mechanisms, determine crowding. When the vernier groups with the flankers (same length lines) crowding is strong. When the vernier ungroups from the flankers grouping is weaker (shorter or longer lines). Here, we applied this paradigm to schizophrenia patients. Patients performed worse compared to controls in all conditions. However, when we discounted for this main effect of group, patients performed similarly as controls, indicating that contextual modulation is intact.

First, we performed a 2×4 repeated measures analysis with Group as between-subjects factor (patients vs. controls) and Condition as within-subjects factor (unflanked, equal size flankers, short size flankers, long size flankers). There was a significant effect of Condition (F[1,55]=12.6; p≤0.0005), where performance was best in the unflanked condition and worst in the equal sized flankers condition, reproducing previous results. There was no significant interaction. Schizophrenia patients had, on average, higher thresholds than controls in all conditions (F[1,29]=56.3; p≤0.0005). For example, mean thresholds for the unflanked vernier discrimination were 16.2′±8.6 and 27.0′±13.1 in controls and patients, respectively. To account for this base deficit, we normalised performance by dividing the “crowding” threshold for each observer by his/her threshold in the unflanked condition, i.e., we determined performance in terms of threshold elevation. These values were subjected to a 2×3 repeated measures ANOVA with factors Group (patients, controls) and Condition (equal size, short size, long size flankers). There was a main effect of Condition (F[1,42]=11.9; p≤0.0005) but no significant main effect of group (F[1,29]=0.02; p=0.9) and no significant interaction (F[1,42]=0.2; p=0.75). Hence, there is nonspecific deterioration of patients in all conditions. When this effect is controlled for, performance of patients is about on the same level as the one for controls.