Ality scores were lower for patients compared to GW610742 web controls for GD. ANCOVAs (Supplementary Information 5) showed that contextual control and IQ did not explain group differences in intentionality for GD; however, contextual control explained most of the group difference in intentionality for ToM. For controls, intentionality was order SIS3 greater for GD than for R (F(1, 28) = 169.2, p < 0.001) and greater for ToM than for R (F(1, 28) = 224.6, p < 0.001) and GD (F(1, 28) = 27.6, p < 0.001). For patients, intentionality was greater for GD than for R (F(1, 28) = 107.4, p < 0.001) and greater for ToM than for R (F(1, 28) = 164, p < 0.001) and GD (F(1, 28) = 30.4, p < 0.001). Contingency/intentionality scale. Barplots are presented in Fig. 2 and boxplots in Supplementary Information 6. The ANOVA run on the contingency/intentionality scale revealed a significant triple interaction between Condition, Intentionality, and Group (F(2, 112) = 6.8, p[GG] = 0.002). This suggests that patients and controls differed in their descriptions of intentional actions, but not in the same way for R, GD and ToM animations. However, neither the triple interaction between Condition, Contingency, and Group (F(2, 112) = 1.9, p[GG] = 0.16) nor the interaction between Contingency and Group (F(1, 56) = 0.5, p = 0.47) were significant. This suggests that patients and controls did not differ in their descriptions of contingent actions. The quadruple interaction between Condition, Intentionality, Contingency and Group was significant (F(2, 112) = 4.9, p[GG] = 0.011). The post-hoc tests revealed that Group was significant for intentional contingent actions in ToM (F(1, 56) = 5.4, p = 0.024) and GD (F(1, 56) = 12.1, p < 0.001) but not in R (F(1, 56) = 0.1, p = 0.74). Group was also significant for intentional non-contingent actions in ToM (F(1, 56) = 25.2, p < 0.001) but not in GD (F(1, 56) = 2.5, p = 0.12) and R (F(1, 56) = 2.3, p = 0.13). Group was not significant for mechanical non-contingent actions in ToM (F(1, 56) = 0, p = 0.97), GD (F(1, 56) = 0.6, p = 0.43) and R (F(1, 56) = 0.9, p = 0.34). Group was not significant for mechanical contingent actions in ToM (F(1, 56) = 0.2, p = 0.62), GD (F(1, 56) = 1.9, p = 0.17) and RScientific RepoRts | 6:34728 | DOI: 10.1038/srepwww.nature.com/scientificreports/(a) Accuracy scale 5 (b) Intentionality scale2.1.1.*Intentionality [0-5]** *Accuracy [0-2]0.0.RandomGoal directedToM0 RandomGoal directedToM(c) Length scale 4 Controls PatientsLength [0-5]Error bars represent the standard error of the unajusted mean. * represents the significance of statistical tests that were carried out including covariates (p<0.05). Random Goal directed ToMFigure 1. Mean (a) accuracy, (b) intentionality and (c) length of participants descriptions for random, goaldirected and theory of mind animations.(F(1, 56) = 0.3, p = 0.61). ANCOVAs (Supplementary Information 5) showed that contextual control and IQ did not explain group differences. The number of described actions was lower for patients compared to controls for intentional non-contingent actions in ToM and intentional contingent actions in GD. The post-hoc tests for the remaining contrasts are presented in Supplementary Information 7. There was no significant correlation between verbal descriptions of Frith-Happ?stimuli and clinical symptoms (see Supplementary information 8). There was no fixation whose duration was below 100 ms (the shortest fixation duration was 199.8 ms). The ANOVA run on fixation duration.Ality scores were lower for patients compared to controls for GD. ANCOVAs (Supplementary Information 5) showed that contextual control and IQ did not explain group differences in intentionality for GD; however, contextual control explained most of the group difference in intentionality for ToM. For controls, intentionality was greater for GD than for R (F(1, 28) = 169.2, p < 0.001) and greater for ToM than for R (F(1, 28) = 224.6, p < 0.001) and GD (F(1, 28) = 27.6, p < 0.001). For patients, intentionality was greater for GD than for R (F(1, 28) = 107.4, p < 0.001) and greater for ToM than for R (F(1, 28) = 164, p < 0.001) and GD (F(1, 28) = 30.4, p < 0.001). Contingency/intentionality scale. Barplots are presented in Fig. 2 and boxplots in Supplementary Information 6. The ANOVA run on the contingency/intentionality scale revealed a significant triple interaction between Condition, Intentionality, and Group (F(2, 112) = 6.8, p[GG] = 0.002). This suggests that patients and controls differed in their descriptions of intentional actions, but not in the same way for R, GD and ToM animations. However, neither the triple interaction between Condition, Contingency, and Group (F(2, 112) = 1.9, p[GG] = 0.16) nor the interaction between Contingency and Group (F(1, 56) = 0.5, p = 0.47) were significant. This suggests that patients and controls did not differ in their descriptions of contingent actions. The quadruple interaction between Condition, Intentionality, Contingency and Group was significant (F(2, 112) = 4.9, p[GG] = 0.011). The post-hoc tests revealed that Group was significant for intentional contingent actions in ToM (F(1, 56) = 5.4, p = 0.024) and GD (F(1, 56) = 12.1, p < 0.001) but not in R (F(1, 56) = 0.1, p = 0.74). Group was also significant for intentional non-contingent actions in ToM (F(1, 56) = 25.2, p < 0.001) but not in GD (F(1, 56) = 2.5, p = 0.12) and R (F(1, 56) = 2.3, p = 0.13). Group was not significant for mechanical non-contingent actions in ToM (F(1, 56) = 0, p = 0.97), GD (F(1, 56) = 0.6, p = 0.43) and R (F(1, 56) = 0.9, p = 0.34). Group was not significant for mechanical contingent actions in ToM (F(1, 56) = 0.2, p = 0.62), GD (F(1, 56) = 1.9, p = 0.17) and RScientific RepoRts | 6:34728 | DOI: 10.1038/srepwww.nature.com/scientificreports/(a) Accuracy scale 5 (b) Intentionality scale2.1.1.*Intentionality [0-5]** *Accuracy [0-2]0.0.RandomGoal directedToM0 RandomGoal directedToM(c) Length scale 4 Controls PatientsLength [0-5]Error bars represent the standard error of the unajusted mean. * represents the significance of statistical tests that were carried out including covariates (p<0.05). Random Goal directed ToMFigure 1. Mean (a) accuracy, (b) intentionality and (c) length of participants descriptions for random, goaldirected and theory of mind animations.(F(1, 56) = 0.3, p = 0.61). ANCOVAs (Supplementary Information 5) showed that contextual control and IQ did not explain group differences. The number of described actions was lower for patients compared to controls for intentional non-contingent actions in ToM and intentional contingent actions in GD. The post-hoc tests for the remaining contrasts are presented in Supplementary Information 7. There was no significant correlation between verbal descriptions of Frith-Happ?stimuli and clinical symptoms (see Supplementary information 8). There was no fixation whose duration was below 100 ms (the shortest fixation duration was 199.8 ms). The ANOVA run on fixation duration.
