54, p = .003). Scores on the TASIT were found to be significantly selectively correlated Tanespimycin in vivo with performance on the mentalising task, (rho = .55, p = .002) though not the non-mentalising task (rho = .34, p = .067). In addition, scores on the selected CBI item (‘Appears indifferent to the worries and concerns of family members’) were significantly negatively correlated with performance on the mentalising task (rho = −.6, p = .03), but not the non-mentalising task (rho = −.1, p = .67). There were no correlations of performance on either experimental task with executive function, single-word comprehension, clinical disease duration, years of education, or premorbid

intelligence estimates. Only two control subjects reported prior familiarity with over half the musical examples used; most participants reported no prior familiarity buy Trametinib with the musical examples. Accordingly we did not perform a formal regression analysis of

performance on prior musical familiarity. However, a separate analysis excluding the two control subjects who reported higher prior familiarity with the musical examples yielded identical results with respect to the experimental tasks. ROC curves based on each of the experimental tasks discriminated between bvFTD patients and healthy controls (Fig. 2). No significant AUC difference was found between the mentalising and non-mentalising tasks, however mentalising task performance showed a trend towards greater sensitivity and specificity (AUC coefficient .88 [95% confidence interval (CI): .73,

.95]) compared with the non-mentalising task (AUC coefficient .73 [95% CI: .57, .90]). Further binomial breakdown of the AUCs revealed that a cut-point raw score of 15 on the mentalising task correctly classified 85% of participants as being either a patient or a control, whereas this was reduced to 71% for the non-mentalising task using the same cut-point value. Examining individual subject performance profiles (Fig. 3), five patients showed a clear (>four point) discrepancy in favour of superior performance on the non-mentalising task. However, two patients showed the reverse pattern, with superior performance on the mentalising task. No similarly Protirelin marked discrepancies were seen for individuals in the healthy control group (Fig. 3). SPMs of grey matter volume associated with performance in the mentalising and non-mentalising conditions are shown in Fig. 4; data for local maxima of grey matter change are summarised in Table 2. When assessed separately, performance on the mentalising task was positively associated with grey matter volume in right entorhinal cortex (p < .05 after FWE correction for multiple comparisons within the anatomical small volume of interest). No significant negative inverse associations between performance and grey matter volume were identified.