H trial was presented as text through a series of two screens, the first of which described the short scenario and the second of which asked whether the subject would do the relevant action, requiring a yes/no button press (Figure 1a). Subjects read each scenario and question at their own pace (up to 25 s for the scenario and 15 s to make their choice) and pressed a button to advance through the screens. Between each trial, a fixation cross was displayed for 2 s. At the end of each block, there was an inter-block-interval (IBI) of 16 s to allow the hemodynamic response function to return to baseline. Baseline was defined as the mean signal across the last four images of this 16 s IBI. Neural activity was measured using the floating window LOR-253MedChemExpress APTO-253 method (Greene et al., 2001). This method isolates the decision phase by including the time around the decision8 s before the response, 1 s during the response and 6 s following the responsefor a total of 15 s of recorded activity for every response. The rationale for using the floating window approach is to not only account for the 4? s delay following a psychological event in the hemodynamic response but also to create a flexible analysis structure for a complex, self-paced task. Imaging acquisition MRI scanning was conducted at the Medical Research Council Cognition and Brain Sciences Unit on a Siemens 3-Tesla Tim Trio MRI scanner by using a head coil gradient set. Whole-brain data were acquired with BLU-554 custom synthesis echoplanar T2* weighted imaging, sensitive to BOLD signal contrast (48 sagittal slices, 3 mm-thickness; TR ?2400 ms; TE ?30 ms; flip angle ?788 and FOV 192 mm). To provide for equilibration effects, the first 8 vol were discarded. TSCAN (2014)O. FeldmanHall et al.(F(1,36) ?24.34, P < 0.000). Interestingly, moral scenarios (mean 3.65 s, s.d. ?.14) also took slightly longer to respond to relative to non-moral scenarios (mean 3.43 s, s.d. ?.15), likely reflecting their higher emotional impact (F(1,36) ?5.35, P ?0.027). There was therefore also a significant Difficulty by morality interaction (F(1,36) ?143.14, P < 0.000), reflecting the fact that the moral ifficult scenarios took the longest to respond to. IMAGING RESULTS We contrasted neural activation associated with making a decision for each of the four categories against one another: Easy Moral, Difficult Moral, Difficult Non-Moral and Easy Non-Moral. To explore potential interactions among the four conditions and to verify that overall the current scenarios elicited activations consistent with the moral network described in the literature (Moll, Zahn et al., 2005), we ran a full factorial Morality ?Difficulty ANOVA (Morality ?Difficulty interaction). A whole-brain analysis of the interaction term (thresholded at P ?0.001 uncorrected) revealed a robust network of areas including bilateral TPJ, mid temporal poles, vmPFC, dACC and dlPFC (Figure 2; a full list of coordinates can be found in Table 1). We then examined a priori ROIs (Greene et al., 2001; Young and Saxe, 2009) (thresholded at FWE P ?0.05) to determine if this network specifically overlapped with the regions delineated within the literature. As expected, the vmPFC, ACC and bilateral TPJ ROIs revealed significant activation for the interaction term. The interaction term qualified significant main effects of Morality and Difficulty. Although these activations are suprasumed by the interaction, for completeness, we report then in Tables 2 and 3. As this initial full factorial analysis identified.H trial was presented as text through a series of two screens, the first of which described the short scenario and the second of which asked whether the subject would do the relevant action, requiring a yes/no button press (Figure 1a). Subjects read each scenario and question at their own pace (up to 25 s for the scenario and 15 s to make their choice) and pressed a button to advance through the screens. Between each trial, a fixation cross was displayed for 2 s. At the end of each block, there was an inter-block-interval (IBI) of 16 s to allow the hemodynamic response function to return to baseline. Baseline was defined as the mean signal across the last four images of this 16 s IBI. Neural activity was measured using the floating window method (Greene et al., 2001). This method isolates the decision phase by including the time around the decision8 s before the response, 1 s during the response and 6 s following the responsefor a total of 15 s of recorded activity for every response. The rationale for using the floating window approach is to not only account for the 4? s delay following a psychological event in the hemodynamic response but also to create a flexible analysis structure for a complex, self-paced task. Imaging acquisition MRI scanning was conducted at the Medical Research Council Cognition and Brain Sciences Unit on a Siemens 3-Tesla Tim Trio MRI scanner by using a head coil gradient set. Whole-brain data were acquired with echoplanar T2* weighted imaging, sensitive to BOLD signal contrast (48 sagittal slices, 3 mm-thickness; TR ?2400 ms; TE ?30 ms; flip angle ?788 and FOV 192 mm). To provide for equilibration effects, the first 8 vol were discarded. TSCAN (2014)O. FeldmanHall et al.(F(1,36) ?24.34, P < 0.000). Interestingly, moral scenarios (mean 3.65 s, s.d. ?.14) also took slightly longer to respond to relative to non-moral scenarios (mean 3.43 s, s.d. ?.15), likely reflecting their higher emotional impact (F(1,36) ?5.35, P ?0.027). There was therefore also a significant Difficulty by morality interaction (F(1,36) ?143.14, P < 0.000), reflecting the fact that the moral ifficult scenarios took the longest to respond to. IMAGING RESULTS We contrasted neural activation associated with making a decision for each of the four categories against one another: Easy Moral, Difficult Moral, Difficult Non-Moral and Easy Non-Moral. To explore potential interactions among the four conditions and to verify that overall the current scenarios elicited activations consistent with the moral network described in the literature (Moll, Zahn et al., 2005), we ran a full factorial Morality ?Difficulty ANOVA (Morality ?Difficulty interaction). A whole-brain analysis of the interaction term (thresholded at P ?0.001 uncorrected) revealed a robust network of areas including bilateral TPJ, mid temporal poles, vmPFC, dACC and dlPFC (Figure 2; a full list of coordinates can be found in Table 1). We then examined a priori ROIs (Greene et al., 2001; Young and Saxe, 2009) (thresholded at FWE P ?0.05) to determine if this network specifically overlapped with the regions delineated within the literature. As expected, the vmPFC, ACC and bilateral TPJ ROIs revealed significant activation for the interaction term. The interaction term qualified significant main effects of Morality and Difficulty. Although these activations are suprasumed by the interaction, for completeness, we report then in Tables 2 and 3. As this initial full factorial analysis identified.
