To make all values positive and interpretable, CB-839 we expressed these standardized scale scores as T scores, normed to a mean of 50 and a standard deviation of 10 (within the profile). The WISDM scales (both raw-score and normalized) were compared between groups using multivariate repeated-measures MANOVA, with the scores as dependent variables, and contrasts tested differences in particular scores. As an alternative approach, we also analyzed the rank ordering of WISDM scales within each subject’s profile, using a nonparametric one-way test of differences. This analysis produced essentially identical results, so is not reported here in detail. As shown in Table 2, contrary to our hypothesis,

DS and ITS had similar within-profile standard deviations (scatter). Repeated-measures MANOVA showed a significant group-by-scale interaction, indicating differences in profile shape. These are seen in the standardized profile, shown in Fig. 1a. In between-group comparisons of the standardized scores, DS score higher than ITS (in order of the size of the differences) on Tolerance, Craving, Automaticity, Loss of Control, and Behavioral Choice, while ITS score higher on Social Goads, Cue Exposure, Weight Control, Taste/Sensory Properties, and Positive Reinforcement (and numerically higher scores on

Negative Reinforcement). The groups did not differ on Affiliative or Cognitive Enhancement motives. On higher-order factors, DS scored higher than ITS on PDM, but ITS BAY 73-4506 research buy scored higher on SDM, as seen in Fig. 2a (interaction p < .0001). Comparing the profiles of CITS and NITS showed no differences in profile scatter (Table 2). On contrasts based on standardized scores of individual scales, CITS scored higher in Tolerance and Loss of Control, while NITS scored higher on Positive Reinforcement. However, repeated-measures analysis yielded no significant group-by-scale interactions: the shapes of NITS’ and CITS’ profiles were not reliably different, despite the variations in the significance of differences on particular scales (Fig. 1b). On higher-order factors, CITS

scored significantly higher on PDM, while NITS scored significantly higher on SDM (by non-parametric test). The group-by-scale interaction was significant (p < .05; see Fig. 2b). Because CITS scored intermediate between NITS and DS, and were Sodium butyrate formerly DS, we also tested differences between CITS and DS. On raw scores, DS scored significantly higher on all scales (Table 2). On standardized scores, DS scored higher on all PDM scales, as well as Behavioral Choice, and lower on Social Goads, Cue Exposure, Weight Control, and Taste-Sensory motives, largely paralleling ITS–DS differences (Fig. 1). Previous analyses (Piasecki et al., 2007 and Shiffman et al., 2012b) had demonstrated that ITS are less dependent than DS on multiple measues, including the WISDM. This analysis of the WISDM scales extends prior results by demonstrating differences between DS and ITS in the profiles of smoking motives.