927, .462; standardised coefficients: 1.229, .519 for intensity and location respectively). Separate follow-up univariate ANOVAs on accuracy of intensity

and location judgement, confirmed that this effect was driven by differences in judgements of intensity [F(2, 32) = 4.75, p = .016, Δη2 = .229], not location [F(2,32) = .215, p = .808, Δη2 = .013]. Post-hoc protected comparisons using Fisher’s least significant differences test (LSD) were then used to identify significant differences in intensity judgements between TMS conditions. These showed that participants made greater errors in the intensity discrimination task when TMS was applied over S2 VE 821 (mean 67.8%, SD = 9.1) compared to vertex (mean 74.0%, SD = 8.1; p = .032) and also when TMS was applied over S2 relative to S1

(mean 75.0%, SD = 8.9; p = .004). In contrast, S1 and vertex TMS conditions did not differ (p = .727) (see Fig. 3). Thus, single-pulse TMS over S2 disrupts perception of pain intensity. TSA HDAC TMS might either alter response sensitivity (i.e., loss of information about whether the stimulus was strong or weak) or response bias (i.e., all stimuli perceived as higher or lower intensity). To distinguish between these possibilities, we also analysed our data using signal-detection theory (Green and Swets, 1966). We arbitrarily defined ‘High’ intensity and ‘Distal’ location as the to-be-detected signals. We computed measures of stimulus sensitivity (dprime) and response bias (criterion) for each participant

in each condition. Dprime scores indicate the sensitivity of the participant to the actual intensity or location of the stimulus, while response bias indicates the tendency to respond ‘High’ or ‘Distal’, irrespective of actual intensity/location. The dprime and criterion values for intensity and location judgements were analysed as four dependent variables using MANOVA, as before. The MANOVA again revealed a significant, but now stronger, overall PD184352 (CI-1040) effect of TMS on pain processing [Wilks' Lambda = .530 F(8, 58) = 2.71, p = .013, Δη2 = .272]. The canonical structure (.629, .222, .081, .451 for Intensity dprime, Intensity criterion, Location dprime, Location criterion respectively) suggested that TMS primarily affected sensitivity of intensity perception. Follow-up univariate ANOVA confirmed that effects of TMS were confined to sensitivity of intensity judgements [F(2, 32) = 4.09, p = .026, Δη2 = .204]. There was no significant effect of TMS site when analysing biases in intensity [F(2, 32) = 2.30, p = .117, Δη2 = .126], sensitivity to location [F(2, 32) = .025, p = .975, Δη2 = .002] nor biases in location [F(2, 32) = 2.14, p = .134, Δη2 = .118]. The significant univariate ANOVA on sensitivity in intensity judgement was followed up using Fisher’s LSD. S2 TMS reduced stimulus sensitivity (mean dprime = 1.15, SD = .59) relative to vertex control (mean dprime = 1.57; SD = .52; p = .021) and relative to S1 (mean dprime = 1.56, SD = .59; p = .