Figure 3. Temporal dynamics during observational threat learning.

(a) Brain BOLD responses to the observational US in the Naltrexone as compared to the Placebo group resampled to 1 cubic mm. (b) Blockade of opioid receptors in the Naltrexone group enhances responses towards the observational US in the PAG and midline thalamus as compared to Placebo controls. ‘Obs US' refers to responses to the observational US and ‘no obs US' to responses to CS+ outcomes that are not followed by the US. (c) Temporal dynamics of the midline thalamus (upper) and PAG (lower) indicates decreasing responses to the observational US as a function of learning in the Placebo group and sustained observational US responses in the Naltrexone group, see Supplementary Fig. 3 for specific response to obs US and no obs US trials). (d) PAG responses displayed an increased functional connectivity (PPI) within the superior temporal sulcus (STS) in the Naltrexone, as compared to Placebo, group (unpaired t-test, one-sided (Naltrexone N=22, Placebo N=21), x,y,z(MNI)=46;−50;8; t=3.94, P(FWE, cluster)<0.035). The cluster in the STS is located within a mask (red line) that represents the reverse inference of the terms STS and SOCIAL from the neurosynth data base with a threshold of P(FDR)<0.01. The error-bars denote the standard error of the mean, and T-maps are superimposed on an average structural image with a threshold of P(uncorrected)<0.01 for illustrative purposes. Asterisks indicate significant differences between groups, corrected for multiple comparisons (Bonferroni-Holmes).