Placebo hypoalgesia (top) and nocebo hyperalgesia (middle) are prime examples of the impact that psychological factors have on pain perception. Using functional magnetic resonance imaging of the human spinal cord we could show that placebo analgesia results in a reduction of nociceptive processing (top), whereas nocebo hyperalgesia results in an increase of activation in the spinal cord (middle). Similar to placebo hypoalgesia, we observed a reduction of activation in the spinal cord during a demanding cognitive task (2-back working memory). Both cognitive distraction and placebo hypoalgesia are mediated by endogenous opioids. The mechanism of nocebo hyperalgesia is still unclear.

We put forward the idea that the brain is not passively waiting for nociceptive stimuli to impinge on it but is actively making inferences based on prior experience and expectations (top). The Bayesian formulation within the predictive coding framework can directly account for differences in the magnitude but also the precision of expectations that are known to influence the strength of placebo hypoalgesia. We then experimentally tested these hypotheses and could show (middle) that decreasing the precision of expectations reduces placebo hjypolagesia. Finally, a cued pain paradigm indicated that posterior areas in the insula region such as S2 simply code pain intensity (blue), whereas anterior regions (yellow) represent predictions and prediction errors regarding pain.