Abstract
Perception is the result of both expectation and sensory stimulation. This is reflected in placebo analgesia, where expecting low pain leads a painful stimulus to feel less painful. Yet it is maladaptive for a highly erroneous expectation to result in an unrealistically low pain experience. We hypothesised that in estimating the intensity of a painful stimulus which is preceded by a very discrepant expectation, the perception is influenced less by the expectation. We modelled the reported pain intensity as a function of the prediction error. We used linear mixed modelling on two independently collected pain cueing datasets, the second of which was preregistered (osf.io/5r6z7). Reported pain intensities were best explained by a quartic polynomial model of the prediction error, indicating the influence of expectations on perceived pain decreased when pain was highly discrepant to expectation, suggesting that the size of prediction error has a functional role in pain perception.
