Date Published: June 30, 2017
Publisher: Public Library of Science
Author(s): Abeer F. Almarzouki, Christopher A. Brown, Richard J. Brown, Matthew H. K. Leung, Anthony K. P. Jones, Tim Salomons.
It is well known that the efficacy of treatment effects, including those of placebos, is heavily dependent on positive expectations regarding treatment outcomes. For example, positive expectations about pain treatments are essential for pain reduction. Such positive expectations not only depend on the properties of the treatment itself, but also on the context in which the treatment is presented. However, it is not clear how the preceding threat of pain will bias positive expectancy effects. One hypothesis is that threatening contexts trigger fearful and catastrophic thinking, reducing the pain-relieving effects of positive expectancy. In this study, we investigated the disruptive influence of threatening contexts on positive expectancy effects while 41 healthy volunteers experienced laser-induced heat pain. A threatening context was induced using pain-threatening cues that preceded the induction of positive expectancies via subsequent pain-safety cues. We also utilised electroencephalography (EEG) to investigate potential neural mechanisms underlying these effects. Lastly, we used the Fear of Pain Questionnaire to address whether the disruptive effect of negative contexts on cued pain relief was related to the degree of fear of pain. As predicted, participants responded less to pain-safety cues (i.e., experienced more pain) when these were preceded by pain-threatening cues. In this threatening context, an enhancement of the N2 component of the laser-evoked potential was detected, which was more pronounced in fearful individuals. This effect was localised to the midcingulate cortex, an area thought to integrate negative affect with pain experience to enable adaptive behaviour in aversive situations. These results suggest that threatening contexts disrupt the effect of pain relief cues via an aversive priming mechanism that enhances neural responses in the early stages of sensory processing.
It is well established that the experience of pain is often influenced by expectations of its intensity  Many studies have identified that “positive” expectancies (e.g. expecting pain reduction) are associated with reduced levels of pain and improve treatment outcomes [2–5], whereas “negative” expectancies (e.g., expecting increased pain) have the opposite effect . Similar phenomena are observed with placebo and nocebo manipulations respectively [7,8]. It is thought that such expectancy effects reflect a neural system that uses prior knowledge to make predictions about future sensory inputs, thereby enabling appropriate and timely behaviour [9–12].
The main purpose of this study was to investigate whether, and how, prior threat cues (i.e. negative expectancies) disrupt the pain reducing effect of safety cues (i.e. positive expectancies) on pain perception. The first step was to validate the effect of positive (safety) and negative (threat) cues on medium intensity stimuli using a neutral control cue, which confirmed this effect. We then demonstrated that participants perceived laser stimuli as more painful when safety cues were preceded by threat cues, in comparison to when safety cues were presented alone. In other words, suggestions of pain reduction were less effective when participants had prior negative expectations regarding outcomes (i.e. expected the outcome to be more painful). The prior negative expectancy effect on pain ratings occurred alongside an enhancement of sensory processing, as demonstrated by increased amplitude of the N2 component of the LEP. This effect was most pronounced in individuals with the highest fear of pain and corresponded to enhanced activity in the MCC.
Previous work by Brown et al. found that certainty about forthcoming stimuli influences participants’ perception of the direction of the anticipatory cues . MCC activation is reported in uncertain pain contexts . Accordingly, it may be argued that there is more outcome uncertainty in the prior negative expectancy condition (due to the change of the cue) as compared to the positive expectancy condition, and that certainty in the positive expectancy condition favoured a lower stimulus rating as compared to the prior negative expectancy condition. Precision of predictions is thought to be an essential determinant of the expectancy modulation of pain, according to predictive coding models of pain processing . The resultant effect of reduced certainty in cue-based predictions is a pain experience that is determined more by the bottom-up characteristics of the stimulus (intensity of stimulus) . Such a mechanism would be consistent with our results, as the prior threat condition (contrasting cue condition) produced pain reports that were similar to the control condition. However, it is important to note that participants were informed that the second cue would predict the stimulus and a retrospective questionnaire after the experiment suggested participants on the whole believed that the cues were accurate, suggesting uncertainty may have less of an influence in our study. A further limitation of the current study was a relatively small sample size, which limited our ability to analyse potential gender differences in the effects observed.
Prior threatening cues interfere with the effect of safety cues on pain perception, through a priming effect which enhances early cortical representation of sensory stimuli in the MCC, an area associated with adaptive control in aversive situations. Fearful individuals demonstrate an increased early neural response to threatening predictions, providing neurophysiological insights into the fear-avoidance model. The findings in this study provide important neurophysiological insights into how negative preconceptions may interfere with positive expectations about pain, resulting in a reduced therapeutic outcome. Moreover, the enhanced priming of early sensory mechanisms in fearful individuals identified in our study further strengthens the theory that pain-related fear may be an important target when managing pain. Together these results suggest an assessment of patient’s expectations and feelings towards pain and treatments, before delivering positive information about treatments, may be important in maximising therapeutic outcome. Further studies in patients are required to establish if this approach can provide a method of phenotyping patients’ different styles of cognitive and emotional appraisal of threatening stimuli.