Roy, M., Lebuis, A., Hugueville, L., Peretz, I., and Rainville, P. 2012 European Journal of Pain, 16: 870-877.
Summary
Previous studies show that music has an ability to modulate pain, and this paper aims to elucidate the neurophysiological mechanisms of this. Pleasant music provides a decrease in pain, while unpleasant music excerpts either has no effect or worsens pain. This trend is also seen with emotional inducers such as pictures, films, odours, hypnosis, and emotional sentences. This phenomenon when elicited by pictures is in part mediated by descending pain-modulatory pathways. Descending pathways are white-matter spinal tracts that carry information from the brain to the spinal cord. This was measured via the amplitude of the spinally mediated nociceptive reflex (RIII reflex). Therefore, pleasant images seem to mediate pain by inhibiting the nociceptive pathway at the spinal cord. This paper investigates the theory that pain mediation by music operates via similar pathways. This study also looked to address whether pleasant-relaxing music had a greater analgesic effect than pleasant-stimulating music. This question arises from contradicting theories that arousal either enhances or dampens nociceptive inhibition.
In order to measure the influence of music on pain, RIII reflexes were measured after moderately painful electrical shock. . This study used 30 healthy individuals, with an average age of 24.26 years. After the pin-prick shock, and musical excerpt (pleasant-arousing, pleasant-relaxing, unpleasant, or silence), participants rated their pain on a visual analogue scale from 0-100. In order to measure individual arousal or sentiment brought up by the musical excerpt, participants rated the excerpt on a Self-Assessment Manikin test from 1-9 (1=unpleasant and 9=pleasant, 1= low arousal and 9= high arousal).
In terms of results, there was no difference in pain ratings between silence and pleasant-stimulating music, and only a mild difference between silence and pleasant-relaxing music. There was no statistical difference between the pain ratings after pleasant-stimulating and pleasant-relaxing musical excerpts. Pain ratings were statistically higher during unpleasant music when compared with both types of pleasant music. Intriguingly, reflex amplitudes were at the absolute lowest during silence, which means silence mediated pain better than pleasant or unpleasant music. Pain perception was well correlated with RIII amplitudes, which supports the hypothesis that music-mediated pain alleviation operates in part via inhibition of descending pain-modulatory pathways. However, there was a discrepancy in that participant-rated pain for pleasant (both types) music generally decreased or remained the same in comparison with silent trials, but the RIII amplitudes were statistically higher with pleasant music than with silence.
Comment:
In the summary of this paper, the authors state that this work "reinforces the idea that music is a powerful emotional inducer that can have a strong influence on pain." As much as I applaud the efforts in elucidating the effects of music on pain in humans, this conclusion was not even remotely supported by their data. Their data found no improvement in pain mediation in pleasant music trials, when compared to silence. Additionally, silence produced a stronger analgesic effect than pleasant or unpleasant music trials. This does not support the conclusion that music has a strong influence on pain. There are several factors that would need to be addressed before this claim could be made, including: fear conditioning, the effect of music on clinically relevant intensities of pain, and duration of the musical excerpt.
Firstly, and ironically the authors directly admit this point in the discussion, when you perform consecutive shock trials that are preceded by music, the music begins to serve as a cue for the coming shock. This creates negative anticipation of the event, and serves to make the shock seem more painful. Anticipation is key in many aspects of emotional response, such as appetite, fear, excitement, or even sexual arousal (Baumeister, et al., 2007, http://psr.sagepub.com/content/11/2/167.short). No matter how pleasant the music excerpt was to the participant, repeated trials train the brain to associate music with an upcoming shock, which of course would create anticipation, and a larger negative response. It would have been a much more conclusive experiment model to have extended background music interspersed by random shocks, as opposed to ordered, semi-predictable shocks during short musical excerpts of only 1 minute in length.
The other drawback on this experiment was that it measured the effect of music on pain from a pin-prick shock. This is a start, and experimental design on music and pain in humans is obviously restricted for ethical reasons, but the ability for music to mediate pain associated with a pin-prick is not clinically relevant. The analgesic effect of music is probably substantially altered at a higher intensity or duration of pain. What would have been more interesting would have been to assess patients who were recovering from a commonplace surgery in a single-blind study. One could qualitatively evaluate the pain ratings of patients that had silence, pleasant-stimulating, or pleasant-relaxing music in their recovery rooms. One could inform the patient that the study aims to investigate subtle ambient differences, such as music, wall colour, or artwork and their effect on pain perception, which would allow for patient consent while maintaining the single-blind model. The only challenge with this is that it would be inconvenient and possibly uncomfortable to measure the RIII amplitude of a recovering surgery patient; however, this study validated a correlation between RIII amplitude and skin conductance response (SCR), which is much less intrusive to quantify. That being said, RIII amplitude and SCR may only be applicable to measuring pain associated with a brief, acute shock, and may not be helpful in measuring extended pain during the course of a surgical recovery.
Lastly, Jourdain's book elaborates on the point that music is so powerful in part because the listener builds anticipation of chord resolution throughout the piece. Beethoven's Symphony 7 allegretto, arguably one of the most emotionally riveting compositions in history, does not elicit its strong arousal or pleasurable response in 1 minute. In order for music to elicit its maximum potential for pleasure, and potentially analgesic effects, it requires preparation time and context in order for the listener to build necessary anticipation. Using a 1 minute musical except is in many ways like measuring the satiation potential of gourmet food by evaluating a microwave dinner.
On the other hand, pain studies in relation to music is an extremely difficult subject matter to approach, because you can't use an animal model, and you also can't intentionally subject human participants to clinically relevant levels of pain. That being said, this study was a reasonable start. Also, I feel the comparison of RIII amplitudes with the patients' qualitative pain scoring was an elegant solution to assessing pain levels as accurately as possible, while simultaneously supporting their hypothesis of descending pain-modulatory pathway involvement.
In summary, there was significant room for improvement, and this study did not accurately address the question of the use of music as a complementary analgesic, nor did it solve the debate on whether pleasant-stimulating music is of greater or lesser benefit than pleasant-relaxing music. However, it did confirm the hypothesis that music-mediated pain relief involves descending pain-modulatory pathways in the spinal cord.