Placebo analgesia occurs when a subject experiences a reduction in pain from a pharmacologically inert substance because the subject believes or expects it to be an effective pain reliever (due, for example, to conditioning or verbal suggestion; e.g. Hoffman, Harrington & Fields, 2005; Price, Finniss & Benedetti, 2008). Placebo analgesia has been demonstrated in placebo arms of clinical trials, as well as in laboratory studies of healthy volunteers (Vase et al. 2009). Basic research on mechanisms of placebo analgesia is important to develop a better understanding of both pain processing in general and how pain can be controlled. In addition, learning more about the mechanisms of placebo analgesia may inform efforts to more effectively harness top-down modulation of pain in clinical treatment. Research on the placebo mechanism is often confounded by the limitations of methods available to measure pain. Because pain is a subjective experience, placebo analgesia effects are almost always quantified using verbal or written perceptual reports. As such, they are susceptible to conscious and unconscious response bias related to demand characteristics of the experiment (Zellner et al. 2004). In other words, subjects may report reduced pain because they believe they should be experiencing reduced pain, not because their sensory pain experience has changed. Placebo responsiveness is only observed in a subset of individuals, further supporting the idea that response bias may contribute to the effect. Studies that have tried to determine how many people are actually placebo responders have used a variety of criteria and have produced a variety of estimates ranging from only 25% to as many as 72% (Benedetti 1996; Domnick, Lorenz & Hauck, 2011; Hoffman et al. 2005; Kong, Fufa et al., 2005; Price et al. 2008; Wager et al., 2004). Such estimates are particularly difficult, since there currently is no objective measure of pain nor any agreed-upon standard for demonstrating pain reduction. Some would like to use brain imaging as an objective measure, but imaging can only determine if a persons brain response to pain is typical, not whether the person is really experiencing pain or the amount of pain the person is experiencing. The current study attempts to provide an objective measure of placebo analgesia, to determine if in fact the effect truly alters perception in at least a subset of individuals, independent of experimental demand characteristics. The methods are based on our knowledge that experimental subjects can discriminate large differences in cutaneously applied painful temperatures better than they can discriminate smaller differences (Bushnell et al. 1983). We hypothesize that if the placebo effect reduces the painfulness of a heat stimulus on skin treated with a topical placebo analgesic, then the ability to discriminate between pain stimuli delivered to this treated skin versus stimuli delivered to untreated skin should be altered. By presenting a mix of varied temperatures to both treated and untreated areas, we can render the subject unable to guess what the correct response is, and thus unable to give responses intended to please the experimenter. Results: Our placebo testing procedure was designed to eliminate most potential sources of response bias. This led to a lower rate of placebo response than in many similar studies. Of the two clear placebo responders identified, one showed strong evidence of the sensory placebo phenomenon and the other showed weak evidence. Of six non-placebo-responders, none showed an effect. Our results suggest that placebo analgesia can alter sensory perception- but that it does not always do so. Further, we report dissociations between subjects overall ratings of the placebo cream effectiveness (subjective placebo effect), average ratings of pain on either arm (perceptual placebo effect), and our sensory discrimination placebo effect.
