<h2>The effects of sham hypnosis techniques </h2> ---------- <h4>Introduction and background</h4> <p>Take-up of complementary and alternative medicine (CAM) approaches is increasing both in the EU and the US with a prevalence as high as 86% in some countries (Eardley, et al., 2012; Clarke, et al., 2015). This growing demand for complementary treatments is coupled with rising scientific interest in the efficacy of various mind-body medicine interventions. A very promising treatment modality is medical hypnosis, the usefulness of which is demonstrated by clinical trials in several medical areas (Kekecs, Nagy & Varga, 2014; Kekecs & Varga, 2013). Medical hypnosis is used effectively in the management of both acute procedural pain and chronic pain, as indicated by a long series of clinical trials demonstrating the analgesic effects of the treatment (Elkins, Jensen & Patterson, 2007; Kendrick, et al., 2016). However, so far most clinical trials assessing the effectiveness of hypnosis provide little information about how the intervention might work or about the underlying mechanisms that elicit the treatment effect.</p> <p>In order to be able to optimize a particular intervention, it is important to identify the underlying mechanisms, and the active components of a treatment that evoke this mechanism. One of the important mechanisms often proposed to be involved in hypnosis interventions is response expectancy; an expectation by the recipient of the treatment that the treatment will elicit the desired response (e.g. decrease in the symptom). In fact, some researchers posit that this is the main mechanism driving hypnotic treatment effects (Braffman & Kirsch, 1999; Montgomery, Weltz, Seltz & Bovbjerg, 2002; Kirsch, 1994). </p> <p>Some may feel that such a hypothesis, if found to be true, would question the use of hypnosis altogether, since response expectancy is not linked to any specific treatment component to hypnosis. However, growing evidence suggests that not all interventions are equally effective in eliciting expectancy effects. For example a recent study indicates that placebo acupuncture is almost twice as effective in reducing pain as a placebo pill (Linde, Niemann & Meissner, 2010). It is similarly possible that hypnosis has a unique capacity to evoke certain types of response expectancies very effectively. So it might turn out that expectancies evoked by hypnosis are specific to the treatment, thus, validating its use.</p> <p>Others have argued that there are other important specific effective components to hypnosis interventions, such as the hypnosis induction, or the altered state of consciousness characterizing hypnosis (Kihlstrom, 2005; Spiegel, 2005). The resolution of the debate surrounding the mechanisms is hindered by a lack of methods to effectively control for expectancy evoked by hypnosis. There is a crucial need for a control condition that can evoke the same response expectancies as hypnosis, and matching other common non-specific characteristics of hypnosis interventions, such as relaxation, while not containing specific components of hypnosis, such as focused attention. Such a control condition might be called „sham hypnosis”. </p> <p>Several different sham hypnosis control conditions have been used in previous laboratory studies (Baker & Kirsch, 1993; Council, Kirsch, Vickery & Carlson, 1983; Glass & Barber, 1961). For example, Glass and Barber (1961) used an inert pill as a placebo hypnosis induction technique, while Council and colleagues used placebo biofeedback claimed to be able to induce hypnosis (Council, Kirsch, Vickery & Carlson, 1983). Similarly, a more recent study showed that simply stating that “we will begin the hypnosis procedure now”, and skipping formal hypnosis induction altogether can produce similar reported hypnotic depth and responses to suggestions as a procedure with formal hypnosis induction (Krystek & Kumar, 2016). There has been some prior work in the literature in developing such a sham or placebo control condition for clinical hypnosis studies as well (Kendrick, Koep, Johnson, Fisher & Elkins, 2013, Sliwinski & Elkins, 2017; Barton, Schroeder, Banerjee, Wolf, Keith & Elkins, 2017). In these studies researchers used a so-called “white noise hypnosis” procedure consisting of white noise played to the participant described as hypnosis induction. However, it is not clear yet whether this technique can match the expectancy of pain relief evoked by conventional hypnosis, and whether there are other similar sham procedures that can be equally or even more effectively used to evoke expectancy of hypnotic responses.</p> <p>Depending on theoretical alignment, the concept of sham hypnosis may not be meaningful, since according to the socio-cognitive theory of hypnosis, “it is not the procedural conditions per se that are important but whether or not the subject perceives them as part of a context that is 'appropriate' for displaying hypnotic behavior” (Sheehan & Perry, 1976). Kirsch, for example, claims that the only common ingredient of different hypnosis induction procedure is the label „hypnosis”, and that hypnotic inductions are expectancy manipulations, similar to placebos, because their effects are independent of any specific component or ingredient (Lynn, Kirsch & Hallquist, 2008). According to this theory, any sufficiently credible procedure that is labelled as hypnosis and is believed to be hypnosis by the subject is in fact true hypnosis, and is capable of evoking hypnotic responses. Accordingly, most prior laboratory studies contrasting conventional hypnosis procedures with sham hypnosis procedures did not find significant differences between the responses elicited by the two types of procedures. However, one study found statistically significantly lower behavioral responses to hypnotic suggestions when the suggestions were given after the sham procedure compared to if they were given after the conventional hypnosis procedure (Council, Kirsch, Vickery & Carlson, 1983). Also, descriptive data of similar studies suggests that hypnotic responses elicited by the sham inductions are slightly lower than those elicited by the conventional inductions (Baker & Kirsch, 1993; Council, Kirsch, Vickery & Carlson, 1983; Glass & Barber, 1961, Krystek & Kumar, 2016). This difference does not reach statistical significance in most individual studies, but the pattern is consistent across several studies. This may be just the result of sampling or measurement error, but it might also mean that there is a small difference between the sham and conventional induction procedures, but prior studies were underpowered to detect this small effect. Another limitation of prior research is the use of statistical tests capable only to reject the null hypothesis, but not capable to reject the alternative hypothesis. </p> <p>If there is a difference between sham and conventional procedures, the prediction of the socio-cognitive theory would be that the credibility of sham procedure or the response expectancy evoked by the sham procedure was not high enough, and a more credible procedure could produce undistinguishable hypnotic responses to a conventional induction. Another plausible explanation of a difference might be that the effects of conventional inductions are only partially driven by response expectancy, and that there are other mechanisms at play as well in eliciting the hypnotic responses. These differences might manifest in behavioral, subjective or psychophysiological factors. Previous studies mostly focused on contrasting behavior between sham and conventional hypnosis, and the subjective experiences associated with sham hypnosis are largely neglected.</p> <h4>Fundamental research</h4> <p>In order to be able to resolve the above mentioned gaps in the field, more fundamental research is required. Precise information is needed about the credibility of several different sham procedures, and we need to identify procedures which can evoke comparable expectancy to conventional hypnosis, and which are at least as credible as conventional techniques. We also need to contrast the phenomenal changes elicited by sham and conventional induction techniques to be able to identify potential differential effects of conventional induction over sham induction. Finally, we need to use appropriate statistical techniques which can contrast the null model and the alternative model. That is, no or negligible difference between sham and conventional procedures vs. meaningful difference between the procedures.</p> <p>The purpose of our project is to assess the validity of different sham hypnosis procedures designed to be viable for both clinical and laboratory research. Furthermore, we will contrast the subjective hypnotic experiences and pain reduction evoked by these sham procedures compared to a conventional hypnosis procedure. </p> <h4>Preliminary data</h4> <p>We have used three different sham hypnosis procedures in the previous stage of the project (“white noise hypnosis”, “subliminal hypnosis”, “embedded hypnosis”), and gathered preliminary data about their effectiveness in evoking different response expectancies compared to a conventional hypnosis procedure. In this online survey study we asked participants to read a brochure of a dental clinic (which was actually designed by us) offering hypnosis as one of its methods for analgesia. The brochure contained a general description of hypnosis-induced analgesia, and a description of two different hypnosis induction techniques. One of the described techniques was always a conventional relaxational hypnosis technique commonly used in clinical practice, while the other was randomly one of the three sham techniques. After reading the brochure, participants had to indicate how effective they think each of the two techniques would be for pain reduction and for inducing hypnosis for them. The observed difference between the expected effectiveness of the placebo techniques and the conventional hypnosis technique was negligible (see figure 1). The Bayesian statistical analysis also supported the null hypothesis compared to the alternative, indicating that there might be no difference between sham and conventional hypnosis-evoked response expectancies. Our later exploratory analyses revealed that even though there was no difference between the sham and conventional hypnosis evoked expectancy ratings within participants, there were differences between the three arms of the study. Specifically, participants who got the “white noise hypnosis” description reported the lowest expectancy rating both for the sham and the conventional hypnosis, while people who got the “embedded hypnosis” description reported the highest expectancies again, for both sham and conventional hypnosis. These results indicate that it is plausible that several different sham hypnosis techniques would evoke similar expectations as conventional hypnosis. We also found a very slight preference for the conventional induction over the sham induction in a forced choice situation (57% vs. 43%). However, so far we only tested the expectancies evoked by the description of these techniques, and we do not know whether this high level of expectancy would be maintained during and after the actual administration of these techniques. It is also clear from our data that, unexpectedly, the expectancy ratings of the conventional hypnosis technique was influenced by not only the description of that technique, but also by the description of the sham technique. Thus, it would be important to assess the expectancy evoked by the techniques on their own, without getting information about the other technique. Furthermore, the online study only looked at expected effectiveness. In order to assess the usefulness of these sham techniques in clinical research, it is important to establish their actual effectiveness in reducing pain, inducing hypnosis-like experiences. </p> <h4>Specific aims </h4> The primary goal of the project is to enable a better understanding of the role of response expectancy in hypnosis. To reach this goal, we aim to: <br /> 1. Explore the expectancies and subjective experiences evoked by different sham hypnosis induction procedures and conventional hypnosis induction. <br /> 2. Identify the sham hypnosis procedure that is most effective in evoking hypnosis-like response expectancy. <br /> 3. Contrast the effectiveness of suggestions for pain reduction when coupled with sham and conventional hypnosis induction. <br /> 4. Identify the experiences (if any) that are specific to the conventional hypnosis procedure and that are specific to hypnosis expectancy. <br /> <h4> Significance </h4> <p>This project is highly innovative, since this is the first study contrasting multiple sham hypnosis induction procedures, which will help to select the optimal procedure to be used in later studies. Future research will greatly benefit from these findings, since it will help researchers to select the procedure that is most credible to the participants and which evokes the highest response expectancy.</p> <p>Furthermore, this is the first study using a 2x2 placebo design with hypnosis. That is, a design where a sham and an active treatment is used, with each either being described as the active or as the sham treatment. This approach lets us distinguish the effect of expectancy and the pure effect of the sham and conventional induction procedures without the effect of response expectancy. Thus, enabling us to test an important prediction of the socio-cognitive theory of hypnosis, according to which hypnosis inductions should have no hypnotic effect without being described as hypnosis.</p> <p>Sham hypnosis protocols have not been validated before in terms of evoked hypnosis experiences, so we have no reliable information about what are the experiences that we can expect in different sham hypnosis protocols. The present project will provide valuable information about the phenomenology of sham hypnosis, which is invaluable for researchers looking for viable control conditions in the study of the phenomenology of conventional hypnosis.</p> <p>There is currently no sham hypnosis condition which would be validated for use in electrophysiological studies of hypnosis, so we have no prior knowledge about the psychophysiological differences (if any) between the sham and the conventional hypnosis induction conditions. Our project will be the first to yield such data to the field, potentially providing a new control condition to researchers looking for the psychophysiological correlates of hypnosis, which would control for the expectancies evoked by undergoing hypnosis.</p>