On the face of it, a person’s experience during hypnosis is quite amazing. Many people report that ordinarily unremarkable actions such as lifting a hand have a startling involuntary quality; they experience unusual changes in sensations such as pain; they see things that are not present, and do not see things that are present; they feel that time is slowing down or speeding up; and they display the curious inability to recall the events of hypnosis when amnesia is suggested to them. Indeed, from a social cognitive perspective, hypnosis might well be defined as a situation in which people respond to imaginative suggestions of this sort, regardless of whether a formal hypnotic induction or even the term ‘hypnosis’ have been used (Hilgard, 1973; Kirsch, 1997, 2003). These sorts of changes in thoughts, feelings, actions and sensations appear to be so out-of-the-ordinary that they clearly imply that something extraordinary happens during hypnosis (eg Hilgard, 1965; Spiegel and Spiegel, 1978; Erickson, 1980; Edmonston, 1981; Spiegel, 1998). Indeed, for more than 200 years, the dominant view of hypnosis is that it activates special abilities, produces a trance or profoundly altered state of consciousness and compromises a person’s sense of agency. For more than a century, it was assumed that a trance or altered state of consciousness is responsible for the seemingly remarkable effects of hypnosis. In these early theories, it was assumed that the trance state was a necessary precursor for the production of hypnotic experiences and behaviors, so much so that the presence of a trance was inferred from the presence of the behaviors. Thus, hypnotized subjects showing hallucinations were assumed to be in a deep trance, whereas those who only responded to easier suggestions were assumed to be in a lighter trance. The traditional trance explanation of hypnosis was challenged by the results of the first sustained series of experimental studies of hypnosis (Hull, 1933). In Hull’s studies, the exact same hypnotic suggestions were given with and without induction of hypnosis. These experiments revealed that the effect of inducing hypnosis is relatively small—‘probably far less than the classical hypnotists would have supposed had the question ever occurred to them’(Hull, 1933, p. 298)—and that hypnotic and nonhypnotic suggestibility is very highly correlated. Hull’s research has since been replicated in several other laboratories (Weitzenhoffer and Sjoberg, 1961; Barber and Glass, 1962; Hilgard and Tart, 1966; Tart and Hilgard, 1966; Braffman and Kirsch, 1999, 2001), all using the same basic design and yielding the same basic results. In all of these studies, waking or nonhypnotic suggestibility was defined as