Figure 7.1
MEG scans, showing activity across the brain while listening to a beep and estimating its duration.
7
Hallucinations in Hypnosis
Abstract
The nature of the hypnotic experience is discussed, drawing parallels between hypnotically induced hallucinations and those of conditions such as schizophrenia. A particularly striking parallel is highlighted: all conditions that provoke hallucinations seem to cause time distortion. Consideration is given to the processes that may underpin changes in time perception, and it is proposed that they may provide a clue to the mechanisms of hallucination.
Arguably the most intriguing aspect of hypnosis is its ability to produce hallucinations. It does not achieve this for everyone, but those who are highly hypnotically susceptible appear capable of hallucinating vividly. In other words, they report convincing sensory experiences that do not reflect reality as perceived by others present. An example of this would be the volunteers at hypnotic stage shows who gaze goggle-eyed at the audience after being told by the stage hypnotist that they have x-ray vision and can see everyone without clothes. As is intended, many in the audience are amused and accept at face value the notion that the hypnotized person really “sees” a naked audience. However, the more perceptive ask how we know that the person is not acting. This is a good question, and it is one that has been so difficult to answer with certainty that psychologists researching hypnotic phenomena have looked to hypnotic behavior as a source of data, rather than self-reported experiences. Before we allow ourselves the luxury of exploring the inner, hard-to-verify world of the hallucination, we must follow the behavioral route in an attempt to determine whether hypnosis is really all that it seems.
Hypnotic behavior can look as dramatic as the hallucinating. Turning to stage hypnotists once more for an example, we find them telling a victim that he or she is absolutely rigid, like a plank of wood. To prove that the suggestion has worked, the human plank is picked up and placed between two chairs, one under the head and one under the feet; there is nothing in between. Clearly there is no acting in this case, but should we be impressed? The answer is that we should not, because the hypnosis was completely redundant; anyone who is not overweight and does not have spinal problems is capable of performing this feat without any hypnosis whatsoever. The human plank is of little use to a researcher investigating hypnosis.
So what sort of behavior is generally addressed by researchers? Typically it is rather unremarkable. Perhaps the best place to look for examples is within the battery of tasks used to assess a person’s hypnotizability. Because we are not all equally susceptible, it is common practice in research to include a standardized susceptibility test, along with whatever tasks are involved in the principal study. Some examples of the components of a test are as follows. The subject is told that his or her arm is going to feel lighter and lighter, and that it will gradually start to float up. For those who pass this component, that is exactly what happens. They are told that they will be unable to open their eyes, then challenged to try. The susceptible fail. They are instructed that, at the end of the session, they will be unable to recall all the tests that they were given, but upon a special instruction, the memories will all return. Again, the susceptible display a memory block when first asked to write down all that they recall, but they then recover after the instruction to remember.
One cannot avoid noticing that this list of tests consists entirely of activities that an actor could simulate; we seem to be back with the original problem. The obvious solution would be to find some behavior that an actor cannot reproduce, but unfortunately that has proved very difficult. A moment’s consideration shows that this must inevitably be so. A hypnotic induction often involves instructions for relaxation, followed by requests to engage in guided visual imagery. The latter might entail trying to imagine walking through a wood or perhaps going down steps into a sunken garden. The idea of descent is popular, as it leads naturally into the notion of going deeper into hypnosis. These procedures are all very pleasant and relaxing, but nothing is really done to the participant; it is not like an experiment in which the effects of a drug are monitored. Without some significant modification to a person’s physiology, such as a drug might bring about, it is obvious that they will continue to have access only to their normal repertoire of behavior. Thus it follows that the behavior could be acted. Perhaps the most curious aspect of this is that hypnosis researchers continue to employ these easily feigned tasks as a means of deciding how hypnotically susceptible their participants are; researchers might in reality be determining how well they can act. It is as if there is a tacit agreement between researcher and experimental participant, in which the participant undertakes not to act; participants will produce only the behavior that feels as if it is happening by itself (Naish, 1986).
It is difficult for the scientific approach to embrace notions that depend on trust and unverifiable experiences. As a consequence, for much of the latter part of the twentieth century, the majority of researchers adopted a skeptical stance with regard to hypnosis. Many used concepts from the field of social psychology to explain hypnotic behavior. Thus Wagstaff (1981) pointed out that as social animals, humans are disposed to compliance; in other words, we tend to go along with what is expected of us. Wagstaff’s position was that people were well aware of what the hypnotist expected. Moreover, they would come to the experiment believing that hypnosis existed and that many people were responsive. So, perhaps out of embarrassment at appearing to fail, they acted the part. Of course, not everyone would be so compliant, but then not everyone passes tests of hypnosis. Where this account may seem to stretch credulity is when hypnosis is used to control pain (e.g., Liossi & Hatira, 2003), but Wagstaff proposed that this was achieved simply by trying to think of something else, such as the woodland walk or garden, used in the induction.
One of the most prolific researchers in what might be called the “debunking” school of hypnosis was the Canadian Nick Spanos, whose research career and indeed life were sadly cut short in a flying accident. In a long series of experiments (e.g., Spanos & Burgess, 1994), he set out to show that so-called hypnotized people did precisely what they believed hypnotized people did. For example, subjects were given a list of words to remember, containing both concrete words (e.g., table, horse) and abstract words (e.g., democracy, hope). They were given the false but plausible story that concrete words are stored in one hemisphere of the brain, while abstract are retained in the other. They were further informed that hypnosis would produce amnesia for material stored in one of their hemispheres. Finally, after hypnosis, they were asked to try to recall as many words from the list as possible. Importantly, half the volunteers were told that concrete words were stored in the left hemisphere, while the rest were informed that the abstract words were held on that side. Each group was then halved again (producing four subgroups in all), with one set being told that they would suffer amnesia in their left hemisphere, and the other that this would occur in their right. The reason for this elaborate approach is that it covers the possibility that there might just possibly be something to this kind of story. However, if there were some element of truth, it could only be correct in one of the stories, so there should not be equal levels of amnesia in each condition. In fact, the results showed that one set of words was forgotten in every combination, and it was always the set that should have become inaccessible if the cover story had been true.
Clearly there could be no physiological explanation for the amnesia results, so Spanos concluded that subjects behaved strategically, carefully adjusting their responses to match the expectations. The closest Spanos came to allowing for a kind of reality to the hypnotic experience was to suggest that some experimental subjects may manage to imagine vividly and hence almost come to believe in their experiences, rather like an actor becoming lost in a role. Importantly, this account does not require that some special change of brain state takes place to achieve the behavior and engage in the imagination. Consequently researchers suggested that it was unnecessary to dignify the proceedings with a special label, “hypnosis.” In fact, many felt that the term was counterproductive, since in the public perception it carried so many false beliefs and expectations. Kirsch (1997) has claimed that there is a circularity in the use of the label, because it is employed as both an explanation of the condition and a test for it. Thus: Why does the person perform this behavior? Because he or she is hypnotized. How do we know he or she is hypnotized? Because of the behavior exhibited. This is like explaining a patient’s elevated temperature by saying that it results from the patient’s having a fever. In medicine we look for a further explanation, such as having a virus. In hypnosis it has proved extremely difficult to locate the equivalent of a virus; many have concluded that there is none.
3 Subjective Experience Revisited
Earlier it was shown to be inevitable that hypnotic behavior will be mundane. Similar logic shows that people must carry out the behavior they believe to be hypnotic. Since being relaxed and imagining a woodland walk is unlikely to induce any behavior other than passivity, it follows that the only activity observed will normally be associated with specific suggestions to do something. If, for example, the suggestion is that an arm is getting lighter, then the only reasonable response (other than inactivity) is for the arm to rise. When it does so, the subject is clearly doing what he or she believes is expected, but that does not demand an explanation in terms of deliberate compliance. It remains possible that some process, as yet undefined, but we will call it hypnosis, has enabled this action to feel to the participant as if it “happened by itself.” That, in fact, is the description often used by hypnotized people.
Consideration of the hemispheric amnesia experiment suggests that if the putative hypnotic process exists, it must be complex and subtle. Intuitively, it might be feasible to ignore words in one hemisphere (if they really were stored like that), but since the results demonstrate a strategic excision of just the right words, whatever the combination of instructions, it follows either that this is a deliberate and conscious act or that hypnosis is far more than a simple mental shift. It is not surprising that many researchers have preferred Occam’s approach and opted for the simple explanation. However, one difficulty for this stance is that it ignores a wealth of subjective reports. True, such reports are difficult to verify and often anecdotal, but some are compelling.
An impression of the subjective experience can be gained from the case of Valerie, who was a moth phobic. Hypnosis is an effective vehicle for the treatment of phobias, because a susceptible person is able to visualize the phobic object remarkably vividly, as if it were really there. The advantage of this scenario is that there is no need to acquire a real example of, in this case, a moth. Moreover, the “hypnotic moth” will do as the hypnotist instructs and not flutter unpredictably, which is the sort of behavior that phobia sufferers particularly dislike. With this level of control, it is possible, step by step, to familiarize the patient with the frightening situation, gradually engaging with it more and more closely. What is learned in hypnosis transfers effectively to real life. Valerie had coped (in hypnosis) with seeing a moth fluttering outside the window. She had then become relaxed about seeing one flying around a light in the room while she stood at a safe distance in a darkened corner, assured that the moth would not come there. It seemed to the author that it would be appropriate for Valerie to make physical contact with a moth, although she was still not ready to have one fly around her. She was asked whether she could cope with her hand brushing against a dead moth on the windowsill while she was dusting at home. She agreed to try, and with no more guidance than indicated in the previous sentence, she proceeded to give a commentary on what she was doing. What she was actually doing was sitting in a chair, with her eyes shut, but she described first gathering up the children’s toys, then dusting the piano. She then announced that she would go and dust the windowsill. After a pause, she said in a normal voice, “What’s this, an old leaf?” This was quickly followed with a gasp of apparent shock and “Oh no, it’s a moth!”
As an anecdote, this story can be used only illustratively, but it highlights some intriguing issues. First, the experience appeared to have a level of reality that one would not associate with mere imagination; it would seem to be better described as a hallucination. However, this reality, which included apparent shock and surprise, was actually sketched out in advance. How can someone be surprised by their own script, and why did Valerie first describe the moth as a leaf? True, moths may be leaf shaped, but it had not been suggested to her that she would first believe the moth to be a leaf. It is as if Valerie set the story in motion but was then taken by surprise at what unfolded. The apparent dissociation between instigation and registration of events that this implies is quite remarkable; it is not an account that should be accepted without more formal demonstration.
Haggard, Cartledge, Dafydd, and Oakley (2004) performed an ingenious experiment, investigating people’s time to register their own movement. The movement required was in a finger that, rather like the “light arm,” could be made to lift as if all by itself. Of course, the experimental subject could also move the finger voluntarily, and additionally a lever arrangement enabled the experimenter to lift the finger. There were thus three ways in which the finger could begin to move; in every case, the participant had to indicate as quickly as possible the precise moment when he or she detected the movement. When the movement was self-initiated and deliberate, the response was very quick, but when the experimenter caused the finger to move, a small delay occurred between the movement and its detection. These results were expected; the interesting question was how quickly the subject would register the “moving by itself.” In truth, this movement was self-initiated, so a quick response would be predicted. On the other hand, if the movement felt automatic and came as a surprise, then its detection should come later, just as when movement was induced by the experimenter. The result was a slow response, precisely as if the subject did not know that the finger was going to lift. This finding serves as strong supporting evidence for subjects’ claim that their subjective experience is of the finger moving by itself.
The hallucinations described in the previous section, whether seeing a moth or feeling that actions take place outside one’s volition, occur in response to suggestions. As a result, these distortions to reality may serve to convince a skeptic that hypnosis “works,” but they do not reveal how the experiences are generated. It would be helpful to examine a distortion that occurs simply by virtue of being hypnotized, without any specific suggestion being required. A modification to perception that is intrinsic to hypnosis is more likely to offer clues about the changes in mental processing that are brought about by the hypnotic process itself. There is a candidate for a nonsuggested distortion: time perception. As will become clear, the phenomenon of time distortion is of particular relevance to the processes of hallucination, so we will look at timing effects in some detail.
After a session of hypnosis, the majority of people underestimate its duration, not uncommonly by as much as 50 percent. Although this effect was first reported around thirty years ago (Bowers & Brenneman, 1979), the precise mechanisms underlying the changed perception have remained elusive. A particular puzzle has been the relationship between the extent of the distortion and a person’s hypnotic susceptibility. It is reasonable to suppose that if the underestimate is caused by hypnosis, then those who are more hypnotized will experience a greater level of underestimation. Surprisingly, neither the original Bowers and Brenneman study nor many subsequent studies (for a review, see St. Jean, McInnis, Campbell-Mayne, & Swainson, 1994) have managed to demonstrate such a relationship. Nevertheless, more recently (Naish, 2007) I have shown that whereas highly susceptible subjects reduced their estimations during hypnosis on average by 24 percent, subjects of very low susceptibility actually increased their judgments by 15 percent. This suggests that, at least for people from the extremes of hypnotic susceptibility, there are quite different processes taking place. However, it does not shed light on exactly what takes place.
One of the more plausible explanations to emerge attributed the underestimation to mental workload (St. Jean et al., 1994). These researchers observed that time seems to fly when one is engrossed in a task, so they suggested that the mental activity required to keep oneself in the hypnotic condition acted as just such an engrossing process. One of the experiments conducted by St. Jean’s group was taken to support this hypothesis. Hypnotized subjects either listened to a story (considered a low-workload task) or listened to a story while at the same time solving word puzzles (higher workload). In both cases, they were required at the end to say how long the story seemed to have lasted. In the low-workload condition, people’s estimates averaged 63 percent of the actual duration, while under the high-workload conditions, the estimate fell to 43 percent. This change was as would be predicted on the basis of workload, but since the subjects were hypnotized throughout, it was not possible to reach any conclusions about the impact of hypnosis. However, the researchers then conducted a more pertinent experiment, in which some subjects were tested in hypnosis and others without hypnosis. Incidentally, although it is now known that hypnosis cannot in any sense be likened to sleep, it remains a convenient shorthand to refer to nonhypnotized people as “waking.” St. Jean et al. report that, in the waking state, when people were engaged in a low-workload task, the estimated duration was 92 percent of the true time. In hypnosis this fell to 80 percent, which is not a particularly large reduction. The corresponding figures during high workload were 45 percent for waking and 64 percent for hypnosis. Curiously, adding hypnosis, far from increasing the distortion, seemed to have reduced it, so the notion that hypnosis acts as a source of high workload is not supported. A possible explanation for this strange finding is that different people were used for each section of the study. It may have been that, by chance, those who tended to experience little time distortion were picked for the high-workload-plus-hypnosis group. Since even without hypnosis, people vary in their time estimation accuracy, it follows that the most useful results will be obtained by testing the same person both with and without hypnosis. As explained in the next section, this I have done.
5 Hypnosis and the Internal Clock
It is widely presumed that our ability to make timing judgments is based on the existence of some form of inner clock. This is assumed to produce signals (“ticks”) at a steady rate, which are registered by a mechanism that accumulates ticks over a period of time. The accumulated value represents the duration. The exact nature of the ticking clock is unclear, but evidence suggests that it is linked to the rate of some form of neural oscillator. Fox, Bradbury, Hampton, and Legg (1967) demonstrated that time estimates increased in a person with a fever; presumably the raised temperature led to neurochemical reactions taking place more quickly and hence caused the oscillator to run faster. This in turn would produce more ticks in a given period, so the period would seem relatively long. Patients suffering from Parkinson’s disease (PD) exhibit time estimation problems. They are known to have depleted levels of the neurotransmitter dopamine, the deficiency appearing in the brain’s basal ganglia. This complex lies in a deep region of the brain, serving as a relay between the cortex and the outside world. The basal ganglia receive incoming information from the senses and also direct signals to the muscles. Since PD patients make inaccurate time estimates, researchers have suggested that the site of the clock lies in the basal ganglia (Harrington, Haaland, & Hermanowicz, 1998; Smith, Harper, Gittings, & Abernethy, 2007). We will return to the experience of PD patients, but for the present it suffices to note that, like hypnosis subjects, they experience time distortion, which is presumed to result from an imprecise oscillator in the basal ganglia, producing less-than-constant “ticks.”
We will now revisit the issue of workload. As explained, St. Jean et al. (1994) proposed that the workload associated with maintaining the hypnotic state was the cause of the mistiming. This idea was derived from outside the field of hypnosis, where it was known that additional tasks made time periods seem shorter. Researchers in the field of time perception assume that the impact of increasing workload comes about through the reduction in resources available to count the clock tick (e.g., Brown & Boltz, 2002). If a proportion of ticks was missed, then the accumulated score at the end of a timed period would be less than normal, hence leading to the perception that less time had passed.
Although the timing effects associated with manipulations of workload and attention to other tasks are plausibly related to the number of timing units counted or missed, this is not the only possible cause of timing errors. By analogy with PD, we have another candidate for bringing about changes: the tick rate itself. If this were to change, then the number of ticks counted in a given period would also change, giving rise to changed estimates of duration. What would be the observed effects if hypnosis caused an inner clock to run more slowly? With fewer ticks being counted per unit of real time, the overall number accumulated over a session of hypnosis would be relatively small, leading the subject to conclude that a shorter length of time had passed. That, of course, is the usual observation. A further prediction can be made. If it is true that the clock ticks slowly, a subject waiting for a period of time to pass will wait too long. Suppose, for example, that a subject tries to wait two minutes before carrying out an action. From previous experience, he or she will have in mind what two minutes feels like, presumably based on some representation of the tick accumulation value for this duration. If the clock begins to tick more slowly than usual, then inevitably the person has to wait longer for the appropriate tick value to accumulate.
The “waiting two minutes” idea has been tested (Naish, 2001) by asking hypnotized participants to indicate when they believed two minutes to have passed. They did indeed overshoot, producing an average duration 21 percent longer than a true two minutes. In contrast, at the end of the hypnosis session, their estimate of the overall time taken was only 64 percent of the actual 35 minutes or so that had passed. This comparison of prospective time judgment (i.e., waiting for a future time) with retrospective estimation was made without a comparison with waking judgments, so a more formal evaluation was developed (Naish, 2006). This produced a prospective time estimate that was 60 percent longer in hypnosis than when waking. The retrospective judgment, made by the same people, was 32 percent shorter in hypnosis than in the waking condition.
Unlike many other changes apparently caused by hypnosis, there is no likelihood that this shift in timing is brought about by acting. This position is supported by the findings of Mozenter and Kurtz (1992), who showed that people asked to act as if hypnotized failed to reproduce timing effects. Their failure shows that people do not have accurate expectations about the impact of hypnosis on time perception and so have no idea how to act. Thus the effect can safely be attributed to hypnosis itself. As indicated, this may very well result from the slowing of a clock, but it should be noted that similar results would be obtained if hypnosis interfered with the counting of ticks. However, this putative counting problem is generally associated with increased workload, and as already explained, it seems unlikely that workload is the cause of hypnotic time slowing.
6 Timing Accuracy and Hallucinations
It may seem strange to spend so long in discussing the effects of hypnosis on time perception when we have the far more interesting topic of hallucinations to address. However, the slowing-clock effect is important in being directly attributable to the effects of hypnosis; it confirms that some neural processes are actually changed by being hypnotized. Moreover, as will be shown, the slowing-clock effect has direct links with hallucinating, both in hypnosis and in other conditions.
As mentioned earlier, hypnotic inductions frequently include visualization of a relaxing scene, a visualization that acquires hallucinatory qualities for the hypnotically susceptible. Frequently the scene selected is of a beach, and an induction with this image was incorporated into a time estimation experiment (Naish, 2003). Participants were shown a picture of a particular scene (a tranquil bay, harbor jetty, fisherman) before commencing hypnosis. After a period of relaxation, with their eyes closed, they were asked to “take themselves” to that scene and to imagine themselves waiting to meet a friend there, who was due in precisely five minutes. In the meantime, they were welcome to do as they pleased at the beach. They were given a stopwatch, told to start it at the beginning of the waiting period, then stop it when they believed the friend was due. Participants had eyes closed throughout and were not permitted to view the stopwatch to check on the time. Once they had indicated that the time was up, participants were brought out of hypnosis and asked about the vividness of their experience and also the extent to which the actual outside world intruded (sounds of people outside, for example). They rated vividness and intrusion on seven-point scales, so that a person who became “lost in the scene” might score 7 for vividness and 0 for intrusion; participants from the other end of the spectrum rated vividness very low and awareness of the outside world rather high. The two scores were combined into a composite, which can be treated as a measure of “detachment.” Participants scoring high on detachment became extremely involved in the scene; one woman said that she decided to sunbathe on the harbor wall but after a while found the concrete to be very hard.
There was a statistically significant correlation between detachment scores and the length of time that had actually passed when participants decided that five minutes had elapsed (r = 0.75, p < 0.001). Some participants who scored very low on detachment pressed their stopwatches after only three minutes, whereas those who were successful in generating a thoroughly realistic world did not judge that the five minutes were up until seven minutes or more had actually passed. It is clear that time dilation is strongly associated with the presence of a convincing hallucination.
In section 5, I mentioned that patients with Parkinson’s disease exhibit a deficit in the ability to make accurate time estimates. This problem has received a great deal of research interest, and it is now well established that the nature of the inaccuracy is broadly equivalent to a slow-running clock (see Smith et al., 2007, for a review). Typically, if required to depress a button for an indicated duration, patients produce an interval that is too long. Moreover, if the task is repeated a number of times, responses are found to exhibit a greater-than-normal inconsistency. Overshooting of target interval and increased variance in the attempts is exactly the response pattern exhibited by hypnotized people who are highly hypnotically susceptible (Naish, 2008). The similarity between hypnosis and PD is thus quite striking, but there is another parallel, which to many is even more surprising: PD patients can experience hallucinations (Fénelon, Mahieux, Huon, & Ziégler, 2000). In addition, the evidence strongly suggests that patients who suffer from hallucinations may also be less accurate in their time estimation (see Maravic da Silva, this volume).
The condition most commonly associated with hallucinations is schizophrenia. The auditory hallucinations that are often considered diagnostic of schizophrenia are sufficiently similar to those that can be induced in hypnosis that Szechtman, Woody, Bowers, and Nahmias (1998) used hypnosis-induced auditory hallucinations as an analogue to those experienced by schizophrenia patients. This approach was taken so that brain scanning could be carried out during hallucination; it was considered methodologically complex to obtain scans of a patient when “hearing voices.” The experiencing of hallucinations is not the only parallel between the medical condition and hypnosis; patients report feeling that their own actions occur as if controlled from outside, just as the hypnotized person feels that movements happen by themselves. In fact, the similarities between hypnosis and schizophrenia are such that there is a correlation between hypnotic susceptibility and a measure known as schizotypy (Gruzelier et al., 2004). Schizotypy scales assess the extent to which members of the nonpatient population experience unusual perceptions, such as thinking that they heard someone speak when they knew that no one else was present. People prone to these kinds of experience tend also to be more hypnotically susceptible. There is one more parallel between hypnosis and schizophrenia: patients exhibit time perception difficulties (Elvevåg, Brown, McCormack, Vousden, & Goldberg, 2004).
7 The Senses, Consciousness, and the Clock
It is well established that we become consciously aware of only a small fraction of the incoming stimuli that our brains process (see Naish, 2012, for an overview). Not only is our awareness restricted to a small subset of all that we could in principle monitor, but our conscious experience exceeds what we actually know. This is strikingly demonstrated in the phenomenon of change blindness (e.g., Simons & Rensink, 2005). Two pictures are alternated; they are broadly identical, but a significant feature differs between the pictures, and the viewer is required to identify the change (see, e.g., http://www.usd.edu/psyc301/Rensink.htm). As a participant views either picture, he would claim that he has a full awareness of the scene. Nevertheless it is clear that he does not, since he is likely to require a protracted inspection time before he is able to find the changing feature. Observations such as these suggest that conscious experience is best described as providing an approximate representation of just some of our environment. The existence of this less-than-total identity between reality and experience suggests a system with the potential to become still further detached and thus produce unreal hallucinatory experiences.
Gray (1995) proposed a model of consciousness that accounted both for the selection of material to reach awareness and also for the unreal experiences of schizophrenia. He suggested a cyclic system that repeatedly took “snapshots” of the environment and compared these with what had been predicted on the basis of the previous sample. Discrepancies between observation and prediction would have two results; the predictive system would be updated, and the unexpected observations would probably be selected for attention and awareness. Following this, the cycle would be repeated. In schizophrenia, Gray suggested, there was a problem with the comparator, resulting in many predictable events being treated as if unexpected, and hence requiring attention. Even the patient’s inner voice (when talking to himself or herself covertly) could be treated as unexpected and consequently attributed to some external source; hence the phenomenon of hearing voices.
Gray’s (1995) model included an estimate of how frequently the test-and-predict cycle would be repeated; he suggested ten times per second, or 10 Hz. A series of experiments by Treisman and colleagues (Treisman, Faulkener, Naish, & Brognan, 1990; Treisman, Faulkner, & Naish, 1992; Treisman, Cook, Naish, & MacCrone, 1994) had led us to conclude that the “inner clock” ticked at approximately 12 Hz, which is sufficiently close to Gray’s estimate to raise the possibility that they may be one and the same system. The brain certainly produces electrical activity in this frequency range; such activity is called alpha waves, and it has been suggested that they derive from the thalamus, a brain structure that is one of the basal ganglia. Gray had suggested that the driver of his proposed circuit may be the thalamus, and as detailed earlier, the basal ganglia are very much implicated in timing. Thus there may be a link between the system that gives rise to conscious experience and the mechanisms of timing.
For consciousness to be associated with a neural timekeeper is not implausible, because it would serve to keep a number of complex circuits in synchrony. For a simple system, timing has no relevance. For example, we are able to enjoy a stable image of the world while running or shaking our head. This is achieved by the eyes swiveling by just the right amount in the opposite direction to the head movement. This happens automatically, with nerve fibers making a direct link from the balance organs of the inner ear to the muscles of the eye. Only one response is required for a given signal from the sense organ, so the brain does not need to be involved. However, as soon as a range of signal and response combinations is required, it becomes necessary to introduce a brain into the loop.1 Moreover, if some responses are complex, it becomes necessary to monitor that they have succeeded in reaching the required goal; in other words, a closed loop is required, which transmits control signals and receives feedback signals. The return data would have to be compared with the expected state, the expectation signal being made available at just the right moment for the arriving feedback. Any mismatch would require correcting signals to be transmitted; it might also necessitate the process being brought into conscious awareness. The concept of noting mismatches begins to sound similar to Gray’s account and suggests a level of complexity that may require precise timing. Nevertheless to deal with this kind of feedback may be relatively simple and amenable to automatic control. Real complexity comes with multimodal signals, that is, signals from across the senses.
One of the impressive features of our sensory system is that we have unitary experiences. When we identify an object, we become aware of a whole range of attributes, such as its shape, color, position, direction of movement, and whether it is the source of a sound. All these features are processed in different brain regions, yet our conscious experience is of a single complex stimulus. To unite the correct features from among all the potential candidates appears to require conscious attention and probably involves synchronization of the disparate neural signals via gamma wave oscillations (Doesburg, Roggeveen, Kitajo, & Ward, 2008). Gamma waves are cyclic electrical activity in the brain with a frequency of approximately 40 Hz. They occur across the brain, but associated regions can become synchronized, like a subset of musicians following a conductor while most are playing to their own time. It should be noted that gamma waves have a frequency approximately four times higher than alpha waves. Earlier, based on Gray (1995), I suggested that the data-gathering cycle may run at the alpha rate of about 10 Hz. The faster gamma waves would have to act within that cycle to bind together the elements of each snapshot.
Not only does attended material stand out through a process of gamma synchronization, but additionally its level of activity is enhanced, while that of other, unattended material is inhibited (Buehlmann & Deco, 2008) as if the musicians who do not follow the conductor are required to play more quietly. The role of conductor in the brain appears to be represented by the prefrontal cortex; it is extensively interconnected with more posterior regions, that is, the areas associated with the analysis of incoming signals and the generation of responses.
To summarize the foregoing, successful interaction with our environment requires effective meshing of a number of circuits. Incoming information from the senses is scanned to determine whether a response is required, and sensory information about the responses themselves is also monitored to ensure that they are carried out successfully. The selection and control of actions require a memory system that can supply information about expected states or outcomes. This information is used to classify incoming material into that which requires attention, and so needs to be made more salient, and that which is as expected, and hence can be inhibited so as not to reach conscious awareness. These processes can succeed only if they take place in synchrony. In the central processor of a computer, the various processes are synchronized by a clock; it is possible that a similar mechanism is used in the brain. Moreover, the oscillatory processes that synchronize the neural circuits may also serve as the basis of the “inner clock.” If some kind of disruption affected any of the circuits, it is reasonable to suppose that conscious experience would be changed, perhaps resulting in hallucinations. Additionally, the accuracy of the clock would be compromised. We will now consider whether there is evidence to support this proposal.
A simple account (Naish, 2003) can explain hypnotic time changes; it is based on two topics discussed so far. These are the slowing clock of people who are able to generate a convincing scene (sec. 6) and Gray’s (1995) “consciousness cycle” (sec. 7). The consciousness cycle hypothesizes that information is gathered in snapshots, at a rate of about ten times per second, and this rate may also serve as the basis of the clock. If a person ceases to monitor reality and instead generates his or her own inner world, then there is perhaps less need continually to update the information. If the world is self-generated, it will not change unpredictably and so can be scanned at a slower rate. If that happened, it would follow that the clock would run more slowly, which is exactly what is observed. Thus the more successful the generation of an imagined world, the slower the clock.
I have highlighted the parallels between hypnosis and conditions in which the brain is clearly malfunctioning. For example, in PD the malfunction (or at least a significant part of it) is a depletion of a neurotransmitter. The deficit is found at the heart of the putative clock, which impairs its effectiveness for making accurate time judgments. We will consider later why this might in turn give rise to symptoms such as hallucinations, but first it is worth noting the proposed causal sequence: a timing problem, it is suggested, may be the cause of changes in conscious experience.
In the earlier account of hypnosis, the causal chain was inverted: experience changes first, then as a result so does the clock. Conceptually it is more plausible to propose that a fault in the system gives rise to a change in experience, rather than what is in effect the reverse. For a start, we might wonder what causes the changed experience in hypnosis when there is no formal fault in the system. Second, we may well ask how the altered experience would alter a neural clock rate; the suggestion smacks of a Cartesian dualism, with a disembodied (changed) mind acting to change physical circuitry. Dualism it may be, but this does at least remind us of moth-phobic Valerie’s account (sec. 3), where she seemed surprised by her own script. A way out of this conundrum would be to characterize the hypnotized brain as being like a damaged brain. Given the introductory sections of this chapter, which stressed the rather mundane nature of hypnosis, the idea of it causing alterations that are the equivalent of brain damage may stretch credulity. Nevertheless evidence suggests that this may be the case.
There is a region of the brain referred to as the cingulate cortex (sometimes cingulate gyrus). As part of the cortex, it lies on the surface of the brain, but it is hidden on the inner surface of each hemisphere, where the two touch. The cingulate region, particularly the anterior part, appears to be involved in processes linked to “paying attention” and is also active when perceived events are not as expected. It may serve as a modulator of the extent to which the prefrontal cortex is able to act on other regions of the brain. It has long been known (Whitty & Lewin, 1957) that damage to the anterior cingulate cortex (ACC) has an impact on what is perceived as real; the authors referred to this as “vivid day-dreaming.” Patients who had received an anterior cingulectomy appeared to have difficulty in distinguishing between the real and the imagined. This does not sound very different from the experience of a person deeply hypnotized, where scenes that in nonhypnotized people would be no more than imagined appear to take on a more vivid reality. In view of this, it is particularly interesting to note that brain scanning of people undergoing hypnosis frequently reveals a change in activity in the ACC. It is thus possible, in principle, that hypnotizable people are able voluntarily to modify the activity of a brain region to produce effects analogous to those who have suffered injury to the area. I am not aware of any studies reporting the effects of a cingulectomy on time judgments, so it is not possible to take these parallels any further.
Interestingly, Stebbins et al. (2004) report a reduction in activity in the cingulate gyrus in PD patients who experience hallucinations, but not in patients who are free from hallucinations. Moreover, Shergill, Brammer, Williams, Murray, and McGuire (2000), using a functional magnetic resonance imaging (fMRI) study, concluded that the ACC is one of a number of regions involved in the generation of auditory hallucinations in schizophrenia. Shergill and colleagues also point out that normal behavior and experience require accurate prediction (as described in sec. 7) and go on to show (Shergill, Samson, Bays, Frith, & Wolpert, 2005) that there are sensory prediction deficits in schizophrenia. Subsequently Shergill and collaborators (2007) have used more advanced fMRI techniques to demonstrate abnormalities in the neural tracts linking frontal regions of the brain with the more posterior regions associated with perception and behavioral control. Similarly, Lawrie et al. (2002) concluded that there was a reduction in effective connectivity between frontal regions and the more posterior temporal region. The picture that emerges is of hallucinations deriving from self-generated behaviors (such as self-speech) that, as a result of failed inhibition, are interpreted as unexpected and hence of external origin. The findings of Stebbins et al. (2004) may support a similar account of the hallucinations in PD. In addition to the cingulate effects reported in their study, it also revealed a general shift in activity away from the more posterior visual-processing areas of the brain, toward the frontal regions associated with attentive processes. This implies a reduction in the use of sensory input to determine what is perceived, but an increase in the use of predictive mechanisms, as it were to fill the gap.
A failure to inhibit what should have been expected sensory activity has also been demonstrated in hypnosis. In section 3, I described the study by Haggard et al. (2004); participants were relatively slow to detect movements when these were experienced as “happening by themselves.” Blakemore, Oakley, and Frith (2003) used PET scanning, while subjects produced either voluntary or hypnotically induced movements. The resultant activity in the parietal region of the brain was considerably greater when the movements were experienced as nonvoluntary. As in the case of schizophrenia, the increased activity was presumably the result of failure to inhibit what should have been a predicted detection of movement.
9 Top-Down Processing and Hypnosis
Predictive processes can have two roles. Their value in the inhibition of expected information has been described, but prediction is also important in perceptual analysis itself, and here its function is more likely to be excitatory than inhibitory. I indicated in section 7 that conscious awareness was not a faithful and complete representation of external reality. Part of what is actually experienced is a result of the brain’s imposition of an interpretation of reality on the incoming data stream. Gregory (1980) described this as analogous to hypothesis testing. At a higher level of processing, a possible interpretation of the stimulus begins to emerge, and this “hypothesis” is tested against the incoming data by carrying out further analysis. The influence of more advanced stages of processing on earlier analysis is referred to as top-down processing. It is now known that the top-down influence is considerable, to the extent that earlier stages of processing are not simply rechecked: their activity is actually modified. Of course, the results of the modified analysis then feed forward to the higher levels, which may in turn cause an alteration to the top-down signals. This circular mechanism has been termed reentrant processing (e.g., Di Lollo, Enns, & Rensink, 2000), and it implies that perception is best conceived as a cyclic, rather than linear, sequence of processes.
Such is the power of top-down processing to modify input that it can result in activity where none should be present. Thus Calvert et al. (1997) demonstrated that activity occurs in the auditory cortex when someone views a silent video of a speaker. The expected speech sounds, associated with the lipreading, result in the generation of auditory activity. An observation such as this perhaps makes it a little less surprising that a schizophrenia patient may hear a voice.
Hypnosis appears to be a remarkably potent vehicle for generating activity where none would have been expected. I mentioned Szechtman et al. (1998) in section 6 in the context of auditory hallucinations. They used PET scanning with hypnotized subjects who expected to hear speech but were actually in a silent environment. Although these participants did not even have silently moving lips to watch, they nevertheless produced neural activity equivalent to that produced when speech was present. These effects are not confined to the auditory domain. Kosslyn, Thompson, Costantini-Ferrando, Alpert, & Spiegel (2000) were able to create a color response in the visual cortex when their hypnotized participants were actually viewing black-and-white stimuli.
I have shown a shift in the way faces and facelike material are processed. In an unpublished pilot study, working with Sven Braeutigam at the University of Oxford magnetoencephalography (MEG) facility, participants were scanned while viewing a sequence of visual stimuli. These comprised a mix of three kinds of pictures: photographs of faces, so-called Mooney faces (simplified, cartoonlike pictures), and nonface patterns. Some of the Mooney faces were very facelike, while others showed little resemblance to a face. Thus the stimuli covered a range from clear faces to obvious nonfaces, and as each was displayed, participants were required to indicate whether the picture showed a face or not. The testing took place in three phases: waking, in hypnosis, and in hypnosis, together with the suggestion that the faces would stand out far more clearly.
The proportions of face and nonface judgments made by participants did not differ between waking and simple hypnosis, but following the suggestion for greater clarity, the subjects judged far more of the stimuli to be faces and made correspondingly fewer nonface judgments. Participants appeared to be seeing faces in stimuli that would previously have been rejected. The postexperiment debriefing confirmed that participants’ experience was of seeing more faces. The scanning results endorsed this. There is a region of the brain (the fusiform gyrus) that produces a strong response to face stimuli and far less to other visual material. Following the “enhancing” suggestion, stronger signals were detected in the fusiform region; hypnosis had modified the brain’s response.
In the context of hypnosis, negative hallucinations should be considered, since using hypnosis in this way is not unusual. A negative hallucination is the failure to perceive something, when there is clearly sensory input to indicate its presence. The use of hypnosis to control pain is well established (e.g., Liossi & Hatira, 2003) and is, in effect, an example of a negative hallucination. Faymonville et al. (2003) obtained a 50 percent reduction in the reported intensity of experimentally induced pain in hypnotized subjects. Scanning revealed changed activity across extensive neural circuits, including the cingulate region. Not surprisingly, given the potential of top-down effects discussed earlier, hypnosis can also facilitate the production of pain. Derbyshire, Whalley, Stenger, and Oakley (2004) demonstrated the generation of pain in the absence of a noxious stimulus and showed changes in a number of brain regions, again including the cingulate.
In section 7, I explained that synchronized gamma-band oscillations appear to be involved in enabling an awareness of composite stimuli, uniting information from across different brain regions. De Pascalis, Cacace, and Massicolle (2004) showed that when highly hypnotizable subjects responded to analgesia suggestions (i.e., ceased to feel a painful stimulus), gamma synchrony was lost across the regions normally active during an awareness of pain. De Pascalis (2007) draws on a number of sources to suggest an important role for the modification of gamma activity during altered states of consciousness, including the appearance of unusual gamma activity in schizophrenic patients (e.g., Haig et al., 2000). The nature of the relationship between gamma synchronization, inhibition, and top-down processing is far from clear; they may be independent processes or components of a unified process. However, all have been implicated in the generation of hallucinations, whether in hypnosis or in conditions of brain malfunction.
10 Summary, Recent Developments, and Conclusions
This chapter has outlined the remarkable versatility of the highly hypnotizable brain. It is an unimpaired brain, yet it is able to generate hallucinatory experiences very much like those of schizophrenia or PD. What is more, the mode of generation seems to involve similar processes, including changed activity in the cingulate region and a reduction of inhibition. We should perhaps not be surprised at these similarities, since we might ask how else a person could experience a hallucination. Ultimately hallucinations must derive from inappropriate brain activity, and that activity will be underpinned by modifications to processes such as inhibition and top-down control. That will be the case whether the changes are the result of disruption to neural pathways or the effects of hypnotic suggestion. Nevertheless it is still appropriate to ask why only some people are able to achieve these hypnotic effects; intuitively, one would expect their brains to differ in some way. Derbyshire (personal communication) has some evidence from MRI scans that highly hypnotizable people have a greater proportion of white matter. White matter in the brain comprises the interconnections between neurons. Conceivably, possession of more interconnections confers a greater ability to self-generate “reality,” rather than relying on external stimuli as the basis. In yet another parallel, Seok et al. (2007) report greater white matter density in schizophrenics who experience auditory hallucinations. Of course, schizophrenic hallucinations come unbidden, whereas in hypnosis they appear to order. However, recall that a correlation exists between scores on schizotypy scales and levels of hypnotizability (sec. 6). It seems entirely possible that the differences between the highly hypnotizable person and the schizophrenic are in degree, with an ever-increasing chance of spontaneous hallucination as the continuum is traversed.
A significant part of this chapter has been concerned with distortion to time perception; it has an important place in the field of hypnotic misperception. Whereas hypnotic hallucinations generally require appropriate suggestions to bring them about, time distortion is exhibited automatically. Moreover, it also occurs in the other hallucination-generating conditions that have been considered. The strong association between time distortion and hypnosis suggests that it may well be a pointer to underlying mechanisms; that it occurs in schizophrenia and PD suggests that the mechanisms of hypnosis may have similarities to those that produce pathological hallucinations.
There is no doubt that the hypnotized brain behaves very differently from the waking brain when performing a timing task. As part of the series of pilot studies I have undertaken at the Oxford University MEG facility, participants were required to listen to a beep of duration in the range of a few seconds. They then made a duration judgment. As generally seems to be the case, judgments under hypnosis tended to be shorter, as if an inner clock were running more slowly. However, of more interest were the patterns of brain activity long before a judgment was made. Figure 7.1 shows a series of scans (composites from three participants) covering approximately the first second from the moment of stimulus onset. Since the shortest beep lasted three seconds, the scans fall within the period when the sound was still present. The scan sequence represents the levels of activity prevailing, every 20 milliseconds following the start of the beep. The views are looking down on the head, with the nose at the top; the red coloration represents the greatest activity. Figure 7.1 shows the development of activity in waking subjects. It can be seen that strong bilateral activation begins about 100 ms after the beep started, this representing the response to the sound. The activity declines, since an unvarying stimulus requires little processing, but it becomes far more marked on the right of the brain, somewhat toward the front. This activity is still present in the last of the scans, almost a second after stimulus onset. It is probably indicative of timing activity; this cannot be formal, subvocal counting, since that would have been revealed in the left (speech-supporting) hemisphere, but may be similar to the activity reported by Lewis and Miall (2006). These researchers concluded that important timing circuits are located in the right prefrontal region.
Figure 7.1
MEG scans, showing activity across the brain while listening to a beep and estimating its duration.
A glance at figure 7.2 reveals considerable differences; these scans show the situation during hypnosis. Activity may begin slightly later, but two far more significant changes occur. The first is the marked reduction of the activity in the region previously associated with timing. Whether or not it is correct to identify this region as central to the timing task, some activity remains, and indeed, the participants continued to produce reasonably good (although shifted) time estimates. The other clear difference in the hypnosis condition is the strong activation on the midline. This we presume to reflect activity in the cingulate, the region so often implicated in hypnosis studies. It is worth emphasizing that figures 7.1 and 7.2 represent the same participants engaged in exactly the same task; it is striking to see such very different patterns of activity. No matter how hypnosis is conceptualized, it is clear that the procedure results in significant changes in brain behavior.
Explanations for the association between mistiming and hallucinating must inevitably be conjectural. The right dorsolateral prefrontal cortex, identified by Lewis and Miall (2006) as being involved with timing, receives neural projections from the basal ganglia, the region more often proposed as the site of the inner clock. If the basal ganglia were the source of ticks and the cortical region counted them, then any disturbance to the basal ganglia and associated circuits (as implicated in PD and schizophrenia) could lead to a change in the rate at which ticks were received. Conversely, if hypnosis led to a reduction in activity in the region that counted ticks (as shown in the MEG data), it is plausible that fewer ticks would be detected, and hence less time would be judged to have passed.
A question of interest to philosophers is whether it is possible for hallucinations to be indistinguishable from reality. In the case of hypnosis, participants do seem always to know that the experience is not real. This may be situation dependent; the person is aware that he or she has agreed to be hypnotized, can hear the hypnotist’s voice, and so on. In spite of the hallucinations being convincing, there remains the knowledge that the broader context offers a satisfactory explanation for the experience.
The foregoing contrasts starkly with the situation when a post-traumatic stress disorder (PTSD) patient experiences a flashback. PTSD develops in approximately 30 percent of people exposed to traumatic events, such as being involved in a road traffic accident in which there were fatalities. A characteristic symptom of PTSD is the vivid reexperiencing of elements of the event. Months after the trauma, the victim can suddenly suffer a brief but vivid hallucination, encompassing all the senses. These experiences can also be summoned from within hypnosis, which is an effective vehicle for treating the condition, but when the hallucinations are spontaneous, they seem completely real. A victim of the London tube bombings of July 2005 informed me that her flashbacks made her feel that she was back in the tunnel, where she was going to die. These hallucinations were so convincingly real that she believed the periods of genuine reality, during which people were reassuring her that she had been rescued and was safe, were only her mind playing her tricks; the underground was the reality. Interestingly (and probably not surprisingly at the end of this chapter), PTSD sufferers experience considerable time distortion for the events, and they tend to be more than averagely susceptible to hypnosis (Yard, DuHamel, & Galynker, 2008).
Since this chapter was first prepared for publication, further developments have thrown more light on the similarity between PTSD and hypnosis. Research into the perceptual style of PTSD patients has shown them to be “global” processors (Vasterling et al., 2004). This term is contrasted with “local” processing. Global processing may be thought of as the mode of visual analysis by which the brain acquires the “big picture” of a visual scene, whereas local processing yields information about the fine detail. Patients are often chronically anxious, as if always looking out for danger. This would be a plausible explanation for a processing style that remains alert to the entirety of a scene, in preference to focusing upon the fine detail of one particular region. Of particular relevance to our topic, in right-handed individuals global processing appears to be handled principally in the right hemisphere of the brain, while the left deals with the finer details. PTSD patients have thus shifted the emphasis in hemispheric activity toward the right.
I have shown (Naish, 2010) that right-handed individuals who are highly hypnotizable are more than averagely left hemisphere biased while they are unhypnotized. However, they shift strongly from left to right hemispheric emphasis during hypnosis. A possible conclusion to be drawn from the parallels is to suggest that PTSD patients readily generate flashbacks, because their hemispheric imbalance is like that of a highly hypnotized person, who of course easily experiences hallucinations. As explained in section 6, there is a correlation between hypnotic susceptibility and schizotypy. People who score high on schizotypy scales show greater right hemisphere activity (Kelley, 2011). Moreover, it has been observed that schizophrenia patients frequently show unclear hemispheric dominance, that is, they are neither strongly right nor thoroughly left handed. On this basis Caligiuri et al. (2005) proposed that the right hemisphere in these patients is insufficiently controlled by the left, and as a result generates hallucinations. It is possible that the hemisphere is better able to initiate top-down generation of hallucinations, since these would be more readily produced via a global-first strategy.
There is a further possible link between trauma, hypnosis and schizophrenia. Schizophrenia and other forms of psychosis in adults are strongly associated with childhood trauma, frequently sexual abuse (see, e.g., Hammersley et al., 2003). Similarly, a larger than expected proportion of highly hypnotically susceptible adults report abusive childhoods (Barber, 1999). Bearing in mind that PTSD victims hallucinate (flashbacks) and exhibit a right hemisphere emphasis, we may hypothesize that trauma and abuse in childhood has a similar effect. If the trauma is prolonged or severe it may lead some victims to become permanently right-hemisphere leaning, or unusually labile in the relative bias between the hemispheres. In adulthood such an individual may experience unpredictable hallucinations (in psychosis) or be able easily to generate them to order (in hypnosis). Finally, a shift to the right hemisphere, introducing more activity than is normal, might very well interfere with the working of the clock circuitry located there (Lewis & Miall, 2006).
What are we to conclude from the wide-ranging material discussed here? It is a disparate collection, yet with some remarkable parallels. Having explored a number of those parallels with other hallucination-generating conditions, I will confine my concluding remarks to hypnosis itself. People sometimes ask me whether I believe that it is possible to hypnotize animals. I have to answer that I do not, at least not in the sense that I understand hypnosis. It must take a remarkably advanced and versatile brain to be able voluntarily to deceive itself into experiencing a hallucination.
References
Barber, T. X. (1999). Hypnosis: A mature view. Contemporary Hypnosis, 16, 123–127.
Blakemore, S.-J., Oakley, D. A., & Frith, C. D. (2003). Delusions of alien control in the normal brain. Neuropsychologia, 41, 1058–1067.
Bowers, K. S., & Brenneman, H. A. (1979). Hypnosis and the perception of time. International Journal of Clinical and Experimental Hypnosis, 27, 29–41.
Brown, S. W., & Boltz, M. G. (2002). Attentional processes in time perception: Effects of mental workload and event structure. Journal of Experimental Psychology: Human Perception and Performance, 28(3), 600–615.
Buehlmann, A., & Deco, G. (2008). The neuronal basis of attention: Rate versus synchronization modulation. Journal of Neuroscience, 28, 7679–7686.
Caligiuri, M. P., Hellige, J. B., Cherry, B. J., Kwok, W., Lulow, L. L., & Lohr, J. B. (2005). Lateralized cognitive dysfunction and psychotic symptoms in schizophrenia. Schizophrenia Research, 80, 151–161.
Calvert, G. A., Bullmore, E. T., Brammer, M. J., Campbell, R., Williams, S. C. R., McGuire, P. K., et al. (1997). Activation of auditory cortex during silent lipreading. Science, 276, 593–596.
De Pascalis, V. (2007). Phase-ordered gamma oscillations and the modulation of hypnotic experience. In G. Jamieson (Ed.), Hypnosis and conscious states: The cognitive-neuroscience perspective. Oxford: Oxford University Press.
De Pascalis, V., Cacace, I., & Massicolle, F. (2004). Perception and modulation of pain in waking and hypnosis: Functional significance of phase-ordered gamma oscillations. Pain, 112, 27–36.
Derbyshire, S. W. G., Whalley, M. G., Stenger, V. A., & Oakley, D. A. (2004). Cerebral activation during hypnotically induced and imagined pain. NeuroImage, 23, 392–401.
Di Lollo, V., Enns, J. T., & Rensink, R. A. (2000). Competition for consciousness among visual events: The psychophysics of re-entrant pathways. Journal of Experimental Psychology: General, 129, 481–507.
Doesburg, S. M., Roggeveen, A. B., Kitajo, K., & Ward, L. M. (2008). Large-scale gamma-band phase synchronization and selective attention. Cerebral Cortex, 18, 386–396.
Elvevåg, B., Brown, G. D. A., McCormack, T., Vousden, J. I., & Goldberg, T. E. (2004). Identification of tone duration, line length, and letter position: An experimental approach to timing and working memory deficits in schizophrenia. Journal of Abnormal Psychology, 113(4), 509–521.
Faymonville, M.-E., Roediger, L., Fiore, G., Delgueldre, C., Phillips, C., Lamy, M., et al. (2003). Increased cerebral functional connectivity underlying the antinociceptive effects of hypnosis. Cognitive Brain Research, 17, 255–262.
Fénelon, G., Mahieux, F., Huon, R., & Ziégler, M. (2000). Hallucinations in Parkinson’s disease. Brain, 123, 733–745. doi:10.1093/brain/123.4.733.
Fox, R. H., Bradbury, P. A., Hampton, I. F. G., & Legg, C. F. (1967). Time judgment and body temperature. Journal of Experimental Psychology, 75, 88–96.
Gray, J. A. (1995). The contents of consciousness: A neuropsychological conjecture. Behavioral and Brain Sciences, 18, 659–676.
Gregory, R. L. (1980). Perceptions as hypotheses. Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences, 290, 181–197.
Gruzelier, J. H., De Pascalis, V., Jamieson, G., Laidlaw, T., Naito, A., Bennett, B., et al. (2004). Relations between hypnotizability and psychopathology revisited. Contemporary Hypnosis, 21(4), 169–176.
Haggard, P., Cartledge, P., Dafydd, M., & Oakley, D. A. (2004). Anomalous control: When “free-will” is not conscious. Consciousness and Cognition, 13, 646–654.http://dx.doi.org/10.1016/j .concog.2004.06.001
Haig, A. R., Gordon, E., De Pascalis, V., Meares, R. A., Bahramali, H., & Harris, A. (2000). Gamma activity in schizophrenia: Evidence of impaired network binding? Clinical Neurophysiology, 111, 1461–1468.
Hammerseley, P., Dias, A., Todd, G., Bowen-Jones, K., Reilly, B., & Bentall, R. P. (2003). Childhood trauma and hallucinations in bipolar affective disorder: Preliminary investigation. British Journal of Psychiatry, 182, 543–547.
Harrington, D. L., Haaland, K. Y., & Hermanowicz, N. (1998). Temporal processing in the basal ganglia. Neuropsychology, 12(1), 3–12. http://dx.doi.org/10.1037/0894-4105.12.1.3
Kelley, M. P. (2011). Schizotypy and hemisphericity. Psychological Reports, 109, 533–552.
Kirsch, I. (1997). Suggestibility or hypnosis: What do our scales really measure? International Journal of Clinical and Experimental Hypnosis, 45, 212–225.
Kosslyn, S. M., Thompson, W. L., Costantini-Ferrando, M. F., Alpert, N. M., & Spiegel, D. (2000). Hypnotic visual illusion alters color processing in the brain. American Journal of Psychiatry, 157, 1279–1284.
Lawrie, S. M., Buechel, C., Whalley, H. C., Frith, C. D., Friston, K. J., & Johnstone, E. C. (2002). Reduced frontotemporal functional connectivity in schizophrenia associated with auditory hallucinations. Biological Psychiatry, 51, 1008–1011.
Lewis, P. A., & Miall, R. C. (2006). A right hemispheric prefrontal system for cognitive time measurement. Behavioural Processes, 71, 226–234.
Liossi, C., & Hatira, P. (2003). Clinical hypnosis in the alleviation of procedure-related pain in pediatric oncology patients. International Journal of Clinical and Experimental Hypnosis, 51, 4–28.
Mozenter, R. H., & Kurtz, R. M. (1992). Prospective time estimation and hypnotizability in a simulator design. International Journal of Clinical and Experimental Hypnosis, 40, 169–179.
Naish, P. L. N. (1986). Hypnosis and signal detection: An information processing account. In P. L. N. Naish (Ed.), What is hypnosis? Current theories and research (pp. 121–144). Milton Keynes, Philadelphia: Open University Press.
Naish, P. L. N. (2001). Hypnotic time perception: Busy beaver or tardy timekeeper? Contemporary Hypnosis, 18, 87–99.
Naish, P. L. N. (2003). The production of hypnotic time-distortion: Determining the necessary conditions. Contemporary Hypnosis, 20, 3–15.
Naish, P. L. N. (2010). Hypnosis and hemispheric asymmetry. Consciousness and Cognition, 19, 230–234.
Naish, P. L. N. (2012). Attention. In N. Braisby & A. Gellatly (Eds.), Cognitive psychology (2nd Ed., pp. 33–64). Oxford: Oxford University Press.
Naish, P. L. N. (2006). Time to explain the nature of hypnosis. Contemporary Hypnosis, 23, 33–46.
Naish, P. L. N. (2007). Time distortion and the nature of hypnosis and consciousness. In G. Jamieson (Ed.), Hypnosis and conscious states: The cognitive-neuroscience perspective (pp. 271–292). Oxford: Oxford University Press.
Naish, P. L. N. (2008). Hypnosis: Theory and therapy. Paper presented at the Annual Conference of the British Society of Clinical and Academic Hypnosis, Stansted, UK.
Seok, J.-H., Park, H.-J., Chun, J.-W., Lee, S.-K., Cho, H. S., Kwon, J. S., et al. (2007). White matter abnormalities associated with auditory hallucinations in schizophrenia: A combined study of voxel-based analyses of diffusion tensor imaging and structural magnetic resonance imaging. Psychiatry Research: Neuroimaging, 156, 93–104.
Shergill, S. S., Brammer, M. J., Williams, S. C. R., Murray, R. M., & McGuire, P. K. (2000). Mapping auditory hallucinations in schizophrenia using functional magnetic resonance imaging. Archives of General Psychiatry, 57, 1033–1038.
Shergill, S. S., Samson, G., Bays, P. M., Frith, C. D., & Wolpert, D. M. (2005). Evidence for sensory prediction deficits in schizophrenia. American Journal of Psychiatry, 162, 2384–2386.
Shergill, S. S., Kanaan, R. A., Chitnis, X. A., O’Daly, O., Jones, D. K., Frangou, S., et al. (2007). A diffusion tensor imaging study of fasciculi in schizophrenia. American Journal of Psychiatry, 164, 467–473.http://dx.doi.org/10.1176/appi.ajp.164.3.467
Simons, D. J., & Rensink, R. A. (2005). Change blindness: Past, present, and future. Trends in Cognitive Sciences, 9, 16–20.
Smith, J. G., Harper, D. N., Gittings, D., & Abernethy, D. (2007). The effect of Parkinson’s disease on time estimation as a function of stimulus duration range and modality. Brain and Cognition, 64, 130–143.
Spanos, N. P., & Burgess, C. (1994). Hypnosis and multiple personality disorder. In S. J. Lynn & J. W. Rhue (Eds.), Dissociation: Clinical and theoretical perspectives (pp. 136–155). New York: Guilford Press.
Stebbins, G. T., Goetz, C. G., Carrillo, M. C., Bangen, K. J., Turner, D. A., Glover, G. H., et al. (2004). Altered cortical visual processing in PD with hallucinations: An fMRI study. Neurology, 63, 1409–1416.
St. Jean, R., McInnis, K., Campbell-Mayne, L., & Swainson, P. (1994). Hypnotic underestimation of time: The busy beaver hypothesis. Journal of Abnormal Psychology, 103, 565–569.
Szechtman, H., Woody, E., Bowers, K. S., & Nahmias, C. (1998). Where the imaginal appears real: A positron emission tomography study of auditory hallucinations. Proceedings of the National Academy of Sciences of the USA, 95, 1956–1960.
Treisman, M., Cook, N., Naish, P. L. N., & MacCrone, J. K. (1994). The internal clock: Electroencephalographic evidence for oscillatory processes underlying time perception. Quarterly Journal of Experimental Psychology (A), 47, 241–289.
Treisman, M., Faulkner, A., & Naish, P. L. N. (1992). On the relation between time perception and the timing of motor action: Evidence for a temporal oscillator controlling the timing of movement. Quarterly Journal of Experimental Psychology (A), 45, 235–263.
Treisman, M., Faulkener, A., Naish, P. L. N., & Brognan, D. (1990). The internal clock: Evidence for a temporal oscillator underlying time perception with some estimates of its characteristic frequency. Perception, 19, 705–743.
Vasterling, J. E., Duke, L. M., Tomlin, H., Lowery, N., & Kaplan, E. (2004). Global-local visual processing in posttraumatic stress disorder. Journal of the International Neuropsychological Society, 10, 709–718.
Wagstaff, G. F. (1981). Hypnosis, compliance, and belief. Brighton: Harvester.
Whitty, C. W. M., & Lewin, W. (1957). Vivid day-dreaming: An unusual form of confusion following anterior cingulectomy. Brain, 80, 72–76.
Yard, S. S., DuHamel, K. N., & Galynker, I. I. (2008). Hypnotizability as a potential risk factor for posttraumatic stress: A review of quantitative studies. International Journal of Clinical and Experimental Hypnosis, 56, 334–356.
1. The reader should try shaking his or her head gently from side to side, confirming that it remains possible to read this text. Then move the book from side to side at the same speed and note that the eyes cannot keep up. The neural links in the latter case include the brain and so are too slow for the task.
