click here to download document

ARTICLES

The Neurophysiology of Hypnosis

The existence during hypnosis of a unique neurological state has been hotly contested between two camps, the 'special state' and the 'psychosocial phenomenon'. Those in the first camp point to unique neurological changes that appear during hypnosis, those in the second say that such changes are brought about by suggestion, and are part of our normal experience. The debate continues to this day, and probably will not be resolved to the satisfaction of either side, in the same way that evidence for or against the paranormal only increases the polarisation of that debate. However there is one point upon which all are agreed, that hypnosis is a phenomenon of attention.

The nature of cerebral activity is inhibitory: the brain has more inhibitory than stimulatory neurones so as much as stimulating other brain regions all neural activity is also suppressing activity elsewhere. It is also a long recognised property of all cortical and spinal areas to suppress contra lateral activity (Sherrington 1906), and there is also a general cross modality suppression from paying attention to sensory stimuli (Paus 1997), so that paying attention to one stimulus stimulates blood flow in certain attention related cortical areas (pre-frontal) while decreasing blood flow in occipital visual and imaginative areas. Paying attention also involves inhibition of attention to any other stimulus, unless it is higher in the stimulus hierarchy (Barrios 2001). Attention to a stimulus serves to inhibit extraneous motor activity that would not serve to increase engagement with, or withdrawal from, that stimulus. Once the executive decision has been taken about one particular course of action with regard to that stimulus all other motor intentional choices are suppressed. In the reflex arc the extensors of the thigh are stimulated, but just as significant is the suppression of the flexors, and in walking the activity of the extensor neurones of one leg will inhibit the contra lateral extensor neurones on the other leg. Suppression is equally important in the study of attention. Focussing attention involves suppression of all extraneous attentional activity by concentrating on one specific ‘set’ of cognitive or sensory stimuli, so that ‘when attending to or responding to one stimulus there will be reciprocal inhibition of incompatible stimuli and responses’ (Barrios 2001). Pavlov (1960) recognised that words could also act as conditioned stimuli, as powerful, or more powerful than sensory stimuli through conditioning as borne out by Hudgins (1933). In hypnosis the hypnotist increases attention to stimuli from within the body, related to muscular tension and breathing for instance, and from within the mind with the use of suggestion, which is paying attention to internal cognitive stimuli. (quote on suggestion from Barrios). This is absorption and it is distinct from relaxation. Our patterns of behaviour are determined by unconscious executive decisions throughout our lives, where emotion and motivation are locked in subtle interplay, and the hypnotist seeks to change these by giving power to alternative motor choices (behavioural changes). This can be achieved by powerful conscious or unconscious positive visualisation, which boosts the preferred new choice, and/or the use of relaxation, which boosts the role of the executive in making choices.

There are three sources of information on changes in hypnosis, 1) electrophysiological 2) neuropsychological and 3) cerebral activity studies, as determined by functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) studies. Before exploring these changes it is helpful to review current understanding of normal brain function in the attentional, executive and motor circuits of the brain which is where changes are mostly observed in hypnosis.

The orbito-frontal cortex (OFC) is an important structure in the brain for processing sensory and hypothalamic information and assigning a reward value to the sensory input depending on the current context and the needs of the organism. At a cellular level individual cells in the OFC respond to expectation of nutritional rewards (food/drink) or highly specific expected rewards (i.e. banana, water, orange juice), increasing their activity as the visual cues for the arrival of the reward grow stronger, and with the level of satiety in relation to that specific foodstuff and to the general level of hunger and thirst (Hikosaki 2000). Some cells respond to the arrival of the reward, others to expectation and arrival, and others to the non-arrival of expected reward. Similar networks of cells have been found, responding to social rewards and non-rewards ie positive/negative facial cues (Kringelbach 2003) leading to the hypothesis that such cells respond selectively in relation to face expression specifically when it indicates that behaviour should change in order to maintain such social rewards. Invariant signals from sensory information in the primary sensory cortices of sensation (touch /pain) and gustation (olfaction/taste), after passing through the OFC acquire a new signal power dependent on the internal state of the organism (i.e. level of hunger, level of sexual arousal) (Rolls 2004) From this body of work Rolls proposes a working model for the cells of the OFC with invariant, non-Hebbian synapses for these main modalities (pain touch taste) and Hebbian modifiable synapses for the visual association inputs from the visual cortex (see fig 1). So for example the positive affect from seeing a smile is a result of the previous association of visual stimuli (mothers smile) with the smell, taste and texture of breast milk via the visual hebbian modified connections to the OFC taste olfactory and sensory cells, which are simultaneously responding to their respective other stimuli a. The association of money with food and sexual pleasure gives money a real ‘reward’ value; there are reward cells in the PFC that respond to money rewards (O’Doherty 2001). The resulting cellular activity is then modified by the internal state. Input from the amygdala signals emotional state where this is relevant such in preparedness for withdrawal and input from the hypothalamus signals levels of satiety (sexual/appetitive). Thus the OFC is a key area in our perception of pleasure and pain, with Rolls’ model it is clear how visual associations and triggers may increase our perception of pain. After ascribing the different reward/noxiety values to somatosensory inputs, the OFC then feeds the modified signal: 1) directly into the limbic circuits, the Ventral Tegmental Area (VTA) via glutamatergic projections (the VTA is the key structure in the dopaminergic (positive affect) system increasing activity in the locus coerulus (arousal) and the Nucleus Accumbens (NAc) the ‘pleasure centre’ which itself also receives direct dense glutamatergic projections from the OFC (Phillipson 1985)) and 2) into the executive circuits. The OFC feeds directly into the dorso-lateral pre-frontal cortex (DLPFC) where similar reward cells are situated which also respond to position in the animals’ environment (Hikosaki 2000) to maintain the current object in the field of attention b. The DLPFC then feeds directly into the Anterior Cingulate Cortex ACC, which is situated anatomically and functionally next to the Pre-Motor Cortex. The DLPFC sends stimulatory signals to the ACC and inhibitory signals to the amygdala, so paying attention or the ability to engage working memory directly inhibits amydala activity c

The Anterior Cingulate Cortex (ACC) developmentally has close similarities to the motor cortex in its layers and cell types (Allman 1998) and indeed acts to change motor activity so it is regarded as an area of neo-cortex devoted to the motor expression of emotional and cognitive behaviour. It is divided into dorsal and ventral ACC, and there is strong evidence that these are also functional divisions along cognitive (dorsal) and emotional (ventral) areas, although the areas are probably anatomically interconnected. It also contains certain specialised cells (spindle cells, large very fast responding neurones) not found in any other structure in the brain or in any species other than higher primates (man, chimps gorillas orang-utan) (Nimchinsky 1999). These cells appear at 4 months in the Anterior Cingulate at the time the infant acquires the ability to track objects, reach for them, and smile spontaneously. It is hypothesised that these cells allow primates to respond rapidly to changing social cues. The ACC appears to be involved in monitoring and evaluating the outcomes of actions, responding to events indicating a reduction of reward (Botvinik reference 73) by increasing activation. In one experiment a third of a sample of macaque monkey anterior cingulate cells responded to decreasing reward, but not to continuous levels of reward (Shima 1998). Inhibiting cortical activity in this area led to the monkeys no longer responding to reward. Given the direct communication of the OFC and the DLPFC ‘reward’ and ‘non-reward’ cells with the ACC the cause of this lack of response is clear, the motivation to change behaviour is removed if there is no prefrontal assessment of changing reward values. Focussed concentration increases the cingulate theta signal, which is abolished by anxiety (Gevins 1997, Mizuki 1989). Discrimination of the affect of faces also stimulates ACC activity (George 1993). These functions of the ACC (lack of reward from negative and changing social cues) also relate to cells in the OFC, the ACC presumably is important in selecting the appropriate motor response quickly ie changing face expression and body language which could mean life or death in a social situation. Kerns and colleagues (p7 in botvinik paper) observed that following trials with strong ACC engagement there was relatively strong activation in the DLPFC (get ref) presumably to maintain accurate attentional and motor engagement of rewards. The ACC combines the current emotional and cognitive information, which finally leads to the executive decision and activates the Pre-Motor Cortex, which is where one motor outcome is favoured and others inhibited.

The Ventral Tegmental Area (VTA) is the centre of the dopaminergic pathways in the brain and feeds into the NAc the frontal cortices and the ACC. Ashby (1999) proposed a system where unexpected rewards, through this dopamine circuit can lead to increased cognitive flexibility to see alternate sets and to solve problems (see fig 2).

Neurological changes in hypnosis.

Gruzelier (1998) and colleagues looked at the neuro-physiological changes that accompany hypnosis by comparing the responses of subjects read a story to those given a hypnotic induction. Electro dermal studies showed a diminution of sympathetic responses (electro dermal orienting) under hypnosis implying a disengagement of amygdalo-limbic responding. There was a significant increase in the ability to sort shapes with the left hand under hypnosis suggesting a shift of cognitive processing from the dominant to the non dominant hemisphere, and verbal category and design fluency tasks also showed a dissociation of the left Pre-Fontal Cortex (PFC) and an increase in right sided visual processing due to reducing collateral inhibition of the right. Asymmetry in electro dermal and electro cortical responses to tones and visual sensitivity showed a shift to right sided processing. There was a significant change in the 'Stroop Test', a test of the ability to detect errors and then switch motor responses. In this test the subject presses one button for a right pointing arrow and a different key for a left pointing arrow as long as the arrows are green (congruent trial). However if the colour of the arrow changes from green to red the subject must press the opposite cursor (incongruent trial). This test engages the ACC, which shows increased activity after increasing incongruity relating to more need for monitoring due to this increased incongruity or error detection. Under hypnosis the subjects were faster than controls on the congruent test, but slower than controls on the incongruent tests. There was also a missing EEG event (the error detection signal) in the hypnotised subjects and this signal was traced to the cingulate. Gruzelier draws the implication from this result that the cognitive and emotional divisions of the cingulated are dissociated, as the normal emotional response to errors is not producing the increasing monitoring required perhaps because of the missing event. The lack of error detection and hence conflict monitoring may account for the experience in hypnosis of a diminished tendency to judge monitor and censor, which may be why the hypnotist is able to control thoughts and actions. Gruzeliers work for most people established that the hypnotised subjects could not have been in a psychosocial state (ie behaving as they would expect hypnotised people to behave) and come up with such specific neurological changes c. Based on these experiments Gruzelier hypothesised a 3 stage change in hypnosis involving 1. Activation of the anterior fronto-limbic inhibitory processes 2. Anterior inhibition or disconnection either lateralised to the left or bilateral depending on the processes examined and 3. Involvement of right temporo-posterior processing. All but 3 are supported by subsequent PET studies, subsequently occipito-posterior areas were found more involved in visual dreaming

New data on hypnosis has been gathered using PET scans, and MRI data. Given the deductive reasoning required to use such evidence, it is possible to deduce that an area is active/inactive in a certain state, but given the huge complexity of the brain and its stimulatory/inhibitory interconnections, not what it is doing (i.e. it could be very sensitive but relatively inactive). Given that caveat, there are some consistent changes in hypnosis, although the changes demonstrated vary within the duration of the hypnosis depending on what is being done at the time (such as relaxation absorption or suggestion) as Gruzelier (1998) suggested. There are significant increases in occipital rCBF which is highly correlated with Delta EEG signalling (Rainville 1999) and known to be related to visual imagery and dreaming (Paus 1997). Also Rainville (1999) found consistent decreases in the blood flow to the Right inferior parietal lobe, probably reflecting changes in self other distinction and sense of agency. Specifically suggestion increased activity in the Left DLPFC consistent with lexical and working memory functions engaged by listening, and increased areas in the inferior parietal lobes overlapping the decreases in the Right inferior parietal area. These studies (Rainville 1999, Rainville 2002) have continued to point to the involvement of the anterior cingulate cortex (ACC areas 24, 32 33 & 25). One study done recently on regional cerebral blood flow (rCBF) in hypnosis using PET scans, chose to divide up the state of hypnosis experientially, into 'relaxation' and 'absorption' Rainville (2002 ). Absorption is defined experientially as ‘a shift to the hypnotists voice, and decreased attention to external stimulation' and functionally as a state of 'total attention that fully engages our representational resources and results in imperviousness to distraction events’ (Tellegen & Atkinson 1974). In his study Rainville (2002) found the states of relaxation and absorption were variously engaged. The relaxation phase of the hypnosis showed increases in the perigenual and middle ACC and expected decreases in brainstem activation, while the absorption showed an increase in rostral ACC (area 24) function and thalamic activation. Thus hypnosis is a very dynamic process involving at various times the involvement of a number of different neurological circuits.

Discussion

How can these varied changes be related to the experience of hypnosis? Hypnosis can change the experience of sensory inputs such as pain, and is effective at encouraging problem solving. It is possible that the relaxation element allows the engagement of reward circuits through the activation of muscle relaxation sensory information, which, through the OFC affects the VTA (see fig 1), in turn increasing the flow of dopamine to the NAc and the Rostral ACC (see fig 2). The increase in dopamine from such positive affect also has been shown to increase the ability to switch between alternative responses, identified as a cingulate function, which is markedly impaired in Parkinsonian patients (Ashby). There are many cells in the anterior cingulate that respond either to instantiation of a reward, or a decreasing reward, the ACC has one of the richest dopaminergic innervations in the cortex (Gaspar et al 1989). Suggestion, in the form of Placebo, has been shown to increase the concentration of dopamine in neuronal circuits in Parkinsons. Hypnosis can utilise the placebo effect (which is equally potent as antidepressants in depression) in an honest and probably even more effective fashion than placebo, and such an increase in hypnotic suggestion could allow better function of attention switching and problem solving.

I would hypothesise that in hypnosis the combination of relaxation and absorption may allow stronger activation of the cingulate which is then able to work more effectively on cognitive problem solving using passive imagery and dreaming to solve problems (a solution representing the switch to a more rewarding state) which is normally denied to them due to lack of positive affect. The work of Teasdale and Watkins in depression suggests that the engagement of a solution based cognitive set decreases over generalisation in depression, and my own preliminary studies suggest that hypnosis is a powerful tool in the treatment of depression (unreported data) far more powerful than just the effect of placebo. In a recent paper on the similarities of physical and social pain, it is proposed that there is an overlap in the neural and computational mechanisms monitoring physical and social pain which is evolutionarily adaptive suggesting that the prolonged period of immaturity and the critical need for maternal care in mammalian infants caused pain mechanisms involved in the detection and prevention of physical danger to be co-opted by the more recently evolved social attachment system to detect and prevent social separation (Eisenbenger 2004). The areas involved are the right pre-frontal cortex and the dorsal ACC. It seems increasingly that the ACC contains a generic set of tools for problem solving, which can equally serve physical, conceptual, and social functions. Evolution gives us a huge advantage if we can share tools between functions. If the OFC can have networks of cells in reward circuits which through our executive and affective circuits (ACC, VTA and NAc) combine to maximise these rewards for appetite and money, then as there are also networks serving social rewards it is logical to assume that these circuits will be shared to monitor social pain and solve social problems. The tool we are enabling with hypnosis is thus a ‘reducing reward tool’, which can be engaged in problem solving in these physical, conceptual, and social spheres. In practical terms this corresponds to the hypnotists’ assumption that each person has the tools within them to sort out their own problems. By accessing memories of positive previous solutions (which is inevitable merely by using the word solution) the hypnotist is able to access a mental ‘set’ of conditions in which the ACC was successfully able to solve problems, and the subject is then able to apply these solutions to their current difficulties (unrewarded states) without anxiety, which allows more effective engagement of their ACC cognitive monitoring problem solving state. This is particularly demonstrated in the hypnosis/NLP exercise of collapsing anchors which uses intensive vivid sensory recollection of successful solution states (selected unconsciously based on the current problem) to deal with present anxiety/panic states by neutralising unconscious triggers.

a If for some reason the mother has post natal depression one can visualise the subsequent absence of this association leading to the child later lacking the ability to respond to social cues, so becoming more vulnerable to isolation and depression.

b This enables rewarding stimuli to be maintained in the field of attention, and is proposed as the seat of working memory, through closed loops to the visual and auditory cortices enabling us to hold images numbers and sounds(i.e. phone nos or the beginning of a sentence) while processing other information (rolls reference).

c Amygdalar suppression is an observed phenomenon, working memory tasks can decrease post-traumatic stress (Brewin), and possibly people with good working memory can exclude past negative material by this method. Certainly poor working memory skills are a positive predictor for posttraumatic stress, and posttraumatic stress and depression have many clinical similarities. This is used to good effect in hypnosis and in neuropsychiatry, which use similar techniques (Eye Movement De-sensitisation EMDR and Thought Field Therapy TFT both of which engage DLPFC). Also this is engaged in cognitive tasks (counting backwards) so that induction per se reduces anxiety, and in creative visual tasks (creating a safe place) which author is exploring in a current randomised trial.

c The debate however continues to rage in both camps to this day, and probably will not be resolved to the satisfaction of the 'psychosocial' camp, in the same way that evidence either way in the belief/non-belief debate on paranormal phenomenon serves only to polarise each camp.

Allman J,Hakeem A,Erwin J,Nimchinsky E, Hof P The Anterior Cingulate Cortex: The evolution of an interface between emotion and cognition.

Ashby F, Isen A, Turken A (1999) A Neuropsychological Theory of Positive Affect and Its Influence on Cognition Psychological Review 106:529-550

Barrios A (2001) A theory of hypnosis based on principles of conditioning and inhibition. Contemporary Hypnosis 18(4) 163-203

Gruzelier J Champion A, Fox P et al (2002) Individual differences in personality, immunology and mood in patients undergoing self-hypnosis training for the successful treatment of a chronic viral illness, HSV-2 Contemporary Hypnosis 19/4 2002

Gruzelier J (1998) A working model of the neurophysiology of hypnosis: a review of the evidence Contemoprary Hypnosis (1998) Vol 15 1 3-21

George M, Ketter T, Gill D, Haxby J,Ungerleider L, Hersovitch P, Post R (1993) Brain regions involved in recognising facial emotion or identity: an oxygen-15 PET study. Journal of Neuropsychiatry and Clinical Neurosciences 10:525-535

Eisenbenger N, Lieberman M (2004) Why rejection hurts: a common neural alarm system for physical and social pain Trends in Cognitive Sciences 8:7 pp 294-300

Gevins A, Smith M, McEnvoy L, Yu D (1997) High-resolution EEG mapping of cortical activation related to working memory: difficulty, types of processing, and practice. Cerebral Cortex 7:374-385

Hikosaka K, Watanabe M (2000) Delay activity of orbital and lateral prefrontal neurons of the monkey varying with different rewards. Cerebral Cortex. Vol 10(3) Mar 2000, 263-271

Holroyd C, Coles M, The neural basis of human error processing: reinforcement learning, dopamine, and the error-related negativity, Psychol. Rev. 109 (2002), pp. 679–709

Hudgins C (1933) Conditioning and the voluntary control of the pupillary light reflex Journal of General Psychology 8:4-51

Kerns J et al., Anterior cingulate conflict monitoring and adjustments in control, Science 303 (2004), pp. 1023–1026

Kringelbach M,Rolls E (2003) Neural Correlates of rapid reversal learning in a simple model of human social interaction. Neuroimage 20, 1371-1383

Mizuki Y, Suetsugi M, Imai T, Kai S, Kajimura N, Yamada M (1989) A physiological marker for assessing anxiety level in humans: frontal midline theta activity. Japanese Journal of Psychiatry and Neurology 43:619-626

Nimchinsky E, Gilessen J, Allman D,Perl J, Hof E, Hof P 1999 Spindle neurons of the human anterior cingulated cortex Journal of Comparative Neurology 355:27-37

O’Doherty J, Kringlebach M, RollsE, Hornak J & Andrews C (2001) Abstract reward and punishment in the human orbitofrontal cortex Nature Neuroscience 4, 95-102

Paus T, Zatorre R, Hofle N, Caramanos Z, Gotman J, Petrides M, Evans A (1997) Time related changes in neural sytems underlying attention and arousal during the performance of an auditory vigilance task Journal of Cognitive Neuroscience 9 392-408

Pavlov I (1960) Conditioned Reflexes New York NY: Dover p 407

Phillipson O, Griffiths A (1985) The topographic order of inputs to the nucleus accumbens in the rat. Neuroscience 16: 1550-1560 [Abstract]

Rainville P, Hofbauer R, Paus T, Duncan G, Bushnell M, Price D (1999) Cerebral mechanisms of Hypnotic induction and suggestion Journal of Cognitive Neuroscience 11:1 pp 110-125

Rainville P, Hofbauer R, Bushnell C, Duncan G, Price D (2002) Hypnosis modulates activity in brain structures involved in the regulation of consciousness Journal of Cognitive Neuroscience 146:887-901 Rolls E (2004) The functions of the orbitofrontal cortex Brain and Cognition 55 11-29)

Shima K, Tanji J 1998 Role for the cingulated motor area cells in voluntary movement selection based on reward. Science 282:1335-1338