HomeNewIntroductionQuantum Mind BlogQuantum Mind TheoriesRelated TopicsKey ArticlesReferencesContact UsOnline Book

Emotions

Emotions - Envisaged here as a common neural currency allowing the comparison on unrelated scenarios or courses of action. Includes books and papers by Edmund Rolls, Jaak Panksepp, Antonio Damasio, Hugo Critchley, Leonard Kosiol, Deborah Budding, Jan Lauwereyns and Anthony Grace.

1.) Emotions Explained - Edmund Rolls - Examines the role of the orbitofrontal and the amygdala in emotions

2.) Affective Neuroscience - Jaak Panksepp - Emphasises the subcortical role in consciousness

3.) The Feeling of What Happens - Antonio Damasio - Damasio's take on emotions. Rather conflates consciousness and self consciousness.

4.) Neural mechanisms of autonomic, affective and cognitive integration - Hugo D. Critchley - Discusses basis of subjective emotional experience.

1.)

Emotion Explained

Edmund Rolls

University of Oxford, Dept. of Experimental Psychology

Oxford University Press (2007) ISBN 978-0-19-857004-2

At the beginning of his book, Rolls asks why it is that emotional states ‘feel like something’, and it is admitted that this is part of the larger problem of consciousness. His book looks at the interaction of emotions and cognitive states. Much of the argumentation of the book revolves round a framework of ‘rewards’ and ‘punishers’ A reward is something that animals, including humans, will work for, and a punisher is something they will work to avoid. Learning is implicit in the reward/punisher concept, because an animal has to learn to seek or avoid certain things. Rewards and punishers are referred to as ‘reinforcers’, because they reinforce particular actions or behaviour patterns.

Rolls distinguishes between unlearned or ‘primary’ reinforcers, such as the experience of pain and ‘secondary reinforcers’, which are initially neutral experiences, but are later learned to be associated with pleasant or unpleasant experiences. The unlearned or primary reinforcers are also referred to as ‘unconditioned’ i.e. the animal has not been conditioned by past exposure. When reinforcers are learnt, the behavioural outcome is actions that are described as ‘instrumental’, in that they are goal driven, and may allow an animal to obtain the rewards or avoid the punishers that it has learned about.

The advantage of emotion is that these instrumental actions do not need to be specified by the genes, but can be learned from the outcome of actions. Emotions are thus viewed as states produced by reinforcers. The amygdala, the orbitofrontal, the cingulate cortex, and the basal forebrain including the hypothalamus, are seen as the brain areas most involved with emotions. Primary reinforcers, such as taste and touch, are represented in the orbitofrontal cortex. The brain is thus seen as being organised in two stages, first processing stimuli, so that it is perceived as an object, and then assessing whether it is a reward or a punisher.

Emotional states are usually initiated by reinforcing stimuli present in the external environment. Cognitive processes are often required to determine whether an environmental stimulus is a reward or a punisher. The author argues that cognitive decoding of a stimulus is needed to determine which emotion will be felt. It is argued that the reinforcement value of a stimulus is decoded in the orbitofrontal and the amygdala. These and other brain structures that receive connections from them are regarded as the areas involved in decoding the emotional significance of stimuli.

The author sees learning relative to emotions as a two-stage process. First, a rapid response, as to which stimuli are associated with reinforcers, and then a second slower stage of producing appropriate instrumental actions to achieve a goal. The brain mechanisms that implement emotions are viewed as those that decode stimuli that are reinforcers, and those that stimulate reinforcer learning, plus the mechanisms that relate these results to actions. Different emotions can be understood in terms of different reinforcers. The specifying of emotions leaves the response open to an on-the-spot decision. The alternative to this system would be to specify fixed responses for every possible stimuli. The attempt to do this would lead to an explosion in the number of fixed responses programmed into the animal.

Rolls compares the performance of robots, such as a robot arm in a factory, with the performance of animals. With the robot arm, the action to be performed is decided externally. This decision is with a view to a purpose, such as making a car. The process in an animal derives from evolution, which is blind to purpose. The animals have reward-punisher systems, which allow them to choose a flexible response. The response will take into account the at-this-moment cost of a particular action, whereas the robot arm is fixed by the designer with one response, regardless of subsequently arising conditions. The reinforcer defines a goal, but does not specify a particular action. The action is selected by the animal as appropriate. The system depends on connections from the amygdala and orbitofrontal to the hypothalamus and also to autonomic nuclei in the brain stem. The amygdala and orbitofrontal are particularly involved in learning.

Emotions may trigger recall of existing memories stored in the neocortex, by means of projections from the orbitofrontal and amygdala to the neocortex. A particular mood state in the orbitofrontal or amygdala can become associated with memories stored in the cortex. These systems may also back project onto the hippocampus.

The brain regions that produce the emotional value of stimuli interact with three main output systems. Firstly, there are the autonomic and endocrine systems, secondly, there are unconscious responses, also related to rewards and punishers, and thirdly there is longer-range planning, capable of deferring a short-term reward in favour of a greater long-term benefit.

It is interesting that Rolls emphasises the importance of longer-range planning, since the whole superstructure of argumentation related to claims that Libet’s experiments in relation to readiness potentials prove the non-existence of freewill rests on only looking at trivial actions, such as the flexing of fingers, and ignoring any longer-term planning. Later on in his book, Rolls is seen to have difficulty in reconciling the non-deterministic aspect of his emphasis on long-term planning with the deterministic orthodoxy of neuroscience.

The inferior temporal visual cortex contains the advanced processing stage, at which objects become represented, and from which there are inputs to other parts of the brain, which learn the reward and punisher aspects of the objects perceived. Reward-punisher values are learned, in respect to perceived objects produced by the later stages of processing, rather than the earlier pixels and edges produced by the earlier stages. Outputs from the temporal lobe visual areas go to limbic and related regions, such as the amygdala, and via the entorhinal cortex to the hippocampus. The amygdala has neurons with face-selective responses. The amygdala also projects to the ventral striatum that may be important in behavioural responses to faces. Another structure in the superior temporal sulcus conveys information about facial expressions as opposed to identity. Visual inputs also reach the orbitofrontal from the inferior temporal visual cortex. The orbitofrontal projects back to the inferior temporal visual cortex and to the entorhinal, seen as the gateway to the hippocampus. Damage to the rear part of the orbitofrontal produces emotional change, such as reduced aggression, lack of emotion, impulsiveness and unusual food preferences. The author relates these problems to damage-related failure to learn from non-reward in various situations.

In the orbitofrontal, it has been shown that some neurons respond to primary reinforcers, such as taste, and others to learned secondary reinforcers, such as visual reward stimuli, predicting or estimating a reward value. For instance, in humans, the reward value and also the subjectively reported pleasantness of food is represented in the orbitofrontal. The neurons can reflect the relative preference for different stimuli. It has also been shown that the subjective quality of, for instance odours, can be altered by the top-down modulatory impact of words, while colour is thought to influence olfactory judgement. The orbitofrontal and cingulate are shown to be influenced by cognition to a much greater extent than the amygdala and primary sensory cortex. This is taken to suggest a link between the highest levels of reasoning, and the level at which basic emotional assessment arises. Humans with frontal lobe damage show impairment in tasks that require them to respond to changes in reinforcers in the environment. This could be due to either difficulty in establishing the initial preference, or difficulty in shifting to a new preference, when the system of rewards changes. The orbitofrontal appears particularly important in situations where the message of the reinforcers is suddenly reversed. Difficulty with this reversal procedure tends to correlate with socially inappropriate behaviour. This is suggested to reflect an inability to pick up social reinforcers, and also an inability to assess facial expressions.

The amygdala is slower and less flexible than the orbitofrontal, with the neurons not reversing their response as rapidly. Three groups of neurons in the orbitofrontal provide computation, as to whether reinforcements formerly associated with particular stimuli are still being obtained. These neurons are involved in altering behavioural responses. This rapid reversal of response carries through from the orbitofrontal to the basal ganglia. It is suggested that in the process of evolution the orbitofrontal in primates has come to be further up in the hierarchy than the amygdala, thus allowing greater flexibility in behaviour.

Damage to the amygdala produces a deficit in the ability to learn to associate visual and other stimuli with primary rewards or punishers. The amygdala is subcortical and in the anterior part of the temporal lobe. It receives substantial projections from parts of the temporal lobe cortex involved in the most advanced stages of visual and auditory processing. These inputs could become secondary reinforcers. The neurons involved in forming such associations are in the amygdala and the orbitofrontal rather than the cortex. The amygdala has back projections to many areas of the cortex, and projects to the entorhinal cortex, which provides inputs to the hippocampus. The amygdala is well placed to receive highly processed information from the cortex, and to influence the rest of the limbic system, some parts of the cortex and the autonomic system. It may be involved in learning associations between visual stimuli and rewards. It is also thought likely that the amygdala controls approaches to new objects that may initially inspire fear. Amygdala neurons with reward-related responses responded to novel stimuli. Another group of neurons in the amygdala respond primarily to faces, and these are thought to receive their inputs from the superior temporal sulcus.

Activation of the anterior cingulate has been suggested to be related to pain. This may include social as well as physical pain. The anterior cingulate is connected to the medial and lateral orbitofrontal, the amygdala and the temporal pole cortex, and is seen as being involved in emotional activity. It is suggested to be part of a response-selection system. There is evidence that anterior cingulate lesions produce apathy and emotional instability. Positive stimuli are found to activate the intersection of the medial prefrontal, the cingulate and the orbitofrontal. Electrical stimulation in this region can relieve depression. Different parts of the cingulate appear to relate to pleasant and unpleasant stimuli.

Apart from emotion, current mood can effect the cognitive evaluation of events. For instance, depressed people may prolong their depression by recalling memories stored when they were previously depressed. It is suggested that when memories are stored, part of the context is stored with the memory. The recall of associations between events, where they occur and what was present, is seen as basic to episodic memory. The hippocampus is thought to contain a representation of the reward associations of highly processed stimuli. Some hippocampal neurons respond to spatial position, or where the subject is looking. This may be the basis of remembering which object or person is associated with which place. Object and spatial information may converge in the hippocampus, and it may store information about where emotionally-related events occurred. Memories may also be strengthened by back projections from the orbitofrontal and the amygdala to the parts of the cortex important in the representation of objects.

It may take several stages of processing to decode a primary reinforcer. Secondary reinforcement only occurs at a late stage of neural processing. Primary reinforcers are represented in the orbitofrontal and the amygdala and these areas are also involved in associating secondary reinforcers with primary reinforcers. These parts of the brain are involved in learning and in motivation. Cognitive inputs impacting emotional state are also represented in the orbitofrontal.

Consciousness & Freewill: Towards the end of his book, Rolls tries to tackle the contentious issues of consciousness and freewill. Emotion might be thought to be a particularly important area in regard to consciousness. Not only are we conscious of emotions, but if we were asked what would make us sure there was a difference between a conscious entity and any form of automaton, it might be the indication of preferences in a conscious entity, which in turn depends to a good extent on its emotional response to situations, objects and other entities.

Rolls admits that in teaching students, explanations of the physical instantiation of emotions are seen as leaving something out. It is not quite clear from the text, whether Rolls views this as indicating the lack of intellectual maturity and erudition in the students, or whether he is acknowledging a problem with orthodox descriptions of the brain. However, he does go on to ask what it is that makes the brain feel like something, when information processing is in progress. He acknowledges that it is not a general property of brain processing, because much brain processing is unconscious. He also acknowledges that qualia or subjective feeling, whether of emotions or sensory experience are a core problem.

It sometimes seems in looking at Rolls attempt at consciousness that he would have done better to follow what seem to be his own instincts and knowledge. Instead, as often happens when neuroscientists and artificial intelligence people look at consciousness, lacking confidence in their own thoughts, they resort to the crutch of depending on a philosopher. This usually means a philosopher who sees it as their duty is to prop up the Newtonian world picture, thus creating a circularity with the initial prejudices of neuroscience. Rolls decides to plump for Rosenthal, whose views are close to Dennett. Rosenthal seems to favour the now rather old-fashioned idea that consciousness is particularly related to language, even if language here means some form of mentalese, rather than natural language. Rosenthal seems drawn to the self-reflexive idea of consciousness. If one thing looks at another, and vice versa, consciousness will arise, much as if pointing two video cameras at one another would produce consciousness in one or both of the cameras. This seems the essential basis of the higher/lower order approaches to consciousness.

Rolls derives from Rosenthal various ideas about these hierarchies in thought, and notably the idea of higher-order thoughts about lower-order thoughts. At one point, Rolls opines that it is difficult to conceive that thought of one order about another order wouldn’t feel like something. It is hard to follow him on this, since there seems no reason in principle why a non-conscious computer should not perform processing, with respect to some slightly earlier and slightly less complex segment of its own processing. Indeed, having said that it is hard to conceive that consciousness would not arise in this type of processing, Rolls appears to produce contrary evidence to this within the same paragraph. He points out that amnesiacs that cannot remember what happened, or what they were thinking about, a view minutes ago, nevertheless, appear to be fully conscious. Well, when in trouble with consciousness take shelter in rambling on about evolution, and Rolls just does this. He points out that it is adaptive to reflect on previous events in order to correct errors. Few would be inclined to disagree with this, but it begs the question of whether or why such reflection, which is just another lot of processing, requires consciousness. Rolls does seem to admit that the so called lower-order thoughts or initial input from the external world would have to constitute real or subjective experience, if the so called higher-order thoughts about the inputs were to feel like something. This brings us full circle, because it begs the question of how or why the experience is there in the first or lower-order place. Rolls, in fact, admits in the next paragraph that after pages of rumination about higher and lower-order thoughts, he has given us no explanation of the qualia (subjective experience) of either sensations or emotions. The answer given to this is a non-answer. The sensations and emotions are conscious because that works better for the higher order system. However, the fact that such conscious inputs might be useful for higher order thoughts provides with no physical explanation of how such qualia, something not found elsewhere in the universe, arises.

The author’s own description of brain processes leaves him with something of a problem. Humans will act because of the experience of pleasantness and unpleasantness, and this would appear to make conscious experience causal for behaviour, contrary to the deeply held neuroscience faith in determinism and the consequently illusory nature of free will. Rolls suggests that the language/high order system would look ahead at the possible choices of action, and be able to go several steps out in looking at their consequences. This is more adaptive and flexible than just relying on the initial reward/punisher response delivered by the orbitofrontal. From this point on, the issue is to a great extent evaded. It is pointed out that a lot of behaviour is unconsciously driven, but can be wrongly thought to be consciously based. However, even in the author’s view this does not account for all the actions that are generally agreed to be conscious. No further arguments are presented, and at this point it is very suddenly and surprisingly asserted that the question of determinism or freewill is after all not apparently important, but instead, we are invited to concentrate our attention on the essentially mechanical question of which brain processes are engaged in decision taking and in emotions. The author’s discussion of freewill seems self-contradictory. Behaviour is described as flowing from the higher-order process of choosing between plans of action, but it is still only, it seems, a by-product of this system that it feels like something to be doing the processing. The author is trying to have his cake and eat it. Either the pleasantness/unpleasantness represented in the orbitofrontal and the higher-order thoughts about these are all based on algorithms, in which case it would be unadaptive to invest further energy in generating the consciousness of the processing, or they are not based on algorithms in which case consciousness of emotions is causal.

After all this, however, Rolls admits that he has not really provided an explanation for consciousness. Why should higher-order language processing feel like something, he asks. It just does, he answers, which is not exactly a satisfactory theory of consciousness. In essence, he does not believe in his own theory.

2.)

Affective Neuroscience

Jaak Panksepp

Oxford University Press, (1998) ISBN 978-0-19-517805-0

The main part of this book concentrates on the neurotransmitters and the areas of the brain that instantiate emotions. In this respect, it appears to be a useful textbook, although now possibly dated in being written before more recent offerings that centre on the importance of the orbitofrontal. The material while interesting in detail lacks much feel of how the whole thing would hold together.

As has become the pattern in modern neuroscience books, discussion of consciousness is quarantined in the final chapter, with the rest of the book written as if consciousness did not exist. Only in this last chapter, is it admitted that humans are creatures that feel. The author argues that emotional systems are adaptive, helping to organise behaviour in an adaptive manner. This refers to a series of systems; SEEKING which investigates the environment for food, mates etc; FEAR which precipitates flight from danger; RAGE which encourages attack; PANIC which promotes the seeking of social support; play and lust are also categorised as systems.

The author argues against the view that the subcortical areas of the brain are all completely unconscious. He points out that damage to, or stimulation of subcortical areas, can have a major impact on purposefulness and emotional responses. His comments on split-brain patients are welcome. Too often, it is claimed that in such patients the consciousness of the two hemispheres go their separate ways. What is usually not mentioned is that such an effect is only achieved by carefully contrived experiments, and that in normal circumstance split-brain patients have surprisingly few problems. The effective misrepresentation of the state of these patients often seems to be part of a strategy of belittling the importance of consciousness. Here, the author argues that the normality of the split-brain patients indicates the importance of subcortical connections for consciousness.

Panksepp, however, takes his argument an ambitious step further, by claiming that evidence for the involvement of subcortical areas in consciousness suggests the existence of a “core of being” in the subcortical areas. This is what is more usually known as the ‘self’. In contrast to many mainstream investigators, who tend to deconstruct the self and categorise it as an illusion, Panksepp makes it the basis of his theory of consciousness.

There is suggested to be a basic body image within the brain stem. There might seem to be quite a jump, from merely putting forward evidence that subcortical regions are involved in consciousness, to zooming in on the brain stem, and then endowing this very archaic area of the brain with a body image. Panksepp seems to be falling back on the old argument that because consciousness ceases, if the brain stem is put out of action, the brain stem has to be the sole, or at least the dominant, centre of consciousness. This ignores the basic logical distinction between something being necessary and it being sufficient. Much other evidence suggests the existence of conscious processing in other areas of the brain. Panksepp himself discusses the issues of blindsight in which the conscious as opposed to non-conscious processing appears to be in the cortex. In fact, the whole argument between cortical and subcortical consciousness seems rather arid. There is evidence for consciousness in both areas, and the most likely explanation is that there is an over arching process binding all conscious areas of the brain. However, the possible involvement of the subcortical areas would at least suggest that the existence of consciousness can be pushed back to an early stage of evolution.

Panksepp, however, wants to concentrate on the brain stem and further suggests, without perhaps much foundation, that in this area consciousness is driven by a primitive motor representation rather than by sensory input. This is where the self resides. In respect of the self, the author seems to try to have his cake and eat it. There is a ‘self representation’ that ‘does not observe in the conventional sense’ but ‘is observed by or intermeshed with the higher perceptual processes’. The self representation or ‘self schema’ provides input into the sensory analysers. What does this all actually mean? The author isn’t willing to say that there is self that observes, presumably because the rejection of such ideas is a faith aspect of neuroscience. However, to make his scheme work he needs something that effectively performs the role of an observing self without actually observing. The self is now demoted to a ‘self-representation’ or ‘self-schema’, but one is left suspecting that the ‘self-schema’ has to somehow do the things expected of a self. The self-representation or schema ‘does not observe in the conventional sense’, we are told, which leaves the question as to what it is to observe something in an unconventional sense. Again, it seems to have to function as if it was observing, while not contradicting the faith claim that nothing in the brain ever observes. But perhaps the burden of being a self can be lightened by passing some of the responsibility to the sensory cortex after all. It is suddenly the self-representation or self-schema that is observed by the higher processes. This might seem to go against the rule that nothing in the brain can observe, but maybe the problem has been sufficiently diluted to permit this, by this stage. In any case, this observing by the higher processes is qualified as ‘observed or at least strongly intermeshed’, although the latter would seem to be a much more general process, perhaps covering any form of synaptic contact.

Beyond this point the author seems to resort to the old ‘video camera’ get out on consciousness. Emotional consciousness somehow arises from the interaction of the higher and lower processes. In this type of consciousness theory, if two things interact for long enough, they will produce consciousness, where there was none before, much like two video cameras becoming conscious by filming one another.

The subcortical basis of consciousness is also supposed to solve the ‘binding problem’, the experienced unity of separate processing in the different parts of the sensory cortex. This is superficially plausible given that the subcortical is fundamental relative to other areas of the brain, but in fact it does not provide a physical explanation of how, for instance, the visual and auditory areas of the cortex are bound together in the experienced unity of consciousness. The motor aspect of the self does seem to have a certain intuitive plausibility, but the experience of motion is after all only a sensory input much like other sensory inputs, and it is not clear, why it should be set so much above other senses in the modern brain, even if the sense of movement did originally pre-date other senses in simple and very archaic brains. He argues that mapping in the subcortical areas is more important than the planning and executive relationship between the prefrontal and the motor cortex. These are merely contents of consciousness. His arguments appear less than convincing. Granted the lower areas are more ancient and seemingly must play a role in the coherence of split-brain patients, but for this they only need to be a part of a conscious system, not of necessity the dominant centre. The author does refer to the ability of subcortical damage to remove the ability of animals to exhibit purposeful action or emotional response. It is not clear, however, that such animals lack internal consciousness, and even if they do, this may be just a example of the electric plug or necessary versus sufficient argument. The plug to the mains is necessary for the operation of the television, but it is not sufficient by itself. Brain stem and similar arguments about consciousness hang on the fact of necessity, but do not prove that consciousness does not exist in other parts of the brain.

Finally, Panksepp admits that even granted his subcortical argument, he has not demonstrated the ‘exact manner in which experience emerges from neurodynamics’. Why should the body image in the brain stem, or its interaction with external sensory information, feel like anything. At this point, Panksepp is not alone in retreating into a ‘new mysterian’ position. Humans are deemed unable to verbally symbolise the complexity of the brain. Maybe, but we have mathematics to symbolise complexity, which has worked well in other science, and the results of this can be re-interpreted as analogies, as in popular accounts of quantum theory, relativity and cosmology. Anyway, it is only a dubious assumption that the problem lies in complexity, rather than the failure to grasp the basis of some process. Advances such as understanding gravity or the qauntisation of energy and matter involved bold simple concepts that were afterwards supported by complex calculations.

3.)

The Feeling of What Happens

Antonio Damasio

The most useful function, although not the main purpose of Damasio's latest book, is to lay to rest several of the scientific notions about emotions, language and other aspects of the mind, which had dogged scientific discussion during the 1990s.

However, in its central purpose, of discussing consciousness and its relationship to emotion and the self the book proves disappointing. In the first place, the book tends to identify the self and consciousness as being the same thing, rather than treating the self as only a part of the contents of consciousness. The book has very little on the question of the qualia. Damasio was already aware of this criticism of his approach at the time of writing the book, but sidesteps this criticism with the claim that all consciousness can be seen as consciousness with a sense of self. This may true in a sense most of the time, but it does not get round the experience of qualia, such as the colour red, as something external to us and the self. Nor does it deal with altered states of consciousness in which the sense of self is reported to disappear, but something continues to observe. In the end consciousness is argued to arise from the ensemble of brain functions governing the body, but it is not all clear from Damasio's book, how the non-conscious signaling systems involved would make the jump into conscious, and thus gain a property not detected elsewhere in the universe.

Damasio argues in particular against the popular late 20th century view that consciousness was derived from language. He sees language, comprising words and sentences, as a translation from non-linguistic images that stand for particular entities or events. The original images are seen as having been non-verbal. However there is no image as such for the self or for consciousness, so if these properties were created by language, they would lack the underlying concept of self.

Where Damasio studied patients with speech impairments, he found that the basic thought processes were intact, and that the patients' consciousness of their situation seemed no different from anyone else. Language impaired patients could be attentive and purposeful, and could communicate effectively with sign language. There were also indications of emotional responses. Even where there was complete loss of speech, it was possible to maintain sign communication, and there was no indication of lack of consciousness.

Damasio further thinks that core consciousness is not based on memory, except in respect of a very brief short term memory. He does not think that core consciousness is based on either longer-term memory or language. Core consciousness is retained even when subjects have major brain deficits. In certain types of amnesia knowledge of both the past and the anticipated future is lost, but core consciousness and a sense of self are retained. The core self is seen as a transient moment-to-moment experience, but the autobiographical self is relatively stable. However, core consciousness is felt to be closely related to both primary and background emotions.

The other problem of the book could be felt to be the over emphasis on the distinction between core consciousness found in all developed organisms and the extended consciousness of humans. Core consciousness seems to be a property of all reasonably developed organisms. Where core consciousness is disrupted the structures involved are usually near the midline of the brain in both humans and non-humans. While it seems fair to make a distinction between bare awareness of surroundings and the various achievements of humans, both would seem to be a subset of the same hard problem, by which consciousness is an entirely different property from anything in the inanimate world.

Damasio identifies consciousness as the biological function that allows us to be aware of emotions. This is at least an advance of 20th century brain research, which was prone to marginalise the role of the emotions. Damasio's own research has shown that contrary to 20th century views emotion is essential to both reasoning and decision taking. He views consciousness as providing knowledge of suffering, pleasure, embarrassement, pride and in humans, the adaptively important experience of empathy. Damage in the cingulate cortex, for instance, can disrupt both core and extended consciousness and may result in zombie like behaviour.

He also argues that our existence of what are effectively zombie states, contrary to Dennett's attempt to prove that these cannot exist. He says that in some unusual states, for instance states related to epilepsy, a subject can be unconscious and not capable of retaining memories but still capable of movement and attention to surrounding objects.

Damasio describes himself as having been puzzled as to how the activities of the limbic system, which instantiates the emotions in the brain, came to produce an actual experience for the organism. In principle, the activity in the limbic system and its interaction with the prefrontal could be an entirely non-conscious signalling system, as are many other segments of the brain.

He defines six primary emotions, happiness, sadness, fear, surprise or disgust. However, there are also secondary or social emotions such as embarrassement, guilt and pride, and 'background emotions' such as calm or tension.

He sees the role of emotion as being to create circumstances advantageous to the organism having the emotions. They are generated from an ensemble of sub-cortical regions, starting with the brain stem, and they both regulate and represent body states. He sees the body as a theatre of the emotions, with emotions regulating internal states in preparation for specific physical reactions.

Damasio argues that the emotions are made known or experienced through the sense of self. Therefore, it was important to know how the sense of self was generated. He identifies two aspects of the self, firstly what he calls 'the movie in the brain', and secondly the sense that there is an owner or observer of the movie.

He thinks that consciousness and emotion are inseparable in a way that consciousness and wakefulness are not, and points out that impaired consciousness usually results in impaired emotional response.

Damasio defines core consciousness as providing a sense of self relative to one moment and one place, the here and now. Extended self, as found in humans, is said to provide a sense of self at a point in the history of an individual and relative to an anticipated future. This autobiographical self corresponds to a collection of unique facts characterising a person and depending on an organised memory.

The body state has to be kept within narrow parameters. The brain governs the body state by detecting small changes in the body's chemical profile, and acting to correct these. The most important components for this control are the brain stem, basal forebrain and hypothalamus. Neural maps signal moment by moment the state of the entire organism. Damasio thinks that the roots of the self are in this ensemble of brain structures that govern the body and in his view at the same time generate the representation of being. Like so much of mainstream consciousness theory there is a quality of wooliness. It is actually not at all clear by what physical process this ensemble of signalling systems in the body would make the essential jump from conscious to non-conscious.

Damasio considers that we are not conscious of our feelings, although we tend to view consciousness and feelings as the same sort of thing. He thinks that consciousness is only necessary if feelings are going to influence the subject in the future. Emotion and consciousness are seen as separate although possibly having a common underpinning. Damasio thinks that emotions are part of the process of homeostasis, by which the body maintains its stability. When emotions are actively experienced, the hypothalmus, basal forebrain and brain stem release neurotransmitters in many regions of the brain with important effects, such as pleasant or unpleasant experience.

The emotions are mainly located in sub-cortical areas. These include the brain stem, hypothalamus, basal forebrain, PAG and amygdala. In the cortex, the anterior cingulate and the ventromedial prefrontal are important. Different sites process different emotions. Sadness activates the ventromedial prefrontal, hypothalamus and brain stem, which are however not activated by anger. The amygdala is in the depths of the temporal lobe and relates to fear but not disgust or happiness.

With an emotional response, two types of message are sent out from the brain region. One route is hormones in the bloodstream that act on receptors in cells and the other is nerve signals. In external pain such as heat, the heat activates thin unmyelinated nerve fibres known as C fibres. This may be experienced as pain, and even if the subject is unconscious, there will be reactions.

Work in Damasio's lab has shown emotion to be integral to reasoning and decision taking. This evidence comes from subjects, who displayed normal rationality, until they lost the ability to experience certain emotions as a result of brain damage in the ventral and medial prefrontal and the parietal. These patients could reason logically, but actual decisions were erratic. Reason can be used to control emotions, but reason also needs a degree of emotional input to function properly. Thus contrary to the Dr. Spock image, reasoning for the purpose of action does not benefit from the abscence of emotion.

Damasio also showed that feelings could be unconscious. This was based on controlled tests with an amnesiac subject, who had no memory of what he had done on the previous day. He would be introduced at separate times to two researchers, one of whom would involve him in tedious task and the other who would let the patient do what he wanted. The next day, although he had no memory of the individuals involved, he showed a preference for the researcher who had allowed him to do what he wanted. Most images whether remembered or not are considered to carry some emotional charge.

Damasio thinks that the sense of self has a biological precedent in an entity called the proto-self. This is a collection of neural patterns mapping the physical structure of the organism. The proto-self is conceived as emerging from the various activities of the nervous system. A number of brain regions are considered to be necessary to implement the proto-self, including the brain stem nuclei, the hypothalamus, the basal forebrain, the insula and the medial parietal cortex. Damasio envisages that a limited core consciousness arises from the relationship between organism and object, although it is not clear what it is in this relationship that kick starts the move from inanimate information exchange to the different property of conscious awareness.

4.)

Neural mechanisms of autonomic, affective and cognitive integration

Hugo D. Critchley, Wellcome and UCL P. Journal of Comparitive Neurology, 493, pp. 154-6

INTRODUCTION: This paper discusses evidence for the involvement of bodily responses in brain processes, particularly those related to emotional experience. There does, however, seem to be 'a dog that doesn't bark in the night' somewhere in this paper. There seems to be an unspoken assumption that there is an important distinction between volitional or motivational actions and unconscious activity, and also an assumption that subjective emotions are somehow important to the former. This of course flies in the face of the rigid orthodoxy of psychology and most neuroscience to the effect this distinction is an illusion, and that subjective emotions and other experience are of little scientific relevance. An additional problem in reading this paper is that it is not clear whether the author thinks that all emotional experience is derived from bodily sensations or only some. While the studies discussed in the paper certainly support the latter, the former looks less plausible.

This paper discusses studies that support the view that bodily processes act on brain processes, and are important in the generation of the subjective experience of emotions. Specific brain areas are highlighted in respect to brain-body interaction and emotion. The anterior cingulate cortex is seen as being involved in generating responses by the autonomic (involuntary) parts of the nervous system, while the insula and orbitofrontal cortex are thought likely to map the visceral (internal organs) responses. The ventromedial prefrontal cortex supports states of rest that may serve as a benchmark for more dynamic activity. The interaction of the anterior cingulate, the insula and the orbitofrontal are suggested as possibly being the basis for emotional experience and motivated behaviour. Generation of and subsequent feedback from autonomic processes is suggested to be linked to subjective emotions.

The autonomic nervous system is the mainly regulator of bodily functions, and allows responses to environmental changes. The autonomic system is divided into the sympathetic and the parasympathetic system. The sympathetic system relates to motor action and changes such as heart rate that relate to motor action, and it is thus associated with 'fight and flight reactions'. The parasympathetic system is involved with recuperative processes, such as reducing the heart rate. The sympathetic system originates in the brain stem, and extends down down the spine and utilises adrenaline and noradrenaline as transmitters. The parasympathetic also originates from the brain stem and uses acetylcholine as a transmitter. The sympathetic system is also acted on by nuclei in the hypothalamus. Experiments have demonstrated influences on the autonomic system from the cingulate, the insula and the medial temporal lobe via both the hypothalamus and the brain stem. However, the brain stem's autonomic centres require feedback from the body to maintain homeostasis (stable conditions in the body). This feedback also influences motivational behaviour by conveying information on levels of comfort or discomfort.

In support of Damasio's somatic marker theory, the experience of feedback from bodily states is hypothesised to be the basis of the subjective experience of emotion. This argument seems sound up to a point, but it is difficult to think that external stimuli, especially the more urgent ones, for instances phobic fear reactions, cannot occur without being laboriously processed through internal organs. The same qualification could apply to emotions arising from cognitive activity. Again it seems laborious and maladaptive in terms of use of energy for everything to have to go via the internal organs, before it can be assessed in terms of emotional experience. Another objection dating back to the 1920s is that bodily arousal is too limited in its range to account for all the variations in subjective emotional experience. The impairment of judgment, decision taking and behaviour in patients with orbitofrontal and ventral prefrontal damage is seen as supportive of the somatic marker idea, but at least some of the deficits here can also be viewed as a consequence of impaired communication between the frontal and limbic areas of the brain. The finding that autonomic arousal is reduced in patients with lesions does not seem that surprising, as outward as well as inward signaling is likely to be impaired by the lesions in the brain. In particular, this does not seem enough to support Damasio's rather vague notion of the self arising from representations of the body state. This is not to say that the body plays no part in it, but it would seem to require considerably more evidence to suggest that the body by itself creates the self.

The evidence of studies indicates that the hypothalamus monitors the body, and that sensory information from the body projects to viscosensory cortex in the insula and the orbitofrontal. In primates this information does not go via a structure in the pons area of the brain stem, and this would seem to be indicative of a less automatic system than in other animals. Similarly, sympathetic arousal of skin conductance has been shown to enhance activity in the ventromedial prefrontal, the right anterior insula and the dorsal anterior cingulate.

The dorsal anterior cingulate is seen as a brain area of particular interest because it is involved in attention and cognitively demanding activities. Researchers have considered that it may have a role in executive control and possibly consciousness. There is a correlation between dorsal anterior cingulate activity and task difficulty, and this type of mental stress enhances sympathetic activity. It is claimed that various studies argue in favour of the dorsal anterior cingulate having control over the autonomic system during volitional behaviour, including difficult cognitive activity.

The amygdala and particularly the central nucleus of the amygdala is also seen to produce autonomic arousal when it is receiving emotional stimuli. The dorsal anterior cingulate and the amygdala are often active at the same time. The activity of the dorsal anterior cingulate during volitional behaviour may provide control over the autonomic system. Also visceral and pain stimuli are associated with enhanced activity in the anterior cingulate, insula and thalamus. The dorsal pons in the brain stem and the anterior cingulate are both sensitive to any absence of feed back from the body. The mid-insula and the amygdala are sensitive to autonomic arousal as a result of emotional stimuli, and this is taken to suggest a role for the right insula in the experience of emotion. Activity in the right anterior insula predicts subjective emotional experience and is also connected to visceral activity.

Generation of autonomic responses may originate in the anterior cingulate during volitional and cognitive activity, while the conscious experience of visceral responses and subjective emotion may arise in the anterior insula, especially the right anterior insula. The insula and the anterior cingulate are often active at the same time in the event of pain, threat or attention. The activity of the dorsal anterior cingulate predicts autonomic arousal. Other research emphasises the importance of the anterior cingulate in cognitive control. The anterior cingulate is important for monitoring actions and their consequences and also performance errors. The anterior cingulate and the amygdala generate and map bodily responses that are interconnected with the orbitomedial region. The intensity of bodily responses is suggested to have an influence on subjective emotion.