s

resolutionmini

haut de page

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

haut de page

mise à jour du
22 août 2002
 Nature Reviews Neuroscience
2002;3:655-666
How do you feel ? Interoception: the sense of the physiological condition of the body
AD Craig
Atkinson Pain Research Laboratory, Division of Neurosurgery,
Barrow Neurological Institute, Phoenix, Arizona. USA
Chat-logomini
L'identification d'une architecture neuronale spécifique dédiée à la représentation de l'état fonctionnel interne de l'organisme et à son homéostasie permet de comprendre comment les sensations internes influencent le comportement et inter-réagissent avec les émotions. Cet article propose une vision possible de la perception du bien-être ressenti par le bâillement et de la neurophysiologie de sa contagiosité.

[···] Conclusions and future directions :

The identification of an entire neural system hat can be cogently conceptualized as a representation of the physiological condition of the material body has several fundamental implications. It provides a rational explanation for the long-recognized association of pain, temperature, itch and other feelings from the body, separate from the lemniscal system that represents exteroceptive touch and proprioception.

It incorporates specific labelled lines for several physical conditions that generate distinct feelings, and it substantiates their common integration in the hierarchical homeostatic network. It provides the long-missing peripheral and central afferent complement to the efferent autonomic nervous system. These findings reveal a direct cortical image of the state of the body that differentiates primates from sub-primates neuroanatomically. The size and multiple representations of this interoceptive image seem to differentiate humans from subhuman primates. Finally, these findings signify the cortical representation of feelings from the body as the likely basis for human awareness of the physical self as a feeling entity. This association provides a fundamental framework for the involvement of these feelings with emotion, mood, motivation and consciousness. These concepts emerge directly from the functional anatomy of the lamina I spinothalamocortical system, rather than from preconceived ideas.

It is important to recognize that this neural sensory system is part of an entire network involved in homeostasis; that is, in the autonomic, hormonal and behavioural neural mechanisms that maintain optimal physiological conditions in the body and that respond in an integrated and on going fashion to all interior and exterior environmental challenges, ranging from exercise, dehydration or altitude to injury, sepsis or social interactions.

The organization of this network is focused at the spinal level on cardiovascular and direct endo organ control at the brainstem level on integrated control of fluid, electrolyte, energy, immune and cardiorespiratory balances the forebrain level in sub-primates on neuro endocrine and behavioural control, and in primates in a high-resolution encephalized representation of all aspects of the condition of the tissues.

Nevertheless, it is avertically integrated system, and it is important to recognize that the basis for feelings from the body in humans is this hierarchical association with homeostatic mechanisms.This hypothesis is supported by the close correlation of brainstem activity with these subjective feelings in human imaging studies. This recognition recommends analysis of the interactions of feelings and emotion with many aspects of subconscious homeostasis; for example, in stress or pain or cognitive behavioural research.

In contrast to the many discriminable sensations from the body, the subjective appreciation of visceral sensation is more diffuse, less well localized, and usually below perceptive thresholds. This was one of the main reasons for the long-standing mis-categorization of pain and temperature as exteroceptive rather than interoceptive. Although it would be highly inefficient for gastro intestinal processing to require constant behavioural supervision, this perceptual difference remains to be explained adequately. Notably, many observations indicate that there is opponent processing between parasympathetic and sympathetic afferents that parallels their efferent opponency. For example, there are obligatory mutual inhibitory interactions between spinal and vagal small-diameter afferent activities in the medulla that are essential for cardiorespiratory control. Similarly, vagal afferent activation inhibits both pain sensation and spinal visceroceptive processing.Vagal stimulation can reportedly reduce stress and depression clinically. Similarly, opposing effects on autonomic function have been elicited by stimulation of human insular cortex on both sides, and corresponding cortical lateralization has been observed with micturition and gustation.

Such a basic organization would be parsimonious with many considerations and could explain the perceptual differences, but this certainly needs further study, particularly because of the potential clinical significance.The association of the re-representation of the interoceptive pathway with self-awarenes simplies the existence of neuroanatomically verifiable correlates of conscious behaviour. To this end, we are now comparing the size and cytoarchitectonic differentiation of the thalamic relay VMpo in different primate species. Preliminary observations are supportive; VMpo in the pygmy chimpanzee, which can recognize itself in a mirror, is clearly similar to that in the human, albeit considerably smaller. By contrast, VMpo in the gibbon,which cannot recognize itself, is barely recognizable, like that in the macaque.

Similarly, a unique cytological feature has been described in the ACC and the anteriori nsula of human and higher sub-human primates that is not present in lower animals.Further anatomical analyses of the interoceptive cortex in sub-human primates and of the anterior insula in humans are certainly needed, and correlative imaging and clinical approaches would be most useful.

Finally, this conceptual framework has strong implications for medicine. The integrated neural representation of all aspects of the condition of the body in a system responsible for homeostasis and associated with stress, including a direct cortical image of physical well-being, provides a sound epistemological foundation for integrated approaches to the treatment of pain, metabolic, eating and psychosomatic disorders.

For example, this provides an easy formulation for somatization under emotional stress. Similarly, these considerations imply that mysterious pain syndromes, such as fibromyalgia (deepaches and pains), could be related to homeostatic dysfunction (for example, salt or waterbalance or cardiovascular function), rather than to tissue damage, and this possibility deserves vigorous study.

Consideration of these findings led directly to the recent proposal that the central pain syndrome is a thermoregulatory disorder. The recognition that sensual touch is incorporated into the interoceptive system has strong implications for the neurobiological and health effects of conspecific contact, visitors to zoos will remember that monkeys, chimpanzees and bonobos normally spend an enormous amount of time grooming and cuddling each other, and readers will remember the classic studies by H.Harlow showing the importance of conspecific contact for emotional development. Last, the observation that the neuroanatomical substrate for subjective emotion in humans is based on an abstracted meta-representationof the physiological state of the body, consistent with the conjectures of James andDamasio, provides a basis for the volitional modulation of feelings, emotion and efferent activity affecting the state of the body that is unique to humans, and clearly emphasizes the role of the body's health in human consciousness and interaction.


The 'somatic marker' hypothesis of consciousness

On the basis of neurological analyses of patients with forebrain lesions, Antonio Damasio has advanced the 'somatic marker' hypothesis of consciousness.He proposes that the subjective process of feeling emotions requires the participation of brain regions that are involved in the mapping and/or regulation of our continuously changing internal states, that is, in homeostasis. These feelings help to guide behavioural decisions that affect survival and quality of life by producing a 'perceptual landscape' that represents the emotional significance of a particular stimulus that is being experienced, or of a projected future action by means of a further 'as-if-body loop'mechanism. The feelings are grounded in the body itself, based on multi-tiered and evolutionarily developed neural mechanisms that control the body's state. These feelings distinguish between inner-world representations and outer-world representations, and allow the brain to build a meta-representational model of the relationship between outer and inner entities. So, the representational image of the body's state provides a neural basis for distinguishing self from non-self, and re-representations of this image enable the behavioural neural agent to project the effects of possible actions onto the state of the body, as well as the resultant changes in such feeling states due to interactions with other (external) agents. This hypothesis posits that degrees of conscious awareness are related to successive upgrades in the self-representational maps. The anatomical features of Damasio's hypothesis include a central role for the anterior insular cortex in the representation of such feeling states.