Hunting the ghost: toward a neuroscience of consciousness

Abstract

Consciousness is a term with many meanings. In one sense, we use the term to indicate whether or not an organism is in a conscious state. In this sense, consciousness is what is altered, reduced, or even lost when we faint or undergo deep general anaesthesia. In a second sense, it is a trait, an attribute of a psychological process; we may think, desire, hear, see, and feel consciously, thereby becoming conscious of thoughts, wishes, voices and music, of colours and textures. Local anaesthesia, by abolishing conscious sensations of touch in the affected limb, thus interferes with the consciousness of something. Within trait consciousness, I draw a further distinction between conscious representations and conscious Stoerig: Hunting the Ghost 2 access. Conscious representations are of objects or contents of perceptions, desires, or actions; they are usually phenomenal, like a strawberry that is red, sweet. and heart-shaped and that I see, smell, and desire to eat. To describe the strawberry’s looks and taste or to resist its temptation, conscious access to its representation is required (Block, 1995). This chapter reviews evidence pertaining to the neural basis of state and trait consciousness (Stoerig, 2002). As we (still?) lack an objective marker of conscious representations that is independent of the subjects’ overt behaviour, we can assess conscious representations only when our subjects can access and (verbally or non-verbally) express them. The same applies to the conscious state, because it is only observable conscious access that precludes a diagnosis of unconsciousness. If conscious representations as well as state consciousness are attributed on the basis of evidence for conscious access, the neuronal processes that mediate conscious access in its many forms are likely to contaminate what we learn about those that mediate conscious representations and states. A prominent candidate for mediating conscious access is a network of frontoparietal cortical regions that play an important role in attentional and behavioural selection of incoming and stored information. As these regions allocate processing resources and guide behavioural selection, it is not surprising that they are activated both when vegetative state patients recover and when healthy subjects perform demanding perceptual tasks. Although the frontoparietal network has been implicated in the mediation of the conscious state, of conscious representations, and of conscious access, I argue that these different manifestations of consciousness may well depend on different neuronal processes. Stoerig: Hunting the Ghost 3 Consciousness as a State of an Organism Is it possible to identify a system of neuronal structures that mediate all conscious states and whose destruction or downregulation abolishes consciousness? It should be a non-specific system because we are in a conscious state regardless of whether we listen to a concert, indulge in French cuisine, or suffer from heartburn afterward; we are conscious (‘bei Bewusstsein’) regardless of what we are conscious of. Being in a conscious state is prerequisite to any conscious experience, and alterations in the state cause changes in the experience. The following sections discuss sleep as an example of circadian endogenous state changes; general anaesthesia as an example of chemically induced state changes; and comatose, vegetative, and minimally conscious states as examples of pathology-induced state changes. The purpose is to provide up-to-date information on several aspects of consciousness, as well as their interdependence, and to point out some critical confounds.