A loopy model of mind

Rodney Cotterill's Enchanted Looms makes a most timely and welcome appearance. Yes, he does have his own theory of how the brain generates consciousness, but he puts it in the context of a discursive mechanistic presentation of what is and what is not known about how the brain might do the trick. After all, the brain is a machine and its product is consciousness. There is little use to approach it, as others have done, through metaphors such as ripples in a neural pond, Godelian knots, multiple drafts and computer programs. As followers of the literature will recognise, all of these have been part of the diet of the past ten years. Cotterill accepts the extreme biological, anatomical and biochemical complexity of the brain and examines it with the fine-tooth comb of a systems engineer faced with somebody else's design and no instruction manual.

The book is well structured, moving easily between anatomical detail, functional theories, methodologies and the analyses of others, making it mandatory reading for those considering doing research on this topic. But it is patently written at two levels: the educational, instructive text on the one hand, and a seemingly logical progression towards Cotterill's own thoughts on consciousness on the other. At the former level, the text cannot be faulted for thoroughness and careful coverage. The latter, however, deserves a slightly more critical examination. Cotterill regards the main purpose of consciousness as representing in neural activity the choices of actions available to an individual at any one moment. But the action need not actually be taken, and the key element here is a suppression of action - the "veto-on-the-fly" - mediated by a part of the brain called the anterior cingulate. That is, sensory signals are first projected to the rear parts of the cortex where they are deconstructed into significant features. They then travel to the frontal regions where they can access the action centres (premotor areas) where they are reconstructed in such a way as to facilitate an accurate interaction of the organism with the world. So far, this has the essence of models produced by other workers, notably Francis Crick and Christof Koch in California and Semir Zeki in the UK. However, Cotterill goes further to suggest a closed loop back from the premotor areas through subcortical regions on which the anterior cingulate does its vetoing stuff and then on to the occipital regions again. So consciousness emerges as a sustained readiness of this loop to interact with the world as sensory information is received.

The side of his model that I like is that it uses an engineering knowledge of how complex loops with continually recoded data might work: what makes them stable and how they interact with vast amounts of incoming sensory information. Ultimately, however, models are just hypotheses that have to stand alongside others. The challenge for engineering modellers is to generate clear hypotheses that those who study the brain at close quarters can test. With a topic such as consciousness, this is difficult as the hypothesis has to straddle the reports of an individual (a highly filtered version of what that individual feels) with things that can be measured. Measurement methods are advancing rapidly, but what can be measured today is not yet accurate enough to distinguish the difference in brain activity between, say, one who likes and one who dislikes Jeffrey Archer's novels.

Personally, I would have appreciated Cotterill's assessment of his own model from the perspective of some of the more controversial models in the literature, such as the Penrose and Hamerhoff quantum consciousness argument, Susan Greenfield's ideas on the intensity of consciousness and Steven Pinker's notion of how mind and language work. Some quibbles could also be directed at the publisher. The indexing is poor. Important topics such as "association areas" or "primary visual cortex" do not appear, although they are central to the narrative. Cotterill certainly does not claim to provide the solution to the problem of consciousness. And despite the fact that he will not satisfy those who believe in a mysterious gap between brain activity and inner sensation, I do believe he shows how some of the major pieces of the jigsaw puzzle could fit together.

Igor Aleksander is professor of neural systems engineering, Imperial College, London.