How I know they are talking about me

What is consciousness?" is the question emblazoned on the back cover of this book, edited by the psychologist/philosopher Max Velmans. Back in 1913, the founder of behaviouralism, James Watson, declared that "psychology must discard all reference to consciousness I its sole task is the prediction and control of behaviour and introspection can form no part of its methods". Now consciousness is seen by some as psychology's central concern.

The Science of Consciousness is a collection of nine review articles on the state of play in cognitive psychology. The articles are by leading researchers and are aimed at specialists in the field.

Two contributions by John Kihlstrom and John Gardiner review psychologists' attempts to demarcate the boundary between unconscious processing of subliminal input and conscious states. For example, at a cocktail party, speech from all over the room enters the ear, but if one's name is mentioned at the other end of the room, one becomes alert to the fact immediately. How does the brain select just the appropriate message for conscious attention from this cacophony? The brain must be continually analysing and filtering all the potentially meaningful incoming information at an unconscious level. Psychologists have developed some ingenious methods to explore these and other subliminal "priming" effects on perception, memory and learning.

However, as is clear from these two articles, experiments on implicit learning and memory are scratching around the periphery of consciousness. Furthermore, the results obtained, though generally agreed to show positive priming effects, are still controversial. The main difficulty of interpretation is in deciding just whether and of what, the subject is or is not conscious. One way to finesse these difficulties is being explored and depends on the idea that only conscious representations allow voluntary control; unconscious representations do not. Subjects are asked to avoid reporting perceived cues. If the cue is subliminal and unconscious it has an unavoidable influence on subsequent behaviour. Only when the cue is conscious can subjects employ a strategy to overcome these automatic influences.

Bernard Baars and Katherine McGovern, in their lively article, discuss how consciousness seems to be the gateway through which we gain access to a vast continent of knowledge and skills. For example, a carefully placed needle electrode in your thumb can tap into a single motor nerve. When the signal from that nerve is made conscious by being played back to you as a click through a loudspeaker, you can learn to control the output of the nerve in about ten minutes and, if you are talented, even learn to play drumrolls in about half an hour. To explain this flexibility of conscious processing and its relation to the more automatic unconscious activity, Baars and McGovern investigate a number of models. Each of these are systems of labelled boxes connected by arrows, representing conjectured cognitive "modules" and their interconnections. These constructions have developed into something of an art form among cognitive psychologists. For example, there is Daniel Schacter's "dissociable interactions" model, and Tim Shallice's "supervisory system" model, and Baar's own "global workspace" model. Each are attempts to represent in schematic form what is known about how the mind processes information, on the basis of existing psychological evidence. These models lead us to ask questions about what predictions they make that distinguish them from each other, what the inner language of each proposed module might be, and how these are translated so they can talk to each other; and what reality these cognitive modules have in terms of brain architecture.

Control of a single nerve in your thumb is just one example of how consciousness can have somatic or bodily effects. Many others are reviewed by Anees and Katarina Sheick, and Robert Kunzendorf, and by Patrick Wall in his review of the placebo effect. These articles muster a substantial amount of evidence to try to demonstrate these phenomena, however these authors are the first to point out the existence of any possible methodological flaws. The relevance of this material is mainly to clinical work.

Andrew Young discusses how highly selective aspects of consciousness can be affected after brain damage. For example, he quotes pioneering work by Tore Torjussen who reports patients with right-side visual field damage reporting seeing all of a circle when it was presented so as to fall half in the blind region and half in the sighted region, yet they report nothing if a semi-circle is presented in the blind visual field, and just a semicircle if this is presented in the sighted region. This finding shows that the interaction between stimuli in the blind and sighted parts of the visual field can affect what people report they see in the blind region.

L-3drop = /Benjamin Libet reviews his justly famous work on the timing of conscious experience. He made the remarkable discovery that subjects appear to backdate experiences so as to discount the substantial delay from initial stimulus to consciousness.

L-text = /In an introduction and a concluding article Velmans argues, on a more philosophical level, for the reality of the first-person perspective and its importance for psychology. In particular, he emphasises that the basis of all knowledge lies in subjective experience. If this view is correct, it means that only by taking the first person perspective seriously will psychology gain the understanding of consciousness that it desperately needs. With such an understanding it may at last have a central organising principle.

Paul Caro is honorary research associate and Robert Seymour is reader in mathematics, University College London.