You do not currently have permission to view the full text for this document. Please sign in to an account with an active subscription or contact support.
In “The Unconscious Homunculus” and other works, Francis Crick and Christof Koch make an admirable case for attempting to discover the neural correlates of consciousness (NCC). As they observe, their research complements rather than supplants the sort of work I have done on what might be called the representational correlates of consciousness—the formal organization of the computations and data structures that the NCC instantiates neurally. And I am of course gratified that my work has played a role in their thinking. I will offer here four brief reactions to their paper.
In their conclusion, Crick and Koch say, “It cannot be overstated that Chalmers's hard problem of consciousness is unlikely to yeild to a purely logical or philosophical attack.” I can certainly endorse this. They continue: “Rather, it needs to be approached in a reductionist, scientific manner.” This must be construed with care. If a “reductionist, scientific manner” is meant to exclude every approach but neuroscience, I would disagree. I think the problem needs to be approached from every possible front. There is still plenty of room in this field for careful introspection, for philosophical analysis and thought experiments, for traditional psychological and psychophysical experiments, for computational theory and computational simulation. There is no question that neuroscience brings new and important tools to the task, and that these tools may help us reevaluate our previous thinking; but they do not oblige us to discard the old tools outright.
I imagine Crick and Koch would concur. But others have hastened to demean all techniques of cognitive science that do not make direct recordings of brain activity; and therefore I find it vital-to reemphasize the importance of methodological pluralism, with the goal of eventual convergence.
Crick and Koch allude to the aspects of consciousness I have called “affects” (1987) or “valuations” (1996, 1997). They speak of the