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 target article Crick and Koch explore the premise that a neural correlate of consciousness can be discovered by finding neural activity related to high level (Marr's 21/2D or 3D sketch) representations of stimuli. While this is certainly important information, it is not self-evident that this approach provides the necessary leverage on the question. Neurons can respond to complex stimulus properties and arrangements and still have nothing at all to do (directly) with awareness of that stimulus. Evidence for this is the fact that neurons responding to faces are still active and selective under anesthesia (e.g., Gross, Rocha-Miranda, and Bender, 1972). The logical link that a sufficiently high level neural representation correlates with consciousness-or more precisely visual awareness-depends on the premise that it is that level of representation of which we seem to be aware. This may be true and skepticism should not prohibit further investigation along these lines. But I would like to review briefly a more direct path to neural correlates of consciousness, of which Crick and Koch are certainly already aware (e.g., Crick and Koch, 1998).
Determining how the activity of neurons relates to behavior and inferred cognitive states requires an experimental strategy of dissociations. For example, separating in time the presentation of a stimulus from the time of a motor response allows a neurophysiologist to distinguish neural processes related to sensory processing from neural processes related to response production. To investigate neural correlates of consciousness—or more particularly visual awareness—one needs to dissociate the presentation or appearance of a stimulus from awareness of that stimulus. In other words, the neural correlates of visual