we cannot map out our internal representations since apprehension of the structure of our own minds requires a semantic step i.e some sentential convention. concession of the state-of-affairs is outcome dependent, -- the fact that we have the capacity to think of other conscious beings [which would entail that there is some quality of other minds which lends credence to other-minds -- a fact which Nagel referred to when speaking of referential pathways. i cannot conceive of what it would be like to have 20/1 vision, though i can logically apprehend the notion, and indeed i could try to picture it in my mind. this requires a conception of ones own thought-process in procuring the thought. but i cannot have certainty that that internal representation is in any way accurate to what it would actually be like. like trying to coordinate each individual contraction of my muscles; it's not possible, though i can operate under the delusion that it is.
the same goes, generally, with consciousness, or more generally the structure of consciousness -- how the mind perceives itself. the fact that we are, however, conscious beings means that the very notion becomes a utile theoretical concept; what Sellars called the 'manifest image' as opposed to the 'scientific image.' i think the mechanistic notion derives from a logical error. like a drunkard pretending to be sober, neglecting the slurring of his words.