Representations and two systems hypotheses
Thanks for opening up the blog! Let me introduce myself first. I am a lecturer in philosophy at Linköping university, sharing my time almost fifty-fifty between philosophy and cogsci. Apart from the phil of mind, my interests are mainly in philosophy of language and logic, plus some epistemology. A typical Lemming.
I am currently interested in the notion of mental representations, and especially in the context of two slightly different debates in cognitive science/philosophy of mind. The first is the two visual systems view of visual cognition, the second is the debate about extended cognition. These two areas show that the more usual notion of mental representation have to be revised in some manner, but it’s not really clear how. What I mean is the following.
When developing the two visual systems hypothesis, Milner and Goodale came to postulate two systems for storing and handling visual information, with different neural pathways. We have vision for perception and vision for action. In most normal cases, these systems work pretty well together (if not, such disparities would have been fixed by evolution a long time ago). But they can come apart at times, such as in the Ebbinghaus illusion, where subjects often judge that the circles are of unequal size, yet when asked to grasp the objects, size their grasping motion correctly. The perceptual judgement of inequal sizes doesn’t carry over into the grasping motion. One way (not necessarily the best) to describe what is going on is that the subject harbours two representations of an object in tandem, and that one or the other comes to be utilized, depending somehwat on circumstances (is the subject judging the relative sizes of two objects, or is the subject grasping an object, for instance).
Another kind of case, where we have a similar phenomenon, is when looking at cases of decision-making, where System 1 and System 2 (Kahneman and others) give different verdicts. Our gut thinking tells us this, the reflective processes tell us that. This might again be described by saying that the deciding subject has two representations in tandem, and that one or the other comes to be utilized. Again, this is not necessarily the best characterization of what is going on.
One problem with the explanations that appeal to two different representations, running in tandem, is that this appears to erode the usefulness of appeals to representations. They now come out as too ex post facto to do much useful work. It seems that we in cases of the above kind just find ourselves saying that the subject must have had a certain kind of representation, since the subject did this or that, and that we will always be in a position to say that there was a certain representation of a certain kind. But this is surely much too easy. So I would be interested in knowing if there have been discussions of this problem in the literature.
There is the recent anthology:
Perception, action, and consciousness : sensorimotor dynamics and two visual systems / edited by Nivedita Gangopadhyay, Michael Madary, Finn Spicer. Oxford : Oxford University Press, 2010.
which contains some useful papers, but apart from that I haven’t seen very much.