OPTIMIZATION AND SIMPLICITY - COMPUTATIONAL VISION AND BIOLOGICAL EXPLANATION

David Marr's theory of vision has been a rich source of inspiration, f ascination and confusion. I will suggest that some of this confusion c an be traced to discrepancies between the way Marr developed his theor y in practice and the way he suggested such a theory ought to be devel oped in his explicit metatheoretical remarks. I will address claims th at Marr's theory may be seen as an ''optimizing'' theory, along with t he attendant suggestion that optimizing assumptions may be inappropria te for cognitive mechanisms just as anti-adaptationists have argued th ey are inappropriate for other physiological mechanisms. I will discus s the nature of optimizing assumptions and theories. Considering vario us difficulties in identifying and assessing optimizing assumptions, I will suggest that Marr's theory is not purely an optimizing theory an d that reaction to Marr on this issue prompts interesting consideratio ns for the development of inter-disciplinary constraints in the cognit ive and brain sciences.