David Marr:

VISION (MIT Press, 1982)

(Copyright © 2015 Piero Scaruffi | Terms of use )
Marr thinks that the vision system employs innate information to decipher the ambigous signals that it perceives from the world.
Processing of perceptual data is performed by "modules", each specialized in some function, which are controlled by a central module. In a similar fashion to Chomsky and Fodor the brain contains semantic representations (in particular a grammar) which are innate and universal (of biological nature) in the form of modules that are automatically activated and all concepts can be decomposed in such semantic representations. The processing of such semantic representations is purely syntactic.
The physical signal sent to the world is received (in the form of physical energy) by transductors, which transform it into a symbol (in the form of a neural code) and pass it on to the input modules, which extract information and send it to the central module in charge of higher cognitive tasks.
Each module corresponds to neural subsystems in the brain. The central module exhibits the property of being "isotropic" (able to build hypotheses based on any other available function) and "quinian" (the degree of confirmation assigned to an hypothesis is conditioned by the entire system of beliefs).
The visual system is decomposed in a number of independent subsystems. Such subsystems provide a representation of the visual scene at three different levels of abstraction: the "primal sketch", which is a symbolic representation from the meaningful features of the image (anything causing sudden discontinuities in light intensity, such as boundaries, contours, textures); a 2 and a half dimensional sketch, which is a representation centered on the visual system of the observer (e.g., describes the surrounding surfaces and their properties, mainly distances and orientation) and computed by a set of modules specialized in parameters of motion, shape, color, etc; and finally the tri-dimensional representation, which is centered on the object and is computed by Ullman's correspondence rules.
Marr thinks that one can be at either at three levels of analysis: the computational level (which mathematical function the system must compute, i.e. an account of human competence), the algorithmic level (which algorithm must be used, i.e. an account of human performance) and the physical level (which mechanism must implement the algorithm). Cognitive science should investigate the mind at the computational level.

TM, ®, Copyright © 2015 Piero Scaruffi