Structural Analysis

The study of the human mind embodied in cognitive science strives to understand the nature of the abstract representations and computational processes responsible for our ability to reason, speak, perceive and interact with the world, etc. In addition, a commitment to a materialist resolution of the mind-body problem requires that we search for the manner in which these representations and processes are neurally instantiated in the brain. Given this dual aim, one might proceed in one of two ways: (1) bottom-upwards, commencing with the study of how low-level information and computations are encoded in the neuroanatomy of the brain, in the hope of working upwards toward an understanding of the properties of higher level cognitive processes; (2) from the top-downwards, using behavioral data from a variety of sources to provide an abstract characterization of cognitive processes, and then utilizing these results to guide our search for the neural mechanisms of cognition. At present, our understanding of the neural basis of higher level cognition is virtually negligible. Thus, it is probably fair to say that a contemporary study of any of the domains discussed in section II of this volume may only be carried out in a top-down fashion. It is sometimes claimed that top-down research into cognition is incapable of producing anything but \castles in the sky" having little relevance to human thought. However, a brief look at the history of science shows us that the study of a phenomenon in the absence of an understanding of its underlying mechanisms is by no means novel and has moreover produced signiicant success. One prominent example is Gregor Mendel's study of heredity. By observing the external patterns in which properties of pea plants such as color and height were transmitted from one generation to the next, Mendel was able to deduce the existence of genes and the fundamental laws by which they combine. Of course, Mendel had no conception of the biological character of genes, nor any understanding of the reasons that his combinatorial laws held. Yet, his results have been substantially vindicated and formed the impetus for research into the biological basis of genetic material, culminating in Watson and Crick's discovery of DNA. How might we mimic Mendel's methodology (and success) in the study of the mind? That is, how do we go about building an abstract theory of a cognitive process which can form the basis for subsequent neurological investigations? …