Representation in Cognitive Science

by Nicholas Shea

Our thoughts are meaningful. We think about things in the outside world; how can that be so? This is one of the deepest questions in contemporary philosophy. Ever since the 'cognitive revolution', states with meaning-mental representations-have been the key explanatory construct of the cognitive sciences. But there is still no widely accepted theory of how mental representations get their meaning. Powerful new methods in cognitive neuroscience can now reveal information processing in the brain in unprecedented detail. They show how the brain performs complex calculations on neural representations.

Drawing on this cutting-edge research, Nicholas Shea uses a series of case studies from the cognitive sciences to develop a naturalistic account of the nature of mental representation. His approach is distinctive in focusing firmly on the 'subpersonal' representations that pervade so much of cognitive science. The diversity and depth of the case studies, illustrated by numerous figures, make this book unlike any previous treatment. It is important reading for philosophers of psychology and philosophers of mind, and of considerable interest to researchers throughout the cognitive sciences.

Genres: Philosophy, Psychology

306 pages, Hardcover

Published November 27, 2018

Reviews

[DD · Rating 1 out of 5]

Read this for an undergrad philosophy course on mental content and would not recommend it for that purpose. A lack of prior knowledge in the field of cognitive science (and the philosophy surrounding it) made it very difficult to get through the book. Now, this is not the fault of Shea as he didn't write it for the purpose for which I read this. Although occasionally interesting, in the end I don't think I learnt much of anything.

[Larry · Rating 2 out of 5]

Read this quickly today (i.e. really read the first chapters and flipped through the rest). Basically footnotes to Dretske.

Millikan: content is fixed by representation consumers. Shea emphasizes that partitioning brain dynamics into consumers and producers is tricky (pp. 19, 95).

Two kinds of exploitable relation: correlational information and structural correspondence. (Exploitable: for the performance of teleofunctions)

Four kinds of teleofunctions: natural selection, learning, design, contribution to the persistence of the organism.

Stabilized functions are designed (by one of the first three processes above) to produce outcomes robustly, i.e. reliably and flexibly across a range of circumstances.

[Arun]

Descartes thought mechanism can only go so far and consciousness and free will cannot be explained by it. Mechanism it seem cannot explain linguistic competence. Except now a machine has made it possible what was - "inconceivable to Descartes" [sic]. Computers with NLP/Deep Learning capabilities, google auto-complete in search/gmail that fill in our thoughts even before we could think it!

2020 Lakatos award winner. Open access - https://ora.ox.ac.uk/objects/uuid:bd8...