Deflating Representation
: Computation, Structure, and Content

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

The present work focuses on the notions of representation and computation, and the explanatory role they play in the cognitive sciences. I put forward a deflationary view of representational content, and argue that explanatory internal states in the cognitive sciences are primarily individuated by their computational structure, rather than by content. In Part I, I survey the mainstream accounts of representation and content present in the philosophical literature: functional role semantics, informational semantics, teleosemantics, and structural representation. I also briey examine some of the crucial issues that any satisfactory theory of content has to tackle, with special attention to the problem of indeterminacy of content.

I present and develop, in Part II, a version of the mechanistic view of concrete computation able to account for how cognitive systems compute, and for how to individuate their computational structures. The account avoids pancomputationalism and triviality of computation, yielding a robust, objective theory of computation in physical systems.

With this mechanistic view of concrete computation in hand, in Part III I present my deflationary approach to representation, which shifts much of the explanatory burden in making sense of cognition onto computational structures. I examine interpretational semantics as a promising precursor of my view. I propose several modifications to interpretational semantics, producing a theory close to structural representation, but with marked deflationary leanings. On that basis, two deflationary paths are examined: content pragmatism, and mild realism about content. I provide reasons to prefer the latter approach, though I take both paths to be promising. The resulting deflated notion of representation, wedded to a solid notion of computational structure, is advantageous insofar as it dissolves metaphysical puzzles related to content-fixation and indeterminacy, while preserving a notion of representation robust enough to play an important explanatory role in the contemporary study of cognition.
Date of Award2017
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorNicholas Shea (Supervisor), David Papineau (Supervisor) & Matteo Mameli (Supervisor)

Cite this

'