Abstract
According to Ramsey (Representation reconsidered, Cambridge University Press, New York, 2007), only classical cognitive science, with the related notions of input–output and structural representations, meets the job description challenge (the challenge to show that a certain structure or process serves a representational role at the subpersonal level). By contrast, connectionism and other nonclassical models, insofar as they exploit receptor and tacit notions of representation, are not genuinely representational. As a result, Ramsey submits, cognitive science is taking a U-turn from representationalism back to behaviourism, thus presupposing that (1) the emergence of cognitivism capitalized on the concept of representation, and that (2) the materialization of nonclassical cognitive science involves a return to some form of pre-cognitivist behaviourism. We argue against both (1) and (2), by questioning Ramsey’s divide between classical and representational, versus nonclassical and nonrepresentational, cognitive models. For, firstly, connectionist and other nonclassical accounts have the resources to exploit the notion of a structural isomorphism, like classical accounts (the beefing-up strategy); and, secondly, insofar as input–output and structural representations refer to a cognitive agent, classical explanations fail to meet the job description challenge (the deflationary strategy). Both strategies work independently of each other: if the deflationary strategy succeeds, contra (1), cognitivism has failed to capitalize on the relevant concept of representation; if the beefing-up strategy is sound, contra (2), the return to a pre-cognitivist era cancels out.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Unless otherwise stated, page numbers refer to Ramsey’s Representation Reconsidered.
Incidentally, Ramsey himself acknowledges that Ryder’s notion of cortical representation is model-based (80). He also mentions Grush’s (2004) emulation theory of mental representation as a second nomic deserter. However, as Ramsey mulls over the implications of a non representational psychology, he observes that these model-based nonclassical theories are the exception rather than the rule (223). Another model that would serve equally to illustrate the beefing-up strategy is O’Brien and Opie’s (2004) structuralist theory of mental representation. Their model is cashed out in terms of second-order resemblances; a category that insofar as it comprehends all forms of structural isomorphism would receive Ramsey’s beneplacit. Fortunately, the degree of popularity of a theory sheds little light on the alleged representational status of the explanations it offers, so we need not consider further how exceptional model-based connectionist theories happen to be.
See Calvo Garzón (2003a), and the references therein.
It must be said that the borderline between connectionist and dynamicist models of cognition cannot be drawn easily (Spencer and Thelen 2003). Ryder’s SINBAD model is a dynamic system after all. As Ryder (2006) notes elsewhere: “The type of models the cortex is designed to build are dynamic models. The elements of a static model and the isomorphic structure it represents are constants… By contrast, in a dynamic model the elements in the isomorphic structures are variables. Rather than mirroring spatial structure, a dynamic model mirrors covariational structure” (pp. 125–126). However the divide is drawn between connectionist and dynamicist networks, it’s clear that the beefing-up strategy works in the latter case, if it does in the former class of models.
Other options include the well-known microfeatural descriptions of Churchland (1989), and the more recent clustering approach of Shea (2007). Churchland’s microfeatural rendering of the vehicles of content is fleshed out in terms of hidden patterns of activation, and therefore takes us back to the receptor notion of representation (for a criticism of Churchland’s connectionist semantics, see Calvo Garzón 2003b). On the other hand, although Shea’s clustering approach escapes the receptor notion, and may be a more promising candidate in that sense, it is beyond the scope of this paper to analyze it in detail.
A word of caution is needed here, regarding the dialectics of this paragraph, for Ramsey explicitly considers the dangers of panrepresentationalism to motivate his claim that nonclassical explanations in terms of receptors and dispositions are not genuinely representational. However, his rejection of fictionalism is not meant to directly support the claim that classical explanations in terms of IO- or S-representations are genuinely representational, but only to respond to a possible objection (98ff). Thanks to Bill Ramsey for bringing this point to our attention.
References
Calvo Garzón, F. (2003a). Non-classical connectionism should enter the decathlon. Behavioral and Brain Sciences, 26, 603–604.
Calvo Garzón, F. (2003b). Connectionist semantics and the collateral information challenge. Mind and Language, 18, 77–94.
Churchland, P. M. (1989). A neurocomputational perspective: The nature of mind and the structure of science. Cambridge, MA: MIT Press.
Grush, R. (2004). The emulation theory of representation: Motor control, imagery, and perception. Behavioral and Brain Sciences, 27, 377–442.
Hinton, G. (1986). Learning distributed representations of concepts. In Proceedings of the 8th Annual Conference of the Cognitive Science Society. Hillsdale, NJ: Erlbaum.
O’Brien, G., & Opie, J. (2004). Notes towards a structuralist theory of mental representation. In H. Clapin, P. Staines, & P. Slezak (Eds.), Representation in mind: New approaches to mental representation (pp. 1–20). Amsterdam: Elsevier.
O’Brien, G., & Opie, J. (2006). How do connectionist networks compute? Cognitive Processing, 7(1), 30–41.
Ramsey, W. (2007). Representation reconsidered. New York: Cambridge University Press.
Rolls, E. T., & Treves, A. (1998). Neural networks and brain function. Oxford: Oxford University Press.
Ryder, D. (2004). SINBAD neurosemantics: A theory of mental representation. Mind and Language, 19(2), 211–240.
Ryder, D. (2006). On thinking of kinds: A neuroscientific perspective. In G. Macdonald & D. Papineau (Eds.), Teleosemantics (pp. 115–145). Oxford: Oxford University Press.
Shea, N. (2007). Content and its vehicles in connectionist systems. Mind and Language, 22, 246–269.
Spencer, J. P., & Thelen, E. (Eds.). (2003). Connectionist and dynamic systems approaches to development [Special issue]. Developmental Science, 4(4).
Wittgenstein, L. (2001). Philosophical investigations (G. E. M. Anscombe, Trans.). Oxford: Blackwell (Third edition; first edition 1953).
Acknowledgments
We are grateful to Bill Ramsey for helpful comments and suggestions on a previous version of this paper. Preparation of the manuscript was supported by DGICYT Project HUM2006-11603-C02-01 (Spanish Ministry of Science and Education and Feder Funds).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Garzón, F.C., Rodríguez, Á.G. Where is Cognitive Science Heading?. Minds & Machines 19, 301–318 (2009). https://doi.org/10.1007/s11023-009-9157-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11023-009-9157-3