Academic journal article The Journal of Mind and Behavior

Learning How to Represent: An Associationist Account

Academic journal article The Journal of Mind and Behavior

Learning How to Represent: An Associationist Account

Article excerpt

The computational/representational metaphor for mind is still the central guiding idea in cognitive science despite many insightful and well-founded rejections of it (Chemero, 2009; Clark and Toribio, 1994; Dreyfus, 2007; Freeman and Skarda, 1990; Keijzer, 1998; VanGelder, 1995). There is good reason for its staying power: when we are at our cognitive best, we reason about our world with our concepts, unemotionally, amodally, and according to formal principles. And yet the metaphor leads to a deep puzzle for which there seems to be no plausible solution: How does this representational capacity emerge out of the neural-body substrate that makes us up? The response to this theoretical gulf has been, more or less, two-fold: on the one hand, those focusing on embodiment, environment, and on the dynamic interplay between complex systems, have become suspicious of the starting intuition, that cognition is representational/computational at all; and, on the other hand, those unwilling to let those intuitions go - and this is still the maj ority - have tried to bridge the explanatory gap by pushing the metaphor into service at the level of implementation, the brain. But if this move is mistaken, and I will argue here that it must be, then we are building a house of cards: we need a broader theoretical framework that integrates reductive internalist analyses with embodied, enactive insights. How do we become the representation-using beings we are? This is the foundational question on which we need to build.

More recently, a third way has begun to emerge, from extended mind theory. On this view, our cognitive capacities are partly constituted by the external scaffolding, "widewear" (Clark, 1997), that supports them. We develop algorithms that extend our computational capacities, long division for example, and we offload memory tasks onto our environment: a string around the finger reminds one to pick up the dry-cleaning; a to-do list keeps track of tasks; and Alzheimer's patients achieve independence with the aid of sticky notes. Language itself, Clark and others (Cimatti and Vallortigara, 2015; Deacon, 1998; Donald, 1991; Logan, 2007; Wheeler, 2004) have argued, is a cognitive niche construction, a tool that "transforms ... problem spaces in ways that (when successful) aid thinking and reasoning" (Clark, 2006a, p. 370). On this view, the logic-like structure of language is a cognitive resource that complements the pattern-completing capacities of our brains, allowing us to formulate inferences, make plans, think about tasks in a linear fashion, and so on, all cognitive activities that a human without language could not accomplish.

For reasons that will become clear, hybrid approaches which are grounded in internal representations, even minimally (c.f. Clark, 2006a, 2006b; Rowlands, 2006; Wheeler, 2005), suffer the same problems that afflict the mainstream view: only a wholly externalist account will do. We owe our capacity for representation to our skillful use of language, itself a representation tool. This hypothesis offers up many intriguing research questions,1 but before we can tackle these, we need to establish its possibility: How could we become representation users without ourselves being representors?

The Problem

Setting the Terrain

While the cognitive/non-cognitive divide might be fuzzy, there is much so-called cognitive activity, perhaps a majority of it, that does not necessitate representational description. The constraint "when we are at our cognitive best" was meant to carve off that small subset of activity that does make representational appeal. For example, a full account of the cognitive mechanisms that make it possible for me to imaginatively plan my trip for next week will have to explain how I represent to myself possible future scenarios, how, on this imaginative basis, I make a plan to take a certain set of actions, and finally, how, when the event finally occurs, I act on these plans. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.