Connectionism Need Not Be (Strongly) Associationistic

In a previous post, I briefly discussed the relationship between connectionism and associationism.  Thanks in part to the helpful feedback I received, I have now revised the relevant section of the paper I am working on.  I’d be interested in any additional comments or references that anyone might have.  The text of the paper reads as follows:

The relationship between neurocomputational approaches and associationism is more complex than many suppose.  We should distinguish between strong and weak associationism.  Strong associationism maintains that association is the only legitimate explanatory construct in a theory of cognition [cf. Fodor 1983, p. 27].  Weak associationism maintains that association is a legitimate explanatory construct along with others such as the innate structure of neural systems.  (We should also note that there are different associative mechanisms, some of which are more powerful than others; comparing associative mechanisms goes beyond the scope of this paper.)

To be sure, some connectionists profess strong associationism [e.g., Rosenblatt 1958, p. 387].  But that is beside the point, because connectionism per se is consistent with weak associationism or even the complete rejection of associationism.  Some connectionist models do not rely on association at all—a prominent example being the work of McCulloch and Pitts [1943].  And weak associationism is consistent with many theories of cognition including classicism.  A vivid illustration is Alan Turing’s early proposal to train associative neural networks to acquire the architectural structure of universal Turing machines [Turing 1948, Copeland and Proudfood 1996].  In Turing’s proposal, association may explain how a network acquires the capacity for universal computation (or an approximation thereof), while the capacity for universal computation may explain any number of other cognitive phenomena.

Although many of today’s connectionists and computational neuroscientists emphasize the explanatory role of association, many of them combine association with other explanatory constructs (weak associationism) [cf. Smolensky and Legendre 2006, p. 479; Trehub 1991, pp. 243-5; Marcus 2001, pp. xii, 30].  What remain to be determined are which neural networks, organized in what way, actually explain cognition and which role association and other explanatory constructs should play in a theory of cognition.


  1. John Jones

    I would like to make a philosophical point.

    I take it that the term “explanation” in the context of associationism and connectionism is being used in a technical sense in place of the ordinary language term “description”.

    If associationism and connectionism are descriptive propaedeutics then all well and good. In respect of that, there must be, we presume, a growing need for new mythologies that can bring together the findings taken from many different physical studies of the brain in one, easy to grasp, model. Such mythologies are emergent forms, but their propaedeutic focus is a set of texts, not the brain or cognition.

    As propaedeutic descriptive models, associationism and connectionism don’t bring “explanation” or theory. The models conceptually simplify the wealth of complex physical patterns and quasi-physical agencies that now populate brain-science scholasticism.

    If we want explanations for cognition we would be looking for a causal reason that would explain the relationship of cognitive elements – matter and cognition. This would involve unpacking those terms which meld these incommensurable elements. Terms such as “state”, “process”, “activity” etc. assume causation and the commensurability of the elements of cognition.

    No explanation will be forthcoming. Activity, process, operation, activity/inactivity, emergence, etc., are not identified materially. There is nothing about a physical brain that tells us that it is involved in cognition. Rather, it is cognition that identifies a brain, a brain’s “states”, and other arbitrary structures, right down to the dendrite.

    My point here, in summary, is that models of the brain such as associationism and connectionism are text-simplifiers that serve to simplify the ever-increasing complex pattern of scholastic texts, cross-references, movements, technical language and conceptual mappings that are the brain sciences.

    Meanwhile, the nitty-gritty – the explanation of cognition – goes untouched. In its place we have the repatterning of its conceptual, quasi-material hybrids that go proxy for it in current brain-scholasticim.

  2. Perhaps this is relevant to your paper. Two important studies confirming that the representation of space is an innate property of the mammalian brain were just published in Science. These direct neuro-physiological probes of rat brain lend strong empirical support to the basic claim of the retinoid model which, as such, is a connectionist but not an associative mechanism. Here is an excerpt from the commentary on this recent work:

    Science 18 June 2010:
Vol. 328. no. 5985, pp. 1487 – 1488



    A Kantian View of Space

    Linda Palmer and Gary Lynch

    Department of Anatomy and Neurobiology, University of California, Irvine, CA 92697, USA.

    “How does the brain represent space? Is this representation entirely the result of learning from experience? In his Critique of Pure Reason, Immanuel Kant argued that there must be certain “a priori conditions” of cognition, which could not be derived from experience but must instead be given prior to it. His theory includes two “a priori pure forms” of space and of time, regarded as constraints of thought rather than results of investigation or experience (1, 2). On pages 1576 and 1573 of this issue, Langston et al. (3) and Wills et al. (4) both refer to Kant’s theory and report that critical components of the brain’s spatial representation systems are already in place when an animal first encounters an extended environment. This supports the view that spatial representation indeed includes an innate component prior to experience.”

  3. John Jones

    If our experience informs us of the properties of space, then the assertion that the brain’s representation of space can inform our experience of it cannot be made. In which case an inquiry into the way that “the brain represents space” will be fruitless.

    The Kantian link is this. Kant asks how strong is the assumption that space really is spatial. Kant argues that this assumption can only be made on the basis of our experience. The brain scientist, and scientists generally, will argue that subject to the limitations of our senses, our experience tells us that space, in itself, really is spatial. Kant argues that how a thing is in itself simply isn’t communicable, and that our senses have nothing to contribute toward spatiality per se.

    Just to expand on the Kantian perspective …against Kant, the question “how does the brain represent space” assumes an independent spatiality, where space is an object that has the properties of the its appaearance. Such an object, or perspective, Kant calls “transcendentally real”. An object, such as space, is transcendentally real when the conditions for its existence are found in the object “itself”. That is, for the anti-Kantian transcendental realists space is independently real and is pretty much how we perceive it.

    The brain sciences also are a species of transcendental realism. The brain itself is supposed to have the properties we associate with it. These include perceptions, functions, processes, states, and representations of objects. For the transcendental realist, space is one such object. If the brain has these properties itself then, we assume, we can look for them by studying the physical form of the brain. We can look for spatiality or its representation in the brain. But this falls on Kant’s objection of the incommunicability of transcendentally real objects, and it also falls on the epistemic circularity of assuming space before drawing conclusions as to its representation.

  4. I’ve found this web site useful for my every day life. It is very entertaining and intriguing. Anyway, did anyone view the Australian Open women final currently? That was rather a fantastic match. Maria Sharapova got her butt knocked by a girl out of Belarus who’ll be the world number 1 soon after the win. That’s something. Tomorrow is destined to be a good day for tennis. So whoever’s interested by sure to give it a look free essays collection.

Comments are closed.

Back to Top