Paul Smolensky: Is Perception or Logical Inference Primary?

 

What's at the heart of intelligence or cognition?

According to Paul Smolensky, there are two main rivals: perception and logical inference.

One can immediately ask if there can be such a simple categorisation of something as broad as intelligence or cognition. Indeed at a prima facie level one can say that perception doesn't seem to be cognitive at all. It may be a basis for cognition; though is it actually cognition itself?

Logical inference, on the other hand, is clearly cognitive in nature. Though here again logical inference may utilise perception; just as perception itself can be cognitively enriched.

Perception

Perception is deemed by Smolensky to be “subsymbolic”. Thus if it's subsymbolic mustn't it also be sub-logical or even sub-cognitive? Though that would be, perhaps, to beg the question. Why assume that all cognition, or at least intelligence, must somehow be symbolic in nature? Then again, x's being sub-symbolic isn't the same as it being non-symbolic... or perhaps it is.

In any case, that which is deemed to be sub-symbolic (by Smolensky) is still about the “categorisation of other perceptual processes”. So here is can be seen that perception isn't viewed as being basic. It isn't been said here (as stated earlier) that perception is - in and of itself - a question of logical inferences or cognition. What we have here is the “categorisation” of “perceptual processes”. So there's categorisation which seems to be above about beyond perceptions themselves.

Perception & Evolution

There are various things which work to the advantage of seeing things primarily in terms of perception rather than in terms of logical inference.

One, that logical inference/reasoning came after the “categorisation of perceptual processes” in our evolutionary history. Or as Smolensky puts it:

“An evolutionary argument says that the hard side of the cognitive paradox evolved later, on top of the soft side...”

That must mean that there were cognitive processes which predated the higher processes of logical inference/reasoning and indeed of language-use. Indeed surely this must have been the case. Homo sapiens (or the species which grew into homo sapiens) surely couldn't have been logical reasoners from the very beginning. Homo sapiens (or their forbears) couldn't have started off as language-users or logical reasoners. Indeed all the evidence says that this wasn't - and couldn't have been - the case.

Basically, both language and logical inference must have been built upon such things as the categorisation of perceptual processes (as well as upon much else). Language and logical inference didn't occur ex nihilo.

Smolensky's further point arises from all that's just been said.

If this evolutionary account is correct, then it's not a surprise that

“it is much easier to see how the kind of soft systems that connectionist models represent could be implemented in the nervous system”.

After all, isn't it the case that our nervous system today is basically as it was before we acquired language and the skill of logical reasoning? Thus even though cognition and mentality has changed, our biological hardware hasn't. Thus if our biological hardware predates symbolic processing, then perhaps our models should also do so. Symbolisation and computation may be a part of our cognition; though the biological nervous system that subserves all this was designed (in the evolutionary sense!) for other things.

Connectoplasm & Symbols

Some people believe that connectionists reject mental symbols and everything that goes with them. And, by virtue of that, they also believe that connectionists reject computation – at least as the primary basis of cognition.

In Smolensky's case, that isn't the case. Indeed he talks about “building symbols” out of “connectoplasm”. In his view, symbols arising from connectoplasm is a better idea than symbols arising from... well, he doesn't really say. From the Language of Thought or something similarly symbol-based?

In any event, Smolensky writes:

“With any luck we will even have an explanation how the brain builds symbolic computation. But even if we do not get that directly, it will be the first theory of how to get symbols out of anything that remotely resembles the brain...”

It's clear here that Smolensky is creating a theory (or model) that's biologically feasible; unlike many of the alternatives. Of course it will need to be said exactly how and why it's biologically feasible. Or, in Smolensky's words, we'd need to know “how the brain builds symbolic computation”. In fact we'd need to know exactly what he means by the words “the brain builds symbolic computation”.

In any case, the point to stress here is that Smolensky does believe that the brain builds symbols. So, at the very least, symbols are part of Smolensky's connectionism.

Despite that, elsewhere in the same paper Smolensky de-stresses the importance of symbols. Basically he wants “formal accounts” of “continuous mathematics” rather than the “discrete mathematics” of much “traditional symbolic formalism”. In more detail, Smolensky writes:

“... my characterisation of the goal of connectionist modelling is to develop formal models of cognitive processes that are based on the mathematics of dynamical systems continuously evolving in time: complex systems of numerical variables governed by differential equations.”

There's no mention here of symbols or even of quasi-symbols. In fact this account sounds both mechanical and biological in nature; strange as that may seem. Though why shouldn't the biological also be seen as mechanical or at the least as dynamical? And if we're talking of the mechanical or the dynamical, then it stands to reason that mathematics, “numerical variables” and “differential equations” - rather than symbols – will be primary. Indeed it seems that in even simpler terms, this is more about the measurement of dynamical systems rather than the symbols within a symbol-system (i.e. the mind-brain).

It may appear to the case that causation is also of prime importance. That is, we have the numerical measurements of the interplay between the environment (in terms of input) and a dynamical system which will result in certain internal states and then certain kinds of output.

Reference

Smolensky, Paul, 'The Constituent Structure of Connectionist Mental States: A Reply to Fodor and Pylyshyn' (1988)