I want to work on defining some of the basic concepts, like rule vs. association, and symbolic vs. connectionist. This is partly based on the need to explain them to the general public, and also because I think the terms are often used in confused and confusing ways in the scholarly literature. I’ll also need to present the basics of brain anatomy – Levitin’s This is your brain on music does a good job of this as I remember (January 10th 2016).
The roots of the term “symbolic” are clear in Nilsson’s history of AI. I’ve always found it confusing because “non-symbolic” approaches also use symbols. Perhaps the distinction is that they don’t directly process them. Algebraic may be a better term. And is the alternative paradigm better called statistical than connectionist? (Jan. 11, 2016) (Update May 10, 2016: see Smolensky’s 1988 Proper Treatment of Connectionism on symbolic and subsymbolic). (Update June 9, 2016: see also Touretzky and Pomerlau on the delineation of symbolic approaches).
I think associative is better than statistical. Associative vs. algebraic seems to me now like the best way of presenting the two general models of mind to the general public. These are probably already the terms Marcus uses in the Algebraic Mind, which I need to go back to. I see Christiansen has a review – should read that too (Jan. 13, 2016)
In our recent paper in Cognitive Science (Moreton, Pater and Pertsova 2015), we use the terms “cue-based” and “rule-based”. This paper has lots of references to the use of these two types of model in concept learning and language learning.
From the MIT Press blurb for Gallistel (1990) (thanks to Misha Becker):
How do animals represent space, time, number and rate? From insects to humans, Charles Gallistel explores the sophisticated computations performed in these ubiquitous yet neglected domains of animal learning. He proposes new and imaginative hypotheses about brain and mental processes and provides original insights about animal behavior using a computational-representational framework that is an exciting alternative to traditional associative theories of learning.