What the Mind Does, Not How it Does it
I am tempted to categorize this book under Brain Models or possibly
Psychology. In the Preface and Introduction, Boden makes it clear enough
that the intent is to present models of use to Computational Psychology.
However, as she also points out, the gist is on What the Mind Does,
not How. As a result, the descriptions presented have all the
content of current AI material. This is opposed to most of the other books
here which I have categorized as Brain Models. Those are theories of How
the Brain Works, not What it Does. And Psychology books have descriptions
of how people (and animals) behave. So this is AI (but not without a
significant psychological element).
Do You Need Concepts to Understand a Sentence?
In Chapter 4, Boden presents pro and con agruments
for the autonomy of syntax. To what extent is this psychologically
realistic? Jackendoff argues for psychological reality
presentation of linguistic models, and yet, he denies any need for semantic
information in sentence processing. Specifically, just as in Chomsky's models,
Jackendoff's phonological module presents a fully specified PF to the syntax