The interpretation of Piaget's position about intelligence [i.e. intelligence is entirely embodied in sensori-motor action schemes] is plainly false. Piaget's constructivism excludes the reduction of intelligence to any one structure, just because there is no final structure, no structure d'ensemble - nor could there be one structure of everything. [Piaget, 1975/1985. Equilibration of cognitive structures. Univ. of Chicago Press, p. 142. For commentary, see my book Necessary Konowledge, Erlbaum UK, 1993, p.168ff]

One implication of this view that¡¡"living structures" are always open with the possibility of serial but partial closure is that at least one/some mechanism is internal to the functioning of any living structure [this excludes that mechanism being biology alone. The working name for that mechanism given by Piaget was equilibration which - from his first book in 1918 - is a mechanism constgututed by normative, and not merely causal, properties. By the way, and contrad what you say in your email: this means that "Piaget's best work" wasn't his observation of pre-verbal children.

I don't see how a Piaget-style constructivist model can be instructively clarified by reference to AI models unless these AI models too are internally and normatively characterised as to their own functioning directed on their own improvement

--------------
(This article is from email discussions through owner-piaget-list@interchange.ubc.ca)