ERIC Number: ED355033
Record Type: Non-Journal
Publication Date: 1992
Acquisition of Reasoning: A Computational Model of Strategy Development in Transitive Inference.
Halford, Graeme S.; And Others
This paper describes a computer-simulation model of the way in which basic reasoning processes develop in children. The model, based on PRISM-II programming language, was designed to reflect the manner in which world knowledge can be used to construct strategies for reasoning. The model learns strategies for performing transitive inference by using knowledge that is acquired through play experiences with sets of objects. Five experiments testing predictions from the model are reported. Experiment 1 tested children's ability to recognize ordered sets, an ability required for the development of transitive inference strategies. Experiments 2 and 3 tested the prediction that children would recognize when the order of a set of objects is indeterminate and found that the ability to make this recognition was present from the age of 7 years. Experiments 4 and 5 assessed strategies children used in a varied set of transitive inference problems, and found no qualitative differences in strategies as a function of age. Six appendices making up almost half the document include the set of productions existing before the model begins operation and lists of productions produced during various problem forms. (Contains 75 references.) (MM)
Publication Type: Reports - Research
Education Level: N/A
Authoring Institution: Queensland Univ., Brisbane (Australia). Centre for Human Information Processing and Problem Solving.