ABSTRACT
Recent research on the acquisition of knowledge has suggested that texts should not contain any examples because they may distract the reader from deeply processing the text (LeFevre & Dixon, 1986). On the other hand, it is a generally held belief that good learning materials should consist of general descriptions in the form of text as well as concrete examples. This chapter first discusses the research on learning from text and examples and presents an integrated computational model of such learning. It then reports two experiments in which the information given in the text stood in a more natural relation to the examples than in the experiments of LeFevre and Dixon. The experimental results are discussed in terms of the extended construction-integration model that describes knowledge acquisition from text and examples in a unified way.
