LEARNING BY HYPERTEXT WRITING: USING MULTI-LINEARITY TO FOSTER KNOWLEDGE TRANSFER
 
Stahl, E. and Bromme, R., University of Muenster, Germany
 
We examined hypertext 'writing' by learners and its impact on the flexible use of the acquired knowledge. Our paper focuses on the effects of considering two different reader perspectives during the 'writing' of hypertexts. This research is theoretically based on the 'knowledge transforming model' (Bereiter and Scardamalia, 1987) and the 'Cognitive Flexibility Theory (CFT)' (Jacobson & Spiro, 1995). In the experiment to be reported subjects were asked to construct two hypertext documents by linking prepared text nodes about the topic 'Internet'. In the experimental group (n = 20) two different reader perspectives were given to the hypertext authors: In a first stage, subjects were asked to construct a hypertext for readers that are especially interested in the historical development of the Internet. In a second stage they were asked to do likewise for readers especially interested in the services of the Internet. The control group (n = 20) was asked to find an optimal structure for the contents within two runs but no potential readers perspective was specified. All computer screen activities were videorecorded. Every two minutes subjects were asked about there thoughts. The resulting hypertexts were examined with regard to their structure. Learning effects were examined using five tests, measuring knowledge about the contents of single nodes, their relationships and knowledge transfer. The results confirmed the assumptions of the CFT: For example, the experimental group showed significantly more knowledge in the tests regarding relationships between nodes and knowledge transfer.