Ruey-Yun Horng
Department of Industrial Engineering and Management National Chiao Tung University
Ting-Yu Wu
Department of Industrial Engineering and Management Ming-Hsin Institute of Technology
Collaboration can improve the rule discovery performance. Self-explanation or asking the collaborator for explanations can also increase the number of rules discovered (Wu, Wu, & Horng, 1998). The purpose of the present study was to investigate how collaboration and explanation work to affect one’s reasoning during rule discovery. The data were written protocals produced by 114 college students during hypothesis generation and hypothesis testing in Wu, Wu, & Horng’s study. Results showed that working in dyad may increase the likelihood of using a strategy closer to the ideal hypothesis testing strategy proposed by Bruner, Goodnaw and Austin (1956). However, there was no sign of transfer in the subsequent testing session when participants were working alone. Nevertheless, the collaborative experience still had an effect on the number of rules correctly discovered. The effect of collaboration is very likely due to knowledge sharing between the collaborators. The instruction to engage in explanatory activity may increase the number of rules correctly discovered as well. Moreover, with practice, explanation appears to enable one to extract more features from the observed instances and come up with various approaches to modify a hypothesis when confronted with negative testing results. Thus, explanation appears to implicitly shape one’s hypothesis-testing strategy.
Keywords： scientific reasoning, rule discovery, confirmation bias,