2. THE MECHANISMS OF COLLABORATIVE LEARNING

2.2 The alternative proposal

The second mechanism is close to the former. It is related to what psychologists refer to as the 'confirmation biases': subjects tend to design only experiments that confirm their hypotheses and to disregard any empirical finding that contradicts their hypotheses [4, 5]. The fact that the confirmation bias is reduced in collaborative learning can be explained by the 'conflict' mechanism: if the learners disagree, there are fewer chances that they design an experiment or a data analysis which satisfies one hypothesis to the detriment of the other. A complementary explanation is that the subjects are reluctant to abandon their own hypothesis because they do not have another hypothesis for replacement. The positive effect would then result from the fact that peers may propose an alternative hypothesis.

ICCAI 95 article - 08 FEB 95

[Next] [Previous] [Up] [Top]

Generated with CERN WebMaker