Juicios sobre el conocimiento propio: Buscan- do factores que influyen en su validez

Introduction. Research in the field of metacomprehension often uses the accuracy of judgements of own knowledge as a measure of monitoring which is a central component of metacognition. One aim of this study is to investigate if the accuracy above chance usually found in studies using traditional texts can be replicated with hypermedia. More importantly, the study investigates differences between the accuracy of monitoring in two kinds of tests: a comprehension test and a transfer test. Method. To investigate these questions, 126 university students learnt the basic concepts of operant conditioning presented in a hypermedia system within 30 minutes. Immediately afterwards, a comprehension test and a transfer test were administered. For each test, the students predicted their knowledge just before it was administered and retrospectively judged their confidence regarding each item. Results. The results showed that the students' accuracy of monitoring was above chance. Further, accuracy of retrospective judgements was higher in the comprehension test than in the transfer test, and the predictive judgement was less accurate than retrospective judgements in the comprehension test, but not in the transfer test. Structural equation modelling rendered no evidence that this difference in accuracy between the two tests was due to differences in the use of experience-based cues. Discussion and Conclusion. The results are discussed with regard to the distinction between theory-based and experience-based cues and with regard to the accessibility hypothesis.