Adaptive learning and metacognitive regulation support for iii-structured problem solving processes in web-based learning

Abstract

This research was set up to investigate the effect of adaptive user interface and self regulation learning supports, in a web-based non-collaborative learning system, on the learners' cognition, metacognitive awareness and application of problem solving processes. The research was designed as a mixed study with two quantitative experimental studies and a qualitative cross-case comparative study. Study 1 investigated the effect of adaptive user interface supports on the learner's cognitive performance, measured as test scores, and perceived value of the adaptive features of the learning environment. Study 2 investigated the effect of self-regulation scaffolds on the learner's level of metacognitive awareness and level of application of problem solving processes. The qualitative study further investigated the effects of self regulation scaffolds by using the think aloud method. Results from Study 1 indicated no significant difference in mean test scores between learners using and those not using adaptive features. Regarding perceived value of the adaptive features, both groups "starting with" and "starting without" adaptive features perceived the features as useful. The results show that the order of sequencing course content as well as the strategy of presenting adaptive user interface supports has an impact on learner performance and experience. From Study 2, learners who used the scaffolds had a significantly higher metacognitive awareness and application of problem solving processes, compared to the control group. However, whereas the scaffolds were effective for knowledge of regulation of cognition, they had minimal effect on knowledge of cognition. Hence, there is need for a different set of scaffolds for knowledge of cognition. Study 2 also showed that the effect of scaffolds on the learners' cognitive processes is better measured using the self-report method followed by solving of a task. Keywords: Scaffolding, cognitive load, metacognition, cognition, ill-structured problem solving, adaptive web-based learning, Machine Learning algorithms, instructional design