Paraphrasing and prediction with self-explanation as generative strategies for learning science principles in a simulation

This study examined the incorporation of generative strategies for the guided discovery of physics principles in a simulation. Participants who either paraphrased or predicted and self-explained guided discovery assignments exhibited improved performance on an achievement test as compared to a control group. Calibration accuracy (the correspondence between judgments of performance and actual performance) was also improved for the two generative strategy groups. The thoroughness of generative content and quality of self-explanations significantly predicted test performance. In regards to cognitive load, participants who predicted and self-explained reported significantly higher levels of mental effort, decreased levels of confidence, and higher levels of frustration compared to those in other treatments. The improvement in learning by the two generative strategy groups is consistent with the generative model of learning describing the importance of knowledge construction.