No abstract

Students' flawed conceptions can often be corrected during learning by successive contradictions from text sentences, allowing them to revise or replace their individual false beliefs as they read a text. Although this process supports learning new knowledge at a local level (i.e., individual propositions) it is less effective in facilitating systemic conceptual change.In contrast, constructive processes such as self-explaining can facilitate conceptual change through building a mental model from cumulative revisions of individual false beliefs. In the current experiment, I investigated whether comparing and contrasting examples can achieve the same outcome. Students (n=22) in the compare group were first shown a diagram of with their own flawed mental model, and then asked to compare it with a diagram of the correct model. This condition was compared with self-explaining the correct diagram (n=22), and a control condition in which students simply read the text twice (n=20). Results showed that the compare group performed better than the other two groups on questions requiring deep inference making.iv

Students' flawed conceptions can often be corrected during learning by successive contradictions from text sentences, allowing them to revise or replace their individual false beliefs as they read a text. Although this process supports learning new knowledge at a local level (i.e., individual propositions) it is less effective in facilitating systemic conceptual change.In contrast, constructive processes such as self-explaining can facilitate conceptual change through building a mental model from cumulative revisions of individual false beliefs. In the current experiment, I investigated whether comparing and contrasting examples can achieve the same outcome. Students (n=22) in the compare group were first shown a diagram of with their own flawed mental model, and then asked to compare it with a diagram of the correct model. This condition was compared with self-explaining the correct diagram (n=22), and a control condition in which students simply read the text twice (n=20). Results showed that the compare group performed better than the other two groups on questions requiring deep inference making.iv

No abstract

Research in classrooms has shown mixed evidence for benefits of collaborative learning compared with learning individually. Moreover, laboratory research has shown that individuals working in dyads or groups often perform worse than individuals working alone -a robust finding called the collaborative inhibition effect. Despite these findings, we hypothesize that some classroom activities may afford benefits for collaborative learning over individual learning. We created a classroom writing activity that incorporated features such as shared prior knowledge and error-correction processes, which have been hypothesized to eliminate collaborative inhibition and to support constructive collaboration. Students participated in this activity either individually or in dyads. Results showed that the individuals who worked collaboratively performed equally well as those who worked individually and also showed better learning as measured by performance on a future writing assignment.