No abstract

Studies exploring how students learn and understand science processes such as diffusion and natural selection typically find that students provide misconceived explanations of how the patterns of such processes arise (such as why giraffes' necks get longer over generations, or how ink dropped into water appears to ''flow''). Instead of explaining the patterns of these processes as emerging from the collective interactions of all the agents (e.g., both the water and the ink molecules), students often explain the pattern as being caused by controlling agents with intentional goals, as well as express a variety of many other misconceived notions. In this article, we provide a hypothesis for what constitutes a misconceived explanation; why misconceived explanations are so prevalent, robust, and resistant to instruction; and offer one approach of how they may be overcome. In particular, we hypothesize that students misunderstand many science processes because they rely on a generalized version of narrative schemas and scripts (referred to here as a Direct-causal Schema) to interpret them. For science processes that are sequential and stage-like, such as cycles of moon, circulation of blood, stages of mitosis, and photosynthesis, a Direct-causal Schema is adequate for correct understanding. However, for science processes that are non-sequential (or emergent), such as diffusion, natural selection, osmosis, and heat flow, using a Direct Schema to understand these processes will lead to robust misconceptions. Instead, a different type of general schema may be required to interpret non-sequential processes, which we refer to as an Emergent-causal Schema. We propose that students lack this Emergent Schema and teaching it to them may help them learn and understand emergent kinds of science processes such as diffusion. Our study found that directly teaching students this Emergent Schema led to increased learning of the process of diffusion. This article presents a fine-grained characterization of each type of Schema, our instructional intervention, the successes we have achieved, and the lessons we have learned.Correspondence should be sent to Michelene T.

Betty's Brain is a computer-based learning environment that capitalizes on the social aspects of learning. In Betty's Brain, students instruct a character called a Teachable Agent (TA) which can reason based on how it is taught. Two studies demonstrate the prote´geé ffect: students make greater effort to learn for their TAs than they do for themselves. The first study involved 8th-grade students learning biology. Although all students worked with the same Betty's Brain software, students in the TA condition believed they were teaching their TAs, while in another condition, they believed they were learning for themselves. TA students spent more time on learning activities (e.g., reading) and also learned more. These beneficial effects were most pronounced for lower achieving children. The second study used a verbal protocol with 5th-grade students to determine the possible causes of the protégé effect. As before, students learned either for their TAs or for themselves. Like study 1, students in the TA condition spent more time on learning activities. These children treated their TAs socially by attributing mental states and responsibility to them. They were also more likely to acknowledge errors by displaying negative affect and making attributions for the causes of failures. Perhaps having a TA invokes a sense of responsibility that motivates learning, provides an environment in which knowledge can be improved through revision, and protects students' egos from the psychological ramifications of failure.

No abstract