proach to reading comprehension, which makes the process of comprehension explicit for learners (Palincsar and Brown, 1984). In reciprocal teaching of reading comprehension, for example, teachers model the important elements of comprehension, such as predicting, summarizing, and questioning, and then students begin to take on individual elements of the task. The task is essentially distributed among students, who share responsibility for its completion.
In elementary science classrooms, researchers have attempted to establish classroom versions of scientific communities (e.g., “community of learners” or “learning community” approaches), beginning with the “community of learners” (Brown and Campione, 1990, 1994, 1996) and “knowledge building” (Scardamalia and Bereiter, 1991, 1994) approaches in elementary school classrooms. Core to the approach is the notion that students are working together to build their understanding of answers to questions; the models they build can be revised as new ideas are uncovered through research; proposal and critique are essential to testing ideas in the community; and the teacher’s role is to facilitate this process rather than provide authoritative answers to questions. The classroom designs attempt to realize these goals through various “distributed expertise” activity structures, such as creating research teams pursuing particular topics, jigsaw activities in which representatives of different teams (who have developed different expertise) come together in new teams to pursue new questions, and culminating activities in which teams present and critique one another’s findings.
In general, the empirical research uncovers both the challenges and the promise of these approaches. Students in elementary school classrooms participate successfully in these types of learning communities. They take responsibility and ownership of the questions they pursue, and they exhibit increasing focus on discipline-appropriate peer-to-peer discourse, such as justifying and critiquing ideas and their evidence (Scardamalia and Bereiter, 1991, 1994). However, the classroom interactions are not simple to facilitate, requiring professional development for participating teachers. They also take time to become established and develop into shared classroom norms. These interventions are usually year-long collaborations between schools and researchers.
Herrenkohl and her colleagues have explored using “intellectual roles” to help make tasks more explicit to elementary schoolchildren (Herrenkohl and Guerra, 1998; Herrenkohl et al., 1999). These approaches build on the idea of supporting a task by making the process explicit, through assignment of specific responsibilities or roles for particular individuals. For example, Herrenkohl and Guerra (1998), working with two fourth grade classrooms, identified intellectual roles corresponding to particular aspects of the investigation task, such as making or checking predictions, summarizing findings, and connecting findings to theories. As the investigation proceeded,