It is argued that research on the acquisition of science concepts has rich
implications for the teaching of science and can lead to the development of
useful principles for the design of learning environments. An experimental
project that attempted to use these research-based principles to construct
a learning environment for teaching mechanics to fifth- and sixth-grade st
udents is described. The students were encouraged to take active control of
their learning, express and support their ideas, make predictions and hypo
theses and test them by conducting experiments. They worked in small groups
and presented their work to the classroom for debate. Metaconceptual aware
ness was promoted by encouraging students to make their ideas overt, to tes
t them and compare them with those of other students and to give scientific
explanations. Emphasis was also placed on giving the students the opportun
ity to use models. representational symbols, and measurements. Results show
ed significant differences between the experimental and control groups in p
re-rest-post-test comparisons, confirming our hypothesis that the experimen
tal learning environment would result in cognitive gains for the participat
ing students. Further interview analyses and analyses of the classroom disc
ourse helped to clarify some of the variables contributing to the observed
conceptual change. (C) 2001 Elsevier Science Ltd. All rights reserved.