condition. However, we do not expect the pattern of results to
change substantially as the pure summary condition performed
worse than the drawing condition, even though students had
enough time to apply the strategies. Finally, a homogeneous sample
of students with a limited age range participated in the study.
Further research is required to replicate the pattern of results with
respect to other age groups.
In sum, our results suggest not only that researchers should
consider more specifically how learning strategies affect processes
of coherence building, but also that they should consider the type of
text, in particular the structure of the content examined. When
texts are used that describe spatial relationships between objects,
selection of main ideas and summarization may direct attention to
the text itself and thus may not prove as beneficial. Although we
focused on these kinds of texts, there is evidence that readers use
spatial representations in order to visualize relations other than
spatial relationships. They relate, for example, temporal sequences
of events spatially on a left-to-right axis (Schaeken, Johnson-Laird,
& d’Ydewalle, 1996). Thus, the benefits of visualizations may not
be limited to spatial relationships. On the practical side, the
results demonstrate the specific value of drawing activities for
visualizing to-be-learned content and for improving science text
comprehension.
Acknowledgments
The present publication is based on research projects funded by
the German Research Foundation (DFG; LE 645/9-1 as part of FOR
511). We thank Stephan Dutke for helpful comments on earlier
versions of this manuscript.