Interleaved practice in multi-dimensional learning tasks: Which dimension should we interleave?

Research shows that multiple representations can enhance student learning. Many curricula use multiple representations across multiple task types. The temporal sequence of representations and task types is likely to impact student learning. Research on contextual interference shows that interleaving learning tasks leads to better learning results than blocked practice, but this research has not investigated whether it matters on which dimension we interleave learning tasks. Many educational materials include multiple task types and multiple representations. Should we interleave representations or task types? We conducted a classroom experiment to investigate the effects of interleaving task types (while blocking representations) and interleaving representations (while blocking task types). The participants (158 5th- and 6th-graders) worked with a corresponding version of an intelligent tutoring system for fractions. Our results show an advantage for interleaving task types over interleaving representations. These results extend prior work on contextual interference by showing that this effect is sensitive to the dimension being interleaved. We also extend the literature on learning with multiple representations by investigating the effect of interleaved practice with different representations. The results provide guidance to designers of complex curricula.

► We examine learning with an intelligent tutoring system for fractions.
► The tutoring system includes multiple task types and graphical representations.
► We contrast the effects of interleaving task types versus representations.
► We assess representational knowledge and operational knowledge of fractions.
► Interleaving task types leads to better learning than interleaving representations.