Encouraging the Active Integration of Information During Learning with Multiple and Interactive Representations

Computer-based learning environments commonly comprise various linguistic as well as static and dynamic pictorial representations, frequently combined with the possibility to modify them interactively. While multiple and interactive external representations have the potential to improve learning in specific ways, they also place specific demands on learners. For instance, learners have to process and relate different representations, to control and evaluate their interactions with these representations, and to construct a coherent mental representation. In many cases, learners are not able to meet these demands and suffer from cognitive overload. Taking advantage of cognitive load theory, we try to improve learning with multiple and interactive representations by reducing extraneous cognitive load and by increasing germane cognitive load which is supposed to be related to learning processes. To accomplish this, we encourage learners to actively integrate different representations and to interact with them in a structured and reflective way. We implemented these measures into the statistics learning environment VISUALSTAT and evaluated them experimentally. An analysis of variance revealed (1) that the active integration of different representations improved learning significantly, and (2) that the structured interaction with different representations increased verbal comprehension.