Abstract:
In the current thesis, through a series of three experiments, it was examined how the understanding of dynamic phenomena in the Natural Sciences could be fostered by generally adding visualizations to text, and by using particularly dynamic visualizations (e.g., videos or animations) as compared to static visualizations.
To do so, the physical principles underlying fish locomotion were exemplarily chosen as a learning domain. Based on an analysis of the properties of this domain, it was expected that dynamic visualizations would be better suited than static visualizations, particularly with respect to achieving a deeper understanding of the content (cf. Bétrancourt & Tversky, 2000). However, this analysis also revealed that learning with dynamic visualizations might be hampered due to their high degree of visual complexity (cf. Schnotz & Lowe, 2008). Therefore, based on a literature review, instructional methods were derived to cope with the visual complexity of dynamic visualizations. These methods referred on the one hand to the influence of using spoken text (Study 2), and on the other hand to the influence of highlighting relevant information and relationships between text and visualizations (Cueing; Study 3). These methods were assumed to improve the instructional material in general, and specifically dynamic visualizations compared to static visualizations.
In a first step, in Study 1 it was examined if adding visualizations to text would lead to a better understanding than text alone, as could be derived from the Cognitive Theory of Multimedia Learning (Mayer, 2009). Results confirmed that assumption; learners receiving text and visualizations performed better in pictorial tasks and gained a deeper understanding (as measured by transfer tasks) than learners receiving text alone, indicating that visualizations are crucial for comprehending this domain.
In Study 2 it was investigated whether a superiority of dynamic over static visualizations might be more accentuated when learners did not have to split their visual attention between text and visualizations. For this purpose, a 2x2-design was chosen with type of visualization (dynamic vs. static) and text modality (spoken vs. written) as independent variables. In accordance with the modality effect (e.g., Ginns, 2005), results showed that spoken text lead to better learning outcomes for pictorial tasks and transfer tasks. Moreover, learning with dynamic visualizations led to better performances for transfer tasks than learning with static visualizations. However, other than what was expected, this superiority of dynamic over static visualizations was not more pronounced for using spoken instead of written text.
Study 3 addressed the impact of cueing on learning with dynamic and static visualizations. Since cueing is supposed to reduce the visual complexity of visualizations (e.g., de Koning, 2009), it was expected that the superiority of dynamic over static visualizations would be more pronounced under cued than under non-cued conditions. To ensure that the superiority of dynamic over static visualizations that was found in Study 2 was not restricted to one specific presentation format of static visualizations (namely static-sequential visualizations), a further presentation format of static visualizations was implemented (static-simultaneous). Thus, this resulted in a 2x3-desing with cueing (yes/no) and type of visualization (dynamic, static-sequential, static-simultaneous) as independent variables. Results revealed that learners in the cued conditions performed better for pictorial tasks, but other than what was expected, not better for transfer tasks. Likewise as in Study 2, learning with dynamic visualizations led to a better performance on transfer tasks than learning with static visualizations, whereas there was no effect of varying the presentation format of static visualizations. Contrary to the initial assumption, the superiority of dynamic over static visualizations was not more pronounced under cued than under non-cued conditions.
Taken together, first, adding visualizations to text can be regarded as crucial for understanding the chosen domain. Second, dynamic visualizations seem to be better suited than static visualizations to gain a deeper understanding of that domain. However, neither the text modality nor cueing moderated learning with these different types of visualizations, indicating that the visual complexity of dynamic visualizations might have played a subordinate role for this domain.