Abstract:

Learning by imagining, which refers to learners forming mental images corresponding to learning materials in their minds, is an essential generative learning strategy. Whether the imagination strategy has a positive impact on multimedia learning has yet to be consistent. Generative learning theory holds that the imagination strategy can enhance learners' memory and understanding of learning materials by encouraging them to produce a structured well-connected mental image in mind. However, according to the cognitive load theory, the formed mental images will occupy a large number of cognitive resources, thus leading to cognitive overload and hindering learning. In this study, we conducted a meta-analysis to explore whether the imagination strategy can affect learning performance, time spent, and cognitive load. Based on relevant studies, it was found that the manipulative way of imagination strategy (learning material visibility, timing of use) and the learner’s individual characteristics (learner’s age) might be the factors for the effect of the imagination strategy. So we further explored several moderators (i.e., learning material visibility, the timing of imagination strategy, and learner’s age) that may have contributed to the boundary conditions of the imagination strategy. We found 20 articles met the inclusion criteria and generated 65 effect sizes. The results of the meta-analysis indicated that the imagination strategy could positively facilitate learning performance (g _retention = 0.40, g _{comprehension} = 0.27, g _transfer = 0.43). But there is no significant effect on learning time and cognitive load. This result confirmed the hypothesis of generative learning theory that the imagination strategy can promote meaningful learning by helping learners to attend, organize and integrate relevant materials. Furthermore, moderator analysis found that the learning material visibility moderated the impact of imagination strategy. When the material was visible, the imagination strategy could positively impact learning (g _retention = 0.42). But when the material was not visible, the imagination strategy had a negative impact on learning (g _retention = -0.26). This result could be explained by the cognitive load theory. When learning materials were not visible, learners needed to spend more cognitive resources on recall and retrieval, which might cause cognitive overload. So it could not promote learning and even have a negative impact. In addition, no moderating effect was found on the timing of using imagination strategy and learner’s age. There might be other factors that need further exploration. In conclusion, the imagination strategy plays a positive role in multimedia learning. In educational practice, educators should encourage learners to adopt the imagination strategy for learning. Allowing learners to view the material while imaging will lead to better learning performance. Due to the limited number of articles, this study did not set a more detailed baseline. Future research should focus on boundary conditions, neural mechanisms, and ecological validity to further explore the mechanism of imagination strategies.

Key words: learning by imagining, learning strategy, cognitive load theory, generative learning theory