Learning through the lens of modality : From cognitive learning theory to instructional design: a within-subject study of retention and transfer

Sammanfattning

This thesis examines how instructional modality affects meaningful learning, defined as retention and transfer, using Mayer’s Cognitive Theory of Multimedia Learning (CTML) as an analytical framework. Two research questions guided the study. The first examined how CTML integrated key cognitive learning theories (Baddeley’s Working Memory Model, Paivio’s Dual Coding Theory, and Sweller’s Cognitive Load Theory) to inform instructional design. The second empirically examined whether CTML-aligned multimedia instructions produced different learning outcomes compared to a unimodal text-based transcript. A within-subject design was conducted with 23 participants, where each participant was exposed to both a multimodal video and a verbatim transcript across two content topics. Retention measured how well participants remembered the presented material, while transfer measured their ability to apply it to novel situations. Both outcomes were measured using baseline-adjusted scores, and a paired \textit{t}-test was used to identify if there was a significant difference between the modalities. No statistically significant difference was found between the modalities for either retention or transfer, with effect sizes close to zero. A post-hoc exploratory analysis indicated that age may have acted as a moderating variable, with older participants showing a significant modality preference for the multimodal video on retention, while the younger participants showed no such preference. The findings suggest that instructional modality alone was not sufficient to determine meaningful learning outcomes. Self-paced learning was the most central contextual factor, as the transcript allowed participants to regulate their own cognitive load in a way that the fixed-pace video did not. These results do not contradict prior research on CTML, but instead indicate that modality effects are contextually dependent, and conditional on learner characteristics and the broader instructional context.