In digital learning environments, the absence of social interactions can negatively affect student engagement, motivation, and learning outcomes. One proposed solution is the use of pedagogical agents (PAs) – digital characters designed to enhance social presence, provide instructional support, and foster learning motivation. While previous research suggests that PAs can positively impact learning, the underlying mechanisms of interest and PAs remain unclear.
This study investigates whether a pedagogical agent improves students' retention and knowledge transfer and whether their interest mediates this effect. Specifically, it examines whether a PA that explicitly demonstrates interest in the learning topic increases motivation and engagement, leading to improved learning outcomes.

A 2 (Pedagogical Agent: with vs. without) × 2 (Demonstrated interest: with vs. without) between-subjects experiment was conducted with 132 German secondary school students (Mage = 10.92, SDage = 0.77). Dependent variables were topic interest, cognitive load, situational interest, motivation, and additionally, learning outcomes in terms of retention and transfer were measured. Prior knowledge was included as a control variable to account for baseline differences.

The learning material was designed for 5th and 6th grade students and focused on the topic of "Plastic in the Ocean". The educational videos were approximately 12.5 minutes long. When participants were exposed to a pedagogical agent, it was in the form of an animated, humanized turtle. This digital character was designed to encourage social presence and interaction. The demonstrated interest conditions included verbal cues, such as phrases like "Isn't that interesting?" These cues were intended to express enthusiasm for the topic explicitly, thereby increasing students' motivation and engagement with the material.

A multiple linear regression analysis examined the influence of the pedagogical agent (PA) and topic interest (plastic in the ocean) on information retention. The overall model was significant but explained only a small variance of the dependent variable. While topic interest had a significant positive effect on retention, no significant effect was found for the pedagogical agent.

A second multiple linear regression analysis examined the influence of the demonstrated interest and topic interest (plastic in the ocean) on information retention. The overall model was significant but explained only a small variance of the dependent variable. While topic interest had a significant positive effect on retention, no significant effect was found for the demonstrated interest.

The findings of this study indicate that topic interest is more important for learning success than the integration of a PA, and that topic interest is more important for learning success than the integration of demonstrated interest (by an PA/instructor).

The results have implications for the design of digital learning environments. Integrating PAs with a demonstrated interest in the topic may further enhance student engagement and learning effectiveness if students are generally interested in the topic.

Keywords: e-learning Experiences, Primary and Secondary Education, Pedagogical Innovations.