摘要：Animated pedagogical agents have become popular in multimedia learning with combined delivery of verbal and non-verbal forms of information. In order to reduce unnecessary cognitive load caused by such multiple forms of information and also to foster generative cognitive processing, multimedia design principles with social cues are suggested (Mayer, 2014a). This study presents the design model of pedagogical agent multimedia learning by using four design principles based on social cues. Then reported are the findings of a study examining the effects of the pedagogical agent multimedia learning on perceived cognitive load, situational interest, motivation, and achievement. One hundred twenty seven college undergraduate students enrolled in four Computer literacy classes participated in this study and were randomly assigned to one of the six conditions in which on-screen images (image vs. no image) and narrations of a pedagogical agent (human voice narration, on-screen text, no narration) were presented in different levels. The results indicated that, overall, the presence of images does not significantly affect perceived cognitive load, situational interest, motivation, or achievement. However, the form of narration influenced the four outcome measures differently. The use of human voice narrations presented by a pedagogical agent was effective to reduce the perceived cognitive load compared to on-screen text narration and no narration conditions. Human voice narration by a pedagogical agent was also found to promote learners situational interest, which is negatively correlated to cognitive load. The personalized narration was found to improve learners motivation in terms of relevance and confidence whether presented by a pedagogical agent or in on-screen text although no significant differences were found in the recall test and the comprehension test.
