Effects of emotional tones in computer-based learning: Insights from system-paced and learner-paced experiments

Emotional design has increasingly been a promising avenue in multimedia learning research. Nevertheless, limited attention has been paid to the effects of two important emotional dimensions, which are valence and arousal. In two experiments we examined the effects of valence and arousal of background images in system-paced (Experiment 1) and learner-paced (Experiment 2) multimedia lessons. In both experiments, junior high school students were randomly assigned to one of the four experimental conditions in a 2 × 2 factorial design with valence (positive vs. negative) and arousal (low vs. high) as between-subjects factors. For retention, in Experiment 1 (system-paced), those receiving negatively valenced images significantly outperformed those receiving positively valanced images. However, the results across both experiments did not reveal significant valence × arousal interaction effects. For transfer, the results in Experiment 1 showed that high-arousing images significantly outperformed low-arousing images when positively valenced, while low-arousing images significantly outperformed high-arousing images when negatively valenced. The results in Experiment 2 (learner-paced) showed that negative valenced images outperformed positively valenced images when their arousal levels were high. Apart from learning outcomes, participants’ cognitive load and intrinsic motivation were also measured, but in both experiments no significant differences were found regarding these measures. Implications and future directions were discussed.