Effects of Agent Emotional Support and Cognitive Motivational Messages on Math Anxiety, Learning, and Motivation

Mathematics plays an important role in everyday life. Math anxiety has been one of the major reasons that students do not like math mathematics. Many efforts have been done to alleviate learner's math anxiety with respect to curriculum design, math assessment, classroom culture, and teacher attitudes. In this study, math anxiety from the GED (General Educational Development) population is addressed through emotional and motivational support provided by pedagogical agents. There are two reasons to this approach. One major reason is the inherent social nature of learning and research on learners' emotion and motivation are gaining more and more interest. The other major reason is that the embodiment nature of pedagogical agents have been recognized by researchers and the pedagogical agents become one of the ideal candidates to improve learning and influence learners' emotion and motivation in a computer-based learning environment. In this study, emotional support and cognitive motivational messages were embedded in pedagogical agents to promote learning, alleviate math anxiety, and improve leaner motivation. 109 GED students affiliated with local community college participated in this study. This study used a 2 x 2 factorial design. The two independent variables were emotional support and cognitive motivational messages and each dependent variable contained two levels: presence versus absence. The dependent measure of study includes learning, math anxiety, and leaner motivation. MANOVA analysis revealed a main effect of emotional support. Further univariate analysis showed that students who worked with emotional support agents performed better in learning and had less math anxiety than those students who worked with none-emotional support agents. No significant results were found on the main effect of cognitive motivational messages. This study shows quantitative evidence of the effectiveness of pedagogical agents as a means to deliver emotional support in a computer based learning environment. Further research is suggested to examine the adaptive nature of emotional support and cognitive motivational messages. The effect of emotional support and cognitive motivational messages in non-GED population needs to be tested.