Apply ATI to Support Adaptive E-Learning

Abstract

Most of the existing adaptive e-Learning research mainly focuses on how to use the latest technologies to provide adaptive features to learners. They neglected the importance of support from the educational theories. There is evident that educational theories provide lot of explanations on how can learners achieve better learning outcomes through adaptive way. Adaptive learning in fact is not a new concept in the field. It is originated from the individual differences of learners. Furthermore, certain previous adaptive e-Learning research focuses on using technology to personalize the content presentation style because of the individual differences in learning style. However, it seems that none of them has used adaptive e-learning to deal with the individual differences in aptitude which is more critical relatively. Aptitude Treatment Interaction (ATI) theory addresses the fact that if the instruction methods and the aptitude of individual learner is matched properly, then the learner is having a higher chance to obtain a better learning outcome. Aptitude means ability or talent to learn a specific area of knowledge which is a quite consistent pattern and may take a long time to change or even cannot be changed. The only thing we can change is the instruction method. Snow (1989) suggests providing a highly structured learning environment or learning content to learners who have poor aptitude and vise versa. ATI theory is hard to be operationalized unless we know how to evaluate the aptitude level of learners and to determine what does it meant by appropriate instruction method. We then examined the literatures about curriculum design and found that Biggs’ Constructive Alignment Model is designed for a similar purpose. Biggs suggests teacher to use assessment tasks to evaluate the level of attainment of the learners before the learner could advance to the next level of study. More highly structured learning content could be provided if needed immediately to avoid frustration. We believe that building an adaptive e-Learning based on the ATI theory can help improve the intrinsic motivation of students and result in a better learning outcome eventually. To test our belief, we design an adaptation model, apply this model into an agent-based adaptive e-Learning system and conduct a controlled laboratory experiment. The result is satisfactory. This approach has narrowed down the differences in learning outcome in class. Students who have received the adaptive support have showed a significant improvement in the post-test result when compared to the control group. The written comments by the participants show that the perceived difficulty in the course contents is different among the class. That might be the individual differences in ability.