学習モチベーションの社会構成主義アプローチ～ソーシャル・ネットワーク・サービスにおける多様ピアメッセージの推薦システム

摘要

Contemporary learning theories and their implementations associated with information and communication technologies increasingly integrate social constructivist approaches in order to assist and facilitate the construction of knowledge. Social constructivism also highlights the important role of culture, learning attitude and behavior in the cognitive process. Modern e-learning systems need to include these psychological aspects in addition to knowledge construction in order to connect with long-standing pedagogical issues such as the decrease and lack of motivation for education. Motivation is a central part of educational psychology and plays an important role in Computer-Supported Collaborative Learning (CSCL) environment. A prominent factor of motivation consists in the strong connection between pedagogical goals and purposes for learning because learners want to know the reasons why learning is important for them, to make it more meaningful. However, although pedagogical institutions provide structured curricula with specific outcomes, students are often unable to relate to these goals as they have various conceptual perceptions and learning purposes. This issue has even more consequence in informal and self-regulated learning environments where learners must monitor their own actions, motivation, and goals. Contemporary CSCL applications need therefore to integrate a larger social presence in order to provide more diverse purposes for achieving a shared goal. Current social networking services (SNS) provide a platform where peers can for instance express their passion, emotion and motivation towards learning. This research utilizes therefore this platform to recommend motivational contents from peers for learning motivation enhancement (i.e. learners’ perception of their goal and purpose for learning). The proposed system consists of an SNS platform for learners to 1) express and evaluate their own goals for learning, 2) observe diverse motivational messages expressed by peers who share a same goal and recommended by an LDA-based (Latent Dirichlet Allocation) model, and 3) evaluate their perceptions on motivational attributes after each observation. This platform initially requires a database of messages from peers publicly expressing on SNS their own purposes for learning various subjects. This part of the research focuses on collecting and analyzing messages from Twitter to determine linguistic features used to construct the meaning of expressing diverse learning purposes. The recommender system was implemented as a Web-based application using SNS environment to conduct an experiment over a semester, with students who could observe purposes expressed by other peers. Results compared evaluations from 77 students on motivational attributes before and after observing diverse or similar purposes from peers. Participants who observed diverse purposes significantly and positive enhanced their motivational perceptions, such as on goal specificity, attainability and on the confidence to achieve the desired outcome.