Motivation and Goals

Electronic learning can be quite time consuming and unexciting. Even with some of today’s best computer supported instructional technology, students engage in gaming behaviors associated with less learning. Time spent on task and motivation as key factors of effective learning need to be sustained but contemporary tutoring systems seem to be failing in this respect; all too many students drop out due to low motivation. Can computer tutors build trust and respect with students that would motivate them to learn at all?

This dissertation investigated components of effective learning environments, and proposed an approach to bootstrap a socially intelligent tutoring strategy that operates on top of these components.

• Socially intelligent computer tutor. An automated computer tutor that follows the strategy can select appropriate learning activities for students.
• Balance: individual vs. social, cognitive vs. affective. The method is designed to be able to balance individual and social, and cognitive (on-task) and affective (off-task) activities.
• Long-term motivation. Combined with the off-task dialog facility guided by human wizards, the proposed bootstrapping method is designed to provide novel interaction patterns to students for a relatively long period of time in order to maintain their motivation and increase the time students invest in study.

Results

We present an approach for computer supported education in the form of a socially intelligent learning environment -- Peoplia (available online). It integrates problem solving and instructional materials into individual and group learning scenarios. A Wizard-of-Oz-driven computer tutor accompanies students to maintain their motivation within the learning environment. The agent can hold off-task conversations and guide students to appropriate learning opportunities. Its tutoring strategy is devised by a reinforcement learning control method that operates on socially motivated state and action spaces induced by the human wizard whose interface facilitates rapid prototyping of relevant states and taking appropriate actions.

To make the learning algorithm feasible, states are grouped into equivalence classes according to wizard selected state features, and contextual and linguistic reflection is employed to adjust the immediate action to the current learner's situation. The proposed approach is pedagogically and technologically robust and is well suited for home study, regular classroom use, as well as for both formative and summative assessment.

The learning environment presented was designed to provide students with socially relevant interventions that would motivate learning. The environment consists of a redesign of typical learning components so that diverse social interaction possibilities are available (to the students). Both problem solving and course note facilities are able to operate in individual (for a single student) or social (for group of students) modes.

The core contributions are:

• novel extensions of problem solving and course notes learning activities that enable student to work on personalized content in both individual and group/social mode;
• novel model of user's social context (past events, future goals, etc.) that facilitates knowledge reuse across different students; and
• novel method for bootstrapping a socially intelligent tutoring strategy which can balance individual and group activities, and assemble suitable collaborative groups.

The approach was evaluated by conducting experiments in the domain of middle school mathematics. To evaluate the robustness of assessments, students worked on tasks generated on-the-fly to discourage cheating, while the human wizard judged the answers. The feasibility study of the socially intelligent agent demonstrated that students who engaged with the agent attained higher learning gains and liked the system more. In a collaborative learning experiment, students solved problems in groups more efficiently when being socially motivated. Finally, the bootstrapping of the socially intelligent tutoring strategy was evaluated in simulated student scenarios. Evaluations suggest that our approach for using computers to support students in the learning process is technologically viable.

Conclusion

In this dissertation, we proposed a learning environment that facilitates student learning using a tutoring agent that adheres to a socially intelligent tutoring strategy. The strategy is modeled on the basis of input from human wizards who guide the bootstrapping process.

The proposed approach puts humans in the loop, and requires non-trivial human participation to sustain long-term operation; for example, student answers need to be judged, and the dialog facility requires human wizards to identify previously not encountered interaction patterns. However, the evaluations suggest that the fruits of this approach may be worth it:

• problem solving robustness discourages surface approaches to learning;
• socially engaged students show higher learning gains; and
• the bootstrapping of a tutoring strategy (followed by the human wizards) can find a balance between individual and group/social activities, and learning vs. non-learning activities for students.

Selected publications

Tvarožek, J., Bieliková, M.

Wizard-of-Oz-Driven Bootstrapping of a Socially Intelligent Tutoring Strategy. In ED-MEDIA 2011: Proceedings of World Conference on Educational Multimedia, Hypermedia and Telecommunications, Chesapeake, VA: AACE (2011), pp. 3635–3644.

Tvarožek, J., Bieliková, M.

Feasibility of a Socially Intelligent Tutor. In ITS 2010: Proceedings of Intelligent Tutoring Systems, LNCS, Vol. 6095, Pittsburgh, USA, Springer (2010), pp.423–425.

Tvarožek, J., Bieliková, M.

Enhancing Learning with Off-Task Social Dialogues. In EC-TEL 2010: Proceedings of European Conference of Technology-Enhanced Learning, LNCS, Vol. 6383, Barcelona, Spain, Springer (2010), pp. 445–450.

Tvarožek, J., Bieliková, M.

The Friend: Socially-Intelligent Tutoring and Collaboration. In AIED 2009: Proceedings of Artificial Intelligence in Education, Frontiers in Artificial Intelligence and Applications, Vol. 200, Brighton, IOS Press (2009), pp. 763–764.

Tvarožek, J., Kravčík, M., Bieliková, M.

Towards Computerized Adaptive Assessment Based on Structured Tasks. In AH 2008:Proceedings of Adaptive Hypermedia and Adaptive Web-Based Systems, LNCS, Vol. 5149, Hannover, Springer (2008), pp. 224–234.

to Homepage

to Teaching

to the Top