Authoring intelligent tutoring systems using human computation: Designing for intrinsic motivation


This paper proposes a methodology for authoring of intelligent tutoring systems using human computation. The methodology embeds authoring tasks in existing educational tasks to avoid the need for monetary authoring incentives. Because not all educational tasks are equally motivating, there is a tension between designing the human computation task to be optimally efficient in the short term and optimally motivating to foster participation in the long term. In order to enhance intrinsic motivation for participation, the methodology proposes designing the interaction to promote user autonomy, competence, and relatedness as defined by Self-Determination Theory. This design has implications for learning during authoring.

Publication Title

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)