Social Foundations of Learning


These projects are exploring how to make technologies that enhance the best of social interaction for learning while mitigating the worst.

Teachable Agents


Teachable Agents (TA) is a learning technology that draws on the social metaphor of teaching a computer agent to help students learn. Students teach their agent by creating a concept map that serves as the agent’s “brain.” An artificial intelligence engine enables the agent to interactively answer questions posed to it by traversing the links & nodes in its map. As the agent reasons, it also animates the path it is following, thereby providing feedback, as well as a visible model of thinking for the students. Students can then use the feedback to revise their agent's knowledge (and consequently, their own).

The teaching metaphor enlists fruitful social attitudes during the interaction, including a sense of responsibility for one's agent that appears to motivate students to work harder to organize their understanding. TA has been found to improve children's scientific reasoning in both causal and taxonomic (hierarchical) domains.

Relevant publications: