CLAIR– Collaborative Learning Agent for Interactive Reasoning

project

Collaborative Learning Agent for Interactive Reasoning (Clair) is a conversational agent to foster productive student-student dialogue using reflective prompts (aka ‘talk moves’) deeply rooted in learning sciences.

Clair has a chat-based integration with the Go-Lab chat app. This chat app can be embedded into a digital lesson using the Twente ILS Editor.

Unlike single-user conversational agents, Clair does not react to all messages, which would distract group interaction. Instead, Clair reacts to dialogue situations in which prompting students, asking them to elaborate, provide an example or recap the discussion to benefit their interaction.

Clair is capable of interacting with student dialogues in various languages. Its effects have been evaluated in many studies, including three studies with secondary schools in Brazil and the Netherlands. Check out our recent paper evaluating Clair in the classroom.

If you would like to know a bit more about the inner workings of Clair, check out our recent AIED’24 paper, where we explain Clair’s implementation in more detail.

researchers

Adelson de Araujo 
Pantelis M. Papadopoulos
Susan McKenney
Ton de Jong

Partners

University of Oulu
Taiwan Tech
The Learning Agency

Duration

1 October 2020 – 31 December 2025

Funding

University of Twente
Tools Competition

project website

Additional information about the project can be found here