Hybrid systems combining artificial and human intelligence to develops learner’s self-regulated learning (SRL) skills.
For all their benefits, Adaptive Learning Technologies (ALT) take over all the monitoring and control of someone's learning progress, known as self-regulated learning (SRL). SRL can be explained as understanding and controlling one’s own learning environment, including setting goals and generating motivation. However, even the most sophisticated ALTs are not capable of supporting learners with self-regulated learning.
But what if you could design, develop and evaluate a hybrid ALT system that combined human and artificial intelligence to support young people with their self-regulated learning?
Living Lab researchers involved in the innovative HHAIR (Hybrid Human-AI Regulation) project are attempting to devise such a solution, drawing on an established network of contacts
and specialized expertise.
In the beginning it would, like regular ALT, withhold the regulation from the learner but would gradually re-introduce it as the pupil had developed the necessary skills. In this way HHAIR supports optimized learning and transfer (deep learning) and development of SRL skills for lifelong learning. This project is ground-breaking in developing the first hybrid systems to train human SRL skills with AI.
Above image visualizes the design of the Learning Path App which is part of the HHAIR project. In designing the learning path app, we followed the four phases of the COPES model, which describes the internal regulation processes that learners enact to regulate their learning. In the first phase, the task definition phase, learners develop an understanding of the task. During the goal setting phase, learners set their goals and plan their learning. In the enactment phase, learners execute their plans and control and monitor progress. In the adaptation phase, adjustments are made when progress towards the goals is not proceeding as planned. Click here if you want to read more about the Learning Path App.
HHAIR is funded by the European Research Council Starting Grant.