Adaptive Learning environments based on BCI Methodology

According to Cognitive Load Theory (CLT) (Sweller et al. 1998) the type and amount of cognitive load during learning is crucial for successful learning. CLT recommends that cognitive load should not exceed working memory capacity at any time during learning. Therefore it should be held within an optimal range of learners’ memory capacity. A continuous online measure of cognitive load during learning is necessary to investigate whether learning environments fulfill this criterion. Hence, we aim at developing computerbased learning environments adapting to learners’ individual cognitive load online and assisting individuals to increase their learning success. (Antonenko et al. 2010) suggest that electroencephalogram (EEG)- data might be used to measure instantaneous cognitive load. Thus, the adaptive mechanism of our system is based on Brain-Computer Interface (BCI) Methodology, which is based on EEG-data to diagnose learners’ cognitive load.

There are two main challenges of such a system: First, detection of characteristics in EEG-signals with high consistency over all subjects, which validly and reliably represent cognitive workload states. Second, development of online classifiers based on machine learning algorithms which distinguish between different cognitive workload levels online.

In this presentation we want to demonstrate how to cope with the challenges and illustrate on the basis of first results why continuous EEG-data combined with BCI methodology is a promising approach to develop a computer-based adaptive learning environment which individually supports learners in an optimal way.