Adaptive tutoring systems

As redundant or unsuitable learning material unnecessarily demands learners’ cognitive resources, they can benefit from aligned instructional support, e.g., due to increasing expertise or states of frustration. When examining the current educational literature on tailored instructional support, we can notice that structured and formalized scaffolding procedures are still lacking. We address this gap from several angles.

Affect-adaptive tutoring systems (2021 – present)
Affect-adaptive tutoring systems are capable of detecting and adequately responding to learners’ emotional states by adapting the learning experience, e.g., by providing additional support to mitigate frustration, or accelerate lesson plans to avoid boredom. Accounting for affective responses during training for safety-critical situations may help develop coping strategies and improve resilience. Current research emerging from a dissertation in the SimTech Graduate Academy focuses on developing mechanisms of effective alignment and paves the way for advances in affective computing.

Leveraging Big Data in math education (2021 – present)
Building on the data base of the adaptive math education platform Bettermarks, a collaborative project leverages data mining techniques to investigate effects of teacher-student interactions and individual learning performance on drop-out rates. Emerging evidence can inform strategies and features for sustained use of adaptive learning technologies.

LEGO Mindstorms robot

Guidance fading in a robot construction task (2017 – 2018)
A first step towards a more systematic guidance fading approach emerged from a project on instructional variations in a robot construction task. The task involved building a LEGO mindstorms robot according to a given instruction. Based on a Hierarchical Task Analysis (HTA), we reduced the level of detail in the provided instructions in repeated building steps. The obtained evidence suggests to further explore this approach in alternative task settings.

Related publications

  • Schmitz-Hübsch, A., Becker, R., & Wirzberger, M. (under review). Accounting for interindividual differences in affect-adaptive systems using deep learning techniques.
  • Schmitz-Hübsch, A., Stasch, S.-M., Becker, R., Fuchs, S., & Wirzberger, M. (2022). Affective response categories – Towards personalized reactions in affect-adaptive tutoring systems. Frontiers in Artificial Intelligence, 5, 873056. https://doi.org/10.3389/frai.2022.873056 [PDF]
  • Spitzer, M. W. H., Gutsfeld, R., Wirzberger, M., & Moeller, K. (2021). Evaluating students’ engagement with an online learning environment during and after COVID-19 related school closures: A survival analysis approach. Trends in Neuroscience and Education, 25, 100168. https://doi.org/10.1016/j.tine.2021.100168
  • Esmaeili Bijarsari, S., Wirzberger, M., & Rey, G. D. (2018). Guidance or Setting? Exploring the learnability of computer-based instructions in a construction task. In J. Hartig, & H. Horz (Eds.), 51st Conference of the German Psychological Society. Abstracts (p. 509). Lengerich: Pabst Science Publishers.
  • Esmaeili Bijarsari, S., Wirzberger, M., & Rey, G. D. (2018). Guidance or Setting? Exploring the learnability of computer-based instructions in a construction task. In A. C. Schütz, A. Schubö, D. Endres, & H. Lachnit (Eds.), Abstracts of the 60th Conference of Experimental Psychologists (p. 69). Lengerich: Pabst Science Publishers.
  • Esmaeili Bijarsari, S., Wirzberger, M., & Rey, G. D. (2017). Lernförderliche Gestaltung computerbasierter Instruktionen zur Roboterkonstruktion [Enhancing the design of computer-based instructions in a robot construction task]. In M. Eibl, & M. Gaedke (Eds.), INFORMATIK 2017, Lecture Notes in Informatics (LNI) (pp. 2279-2286). Bonn: Gesellschaft für Informatik. https://doi.org/10.18420/in2017_228