Project AppLeMat

Development of an app for modular learning therapy in maths

In another project (DigitLern), we are already researching the conditions for successful distance learning therapy. This has resulted in the development of a tool that can help learning therapists and children with learning difficulties to provide effective learning therapy at a distance. Even beyond the pandemic period, children with chronic illnesses or children in rural areas with difficult access to learning therapy services and after relocation can benefit and continue to be supported by learning therapy (at a distance).
As part of Learning Therapy programmes, the app will be used to support children with dyscalculia. We want to build on tried and tested concepts while incorporating the latest scientific developments. The app is designed as a digital game for children, which conveys learning content in a playful and interactive way in order to promote knowledge and skills in a targeted way (= serious game).
Even before the Covid-19 pandemic, it had been shown that children can benefit from digital support in maths. However, the tools used should be theoretically sound, valid and evidence based. Our app will be technically accessible and feasible. Elements of learning psychology should be taken into account and combined with playful elements that promote fun and motivation (e.g. through reinforcement systems or feedback on learning progress). Other elements of our app will be:

• Adaptivity: The difficulty of the tasks used adapts to the learning and performance level.
• Instruction: Help structures and didactic elements help the child to better understand the tasks and promote sustainable learning.
• Feedback: Feedback and reward elements can be given directly and immediately, thus indirectly promoting motivation to learn and perform.

The game “Ecki’s Puzzle Cosmos” has now been published and is available free of charge to learning therapy professionals. The project has also been extended for another two years. The maths application developed in the original “AppLeMat” project will be extended to include AI mechanisms such as machine learning. This should enable the application to automatically analyse errors, make individual and dynamic task adjustments and provide personalised feedback based on the identification of a specific learning or ‘player’ profile.

In addition, we are investigating whether the AI-based design of the intervention (algorithmic selection of tasks based on a learning profile identified by network analyses) offers an advantage over the manual compilation of tasks by learning therapists. To the best of our knowledge, there are currently no adequate and scientifically evaluated support programmes for mathematical skills in German-speaking countries that use AI mechanisms to optimise children’s learning and the learning environment. This needs to be researched as part of the extension project.