With Henk Aarts and Hans Marien (both Social and Behavioural Sciences), and Mehdi Dastani and Marijn Schraagen (Information and Computing Sciences); the aim of this project is to design and develop AI tools for detecting and preventing low literacy in Dutch children, and for providing adequate and personalized support for improving literacy (with the Foundation for Open Speech Technology, Fontys University of Applied Sciences, Royal Library, Stichting Lezen, and other potential partners.
Philosophers have questioned the very existence of weakness of will. Yet, many of us perceive it as a concrete phenomenon causing considerable psychological discomfort. We do not always follow up on our intentions, are not always faithful to our plans and generally sometimes find ourselves doing things we did not want to do. One very interesting way in which we try to increase the success rate of our intentions is by orchestrating our physical and social environments in ways that help us realise our proximal and distal intentions: Heavy drinkers or smokers ask their friends to remind them about their intentions to quit, obese stick notes on the fridge reminding them of what they should not eat, and procrastinators ask their colleagues to coach them to deliver in time. However, often the environment that is doing the scaffolding is either passive or human. We explore how modern artificial intelligence can be used to enhance a subject’s goal autonomy by monitoring its behaviour, learning its habits and providing reports on how some of them stand in the way of reaching predefined goals. Candidates for a language in which such reports and goals could be stated, are studied in philosophical logic. This project aims (1) to provide a logic model for intention formation and realization; and (2) to empirically test how this model is supported by personalizing scaffolding knowledge to a human subject through modern machine learning techniques. This positions this research right at the intersection of philosophy, AI and psychology.