One of the core assumptions in psychology is that children's cognition is shaped by their environment. By interacting with the world and the people around them, children acquire the psychological abilities that allow them to relate to others and become functioning members of society. This view implies that variation in the developmental environment of children should show itself in the way that their cognition develops.
Even though this idea is so central to the field, it is rarely tested empirically. The main reason for this gap is a lack of data about what children’s day to day experience is like. That is, we do not have a good understanding about the input that generates the development we observe.
To close this gap, we want to develop new methods that allow us to quantify children's lived experience. We try to leverage recent advances in machine learning, computer vision and computational modeling to develop tools and pipelines that a) continuously sample children’s day to day experience, b) filter and process the raw data and organise it into meaningful psychological units and c) predict change in cognitive development in relation to these units.