Wouter van den Bos has a broad background in both neuroscience and developmental psychology, and his research investigates the relation between the developing brain and learning and decision-making across adolescent development. To approach these questions, he uses a multi-method approach relying on computational models, game theory, social network analysis, and advanced neuroimaging techniques. Computational models have helped to explain why children, adolescents and adults use different learning strategies. Currently he wants to use these methods to better understand the influence of the social context in which learning is naturally embedded.
My plans for the fellowship period
The shift towards the use of more technology in education has often been paired with the promise of offering individually tailored learning experiences that optimize learning efficiency. However, a significant amount of learning in, and outside, schools is based on social learning. The transmission of information through social learning lies at the root of human culture. Thus, although the benefits of individualized learning are clear, one important challenge remains to integrate the social aspects of learning. To do so we need a firm understanding of how social learning works and how it develops across development. The aim of this project is to gain more insight in social learning in the classroom setting, and how this may change with age, specifically how social learning strategies develop from childhood to late adolescence (6-18 years).
For this we will develop novel learning tasks that are more tuned to the classroom learning (e.g. word-pronunciation mapping, and artificial grammar learning), as well as computational models that integrate social information. In addition, we will rely on social network analyses to map the social connections within classrooms. This will allow us to do fine-grained studies of how social learning is affected by who the demonstrator is.
How will my work change children’s and youth’s lives?
Our results will inform us how social learning strategies change with age, and to identify if potential windows for teaching better social learning strategies (e.g. whom to learn from, and when best to seek social information). To maximize the translational potential of the findings the research will be conducted in the classroom environment, and novel experimental tasks are designed to closely mimic real-world learning (e.g. word-pronunciation mapping, and artificial grammar learning). Our findings will have consequences for the development of new technological interventions and teaching methods, specifically those that aim to integrate social aspects (e.g. teachable agents or co-learning).
Moreover, social learning strategies are highly dependent on the social structure and in our previous research we found that the classrooms network structure differs significantly for children and adolescents. Building on these findings we hypothesize that increased social cohesion (connectedness) in the classroom is related to an increased tendency to favor social learning over individual learning. Thus, our research will also provide novel insights in relationship between classroom structure, social cohesion and overall academic performance.