My name is Anna de Koster and I am a PhD student at the University of Groningen. I am part of the Semantics and Cognition group of the Center for Language and Cognition Groningen (CLCG).
Besides this I’m also affiliated with the cognitive modelling group at the department of Artificial Intelligence.
My supervisors are Prof. Dr. Petra Hendriks and Dr. Jennifer Spenader (Artificial Intelligence). I’m interested in language development and my research focuses on Quantification and Distributivity.
More information about my PhD project can be found under ‘Research’.
Acting individually or together? An investigation of children’s development of distributivity.
This project (October 2016 – October 2020) is funded by the NWO programme “PhDs in the Humanities”.
When we talk about group actions, each member of a group can act individually or the group can act together. Consider the following sentence:
“The boys are pushing a car.”
Are the boys pushing one car together, the collective interpretation, or are they each pushing their own car, the distributive interpretation? Both are possible, but adults generally prefer collective interpretations, while children seem to prefer distributive interpretations. What causes children to have different preferences and what factors affect the development?
In this project we will investigate children’s transition to adult distributivity preferences. Three phases look at each major step: childhood, adulthood and the transition.
Phase 1 investigates a recent proposal that children’s non-adult preferences arise from a confusion about sets. This confusion is argued to stem from children’s limited cognitive resources. We will test this account with children and adults. Phase 2 experimentally investigates a recent account that adult preferences are derived via implicatures, and directly compares distributivity processing with implicature processing. Finally, Phase 3 takes an interdisciplinary approach by using computational cognitive modelling to simulate the development of children’s intuitions into adult preferences. Computational modelling allows precise examination of how each part of the theoretical model influences interpretation by allowing the same experiments to be simulated and to be compared to human results.