Modeling social cognition and its neurologic deficits with Artificial Neural Networks

Artificial Neural Networks are computer models that are loosely inspired by the functioning of the human brain. They are now the start-of-the-art method for tackling a variety of AI problems, and are becoming an increasingly popular tool in neuroscientific studies. However, both domains pursue different goals: in AI model performance is key and brain resemblance is incidental, whilst in neuroscience the aim is chiefly to better understand the brain. This PhD aims to build a bridge between both domains by developing novel ANN building blocks and architectures with a focus on better understanding social cognition within the human brain. In a first phase, the focus lies on control group behavior. Later phases focus on studying how the developed model(s) can be altered in a controlled way to mimic clinical conditions observed in the human population, in particular Autism Spectrum Disorder and Frontotemporal Dementia.