Fedor Anna

Anna Fedor: My main interests are cognition and evolution. I experimented with gibbons to test their object permanence abilities for my MSc thesis and tested human participants on artificial grammar learning tasks for my PhD thesis on language evolution. After finishing my PhD work, I left Eötvös University for my first postdoc job in London: we built an artificial neural network model of vocabulary acquisition and tried to predict which therapy is best for children with language related difficulties. For my second postdoc I moved to Munich to explore the Neural Replicator Hypothesis. We tested human participants on insight tasks to investigate problem solving processes and built a neural network model that implemented evolutionary processes and solved insight tasks similarly to humans.

Research at iASK

My current research topic explores the hypothesis that new ideas and candidate solutions to a problem are generated by an evolutionary process in the brain. Recently, we implemented Darwinian Neurodynamics in artificial neural networks, which consists of thousands of attractor networks that are inter-connected and we showed that it can solve problems similarly to humans. The model also produced some interesting predictions about human behaviour that I would like to test. My plan is to pilot this experiment during my stay at iASK.