Tal Golan


I am interested in improving the experimental interface between artificial neural network models and cognitive neuroscience, using visual object recognition as a model system. If deep neural networks are our new psychological hypotheses, what should hypothesis testing look like? My postdoctoral research in the Kriegeskorte lab is focused on interrogating human behavioral and neuronal visual function by controversial stimuli, which are images optimized to produce conflicting responses in different deep neural network models.