We offer a theoretical perspective that ties choice biases to specific decision computations and shows how time-resolved neural evidence can distinguish them.
We map how different learning rules shape the representations that emerge in deep networks.
Showing all publications