Conference on Cognitive Computational Neuroscience

The meteoric rise in the adoption of deep neural networks as computational models of vision has motivated endeavors to ``align” these models with humans. One dimension of interest includes behavioral choices, but moving beyond characterizing choice patterns to capturing temporal aspects of visual decision-making has been challenging. Here, we sketch a general-purpose methodology to construct computational accounts of reaction times from a stimulus-computable, task-optimized model. Specifically, we consider a perceptual-grouping task and leverage insights from subjective logic theory to train a recurrent neural network vision model equipped with equilibrium dynamics. We introduce a novel metric summarizing evidence accumulation in our model. Our metric aligns with patterns of human reaction times for stimulus manipulations designed to probe the spread of visual attention. This work paves the way for exploring the temporal alignment of model and human visual strategies in the context of various other cognitive tasks toward generating testable hypotheses for neuroscience.