Neurocomputing or cognitive computational neuroscience (CCN) modeling is a recent field of research that focuses on explicitly modeling the relation between brain and cognition using mathematical models and computer simulations. For instance, the figure above shows two neurons that are connected serially. The leftmost neuron receive an input of electrical current and activates the rightmost neuron. The figure above each neuron shows the voltage (blue) and the corresponding built-up of neurotransmitter (green) that will drive activation in the next neuron. In a complete model, several neurons and brain areas are modeled, and the network is used to account for neurological (e.g., single-cell recordings, lesion, etc.) and behavioral (e.g., accuracy, reaction time, etc.) data. Example CCN models include FROST, a model of working memory maintenance, and SPEED, a model of procedural and rule-based categorization automaticity. A tutorial on CCN modeling can be found here.

In addition to the applications described above, CCN modeling can be used to explore neuropsychological disease, brain insults, and normal aging. For instance, we have done some previous work on exploring the role of dopamine, an important neuromodulator, on cognition using CCN models. Cortical dopamine elevation in a CCN version of COVIS allowed to reproduce some effects related to positive affect (e.g., good mood) while dopamine depletion in both the prefrontal cortex and the basal ganglia reproduced cognitive deficits observed in normal aging and Parkinson's disease patients.