The approach is to bi-directionally interface with the intact brain at the spatiotemporal resolution of small groups of neurons or even individual neurons in freely-moving mice and rats. Rodents may be considered as having inferior cognitive capabilities, yet work in the lab showed that they can be taught complex tasks previously carried out only by primates/humans. Thus, the approach actually allows “erasing” and “writing” individual spikes during specific times, often in a closed-loop, while observing the effects of these manipulations on network activity and cognitive behaviour.
Cognitive processes are an outcome of the concerted activity of many neurons. About half of the connections that a given neuron makes are with its immediate neighbours. The goal of the work in the Stark lab is to understand the local network mechanisms that underlie “simple” cognitive processes: the emphasis is on short-term (seconds) and reference (minutes-hours) memory. Stated differently, the objective is to understand the “language” by which neurons interact within a local circuit to generate these functions.