 The researchers found that a subset of neurons in a spiking network could be trained to mimic the activity of neurons in the motor cortex during a decision-making task. This was achieved by creating strong synaptic connections between these neurons, regardless of whether they were part of a trained or untrained group. These strong synapses allowed for the spread of task-related activity throughout the network, even in untrained neurons. The researchers also used optogenetics to observe how this spread of activity affected the network's dynamics. They found that the motor cortex was strongly coupled, suggesting that this mechanism may apply to other cortical networks as well. This article was authored by Christopher M. Kim, Arsene Finkelstein, Carson Seachow, and others.