Pulse Packets: Synchronized Activity of Neurons and Neural Networks

Recently, evidence is accumulating that cortical neurons in vivo are capable of producing action potentials with high temporal accuracy. In recordings of multiple single-neuron activity in behaving monkeys, precisely timed action potentials have been systematically related to stimuli or behavioural events, indicating that these instances of precise spike timing play a functional role. Independent evidence for precise spike timing in cortical neurons came from intracellular recordings in vitro.

But, can an instance of synchronous spiking, once it has occurred, be successfully propagated by subsequent groups of cortical neurons? Under which input conditions can a group of cortical neurons engage in precisely coordinated spike timing, and are such conditions feasible in the cortical network? How can we clarify and quantify the notions of ‘well-timed’ and ‘reliable’, which gained such prominent role in the ongoing debate on temporal coding in the brain? Clearly, these questions must be resolved to decide whether cortical computation on the basis of precise spike timing is possible.

In a series of model studies we are exploring the mechanisms underlying the rapid synchronizations of cortical spiking activity. Specifically we focus on the conditions under which such synchronous volleys of action potentials can propagate reliably through the cortical network. Other questions involve: the functional role of inhibition, interactions with background activity, and the influence of synaptic dynamics. The theoretical approaches we apply in this research combine analytical calculations with large-scale network simulations (see also NEST / SYNOD).