| Contents |
| We discuss a neuronal network model capable of learning the timing of a sequence of events, each of which lasts milliseconds to seconds. Short term facilitation is a second-timescale accumulation process that controls switching between events. Long term plasticity allows the network to learn event timings quickly and accurately. Time scale separations, between the plasticity processes and neuronal activity dynamics, allow us to describe how long term plasticity parameters should depend upon short term facilitation parameters for the network to be able to learn \textbf{any} sequence of timings. |
|