A neuronal learning rule for sub-millisecond temporal coding

Abstract

A paradox that exists in auditory and electrosensory neural systems is that they encode behaviorally relevant signals in the range of a few microseconds with neurons that are at least one order of magnitude slower. The importance of temporal coding in neural information processing is not clear yet. A central question is whether neuronal firing can be more precise than the time constants of the neuronal processes involved. Here we address this problem using the auditory system of the barn owl as an example. We present a modelling study based on computer simulations of a neuron in the laminar nucleus. Three observations explain the paradox. First, spiking of an 'integrate-and-fire' neuron driven by excitatory postsynaptic potentials with a width at half-maximum height of 250 micros, has an accuracy of 25 micros if the presynaptic signals arrive coherently. Second, the necessary degree of coherence in the signal arrival times can be attained during ontogenetic development by virtue of an unsupervised hebbian learning rule. Learning selects connections with matching delays from a broad distribution of axons with random delays. Third, the learning rule also selects the correct delays from two independent groups of inputs, for example, from the left and right ear.

Time scales of memory, learning, and plasticity

“…An influence of the temporal signal order on plasticity was proposed by Gerstner et al (1996) and experimentally confirmed by Markram et al (1997), Magee and Johnston (1997), and Bi and Poo (1998) [although Levy and Steward (1983) had already discovered this phenomenon]. These authors found that for the expression of LTP or LTD not only the activity as such matters but also the timing.…”

Active control of spike-timing dependent synaptic plasticity in an electrosensory system

“…Excitatory synapses from adaptable inputs are represented by weighted excitatory postsynaptic potential waveforms, (t n ). The weighting factors, w A (t n ), change in t n as the system adapts due to a learning rule that depends on the relative timing of pre-and postsynaptic spikes [5]. The contribution of adaptable inputs to the neuron's membrane potential is given by, V A (t n ) = m w A m (t n ) (t n − t m ).…”

Numerical modelling of plasticity induced by transcranial magnetic stimulation

“…where T is a suitably chosen timescale, the angle brackets show a temporal average over t ′ centred around time t, and H ab (τ ) describes the STDP function, fitted with two exponentials which depend on the post-synaptic neuron [1,20,13,14]:…”

“…By the same token, the method will be appropriate to cases where large numbers of neurons are stimulated simultaneously, as in TMS. Recently, Robinson [36] has assembled a simple numerical model of spike-timing dependent plasticity (STDP) [1,20,13,14] within neurons that builds upon the platform of neural field modelling. This allowed a first application of a neural field model to TMS.…”