Neuronal excitability
|
In nonlinear
systems the presence of noise often facilitates the detection and
transmission of weak input signals. Sensory neurons exploit this role of
noise for encoding and transmitting weak external stimuli in sequences of
spikes. Although the statistical properties of the spike sequences have been
extensively investigated, the way in which single neurons encode information
remains unclear. Single neurons can encode information in the spike rate
(“rate coding”) or in the timing of the spikes (“temporal coding”). We are
interested in studying temporal correlations in spike timing. We use simple
models of spiking neurons, such as the FitzHugh-Nagumo
(FHN) model or the integrate-and-fire model, and analyze
inter-spike-intervals (ISIs) correlations when the neuron is under the
influence of noise, a subthreshold external input, and is coupled to other
neurons. We are also
interested in the comparison of the statistical properties of neuronal ISI
sequences and those of the experimentally recorded optical spikes emitted by a
semiconductor laser, which resemble neuronal spikes. Establishing a
connection between these different dynamical systems can offer new
perspectives in both, photonics and neuroscience. Laser-based photonic
neurons can provide a novel, inexpensive and controllable experimental set up
for improving our understanding of neuronal activity. On the other hand,
laser-based photonic neurons can be building blocks of neuro-inspired,
ultra-fast optical computing devices. Recent
publications: Emergence
of spike correlations in periodically forced excitable systems J. A. Reinoso, M. C. Torrent,
C. Masoller Phys. Rev. E. 94, 032218 (2016). Analysis of
noise-induced temporal correlations in neuronal spike sequences J. A. Reinoso, M. C. Torrent,
C. Masoller Eur. Phys. J. Special Topics 225, 2689–2696 (2016). Funding: Complex
physical and biophysical systems: towards a comprehensive view of their
dynamics and fluctuations (ComPhysBio, FIS2015-66503-C3-2-P)
and ICREA ACADEMIA |
Neuronal spikes simulated with the
stochastic FitzHugh-Nagumo model. |
Back to
Cristina Masoller’s web page