Accueil

A new model of background noise present in the nervous system could help better understand neuronal signalling delay in response to a stimulus. The authors present a biologically accurate model of the underlying noise present in the nervous system, which has implications for explaining how noise, modulated by unreliable synaptic transmission, induces a delay in the response of neurons to external stimuli as part of the neurons coding mechanism.

Neurons communicate by means of electrical pulses, called spikes, exchanged via synapses. The time it takes for brain cells to first respond to an external stimulus with an electric signal —commonly referred to as fist-spike latency—is of particular interest to scientists. That is because it is thought to carry much more neural information than subsequent serial spike signals.

The authors analyse the presence of noise in the nervous system detected through changes in first-spike latency. The noise is due to the large number of incoming excitatory and inhibitory spike inputs bombarding synapses. Previous attempts at noise modelling used a Gaussian approximation. Now, the authors have devised a noise model that is closer to the biological reality.

It is shown that there is a relation between the noise and delays in spike signal transmission, caused by unreliable synapses. Yet, synaptic unreliability could be controlled by tuning the incoming excitatory and inhibitory input signalling regime and the coupling strength between inhibitory and excitatory synapses. Ultimately, this could help neurons encode information more accurately.

Energy flowing from large-scale to small-scale places may be prevented from flowing freely in specific conditions, similar to those found in disordered solids. In the present article, the author presents an exception he found in a model of turbulence, indicating that there are energy flows from large to small scale in confined space. Indeed, under a specific energy threshold, there are no energy flows, similar to the way electron currents and energy spreading are stopped in disordered solids.

The author relies on numerical simulations to study a kind of turbulence—known as Kolmogorov turbulence—that describes how energy flows from large to small scale in a confined space. According to this concept energy is introduced on large scales, e.g. by wind, and it is absorbed on small scales due to energy dissipation. This approach assumes that a small perturbation will make the system evolution chaotic as energy flows from large to small scales.

However, the author finds that a phenomenon normally observed in disordered metals, called Anderson localisation, which implies that there is no energy flow from one side of the metal to the other, also occurs with the type of turbulences he is focusing on. As a result, energy flow from large scale to small scale does not happen under specific circumstances where the energy level is below a certain threshold level. This result is in keeping with our intuitive experience of a small wind not creating a storm, and that wind needs to reach a certain threshold before a storm can be created.

Controlling the plasma in magnetic fusion experiments remains a major challenge, in particular, with the advent of large-scale facilities such as the ITER tokamak (presently under construction in Cadarache, France). In order to support the operation of the machine, an extensive set of measurements is planned. Passive spectroscopy is a convenient diagnostic tool since it is non-intrusive and quite easy to implement experimentally. For instance, the hydrogen Balmer a line (3 ⇒ 2 transition, visible range) is considered as a way to measure fluxes of the hydrogen isotopes (H, D, T) in the divertor region (see “Progress in the ITER Physics Basis”, Nucl. Fusion, special issue, 2007).

The plasma density in the ITER divertor will be large enough to make Stark broadening observable on the spectral lines of hydrogen isotopes. Such neutral particles survive in the cold and complex edge plasma and their spectra provide invaluable information on its conditions. We have shown that the case where the Stark perturbation can be associated with a series of binary collisions with ions (impact approximation, see the work of Hans Griem, Plasma Spectroscopy) may be adapted to conditions foreseen in ITER. This model is based on an estimation of the S-matrix for atom-perturber collisions using a series expansion for large impact parameters (weak collisions) and, on the other hand, using a cut-off accounting for the oscillating behaviour of the wave-function in the case of small impact parameters (strong collisions). Confrontations with computer simulations indicate that the model is a good candidate for accurate diagnostics in the ITER plasma.