Benefits of noise in the transmission and processing of time-dependent stimuli by populations of sensory neurons


Populations of sensory nerve cells (neurons) fire trains of action potentials (spikes), which encode information about time-dependent stimuli that are relevant to the organism. These neurons are often subject to considerable intrinsic noise and/or to heterogeneity within the network. It is generally poorly understood why some populations of neurons show thus a high degree of intrinsic noise and heterogeneity and some do not. From theoretical studies it is known that the transmission of signals in nonlinear systems can benefit from noise by means of different forms of stochastic resonance, This effect has also been found in neural systems, which are stimulated by a mixture of signal and noise. In neural populations the effect of suprathreshold stochastic resonance (SSR) has a particular potential importance. SSR has been studied predominantly in non-dynamical threshold systems but has not been demonstrated experimentally (a direct proof is not possible because the intrinsic noise cannot be varied in neural systems). The theory of this effect in threshold systems developed so far, however, does not describe effects of stimulus correlation (signal bandwidth) and amplitude, and of population heterogeneity on the transmission of information in populations of spiking neurons. Furthermore, existing theories focussed on the summed population activity although in many situations the synchronous activity is the more meaningful measure for the information flux through the system. The synchronous activity is particularly interesting because, as shown recently in experiments and theory, it can be used for simple forms of signal processing as, for instance, information filtering. In the proposed project, we want to develop a theory for a heterogeneous population of spiking neurons, which is stimulated by a mixture of a common bandpass-limited signal and intrinsic neural noise. This theory should be used to study conditions under which intrinsic noise and parameter heterogeneity can be beneficial for the transmission and processing of a common sensory stimulus. Theoretical predictions will be tested in tight collaboration with an experimental group performing experiments on a biological model organism, the weakly electric fish. Our results will also contribute to a general theory of neural signal processing with applications, for instance, also in cortical networks.


Principal Investigators
Lindner, Benjamin Prof. Dr. (Details) (Theoretical Physics / Theory of Complex Systems and Neurophysics)

Duration of Project
Start date: 03/2014
End date: 07/2017

Publications
2015
D. Bernardi and B. Lindner: A Frequency-Resolved Mutual Information Rate and its Application
to Neural Systems J. Neurophysiol. 113 , 1342 (2015)
A. Kruscha and B. Lindner: Spike count distribution in a neural population under weak common
stimulation Phys. Rev. E 92 , 052817 (2015)

2016
A. Kruscha and B. Lindner: Partial synchronous output of a neuronal population under weak
common noise: analytical approaches to the correlation statistics Phys. Rev. E 94 , 022422 (2016)
B. Lindner: Mechanisms of information filtering in neural systems IEEE Transactions on Molecular,
Biological, and Multi-Scale Communications 2, 5 (2016)

2017
J. Grewe, A. Kruscha, B. Lindner, and J. Benda: Synchronous Spikes are Necessary but not
Sufficient for a Synchrony Code PNAS 114, E1977 (2017)

2018
M. Beiran, A. Kruscha, J. Benda and B. Lindner Coding of time-dependent stimuli in homogeneous and heterogeneous neural populations J. Comp. Neurosci. 44 , 189 (2018)

Last updated on 2020-18-03 at 23:14