Single-Neuron Properties and Network Dynamics: A new computational model shows when and how single neurons affect the activity of the entire network
Click to enlarge. Caption see below.
In a computational model they have studied the interaction between individual neuronal properties and the population dynamics in a neuronal network composed of these neurons. Their results have now been published in Scientific Reports, the online open-access journal of Nature Publishers.
In response to stimulation, neurons generate action potentials, also called spikes. These spikes are the signals neurons use to communicate with each other. Therefore, the temporal patterns of spiking activity are crucial to understanding how the brain works. These spiking patterns are just as numerous as the neuronal types: among others, their variety ranges from irregular single spikes, through rhythmic patterns of spikes, to actual bursting of spikes. Ajith Sahasranamam and colleagues now investigated the effect of changes in single neuron spike patterns on the activity of a network of such neurons. For this purpose, they devised a computational model of a neuron that replicates its bursting behavior without affecting any of its other properties.
“Our model conveyed that the bursting of single neurons had a significant effect on the population dynamics in the transition zone between the oscillatory and the aperiodic state of the network”, Sahasranamam explains. Network oscillations represent the synchronized activity of large numbers of neurons. By contrast, in aperiodic states individual neurons fire irregularly. “In the transition zone, bursting in the oscillatory network state destroys the oscillations, whereas in the aperiodic regime it lowers the threshold for the induction of oscillations,” he adds. “However, if the network states are completely oscillatory or fully aperiodic, the presence of bursting neurons hardly makes a difference.” These dynamics may also apply to other types of neuron properties.
Interestingly, recent experiments have shown that these dependencies also work the other way around: the bursting activity of single neurons depends on the activity of the network itself. “When we incorporate this dependence into our model, we find that the network dynamics exhibits hysteresis, that is, a dependence on its recent history. The response of the network does not only depend on its current input, but also on the previous states of the network activity,” says Sahasranamam.
Bursting activity and network hysteresis may have several important implications, both in normal physiological conditions and in pathology. For instance, they may play a role during development, given the fact that in newborns almost eighty percent of the neurons are bursting. Also, recent findings have indicated that the proportions of bursting neurons in Parkinson’s disease and epilepsy were altered. “Therefore, we can imagine that controlling the bursting activity in appropriate brain regions could offer a potential treatment for such diseases,” he concludes.
Caption:
The mexican wave serves as a metaphor for the transition zone between the oscillatory and the aperiodic state of a neuronal network: While it does not make a difference whether individual audience members remain seated within the wave or are standing outside it, it does make a difference on the edge of the wave.
Image "La Ola 01" Source: „La Ola during the cap Germany vers. Russia, on July 26th 2008 in the Lanxess-Arena of Cologne“ taken by Armin Kübelbeck - https://commons.wikimedia.org/wiki/File:La_ola_01.jpg
Licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license. - https://creativecommons.org/licenses/by-sa/3.0/deed.en
Für deutsche Version bitte hier klicken.
Contact:
Prof. Dr. Arvind Kumar
Bernstein Center Freiburg, University of Freiburg
Dept. of Computational Science and Technology
KTH Royal Institute of Technology, Stockholm. Sweden
Phone: +46 (0)8-790-6224
Fax: +49 (0)761 / 203 – 9559
E-Mail: arvkumar@kth.se
Web: www.kth.se/profile/arvindku/