This title appears in the Scientific Report :
2015
The anatomical origin of locally generated and induced oscillations in a model of the cortical microcircuit
The anatomical origin of locally generated and induced oscillations in a model of the cortical microcircuit
Fast oscillations of the population firing rate in the high gamma range (50-200 Hz), as well as slow firing rate fluctuations are ubiquitous in cortical recordings and have been hypothesized to be generated locally []. Exploiting the recently introduced multi-layered spiking neural network model of...
Saved in:
Personal Name(s): | Bos, Hannah (Corresponding author) |
---|---|
Schücker, Jannis / Diesmann, Markus / Helias, Moritz | |
Contributing Institute: |
Computational and Systems Neuroscience; INM-6 Computational and Systems Neuroscience; IAS-6 |
Imprint: |
2015
|
Conference: | SfN, Chicago (USA), 2015-10-17 - 2015-10-21 |
Document Type: |
Poster |
Research Program: |
Supercomputing and Modelling for the Human Brain Theory of multi-scale neuronal networks Helmholtz Alliance on Systems Biology Connectivity and Activity Connectivity and Activity |
Publikationsportal JuSER |
Fast oscillations of the population firing rate in the high gamma range (50-200 Hz), as well as slow firing rate fluctuations are ubiquitous in cortical recordings and have been hypothesized to be generated locally []. Exploiting the recently introduced multi-layered spiking neural network model of a cortical microcircuit [2] as well as our mean-field theoretical framework, we address the question of the anatomical origin of the observed oscillations by analyzing the spectra generated in the resting state condition as well as under the application of constant and oscillatory input.Deriving the theoretical framework we perform a two-step reduction allowing for an incremental validation of first the prediction of the population firing rates and second the prediction of the population rate spectra. Building on previous work deriving the mapping of populations of leaky integrate-and-fire neurons to a linear rate model [] and the response function of populations of neurons connected by exponentially decaying synapses [1,3], the mean-field framework is applicable to optional circuitries set in the asynchronous irregular regime. In the resting condition the neurons in the model fire irregularly, displaying little synchrony on the population level. Increasing the external input to the excitatory population in layer 5 elicits slow rate fluctuations, reflected as elevated slow frequency components in the population rate spectra. Strengthening the input to the superficial layers triggers population oscillations in the gamma range, while the individual neurons preserve a firing pattern close to irregularity with low rates. We derive a sensitivity measure determining the anatomical connections within the circuit crucial for the generation of the peaks visible in the power spectra as well as their impact on and significance for the frequencies and amplitudes. We identify a sub-circuit located in layer 2/3 and 4 constituting the basis of the high frequency oscillations, while connections within and onto layer 5 determine the existence and strength of slow rate fluctuations. Since the sensitivity measure is derived from the mean-field theory we can analyze the robustness of these findings under changes of parameters in the neuron and synapse model and conclude on the actual contributions of the anatomical connections given their embedding in the full circuit.Exploiting the mean-field framework we generate predictions regarding changes in the power spectra under various stimulation protocols. Analyzing the responses of the circuit to oscillatory input we formulate predictions addressing the susceptibility of individual populations to specific frequencies. |