NETWORK ACTIVITY ANALYSIS OF HYBRID NEURAL CIRCUITS USING PHASE RESETTING CURVES
Document Type
Presentation
Start Date
22-10-2010 10:45 AM
End Date
22-10-2010 12:00 PM
Description
The main goal of this study is to find the general principles leading to synchrony and phaselocked network activity in neural circuits. For networks of intrinsic neural oscillators (repetitively spiking neurons), phase locking can be predicted using a phase resetting curve (PRC) that measures the extent to which a perturbation at a given phase advances or delays the next spike. We use a simple circuit comprised of one model and one biological neuron reciprocally coupled via the dynamic clamp to test our theories. The PRC is measured under the assumption of pulsatile coupling for each isolated neuron. The two neuron circuits exhibit varying degrees of noisy phase locking and/or non-stationarity. Computational results indicate that a discrete map constructed based on the PRCs of the two neurons can qualitatively account for observed firing patterns in which no neuron fires more than twice consecutively. Slipping episodes in which one neuron fires an extra spike can be induced by noise or by a frequency mismatch. Adding noise to the map allows us to develop criteria to differentiate these cases. Further, assuming drift in the period of the biological neuron qualitatively reproduced the observed phase locking activity of circuits that exhibited drift induced bifurcations. We conclude that the measured PRCs can quantitatively predict network activity only when the period of the biological neuron remains relatively constant in the interval between when the PRC was measured and when circuit activity was observed, but can also predict the degree of robustness to noise and robustness to period drift (nonstationarity) expected for a given circuit.
Recommended Citation
Acuthan, Srisairam; Cul, J.; Butera, R. J.; and Canavier, C. C., "NETWORK ACTIVITY ANALYSIS OF HYBRID NEURAL CIRCUITS USING PHASE RESETTING CURVES" (2010). Dr. Joseph M. Moerschbaecher, III Graduate Research Day. 1.
https://digitalscholar.lsuhsc.edu/grad_rs/2010/poster1/1
NETWORK ACTIVITY ANALYSIS OF HYBRID NEURAL CIRCUITS USING PHASE RESETTING CURVES
The main goal of this study is to find the general principles leading to synchrony and phaselocked network activity in neural circuits. For networks of intrinsic neural oscillators (repetitively spiking neurons), phase locking can be predicted using a phase resetting curve (PRC) that measures the extent to which a perturbation at a given phase advances or delays the next spike. We use a simple circuit comprised of one model and one biological neuron reciprocally coupled via the dynamic clamp to test our theories. The PRC is measured under the assumption of pulsatile coupling for each isolated neuron. The two neuron circuits exhibit varying degrees of noisy phase locking and/or non-stationarity. Computational results indicate that a discrete map constructed based on the PRCs of the two neurons can qualitatively account for observed firing patterns in which no neuron fires more than twice consecutively. Slipping episodes in which one neuron fires an extra spike can be induced by noise or by a frequency mismatch. Adding noise to the map allows us to develop criteria to differentiate these cases. Further, assuming drift in the period of the biological neuron qualitatively reproduced the observed phase locking activity of circuits that exhibited drift induced bifurcations. We conclude that the measured PRCs can quantitatively predict network activity only when the period of the biological neuron remains relatively constant in the interval between when the PRC was measured and when circuit activity was observed, but can also predict the degree of robustness to noise and robustness to period drift (nonstationarity) expected for a given circuit.