Response of a Hodgkin-Huxley neuron to a high-frequency input

Sensory processing by motoneurons: a numerical model for low-level flight control in flies

Rhythmic locomotor behaviour in animals requires exact timing of muscle activation within the locomotor cycle. In rapidly oscillating motor systems, conventional control strategies may be affected by neural delays, making these strategies inappropriate for precise timing control. In flies, wing control thus requires sensory processing within the peripheral nervous system, circumventing the central brain. The underlying mechanism, with which flies integrate graded depolarization of visual interneurons and spiking proprioceptive feedback for precise muscle activation, is under debate. Based on physiological parameters, we developed a numerical model of spike initiation in flight muscles of a blowfly. The simulated Hodgkin–Huxley neuron reproduces multiple experimental findings and explains on the cellular level how vision might control wing kinematics. Sensory processing by single motoneurons appears to be sufficient for control of muscle power during flight in flies and potentially other flying insects, reducing computational load on the central brain during body posture reflexes and manoeuvring flight.

Effect of autaptic activity on the response of a Hodgkin-Huxley neuron

An autapse is a special synapse that connects a neuron to itself. In this study, we investigated the effect of an autapse on the responses of a Hodgkin-Huxley neuron to different forms of external stimuli. When the neuron was subjected to a DC stimulus, the firing frequencies and the interspike interval distributions of the output spike trains showed periodic behaviors as the autaptic delay time increased. When the input was a synaptic pulse-like train with random interspike intervals, we observed low-pass and band-pass filtering behaviors. Moreover, the region over which the output ISIs are distributed and the mean firing frequency display periodic behaviors with increasing autaptic delay time. When specific autaptic parameters were chosen, most of the input ISIs could be filtered, and the response spike trains were nearly regular, even with a highly random input. The background mechanism of these observed dynamics has been analyzed based on the phase response curve method. We also found that the information entropy of the output spike train could be modified by the autapse. These results also suggest that the autapse can serve as a regulator of information response in the nervous system.

Bistability and resonance in the periodically stimulated Hodgkin-Huxley model with noise

We describe general characteristics of the Hodgkin-Huxley neuron's response to a periodic train of short current pulses with Gaussian noise. The deterministic neuron is bistable for antiresonant frequencies. When the stimuli arrive at the resonant frequency the firing rate is a continuous function of the current amplitude I(0) and scales as (I(0)-I(th))(1/2), characteristic of a saddle-node bifurcation at the threshold I(th). Intervals of continuous irregular response alternate with integer mode-locked regions with bistable excitation edge. There is an even-all multimodal transition between the 2 : 1 and 3 : 1 states in the vicinity of the main resonance, which is analogous to the odd-all transition discovered earlier in the high-frequency regime. For I(0)<I(th) and small noise the firing rate has a maximum at the resonant frequency. For larger noise and subthreshold stimulation the maximum firing rate initially shifts toward lower frequencies, then returns to higher frequencies in the limit of large noise. The stochastic coherence antiresonance, defined as a simultaneous occurrence of (i) the maximum of the coefficient of variation and (ii) the minimum of the firing rate vs the noise intensity, occurs over a wide range of parameter values, including monostable regions. Results of this work can be verified experimentally.

Response of Morris-Lecar neurons to various stimuli

We studied the responses of three classes of Morris-Lecar neurons to sinusoidal inputs and synaptic pulselike stimuli with deterministic and random interspike intervals (ISIs). It was found that the responses of the output frequency of class 1 and 2 neurons showed similar evolution properties by varying input amplitudes and frequencies, whereas class 3 neuron exhibited substantially different properties. Specifically, class 1 and 2 neurons display complicated phase locking (p : q, p > q, denoting output action potentials per input spikes) in low-frequency sinusoidal input area when the input amplitude is above their threshold, but a class 3 neuron does not fire action potentials in this area even if the amplitude is much higher. In the case of the deterministic ISI synaptic injection, all the three classes of neurons oscillate spikes with an arbitrary small frequency. When increasing the input frequency (both sinusoidal and deterministic ISI synaptic inputs), all neurons display 1 : 1 phase locking, whereas the response frequency decreases even fall to zero in the high-frequency input area. When the random ISI synaptic pulselike stimuli are injected into the neurons, one can clearly see the low-pass filter behaviors from the return map. The output ISI distribution depends on the mean ISI of input train as well as the ISI variation. Such different responses of three classes of neurons result from their distinct dynamical mechanisms of action potential initiation. It was suggested that the intrinsic dynamical cellular properties are very important to neuron information processing.

Multimodal transition and stochastic antiresonance in squid giant axons

The experimental data of Takahashi [Physica D 43, 318 (1990)], on the response of squid giant axons stimulated by periodic sequence of short current pulses is interpreted within the Hodgkin-Huxley model. The minimum of the firing rate as a function of the stimulus amplitude I0 in the high-frequency regime is due to the multimodal transition. Below this singular point only odd multiples of the driving period remain and the system is sensitive to noise. The coefficient of variation has a maximum and the firing rate has a minimum as a function of the noise intensity, which is an indication of the stochastic coherence antiresonance. The model calculations reproduce the frequency of occurrence of the most common modes in the vicinity of the transition. A linear relation of output frequency vs I0 above the transition is also confirmed.