TL;DR
This paper investigates how periodic external signals can synchronize or desynchronize populations of biological "bursting" neurons. By developing a mathematical framework based on discontinuous circle maps, the authors show that weak, slightly slower-than-natural input signals can effectively desynchronize these neurons, a phenomenon with implications for understanding neural network dynamics and therapeutic interventions.
What Problem Does This Paper Address?
Many biological systems, particularly in the nervous system, rely on rhythmic activity generated by neurons that "burst" – producing rapid sequences of electrical spikes followed by periods of silence. Understanding how these bursting neurons interact and synchronize is crucial for comprehending normal brain function and pathological conditions like Parkinson's disease, where abnormal synchrony is observed. A key question is: what kind of external input signals will cause these neurons to fire in unison (synchrony) versus independently (desynchrony)? This paper tackles this by examining the response of uncoupled bursting neurons to common periodic pulsatile inputs, aiming to predict and explain the conditions that lead to either entrainment or separation of their firing patterns.
Methods
The authors employ a multi-pronged approach:
- Phase Reduction: They adapt a standard technique called phase reduction, which simplifies the complex dynamics of bursting neurons into a one-dimensional "circle map." This map captures how the timing of an external input (a "kick") shifts the neuron's internal phase.
- Normal Form Model: They utilize a simplified mathematical model of elliptic bursting neurons (Izhikevich's normal form) for analytical tractability. This model captures the essential features of bursting, including the slow passage through a Hopf bifurcation, which is critical for the observed phenomena.
- Discontinuous Circle Maps: The core of their theoretical framework is the derivation of a "kick map" that is discontinuous. This discontinuity arises from the specific dynamics of elliptic bursters and is crucial for explaining complex behaviors.
- Analytical and Numerical Analysis: They combine analytical derivations of the kick map with numerical simulations of both the simplified model and a more biophysically detailed neuron model.
- Noise Modeling: They extend their analysis to include the effects of stochastic noise, investigating how it impacts the phase reduction and the resulting dynamics.
Datasets
Not discussed in the paper.
Key Insights and Ideas
- Weak Inputs Drive Desynchrony: Surprisingly, weak periodic inputs with frequencies slightly slower than the natural bursting frequency are most effective at desynchronizing a population of neurons. This is a counter-intuitive finding, as one might expect stronger inputs to have a more significant effect.
- Slow Passage is Key: The "slow passage effect" through a subcritical Hopf bifurcation is a critical mechanism. External inputs can alter the timing of this slow passage, leading to either entrainment or desynchrony.
- Discontinuous Circle Maps Capture Rich Dynamics: The derived discontinuous circle maps accurately predict a wide range of behaviors, including chaotic dynamics and phase locking, which are not easily captured by simpler models.
- Noise Modulates Dynamics: While noise can shorten slow passages and alter the shape of the kick map, it doesn't fundamentally change the principle that weak inputs can lead to desynchrony. However, the mechanism of desynchrony can shift from chaotic dynamics to high-period orbits combined with noise.
- Therapeutic Implications: The ability to desynchronize neurons with specific input patterns has direct relevance for therapies like Deep Brain Stimulation (DBS) for Parkinson's disease, where abnormal synchrony needs to be mitigated.
Key Results
- The authors derive a discontinuous circle map that accurately models the response of elliptic bursters to periodic pulsatile inputs.
- They identify parameter regimes (specifically, weak kicks with frequencies slightly slower than the natural burst frequency) that lead to population desynchrony.
- They demonstrate that their theoretical framework, based on the kick map, predicts the behavior observed in both a simplified normal form model and a more biologically realistic neuron model.
- They show that noise can alter the dynamics but generally preserves the desynchronizing effect of weak, slow inputs.
- An example is presented showing how carefully timed weak kicks can compete with a strong synchronizing input to desynchronize a population.
Limitations and Future Work
- Model Simplifications: While the normal form model is analytically tractable, it is a simplification. The authors acknowledge that more complex models might exhibit additional features.
- Generalizability: While the authors argue their findings are general for elliptic bursters, applying the framework to other bursting types or more complex network structures would be valuable.
- Parameter Space Exploration: The paper focuses on specific parameter regimes. A more exhaustive exploration of the parameter space could reveal further interesting dynamics.
- Biological Realism: The application to DBS is a brief demonstration. A more detailed, biologically complete model incorporating lateral connections and specific network architectures would be a next step.
- Arbitrary Inputs: The current framework primarily deals with periodic inputs. Extending it to analyze arbitrary sequences of inputs (non-periodic, varying amplitudes) is a promising future direction.
Summary updated