Mechanisms explaining transitions between tonic and phasic firing in neuronal populations as predicted by a low dimensional firing rate model

Several firing patterns experimentally observed in neural populations have been successfully correlated to animal behavior. Population bursting, hereby regarded as a period of high firing rate followed by a period of quiescence, is typically observed in groups of neurons during behavior. Biophysical membrane-potential models of single cell bursting involve at least three equations. Extending such models to study the collective behavior of neural populations involves thousands of equations and can be very expensive computationally. For this reason, low dimensional population models that capture biophysical aspects of networks are needed. \noindent The present paper uses a firing-rate model to study mechanisms that trigger and stop transitions between tonic and phasic population firing. These mechanisms are captured through a two-dimensional system, which can potentially be extended to include interactions between different areas of the nervous system with a small number of equations. The typical behavior of midbrain dopaminergic neurons in the rodent is used as an example to illustrate and interpret our results. \noindent The model presented here can be used as a building block to study interactions between networks of neurons. This theoretical approach may help contextualize and understand the factors involved in regulating burst firing in populations and how it may modulate distinct aspects of behavior.

💡 Research Summary

The paper addresses a fundamental problem in computational neuroscience: how to capture the transition between tonic (steady) and phasic (bursting) firing in neuronal populations without resorting to high‑dimensional, biophysically detailed membrane‑potential models. The authors propose a two‑dimensional firing‑rate model that abstracts the collective dynamics of thousands of neurons into two state variables: the mean firing rate r(t) and an adaptation variable a(t) representing slow negative feedback (e.g., calcium‑dependent potassium currents). The governing equations are

 dr/dt = −r + S(w r − g a + I_ext − θ)

 da/dt = (r − a)/τ_a,

where S is a sigmoidal activation function, w is self‑excitation, g scales the adaptation feedback, I_ext denotes external drive, θ is a threshold parameter, and τ_a is the adaptation time constant. This “slow‑fast” architecture yields a system that can generate both stable fixed points (tonic firing) and limit‑cycle oscillations (bursting) depending on parameter values.

Using bifurcation analysis, the authors map how variations in I_ext and θ move the system across Hopf and Saddle‑Node on Invariant Circle (SNIC) bifurcations. At low I_ext the system possesses a single stable fixed point, corresponding to tonic firing. As I_ext crosses a critical value, a Hopf bifurcation destabilizes the fixed point and a stable limit cycle emerges, producing rhythmic bursts. Decreasing θ can trigger a SNIC bifurcation, leading to an abrupt onset of high‑frequency firing. The adaptation time constant τ_a controls burst duration: larger τ_a yields slower accumulation of a, prolonging the burst, whereas smaller τ_a quickly suppresses r, shortening or eliminating bursts.

To ground the model in biology, the authors fit the parameters to the firing patterns of midbrain dopaminergic neurons (DANs) in rodents, a well‑studied system that exhibits burst‑pause dynamics during reward‑related behavior. They interpret NMDA receptor activation as an increase in I_ext and D2 autoreceptor activation as a reduction in θ (or an increase in g). Simulations show that these manipulations reproduce experimentally observed transitions: NMDA‑driven bursts arise via Hopf bifurcations, while D2‑mediated suppression shifts the system back to the tonic regime. Parameter sweeps demonstrate quantitative control over burst frequency, amplitude, and inter‑burst interval, illustrating the model’s capacity to encode behavioral state changes.

The discussion highlights several strengths: (1) computational efficiency—only two differential equations are needed to emulate population‑level bursting; (2) analytical tractability—standard dynamical‑systems tools (nullclines, Jacobian eigenvalues, continuation methods) provide clear mechanistic insight; (3) extensibility—additional populations or coupling terms can be added with minimal increase in dimensionality, enabling the construction of larger network models. Limitations are also acknowledged: the model abstracts away spike‑time variability, ignores heterogeneous connectivity, and lacks direct parameter estimation from electrophysiological data. Future work is proposed to incorporate stochastic noise, heterogeneous subpopulations, and to couple multiple such modules to study cortico‑striatal‑midbrain loops.

In conclusion, the study demonstrates that a low‑dimensional firing‑rate framework, grounded in bifurcation theory, can faithfully reproduce the essential features of tonic‑to‑phasic transitions observed in real neuronal ensembles. By linking specific biophysical modulators (e.g., NMDA, D2 receptors) to model parameters, the work provides a bridge between experimental neurophysiology and theoretical analysis, offering a versatile building block for larger-scale models of brain function and for the design of neuromorphic systems that require burst‑based coding.