On the role of oscillatory dynamics in neural communication

Resum

In this Thesis we consider problems concerning brain oscillations generated across the interaction between excitatory (E) and inhibitory (I) cells. We explore how two neuronal groups with underlying oscillatory activity communicate much effectively when they are properly phase-locked as suggested by Communcation Through Coherence Theory. In Chapter 1 we introduce the Wilson-Cowan equations (WC), a mean field model describing the mean activity of a network of a single population of E cells and a single popultation of I cells and review the bifurcations that give rise to oscillatory dynamics. In Chapter 2 we study how the oscillations generated across the E-I interaction are affect by a periodic forcing. We take the WC equations in the oscillatory regime with an external time periodic perturbation. We consider the stroboscopic map for this system and compute the bifurcation diagram for its fixed and periodic points as the amplitude and the frequency of the perturbation are varied. From the bifurcation diagram, we can identify the phase-locked states as well as different areas involving bistablility between two invariant objects. Chapter 3 exploits recent techniques based on phase-amplitude variables to describe the phase dynamics of an oscillator under different perturbations. More precisely, the applications of the parameterization method to compute a change of variables that describes correctly the dynamics near a limit cycle in terms of the phase (a periodic variable) and the amplitude. The computational method uses the Floquet normal form to reduce the computational cost. This change provides two remarkable manifolds used in neuroscience: the sets of constant phase/amplitude (isochrons/isostables). Moreover, we compute the functions describing the phase and amplitude changes caused by a perturbation arriving at different phases of the cycle, known as Phase and Amplitude Response Curves, PRCs and ARCs, respectively. The computed parameterization provides also the extension of these curves outside of the limit cycle, defined as the Phase and Amplitude Response Functions, PRFs and ARFs, respectively. We compute these objects for limits cycles in systems with 2 and 3 dimensions. In Chapter 4 we apply the parameterization method to compute Phase Response Curves (PRCs) for a transient stimulus of arbitrary amplitude and duration. The underlying idea is to construct a particular periodic perturbation consisting of the repetition of the transient stimulus followed by a resting period when no perturbation acts. For this periodic system we consider the corresponding stroboscopic map and we prove that, under certain conditions, it has an invariant curve. We prove that this map has an invariant curve and we provide the relationship between the PRC and the internal dynamics of the curve. Moreover, we link the existence properties of this invariant curve as the amplitude of the perturbation is increased with changes in the PRC waveform and with the geometry of isochrons. Furthermore, we also provide algorithms to obtain numerically the PRC and the ARC. In Chapter 5 we study the dynamics arising when two identical oscillators are coupled near a Hopf bifurcation, where we assume the existence of a parameter uncoupling the system when it is equal to zero. Using a recently derived truncated normal form, we perform a theoretical dynamical analysis and study its bifurcations. Computing the normal form coefficients in the case of 2 coupled Wilson-Cowan oscillators gives an understanding of different types of behaviour that arise in this model of perceptual bistability. Notably, we find bistability between in-phase and anti-phase solutions. Using numerical continuation we confirm our theoretical analysis for small coupling strength and explore the bifurcation diagrams for large coupling strength, where the normal form approximation breaks down. We finally discuss the implications of this dynamical study in models of perceptual bistability.