Asya Shpiro

Balance between noise and adaptation in competition models of perceptual bistability

Perceptual bistability occurs when a physical stimulus gives rise to two distinct interpretations that alternate irregularly. Noise and adaptation processes are two possible mechanisms for switching in neuronal competition models that describe the alternating behaviors. Either of these processes, if strong enough, could alone cause the alternations in dominance. We examined their relative role in producing alternations by studying models where by smoothly varying the parameters, one can change the rhythmogenesis mechanism from being adaptation-driven to noise-driven. In consideration of the experimental constraints on the statistics of the alternations (mean and shape of the dominance duration distribution and correlations between successive durations) we ask whether we can rule out one of the mechanisms. We conclude that in order to comply with the observed mean of the dominance durations and their coefficient of variation, the models must operate within a balance between the noise and adaptation strength—both mechanisms are involved in producing alternations, in such a way that the system operates near the boundary between being adaptation-driven and noise-driven.

Self-consistent treatment of nonequilibrium spin torques in magnetic multilayers

It is known that the transfer of spin angular momenta between current carriers and local moments occurs near the interface of magnetic layers when their moments are noncollinear. However, to determine the magnitude of thetransfer, one should calculate the spin transport properties far beyond the interface regions. Based on the spin-diffusion equation, we present a self-consistent approach to evaluate the spin torque for a number of layered structures. One of the salient features is that the longitudinal and transverse components of spin accumulations are intertwined from one layer to the next, due to the presence of the much longer longitudinal spin-diffusion length and thus, the spin torque could be significantly amplified with respect to treatments which concentrate solely on the transport at the interface. We conclude that bare spin currents do not properly estimate the spin angular momentum transferred between the magnetic background; the spin transfer that occurs at interfaces should be self-consistently determined by embedding it in our globally diffuse transport calculations.

Mechanisms for Frequency Control in Neuronal Competition Models

We investigate analytically a firing rate model for a two-population network based on mutual inhibition and slow negative feedback in the form of spike frequency adaptation. Both neuronal populations receive external constant input whose strength determines the systems dynamical state-a steady state of identical activity levels or periodic oscillations or a winner-take-all state of bistability. We prove that oscillations appear in the system through supercritical Hopf bifurcations and that they are antiphase. The period of oscillations depends on the input strength in a nonmonotonic fashion, and we show that the increasing branch of the period versus input curve corresponds to a release mechanism and the decreasing branch to an escape mechanism. In the limiting case of infinitely slow feedback we characterize the conditions for release, escape, and occurrence of the winner-take-all behavior. Some extensions of the model are also discussed.

Maximum alternation rate in bi-stable perception occurs at equidominance: experiments and modeling

We present ambiguous stimuli, such as a pair of superimposed gratings exhibiting bi-stable depth ordering and binocular rivalry stimulus, to human subjects and measure the amount of time each of the two possible percepts dominates the other. Parameters of the stimuli are manipulated in order to strengthen gradually one of the percepts compared to the other, that is, to increase the fraction of time that this percept is dominant. For some choice of the parameters, the two percepts become equidominant: they dominate for the same fraction of time, or, equivalently, their mean dominance durations are the same. When the parameter which controls the strength of the percepts is varied from the equidominance value, its effect on the mean dominance duration for each percept is not the same. For the percept which becomes stronger, the mean dominance duration increases greatly in comparison to its value at equidominance, while for the percept which becomes weaker, the mean dominance duration is only slightly reduced. This result implies that as a function of the parameter that controls percept strength, the alternation rate between the two interpretations reaches a maximum at the equidominance point. We show that these features naturally arise in a bi-stable energy-based attractor model [1] where parameter manipulations produce symmetrical deformations of the energy landscape. Based on this formalism, we construct a two-population rate-based model with divisively normalized inputs, which produces behavior qualitatively similar to the one observed experimentally. A general class of neuronal competition models which describe rivalry during ambiguous stimulus presentation [2,3] exhibits the described behavior as well, as long as divisively normalized inputs are introduced. We calculate the entropy of a binary system in which probabilities are equal to the fraction of dominance of each state and show that it correlates with the alternation rate between the states in both experiments and models, suggesting that the alternation rate is a reflection of the uncertainty present in the system due to ambiguous stimulation.

Magnetization switching induced by spin polarized currents

Summary form only given. When a spin polarized current moves across a magnetically inhomogeneous structure, a spin accumulation will be produced in order to maintain a continuous flow of the spin current. This phenomenon has been identified as the primary origin of the CPP (currents perpendicular to the plane of the layers) giant magnetoresistance in magnetic multilayers. By considering the exchange interaction between the spin accumulation and the local magnetization, we have found that the spin accumulation is also a driving force for the current induced magnetization switching.

Transport in Metallic Multilayers with Both Specular and Diffuse Scattering at Interfaces

By using a rather simple model potential we calculate the influence of specular and diffuse scattering at interfaces on the resistance for currents perpendicular to the plane of the layers (CPP) in multilayered metallic structures. We find that if one maintains the same interfaces between layers, their contribution to the total resistance of the system depends on the thickness of the intervening layers. Their contribution is constant only if the distance between interface is larger than the electron mean free path in the bulk of the layers. The error incurred by neglecting the dependence of the interface resistance on layer thickness depends on the amount of diffuse scattering at interfaces and the height of the potential step at the interface. For a Fe-Cr or Co-Cu multilayer consisting of three or five layers with realistic interfaces where both specular and diffuse scattering may occur, the error reaches 15-20%; this exceeds the experimental error. We suggest systems where this effect may be observed.