Peter Loxley

Peter Loxley

Postdoctoral Research Fellow (current)
Center for Nonlinear Studies
Los Alamos National Laboratory
NM 87545
USA
Email: loxley@lanl.gov
CNLS web page

Postdoctoral Research Fellow (2005-08)
School of Physics
The University of Sydney
NSW 2006
Australia

Education

PhD (Physics), University of Western Australia, WA, Australia, 2005.

BSc Hons. (Physics), University of Sydney, NSW, Australia, 1998.

BSc (Chemistry) Murdoch University, WA, Australia, 1996.

Research

My background is theoretical physics, and my current research involves analysis of nonlinear phenomena in physical and biological systems. This includes the role of solitons, phase transitions, and competitive dynamics with feedback. Tools I find useful for these investigations include: field theory, statistical mechanics, energy descriptions and variational approaches, dynamical systems theory.

Condensed Matter Physics

Ferromagnetic systems of restricted dimension provide an ideal opportunity to investigate fundamental phenomena such as phase transitions. One interesting case includes the decay of a metastable state, where a ferromagnet becomes metastable in an applied magnetic field and magnetization reversal takes place via thermal activation over a barrier. In a 1D uniaxial ferromagnet the nucleation of a soliton-antisoliton pair provides the lowest activation barrier for reversal. However, in many realistic systems, it is known that nucleation takes place at a nonuniformity or defect, such as an irregularity in a sample. In collaboration with Robert Stamps at UWA, I have derived formulae for the rate of reversal due to the nucleation of solitons at various simple magnetic nonuniformities. I have shown that some nonuniformities can significantly lower the activation barrier for reversal, and can also modify the reversal rate prefactor.

Selected Publications:

P. N. Loxley, Phys. Rev. B 77, 144424 (2008):
"Rate of magnetization reversal due to nucleation of soliton-antisoliton pairs at point-like defects".
P. N. Loxley and R. L. Stamps, Phys. Rev. B 73, 024420 (2006):
"Theory for nucleation at an interface and magnetization reversal of a two-layer nanowire".
P. N. Loxley and R. L. Stamps, IEEE Trans. Mag. 37, 2098 (2001):
"Theory of domain wall nucleation in a two section magnetic wire".

Neural Networks of Visual Perception

The processing of visual information by the brain is thought to take place through various nonlinear interactions and feedback loops. For example, dynamics of the perceptual reversal of ambiguous figures such as the Necker cube can be modeled using a neural network with a form of competitive dynamics known as rivalry. In a winner-take-all process a single percept becomes dominant and suppresses all other percepts. During rivalry, it is generally thought that neurons encoding the dominant percept eventually fatigue after some period of time, allowing a previously suppressed percept to become dominant. Model predictions of this dynamic switching successfully match measurements of perceptual dominance durations from psychophysics experiments. In collaboration with Peter Robinson at USyd, I have developed an energy approach for understanding rivalry dynamics, and have investigated plausible biophysical mechanisms for the switching process. I have also developed a soliton model that describes the transition between binocular fusion and binocular rivalry.

Selected Publications:

P. N. Loxley and P. A. Robinson, Phys. Rev. Lett. 102, 258701 (2009):
"Soliton Model of Competitive Neural Dynamics during Binocular Rivalry".
P. N. Loxley and P. A. Robinson, Phys. Rev. E 76, 046224 (2007):
"Energy approach to rivalry dynamics, soliton stability, and pattern formation in neuronal networks".
P. N. Loxley and P. A. Robinson, Biol. Cybern. 97, 113 (2007):
"Spike-rate adaptation and neuronal bursting in a mean-field model of brain activity".

Last changed 23/6/2009