You are here: MIMS > EPrints
MIMS EPrints

2006.27: Nonlinear time series analysis of jerk congenital nystagmus

2006.27: O. E. Akman, D. S. Broomhead, R. A. Clement and R. V. Abadi (2006) Nonlinear time series analysis of jerk congenital nystagmus.

Full text available as:

PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
411 Kb

Abstract

Nonlinear dynamics provides a complementary framework to control theory for the quantitative analysis of the oculomotor control system. This paper presents a number of findings relating to the aetiology and mechanics of the pathological ocular oscillation jerk congenital nystagmus (jerk CN). A range of time series analysis techniques were applied to both recorded jerk CN waveforms and simulated waveforms produced by an established model in which the oscillations are a consequence of an unstable neural integrator. The results of the time series analysis were then interpreted within the framework of a generalised model of the unforced oculomotor system.

This work suggests that for jerk oscillations, the origin of the instability lies in one of the five oculomotor subsystems, rather than in the final common pathway (the neu- ral integrator and muscle plant). Additionally, experimental estimates of the linearised foveation dynamics imply that a refixating fast phase induced by a near-homoclinic tra- jectory will result in periodic oscillations. Local dimension calculations show that the dimension of the experimental jerk CN data increases during the fast phase, indicating that the oscillations are not periodic, and hence that the refixation mechanism is of greater complexity than a homoclinic reinjection. The dimension increase is hypothe- sised to result either from a signal-dependent noise process in the saccadic system, or the activation of additional oculomotor components at the beginning of the fast phase. The modification of a recent saccadic system model to incorporate biologically realistic signal-dependent noise is suggested, in order to test the first of these hypotheses.

Item Type:MIMS Preprint
Uncontrolled Keywords:Nystagmus, delay embedding, time series analysis, nonlinear dynamics
Subjects:MSC 2000 > 37 Dynamical systems and ergodic theory
MSC 2000 > 92 Biology and other natural sciences
MIMS number:2006.27
Deposited By:Dr Mark Muldoon
Deposited On:10 March 2006

Download Statistics: last 4 weeks
Repository Staff Only: edit this item