An improved algorithm for the detection of dynamical interdependence in bivariate time-series

 We describe a new algorithm for the detection of dynamical interdependence in bivariate time-series data sets. By using geometrical and dynamical arguments, we produce a method that can detect dynamical interdependence in weakly coupled systems where previous techniques have failed. We illustrate this by comparison of our algorithm with another commonly used technique when applied to a system of coupled Hénon maps. In addition, an improvement of ∼20% in the detection rate is observed when the technique is applied to human scalp EEG data, as compared with existing techniques. Such an improvement may assist an understanding of the role of large-scale nonlinear processes in normal brain function. Received: 28 December 2001 / Accepted in revised form: 2 October 2002 Correspondence to: J.R. Terry (e-mail: J.R.Terry@lboro.ac.uk) Acknowledgements. JRT acknowledges the support of the Royal Society and the London Mathematical Society for jointly funding a trip to the School of Physics at the University of Sydney and The Brain Dynamics Centre at Westmead Hospital, Sydney. The authors thank Professor Peter Robinson for useful conversations and the provision of local facilities within the School of Physics during the visit of JRT. The support of the EPSRC via Grant GR/N00340 is also acknowledged. MJB is a recipient of a University Postgraduate Award from the University of Sydney.