Schiepek Günter, Strunk Guido
Institute of Synergetics and Psychotherapy Research, Paracelsus Medical University, Universitätsklinikum/Christian Doppler Klinik, Ignaz Harrer Str. 79, 5020 Salzburg, Austria.
Biol Cybern. 2010 Mar;102(3):197-207. doi: 10.1007/s00422-009-0362-1.
We introduce two complementary measures for the identification of critical instabilities and fluctuations in natural time series: the degree of fluctuations F and the distribution parameter D. Both are valid measures even of short and coarse-grained data sets, as demonstrated by artificial data from the logistic map (Feigenbaum-Scenario). A comparison is made with the application of the positive Lyapunov exponent to time series and another recently developed complexity measure-the Permutation Entropy. The results justify the application of the measures within computer-based real-time monitoring systems of human change processes. Results from process-outcome research in psychotherapy and functional neuroimaging of psychotherapy processes are provided as examples for the practical and scientific applications of the proposed measures.
波动程度F和分布参数D。正如逻辑斯谛映射（费根鲍姆情景）的人工数据所表明的那样，即使对于短的和粗粒度的数据集，这两种方法都是有效的测量手段。我们将其与将正李雅普诺夫指数应用于时间序列以及另一种最近开发的复杂性度量——排列熵进行了比较。结果证明了这些方法在基于计算机的人类变化过程实时监测系统中的应用合理性。心理治疗中的过程-结果研究以及心理治疗过程的功能神经成像结果作为所提出方法的实际和科学应用示例给出。
