Symbolic dynamics applied to optokinetic nystagmus signals

In this study symbolic dynamics is used to analyse the time evolution of the optokinetic nystagmus (OKN). The inter-saccadic nystagmus differences are transformed into a sequence of three equally numbered letters (symbols) which represents the temporal changes in the inter-saccadic signal: (a) a fall in the temporal change, (b) no changes and (c) a rise in temporal change. The complexity of the data series was then calculated as the entropy of the word length three probability distribution of the symbol sequence. The method was applied to OKN signals from ten healthy subjects and ten patients suffering from vertigo (four tests on each subject) and to 40 artificial white noise data series of the same length as the symbolic representation of the OKN data. Applying Student's t-test showed a statistically significant lower mean entropy value (p<0.05) for the patients.