Statistical evaluation of the numerical models.

Statistical evaluation (Kruskal-Wallis Test, followed by Dunn's Multiple Comparisons Test, * Plight condition Is shown. The lack of statistically significant difference between the model and experimental values was taken as a measure for the ability of the respective models to properly capture the features of the animal IPI distributions. The high and very high significance of the difference between the subtraction and probability model interactions of four CPGs to the experimental data clearly showed the inability of these models to describe the coupling of the pacemakers of T. cystophora adequately. Both types of models with two CPGs produced mean, median and standard deviations which were not significantly different from the experimental data. The hyperpolarizing subtraction model interaction of four pacemakers did so for the mean and median IPIs, while the standard deviation of the IPIs differed in a highly significant way.