Non-linear analysis of the rhythmic activity in rodent brains

This paper discusses the employment of non-parametric non-linear prediction algorithms to investigate non-linear dynamics in the rhythmic brain activity of rats. Three algorithms (Sugihara–May Simplex, K-neighbour and Casdagliʹs) were tested yielding similar prediction results which – when subject to a suitable bootstrap based t-tests – revealed that the theta waves recorded in rat brains cannot have their intrinsic non-linearity dismissed at a significance of 0.05.