Fractal measures in activity patterns: Do gastrointestinal parasites affect the complexity of sheep behaviour?

International audience

anglais. Gastrointestinal nematodes are known to be one of the most economically important parasites in livestock production. In order to test whether fractal analysis of behaviour can be used as a diagnostic tool for detection of infected animals, we investigated fractal patterns in the behavioural activity of domestic sheep (Ovis aries) in relation to strongylid infection. Temporal dynamics in activity patterns of 20 sheep were recorded at high resolutions using tri-axial accelerometer loggers attached to the neck of naturally infected subjects. We measured fractal dynamics in the resultant acceleration time series, divided into periods of activity and inactivity, using several fractal methods and tested the prediction that temporal complexity in the activity patterns of infected control sheep and experimentally dewormed sheep should differ. Detrended fluctuation analysis (DFA) revealed that sheep behaviour sequences were characterized by long-range correlations, meaning that acceleration fluctuations are not random but depend on long-term activity events. Generalized linear mixed models built to test for the effect of deworming on fractal estimates showed that the temporal organization of sheep activity varies with the status of strongylid infection. Our results indicate that sheep treated with anthelmintics exhibited a higher complexity in their activity sequences than parasitized sheep, suggesting that organizational patterns of their behaviour change with gastrointestinal parasite infection. Thus, we provide evidence for the potential utility of fractal methods in behavioural welfare monitoring.