Power law and competitive model for behavioral characterization of incompatible spontaneous behavior in Drosophila melanogaster.

The present report aims to examine models that best describe the temporal structure of walking and immobility in Drosophila melanogaster, using data from a public repository (Maesani et al., 2015). Two candidate models were evaluated: a power-law distribution, which implies criticality in the neural networks underlying these behaviors, and a geometric distribution, which suggests stochastic competition between neural circuits governing two mutually exclusive motor states. The results indicate that both models provide satisfactory fits to the observed spontaneous walking-resting behavior. Both models provide an alternative framework for understanding internal neural mechanisms through the observation and quantification of overt behavior. These findings underscore the importance of testing alternative models to more accurately characterize the underlying dynamics of behavioral organization.