Abstract
1.SummaryThe rapid decrease of light intensity is a potent and natural stimulus of rats’ activity. The nature of this activity, in that, the character of the social behavior and the composition of concomitant ultrasonic vocalizations (USVs) is unknown.This study, using deep learning algorithms, sought to assess the social life of rats’ pairs kept in the semi-natural conditions at the two twilight periods. Over six days, animals were video and audio recorded during the morning and the evening sessions lasting for 20-minutes each. The videos were used to train and use the DeepLabCut neural network examining animals’ movement in space and time. Numerical data generated by DeepLabCut were subjected to the Simple Behavioral Analysis (SimBA) toolkit, to build models of 11 distinct social and non-social behaviors. DeepSqueak toolkit was used to examine USVs.Deep learning algorithms revealed lights-off induced increases of fighting, mounting, crawling, and rearing behaviors, as well as of 22-kHz alarm calls and 50-kHz flat and short, but not frequency modulated calls. In contrast, lights-on stimulus increased the general activity, as well as adjacent lying (huddling), anogenital sniffing and rearing behaviors. The animals adapted to the housing conditions showing decreased ultrasonic calls, as well as of grooming and rearing behaviors, but not fighting.Present study shows lights-off induced increase of aggressive behavior but fail to demonstrate an increase in the positive affect defined by the hedonic USVs. We further confirm and extend the utility of deep learning algorithms in analyzing rat social behavior and ultrasonic vocalizations.Highlights- Rats display a natural increase in activity induced by lights-off stimulus- Deep learning algorithms allow for rapid characterization of social behavior and ultrasonic calls- The darkness increased aggressive behavior, 22-kHz alarm calls and 50-kHz flat and short, but not frequency modulated calls
Publisher
Cold Spring Harbor Laboratory