Using multi-animal tracking to gain deeper insight into rat social relationships

Jose J. Carmon-Sanchez, The University Of Montana

Abstract / Artist's Statement

All animals interact and form relationships with members of the same species, but we still lack an understanding of how relationships are formed and maintained. Here we investigated the behavioral patterns associated with social relationship renewal in adolescent female rats, by examining reunion behavior between two casemates after a period of separation. A crucial component of data analysis was the ability to track the movements of the rat. This is possible due to the advancements in machine-learning video analysis which can be seen with programs such as DIPLOMAT, developed at University of Montana through a collaboration between our lab and members of Computer Science. The tracking provides specific coordinates of certain body parts for each individual rat. Being able to know the precise location of the rats in a 2D arena allows for much more detailed analysis of the rat-to-rat interactions. Previous examination of these data suggest that rats that have been separated have only slight differences in interaction behavior relative to controls (e.g., more sniffing during the first few minutes). Preliminary examination of the multi-animal tracking shows that the software is working well and may offer new ways of understanding how animals approach each other after being separated. This project provides a unique early test of the multi-animal tracking technology and may pave the way for better understanding of social relationships.

 
Apr 21st, 3:00 PM Apr 21st, 4:00 PM

Using multi-animal tracking to gain deeper insight into rat social relationships

UC South Ballroom

All animals interact and form relationships with members of the same species, but we still lack an understanding of how relationships are formed and maintained. Here we investigated the behavioral patterns associated with social relationship renewal in adolescent female rats, by examining reunion behavior between two casemates after a period of separation. A crucial component of data analysis was the ability to track the movements of the rat. This is possible due to the advancements in machine-learning video analysis which can be seen with programs such as DIPLOMAT, developed at University of Montana through a collaboration between our lab and members of Computer Science. The tracking provides specific coordinates of certain body parts for each individual rat. Being able to know the precise location of the rats in a 2D arena allows for much more detailed analysis of the rat-to-rat interactions. Previous examination of these data suggest that rats that have been separated have only slight differences in interaction behavior relative to controls (e.g., more sniffing during the first few minutes). Preliminary examination of the multi-animal tracking shows that the software is working well and may offer new ways of understanding how animals approach each other after being separated. This project provides a unique early test of the multi-animal tracking technology and may pave the way for better understanding of social relationships.