Who's the Boss? Assessing Convergent Validity of Aggression Based Dominance Measures in Male Laboratory Mice, Mus Musculus

Abstract

Aggression among group housed male mice continues to challenge laboratory animal researchers because mitigation strategies are generally applied at the cage level without a good understanding of how it affects the dominance hierarchy. Aggression within a group is typically displayed by the dominant mouse targeting lower ranking subordinates; thus, the strategies for preventing aggression may be more successful if applied specifically to the dominant mouse. Unfortunately, dominance rank is often not assessed because of time intensive observations or tests. Several correlates of dominance status have been identified, but none have been directly compared to home cage behavior in standard housing. This study assessed the convergent validity of three dominance correlates (urinary darcin, tube test score, preputial gland to body length ratio) with wound severity and rankings based on home cage behavior, using factor analysis. Discriminant validity with open field measures was assessed to determine if tube test scores are independent of anxiety. Cages were equally split between SJL and albino C57BL/6 strains and group sizes of 3 or 5 (N = 24). Home cage behavior was observed during the first week, and dominance measures were recorded over the second. After controlling for strain and group size, darcin and preputial ratio had strong loadings on the same factor, which was a significant predictor of home cage ranking showing strong convergent validity. Tube test scores were not significantly impacted by open field data, showing discriminant validity. Social network analysis revealed that despotic power structures were prevalent, aggressors were typically more active and rested away from cage mates, and the amount of social investigation and aggression performed by an individual were highly correlated. Data from this study show that darcin and preputial ratio are representative of home cage aggression and provide further insight into individual behavior patterns in group housed male mice.