Foraging dynamics are associated with social status and context in mouse social hierarchies

Abstract

Living in social hierarchies requires individuals to adapt their behavior and physiology. We have previously shown that male mice living in groups of 12 form linear and stable hierarchies with alpha males producing the highest daily level of major urinary proteins and urine. These findings suggest that maintaining alpha status in a social group requires higher food and water intake to generate energetic resources and produce more urine. To investigate whether social status affects eating and drinking behaviors, we measured the frequency of these behaviors in each individual mouse living in a social hierarchy with non-stop video recording for 24 h following the initiation of group housing and after social ranks were stabilized. We show alpha males eat and drink most frequently among all individuals in the hierarchy and had reduced quiescence of foraging both at the start of social housing and after hierarchies were established. Subdominants displayed a similar pattern of behavior following hierarchy formation relative to subordinates. The association strength of foraging behavior was negatively associated with that of agonistic behavior corrected for gregariousness (HWIG), suggesting animals modify foraging behavior to avoid others they engaged with aggressively. Overall, this study provides evidence that animals with different social status adapt their eating and drinking behaviors according to their physiological needs and current social environment.

Keywords: Association index; Circadian rhythm; Drinking; Eating; HWIG; Mouse social hierarchies; Social dominance; Social hierarchy; Social status.

Figure 1. Frequency of bouts of (A) eating and (B) drinking on the day of video recording across social ranks on the first day of group housing (Day 1) and after social hierarchies were established (Stable).

Social ranks were determined by individual Glicko ratings based on agonistic behavioral data collected until the day of the video recording. Boxplots show median (horizontal bars), interquartile ranges (boxes) and 95% confidence intervals (whiskers).

Figure 2. Total frequency of eating and drinking in 24 h of all individuals observed on Day 1 of hierarchy formation and after stable hierarchies were established.

(A, C) The first half of observation period was in dark cycle (minute 0 to 720) and the rest half was in light cycle (721 to 1,440). Mice eat and drink more frequently in the dark cycle both on Day 1 and after the hierarchies are established. (B, D) There was a significant interaction effect between dark/light phases and stability of hierarchies on eating and drinking frequency. Boxplots show median (horizontal bars), IQR (boxes) and 95% CI (whiskers).

Figure 3. The effect of social status on maximum length of inactivity (quiescence) in eating and drinking by social status group (A) on Day 1 and (B) in stabilized hierarchies.

Boxplots show median (horizontal bars), IQR (boxes) and 95% CI (whiskers). Raw data points of each group are also shown on the right side of each box plot. Significant differences between groups are shown; ^∗pMCMC < 0.05, ^∗∗ pMCMC < 0.01, ^∗∗∗pMCMC < 0.001.

Figure 4. Association indices of dyadic relationships in eating and drinking behavior.

(A) Foraging association indices are not different between those of alpha-other and other-other relationships. Boxplots show median (horizontal bars), IQR (boxes) and 95% CI (whiskers). Raw data points are also shown on the right side of each box plot. (B) The association strength of foraging behavior is negatively associated with the half-weight index corrected for individual gregariousness (HWIG) of agonistic interactions. The red line indicates the fitted trend line.