Table 1.
Behavioral ethogram and categorization of behaviors.
Fig 1.
Duration of aggressive and subordinate behaviors of eventual dominant and subordinate individuals across 5 days.
Grey lines represent each individual and red lines represent mean duration of each group.
Fig 2.
Exemplar temporal pattern of aggressive and subordinate behaviors by the eventual dominant (D) and subordinate (S) males on days 1–5 (Dyad C and Dyad G).
Yellow bars represent individual bursts identified by the Kleinberg’s burst detection algorithm (gamma = 0.3).
Table 2.
Descriptive statistics of aggressive and subordinate behavior by day (medians and IQRs).
Fig 3.
Phi-coefficient of aggressive and subordinate behavior by burst number within each dyad.
Black dotted vertical lines represent when relationships are resolved according to phi-coefficient method. Red dots demonstrate significantly distinct levels of aggression versus subordinate behavior by each partner. Note that we excluded any bursts with less than 6 behavior bouts to determine the relationship resolution but those bursts are still shown in the graph.
Fig 4.
Differences in aggressive and subordinate behaviors between dominant and subordinate males by burst number within each dyad.
The red line indicates the difference of aggressive behavior (dominant male–subordinate male) and the blue line indicate difference of subordinate behavior (dominant male–subordinate male). In difference method, the relationship is considered resolved at a point when the blue line is always below red line for all remaining bursts. Black dotted vertical lines represent when relationships are resolved according to the difference method.
Fig 5.
Relative proportion of time each behavior is exhibited the by eventual dominant (red) and subordinate (blue) mice during each phase. Data are medians ± IQRs. The asterisks indicate p-values from paired Wilcoxon Signed Rank Tests; *: p<0.05, **: p<0.01, ***: p<0.001.
Fig 6.
Kinetogram of transitions between behaviors for (A) dominants pre-resolution, (B) dominants post-resolution, (C) subordinates pre-resolution and (D) subordinates post-resolution. All transitions that occur with a median probability of above 0.075 are shown. Label size represents the relative frequency of each behavior. Line weight represents the relative transition probability. Red lines indicate transitions that occur at rates significantly greater than expected by chance. Abbreviations: A–moving, AG–allogrooming, Bite–biting, Dig–digging, Flee–fleeing, Freeze–freezing, I–inactive, Lun–lunge, Pur–pursuing, R–rearing, SG–self-grooming, Sn-AG–anogenital sniffing, Sn-B–body-sniffing, Sn-F–sniff-following, Sn-H–head-sniffing, SP–subordinate posture, SS–side-by-side contact, TR–tail rattling.
Fig 7.
Forward STTC values with standard errors of nine selected contingencies in DOM→SUB (red) and SUB→DOM (blue) directions in pre- and post-resolution phases. The asterisks indicate significant differences between DOM→SUB and SUB→DOM directions. The differences in FSTTC values between the directions of each dyad were tested using paired Wilcoxon Signed Rank Tests. The asterisks indicate p-values from paired Wilcoxon Signed Rank Tests; *: p<0.05, **: p<0.01, ***: p<0.001.