Fig. 4: Urine mark allocation across scent-marked contexts.
From: Dynamic changes to signal allocation rules in response to variable social environments in house mice
a Total urine marks deposited by winning and losing males in an empty arena and the four urine-marked treatments: self-self (S-S), self-familiar male (S-FM), self-unfamiliar male (S-UM) and familiar male-unfamiliar male (FM-UM). All males experienced an empty stimulus-free arena. Each male also experienced one of the four urine-marked treatments. b Schematic of the urine stimulus components for the empty and urine-marked treatments. Empty trials have “no stimulus” (gray), S-S and S-FM have “no unfamiliar male” urine present (purple), and S-UM and FM-UM trials have “unfamiliar male” urine present (orange). c The difference in total marks deposited by males in the urine-marked trials relative to the empty trials (log-transformed). Urine-marked treatments are grouped as “no unfamiliar male” urine (purple: S-S and S-FM) and “unfamiliar male” urine (orange: S-UM and FM-UM). Post hoc pairwise comparison significance values are indicated at the top of boxplots. One-sample t-tests (deviation from 0) significance values are indicated on the bottom of the boxplots (green). d Urine mark density plots of losing and winning males toward an empty arena, and to trials with no unfamiliar male urine: S-S (light purple) and S-FM (dark purple), and to trials with unfamiliar male urine: S-UM (light orange) and FM-UM (dark orange). e Estimated marginal means plot of mark latency in the empty trials (gray), urine-marked trials with “No Unfamiliar” male urine (purple), and urine-marked trials with “Unfamiliar” male urine present (orange), given the fight outcome and initial signal investment (# Mesh1 marks). a, c Boxplot midline: median, box limits: upper and lower quartiles, whiskers: 1.5× interquartile range, points: outliers. a, c, e Linear mixed models were used to model relationships (M7-M9: Table S4), analyses of variance were used to test for overall effects, and post hoc pairwise comparisons were performed using the emmeans package (significance codes: NS p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001). Dependent variables were logarithmically transformed to meet assumptions for model residuals.
