Table 2. Results from the linear mixed-effects model.

Significant p-values appear in bold. The significance of each level of a categorical variable was evaluated against the reference level (placed in parenthesis) according to whether their 95% confidence intervals (CI) include zero or not. ‘Pregnant_n’ denotes the nth trimester of pregnancy. To highlight that aggression rates can increase due to an increase in interactions of different score, we include a last column with the effect of some of the tested variables on overall adult female aggression rates, based on results of linear mixed effects models from Smit and Robbins, 2024. ‘ns’: non-significant correlation; ‘+’: positive correlation; ‘-’: negative correlation; ‘na’: not tested (see Smit and Robbins, 2024 for details).

Response variable: Interaction score (recipient-aggressor rank difference)						Aggression Rate (from Smit and Robbins, 2024)
Fixed factor	Level	Estimate	95% CI	Chisq	p-Value
Species (mountain)	Western	0.088	[–0.495; 0.671]	0.088	0.767	ns
Reproductive state (Cycling)	Pregnant_1	0.130	[0.101; 0.159]	138.812	<0.001	+*
	Pregnant_2	0.093	[0.063; 0.123]	138.812	<0.001	+*
	Pregnant_3	0.146	[0.114; 0.177]	138.812	<0.001	+*
	Lactating	0.043	[0.027; 0.059]	138.812	<0.001	-
Number of females		–0.011	[–0.018; –0.003]	8.045	0.005	+
Number of males		0.018	[0.010; 0.026]	19.784	<0.001	-
Aggression intensity (Mild)	Moderate	–0.030	[–0.051; –0.009]	8.531	0.005	na
	Severe	–0.014	[–0.036; 0.009]	8.531	0.229	na

^*Pregnancy was not divided in trimesters in Smit and Robbins, 2024.