Fig. 3: Effects of antiandrogen treatment on the behaviour of dominant dams.
Antiandrogen treatment of matriarchs during late-pregnancy (LP) and early postpartum (PP) reduces hourly rates of a initiating food competition, b scent marking, c receiving prosocial overtures and d receiving submissive acts relative to control matriarchs. Dominant control (DC, purple); dominant treated (DT, orange). Mean ± s.e. on the left y axis; raw data on the right y axis (to accommodate zero inflation). Analyses used GLMMs, with two-tailed χ2, based on 32 pregnancies of 22 dominant dams, observed during 1044 focal sessions. e The percentage of adult females evicted also appeared to be reduced in DT relative to DC dams, but analysis was precluded. Lastly, treatment reduced f hourly rates (scans/partner/hr) of dyadic nearest-neighbour (N-N) associations. For the LMM analysis, matriarchs were matched across control and treated conditions in four clans, with sample sizes (n = 110) reflecting all possible adult dyads involving each matriarch, and a null expectation of no relationship between predictor terms and rate of proximity (P values represent two-tailed tests). Accordingly, relative to DC dams, DT dams were in close proximity to another adult clan member during ~1 fewer scan/individual/2 h observation (in a clan with a mean of 14 adults, this value corresponds to a predicted difference of ~8 N-N associations/h). Values in parentheses represent the numbers of a–e pregnancies or f matched clans sampled; significant P values in grey. Source data are provided as a Source Data file.
