An early-life challenge: becoming an older sibling in wild mandrills

Abstract

In monotocous mammals, most individuals experience the birth of a younger sibling. This period may induce losses in maternal care and can be physiologically, energetically and emotionally challenging for the older sibling, yet has rarely been studied in wild primates. We used behavioural data collected from a natural population of mandrills to investigate changes in maternal care and mother-juvenile relationship throughout the transition to siblinghood (TTS), by comparing juveniles who recently experienced the birth of a younger sibling, to juveniles who did not. We found that the TTS was associated with an abrupt cessation of the weaning process for the juvenile, and to a decrease in maternal affiliation. Juveniles' reactions were sex-specific, as males associated less with their mother, while females tended to groom their mother more often after the birth of their sibling. Despite the substantial loss of maternal care, juveniles did not show an increase in conflict or anxiety-related behaviours. This study contributes to explain why short interbirth intervals often pose a risk to juveniles' survival in monotocous primates. Our results contrast existing studies and further highlight the importance of examining the TTS in species and populations with various life histories and ecologies.

Keywords: mother–offspring conflict; mother–offspring relationship; sibling birth; sibling rivalry; transition to siblinghood; weaning.

Figure 1.

Probability to (a) suckle and (b) be carried by the mother during a focal observation according to juvenile’s age and sibling status, using raw data from 4866 focal observations on 191 juveniles. In all panels, “With” refers to juveniles who recently experienced the birth of a younger sibling and “Without” those who did not. For graphical purposes, we pooled values per age class so that “11” represent the mean probability to suckle/be carried for all juveniles aged 11–12 months old, “13”, juveniles 13 and 14 months old etc. In (a), “25” represents the pooled values for all juveniles older than 25 months old as suckling was no longer observed after this age, and in (b), “21” represents the pooled values for all juveniles older than 21 months old as carrying was no longer observed after this age. Dots represent the average raw value for a given age class and vertical bars represent the standard errors.

Figure 2.

Influence of the birth of the younger sibling on mother–juvenile grooming interactions. In all panels, ‘With’ refers to juveniles who recently experienced the birth of a younger sibling and ‘Without’ those who did not. (a) Predicted probability to be groomed by the mother depending on the sibling status during a focal observation. (b) Predicted probability to groom the mother depending on the sibling status and focal juvenile’s sex (females on the left side of the graph, males on the right). Dots are fitted values from the models (obtained with the function ‘fitted’ from the package stats), and boxplots show the median of the distribution of the fitted values (black horizontal bar), the 25th and 75th quartiles (bottom and top of the boxes, respectively) and the whiskers include a maximum of 1.5 times the interquartile range. The effect of the predictor ‘Sibling status’ and the associated p-values are shown. ‘ns’: not significant (p > 0.05), *: p < 0.05, **: p < 0.01, ***: p < 0.001.

Figure 3.

Influence of the birth of the younger sibling on mother–juvenile proximity. In all panels, ‘With’ refers to juveniles who recently experienced the birth of a younger sibling and ‘Without’ those who did not. (a) Predicted probability to be in body contact with the mother depending on the sibling status during a scan observation. (b) Predicted probability to be within 5 m around the mother depending on the sibling status and focal juvenile’s sex (females on the left side of the graph, males on the right). Dots are fitted values from the models (obtained with the function ‘fitted’ from the package stats), and boxplots show the median of the distribution of the fitted values (black horizontal bar), the 25th and 75th quartiles (bottom and top of the boxes, respectively) and the whiskers include a maximum of 1.5 times the interquartile range. The effect of the predictor ‘Sibling status’ and the associated p-values are shown. ‘ns’: not significant (p > 0.05), *: p < 0.05, **: p < 0.01, ***: p < 0.001.