Page 119 - Emotions through the eyes of our closest living relatives- Exploring attentional and behavioral mechanisms
P. 119

                                Self-scratch contagion observed in orangutans
consider yawning and self-scratching to be related to levels of increased arousal, but not aggression. Relaxed contexts were characterized by behaviors such as foraging, resting, or socio‐positive interactions (e.g., grooming). To ensure the reliability of our data, we restricted our data set to the indoor observations and excluded cases for which the expresser and observer were at a greater distance than 10 m.
The care and housing of the orangutans was adherent to the guidelines of the EAZA Ex situ Program. Only observational data were collected, therefore there was no need for the approval of the Ethics Committee of Apenheul. The study complied with the requirements of the Dutch Animal Care and Use Committee and conformed to the American Society of Primatologists Principles for the Ethical Treatment of Non‐ Human Primates.
Relationship quality
Scan‐sampling was performed every 30 min to score allogrooming, contact sitting,
social play, and sexual behaviors (e.g., mounting and genital contact) to calculate 5 relationship quality with a corrected composite sociality index (CSI; Silk et al., 2006).
Relationship quality was based on two levels: kinship and CSI (Demuru & Palagi, 2012;
Palagi et al., 2014). Regarding kinship, only maternal lineages were considered (r =
.5), resulting in four dyads. However, the dyad involving a juvenile male was excluded
from the analyses and only three kin dyads remained. One of these dyads was a
mother that, in the past, already had an offspring and took on the role of surrogate
mother for another juvenile of the same age as her own.
The CSI is a useful measure for scoring how much the positive relationship of a particular dyad deviates from the average of all dyads. We calculated it by summing up the following variables: frequency of grooming for a specific dyad divided by the mean frequency of grooming of all dyads, frequency of contact sit for a specific dyad divided by the mean frequency of contact sitting of all dyads, and frequency of days spent in the same compartment for the dyad divided by the total number of observation days. The summation of these variables was further divided by 3 (as per (Micheletta et al., 2013)). Since group composition for the orangutans was regularly changed and based on the preferences of the orangutans, we corrected for the total number of days spent together per dyad.
With the CSI score, we could identify high and low relationship qualities (Silk et al., 2006). Dyads with CSI scores in the top quartile were considered to have a high relationship quality, N = 5 (Demuru & Palagi, 2012), which included the kin dyads. Because of the low number of kin dyads, we did not separately test the influence of
 117




















































































   117   118   119   120   121