DOMINANCE HIERARCHY AND RECONCILIATION IN A GROUP OF TUFTED CAPUCHINS
INTRODUCTION
Friendly reunions between former opponents were first observed in a group of chimpanzees (Pan troglodytes) living in captivity (de Wall and van Roosmalen, 1979). The authors designated such post-conflict reunions “reconciliations”.
It has been stated that reconciliation between former opponents serves to repair the damage that the previous conflict caused to the relationship. This hypothesis is supported by the finding that reconciliation reduces the likelihood that the aggressor will engage in further aggression toward the victim or toward other animals and restores the mutual tolerance between former opponents in terms of sharing food resources. The most obvious short-term effect of aggression is dispersion. However, the mechanism of reconciliation results in a long-term tendency toward reunion after aggressive conflicts (de Waal, 1993).
The objectives of the present study were as follows: to determine whether there are mechanisms of reconciliation at work in a group of tufted capuchins (Cebus apella); to determine whether such mechanisms, if present, would be limited by operating within a hierarchical system, and, if so, to what extent.
METHODS
Subjects:
A captive group of 18 tufted capuchins housed on an island (25 m × 8 m) in a lake within the Parque Ecológico de Americana (Americana Ecological Park) in the city of Americana, located in the state of São Paulo.
Methods:
Data collection was carried out by two observers. In order to improve their judgment regarding behavioral events, the observers received training, consisting of in situ observation and the viewing of videos showing agonistic/affiliative events, prior to the collection of data. In order to determine the concordance between the judgments made by the two observers regarding the events registered, a test of interobserver agreement was conducted (k = 0.96; p < 0.00001). The group was observed from 7:30 am to 6:00 pm, three days per week for an intermittent period of two years, totaling 424.5 observation hours.
The sampling method consisted of registering all occurrences of conflicts and affiliative interactions during the observation period. In addition, after a conflict had ended, the observers used focal sampling to register any affiliative behavior presented by the aggressor during the first 5 min of the post-conflict period.
A conflict was defined as consisting of one or more agonistic behaviors aimed at another individual, according to the ethogram below:
|
AGONISTIC BEHAVIORS: 1- Threats:
2- Physical aggression:
|
Affiliative behavior was defined as any non-aggressive, non-assertive body contact with another individual.
|
AFFILIATIVE BEHAVIORS
|
A conciliatory gesture (reconciliation) was defined as the first affiliative contact between the aggressor and the victim during the first 5 min of the post-conflict period.
The dominance hierarchy was determined by registering two classes of behavior:
a) confrontations leading to the displacement of one of the individuals involved (the loser);
b) preference of access to food resources.
The results of the registration of these two classes were placed in a dyadic interaction matrix and submitted to ‘cluster analysis’ (distance measurement: Euclidean distance; amalgamation rule: nearest neighbor). The result (excluding infants) can be seen in Figure 1.
|
Key - A1 - Japanese (male) |
RESULTS
The table below summarizes the events observed.
Table 1. Summary of the affiliative and conflict interactions according to dyad ranking group
|
|
||||
| Conflicts | Post-conflict affiliative interactions |
Dispersion interactions |
Affiliative interactions unrelated to conflicts |
|
|
|
||||
| Distant hierarchical status |
422 | 134 | 288 | 172 |
| Mean per dyad n = 62 |
6.81 | 2.16 | 4.65 | 2.77 |
| Adjacent hierarchical status |
85 | 36 | 49 | 32 |
| Mean per dyad n = 14 |
6.07 | 2.57 | 3.5 | 2.29 |
| Equivalent hierarchical status |
44 | 35 | 9 | 15 |
| Mean per dyad n = 6 |
7.33 | 5.83 | 1.5 | 2.5 |
|
|
||||
| Total | 551 | 205 | 346 | 219 |
|
|
||||
When all dyads (regardless of rank) were considered, no significant difference was found between the total number of affiliative interactions that were unrelated to conflicts and the total number of post-conflict affiliative interactions (Wilcoxon matched-pair test: T = 844; Z = 1.31; p > 0.18). In addition, again considering all dyads (regardless of rank), dispersion interactions were found to be favored over affiliative interactions in post-conflict situations (Wilcoxon matched-pair test: T = 724; Z = 4.06; p < 0.000049). Therefore, overall, there seems to be no evidence of a mechanism of reconciliation operating in the group under study.
However, the analysis of the post-conflict affiliative interactions, in which all dyads were considered regardless of rank, revealed a non-uniform distribution (goodness-of-fit test: c² = 206.97; p < 0.000001).
Consequently, in an attempt to explain these results, the dyads were grouped according to the ranking of their components in the hierarchy: those comprising individuals of the same (equivalent) rank; those comprising individuals of ranks separated by only one position (adjacent ranks); and those comprising individuals of ranks separated by two or more positions (distant ranks).
The ranking group to which a given dyad belonged was found to correlate with the type of interaction (post-conflict affiliative or dispersion interaction), as shown in Figure 2 (maximum likelihood test: G = 39.107; p < 0.001).
Comparison of the samples of post-conflict interactions according to the ranking group of the dyads revealed that aggressors involved in conflicts with opponents of distant ranks tended to exhibit dispersion interactions more frequently than affiliative interactions (Wilcoxon matched-pair test: T = 95; Z = 5.65; p < 0.000001).
In contrast, aggressors involved in conflicts with opponents of adjacent ranks presented no preference of interactions in post-conflict situations (Wilcoxon matched-pair test: T = 37; Z = 0.16; p < 0.88). Finally, when the opponents were of the same rank, there was a tendency toward reconciliation after the conflict (Wilcoxon matched-pair test: T = 1; Z = 1.99; p < 0.05).
This pattern was confirmed through an analysis of linear trends, demonstrating a linear increase in the frequency of dispersion interactions for the dyads comprising individuals of distant ranks (regression coefficient, b = 0.237; Z = 6.405; p < 0.00001). A positive linear correlation was found between an increase in the frequency of dispersion interactions and the degree of difference in rank between the individuals involved in conflicts.
CONCLUSIONS
These findings strongly suggest that conciliatory behavior is dependent on dominance hierarchy, and that the way in which this behavior occurs is dependent on the rank held by the individual in that hierarchy. When a conflict occurs between individuals of different hierarchical status, the aggressor, if of dominant status, tends to adopt a despotic attitude toward the subordinate, rather than seeking reconciliation (cf. Aureli et al., 1997). In contrast, when the conflict involves individuals of equivalent status in the dominance hierarchy, reconciliation seems to serve as a means of repairing relationships.
In summary, although tufted capuchins employ the mechanism of reconciliation in order to prevent disintegration of the group, this mechanism is subordinate to a broader organizational form, the dominance hierarchy.
REFERENCES
Aureli, F.; Das, M.; & Veenema, H. C. (1997) Differential kinship effect on reconciliation in three species of macaques (Macaca fascicularis, Macaca fuscata, and Macaca sylvanus). J. Comp. Psychol. 111:91-99.
de Waal, F. B. M. (1993) Reconciliation among primates: A review of empirical evidence and unresolved issues. In: Mason, W. A. & Mendoza, S. P. (Eds.), Primates social conflites (pp. 111-144) SUNY Press: Albany.
de Waal, F. B. M., & van Roosmalen, A. (1979) Reconciliation and consolation among chimpanzees. Behav. Ecol. Sociobiol. 5: 55-66
Kappeler, P. M., & van Schaik, C. P. (1992) Methodological and evolutionary aspects of reconciliation among primates. Ethology 92: 51-69.
