Abstract: The tendency in primate groups for two opponents to affiliate shortly after a fight has been described as dyadic reconciliation. The response has been shown to restore disrupted relationships and curtail ongoing aggression. Rates of self-directed behaviour (e.g. scratching) are positively correlated with anxiety in primates and the rates decline after reconciliation, indicating that the response also functions to reduce postconflict tension. Third parties not involved in an aggressive interaction are also likely to affiliate with one of the combatants subsequent to a fight. Such `triadic' interactions may also promote conflict resolution when, for instance, the relatives of a victim affiliate with their relative's aggressor. Because aggression in a group influences a bystander's behaviour with combatants, we hypothesized that aggression between two animals would also influence a bystander's behaviour with other bystanders. Such `quadratic' postconflict interactions might also function to reduce postconflict tension or occur in patterns among kin subgroups to resolve conflict. We tested for quadratic interactions in an 18-member group of captive hamadryas baboons, Papio hamadryas hamadryas. Immediately following a fight, an uninvolved bystander was randomly selected for observation and its affiliative interactions with other bystanders and its displacement activities were recorded for 3 min. Rates of behaviour during these postconflict periods were compared to rates during 3-min baseline periods not preceded by aggression. Bystanders engaged in quadratic interactions by increasing affiliation with other bystanders following aggression. Bystanders directed affiliation to nonkin bystanders that were their preferred social partners. Displacement activities of bystanders were significantly higher during postconflict intervals compared to baseline intervals, and bystander displacement activity levels before affiliative contact with other bystanders were significantly higher than after contact. Apparently, bystanders become tense or anxious after witnessing aggression and affiliate with preferred partners to reduce the arousal.

Abstract: Horses are well able to form classical and instrumental associations and so the focus of much recent research has been on the stimulus control of instrumental learning. Horses appear to discriminate using spatial cues more easily than other stimulus features, as indicated both by the speed of initial task acquisition and by the extent to which acquired discriminations can be reversed. Phenomena associated with discrimination learning in laboratory animals, including generalisation and peak shift, have been demonstrated in horses. However, the ability of horses to classify stimuli into categories is more controversial. Although there is some evidence that horses may be able to form categories based on similarities in the physical appearance of different stimuli, there is currently no evidence that they are able to develop abstract concepts. Their performance on social learning tasks has also been poor. Few correlations are observed between the learning ability of individual horses on different tasks, suggesting that it may not be possible to classify individual horses as `good' or `poor' learners. Better learning performance by horses that are naturally calm is probably due to reduced interference in the learning process. Correct handling procedures can lower reactivity levels in horses, and may facilitate learning in some circumstances. Future research on equine learning needs to take into account the complex nature of equine social interaction. Studies on the effects of stress on learning, and on social and spatial cognition, are also particularly needed.

Abstract: n many animal systems agonistic interactions may be rare or not overt, particularly where such interactions are costly or of high risk as is common for large mammals. We present a technique developed specifically for resolving an optimized dominance order of individuals in systems with transitive (i.e. linear) dominance relationships, but where not all relationships are known. Our method augments the widely used I&SI method (de Vries, 1998) with an interpolation function for resolving the relative ranks of individuals with unknown relationships. Our method offers several advantages over other dominance methods by enabling the incorporation of any proportion of unknown relationships, resolving a unique solution to any dominance matrix, and calculating cardinal dominance strengths for each individual. As such, this method enables novel insight into difficult to study behavioural systems.

Abstract: According to the socioecological framework, transitivity (or linearity) in dominance relationships is related to competition over critical resources. When a population is structured into groups, the intensity of between- versus within-group competition influences the form and function of its social organization. Few studies have compared the type and relative intensity of competition at these two levels. African elephants have well-structured social relations, providing an exemplary system for such a study. We report on dominance hierarchies among free-ranging elephants and evaluate the factors that drive their socioecological structure to lie in a region of the three-dimensional nepotism/despotism/tolerance space rarely observed among social species; namely, where non-nepotistic, transitive dominance hierarchies within groups emerge despite kin-based philopatry and infrequent agonistic interactions over widely distributed resources. We found significant transitivity in dominance hierarchies between groups. Dominance relations among the matriarchs of different social groups were primarily age based, rather than driven by physical or group size, and group matriarch rank influenced the dominance relationships among nonmatriarchal females in the population. Our results suggest that between-group dominance relationships induce tolerance among group members, which in combination with high group relatedness, reduces the benefits of nepotism. We postulate that cognitive abilities and high risk of injury in contests enhance winner and loser effects, facilitating the formation of transitive dominance relationships, despite widely distributed resources over which infrequent competition occurs. The interplay of cognitive abilities, winner and loser effects, resource distribution, and within- and between-group dominance relationships may produce behaviour in other strongly social mammals that differs from that predicted by a superficial application of current socioecological models.

Abstract: Methods of ranking individuals in a dominance hierarchy that use transitivity of relationships may obscure irregularities. Furthermore, these methods use only a small proportion of the information available from dominance encounters. This paper presents an intuitively appealing and easily implemented alternative to existing methods for ordering dominance data, developed from the work of Batchelder et al. (1992Journal of Mathematical Psychology36, 185-212). The procedure presented here is based on a mathematical model of paired comparisons and it involves only simple estimation procedures. We illustrate its use with data on dominance among red deerCervus elaphus, stags. The results indicate that dominance relationships are well characterized by the scale values that the model provides, and, because the method provides predictions for all pairings of animals, dominance predictions also exist for pairs of animals that have yet to be observed. Moreover, the dominance outcomes predicted by the model using the order scale are highly correlated with actual dominance observations at all levels. Overall, the procedure described provides a solution to the problem of identifying an appropriate order for a near-linear dominance hierarchy.