Experimental Manipulation of Food Accessibility Affects Conflict Management Behaviour in Ravens

Abstract

Conflict management strategies such as reconciliation and bystander affiliation have been described for a variety of species. A common determinant seems to be a 'complex' social life, with individuals relying on affiliate relationships or social bonds. Little is known, however, about the strategic and flexible use of conflict management skills in experimental settings in species other than primates. We here investigated conflict and post-conflict behaviour of ravens by manipulating the accessibility of food and, thus, the likelihood of aggressive interactions while foraging. Specifically, we presented birds with a certain amount of highly preferred food that varied in the number of pieces (one piece, two pieces or, as a control, small pieces matching the number of participating birds) and observed their agonistic behaviour during feeding and their affiliative behaviour afterwards. The results showed that high levels of conflicts during feeding in the 1-piece and 2-piece conditions led to high levels of affiliation after feeding. Depending on the experimental condition, this effect is best explained (a) by the affiliative behaviour of former aggressors (1-piece condition) and (b) by the affiliation directed to the receivers of aggression after feeding (2-piece condition). Those dyads that engaged in allo-preening after feeding also engaged in allo-preening outside the experimental setting, suggesting that socially bonded individuals provided third-party affiliation to victims of aggression. Moreover, socially bonded ravens fed close to each other in the experiment when food was clumped, indicating that they actively coordinated their behaviour when there was a high conflict potential. Taken together, these findings support the assumption that ravens use their social bonds to avoid conflicts by choosing with whom to feed, and to buffer effects of conflicts by engaging in third-party affiliation as post-conflict behaviour.

Reconciliation and third‐party affiliation in carrion crows

“…We tested for the factor time of the day and did not find an effect. Additionally, the results of the present study are very similar to those of our previours study on ravens (Sima et al., ), making it unlikely that crows’ behaviours were affected by different factors (e.g., occurrence of feeding led to decrease in affiliation rates). Furthermore, the crows did not just respond to occurrence of food, regardless of aggression, but showed behavioural differences between the one‐piece and the two‐piece conditions.…”

“…The method used in the current study, however, allows to reinforce conflicts and to increase the chance of post‐conflict behaviours. In addition, our recent study on ravens (Sima et al., ) employed exactly the same method and reported similar findings as described in observational studies by Fraser and Bugnyar (, ) in ravens applying the traditional PC‐MC design.…”

“…We reported similar findings as observational studies on ravens (Fraser & Bugnyar, , ) using the traditional PC‐MC design. In our previous study on ravens (Sima et al., ), we also implemented a control condition during the feeding experiment (food which is not monopolisable). However, in the present study, we omitted the control because of time constraints faced during the rapid development of corvids in the first weeks post‐fledging.…”

“…So far, post‐conflict behaviours have been studied in four different corvid species: Eurasian jays (hereafter called jays; Garrulus glandarius ), jackdaws ( Corvus monedula ), ravens ( Corvus corax ) and rooks ( Corvus frugilegus ). While the use of third‐party affiliation has been shown in all four corvid species (Fraser & Bugnyar, ; Logan, Emery et al., ; Logan, Ostojić et al., ; Seed et al., ; Sima, Pika, & Bugnyar, ), only ravens seem also to show reconciliatory behaviour (Fraser & Bugnyar, ). One possible explanation is that the four corvid species differ in their use of post‐conflict behaviours.…”

“…To study post‐conflict behaviour, we employed a recently established method designed by Sima et al. (), linking the agonistic behaviour shown during the feeding experiment (30 min) with the affiliative behaviour following feeding (20 min). As a control, we used baseline behaviour recorded during daily life observations (individual focal protocols).…”