What makes a cleaner a cleaner ?

In his last presidential address to the Royal Society in 2005, Robert M. May stated that “The most important unanswered question in evolutionary biology, and more generally in the social sciences, is how cooperative behaviour evolved and can be maintained”. My thesis provides a contribution to answering this big question by investigating how one particular species evolved for an ecology that rests heavily upon cooperative interactions: the bluestreak cleaner wrasse Labroides dimidiatus. This species engages in up to 2000 cooperative interactions per day with dozens of other ‘client’ coral reef fishes, thus making it a prime system for studying cooperation between unrelated individuals. Conflicts between cleaners and clients arise because the former prefer to bite clients to eat their protective mucus rather than focusing on ectoparasites. In response to such exploitation, clients use various forms of partner control mechanisms that promote cooperative behaviour in cleaners. As a result, cleaners are known to use a diversity of strategic behaviours to determine when to cheat and when to cooperate. Cleaners’ behaviour thus appears very well adapted to the demands of their peculiar ecology. In my thesis, I used a comparative approach to investigate which traits appear to be associated with the highly social ecology of cleaners, or in other words: What makes a cleaner a cleaner? Through a series of experiments, I compared the cognitive skills, escape performance, foraging ecology and vision of L. dimidiatus with other species of labrids that do not engage in cleaning (or only occasionally). These comparisons allowed me to identify some of the characteristics that set cleaners apart, and thus further our understanding of how social ecology can affect the evolution of a species. In the first chapter, I showed that L. dimidiatus was able to fine-tune its level of cooperation to the specifics of different cleaning-related situations, an ability that was absent in the closely related Halichoeres melanurus. In the second chapter, I investigated whether the demands of a highly social lifestyle led to an overall increase of cognitive performance in L. dimidiatus, or whether performance was tightly linked to ecological demands. L. dimidiatus outperformed five other species of labrids in two ecologically relevant tasks. However, all species performed similarly in a task with little ecological relevance, suggesting that cognition in cleaners is tightly linked to the challenges faced in nature. In the third chapter, results from foraging experiments suggest that L. dimidiatus evolved a foraging position that allows for increased efficiency in cleaning interactions. In comparison with four other species, L. dimidiatus adopted a significantly lower body angle with regards to the substrate when foraging. Furthermore, this species experienced almost no reduction in efficiency when searching for cryptic food items in comparison with conspicuous ones, while the other species all performed worse in the cryptic condition. It thus appears that the peculiar foraging posture of L. dimidiatus is well suited for cleaning interactions. Unfortunately, my current data did not allow assessing whether the visual system also adapted for a low body angle while foraging. Finally, in the last chapter, I asked whether because of the service they provide to predators, a reduction in predation pressures led to the decay of escape performance in L. dimidiatus. Interestingly, measures of escape performance in a controlled laboratory setup showed that L. dimidiatus was among the top performers in comparison with 5 other labrids. These results suggest that the risks associated with cleaning interactions are sufficient to maintain a high escape performance in cleaners, despite their privileged relationship with predatory clients. In summary, a wide range of characteristics appear to be important for cleaning interactions, and species that specialized in this activity seem to have undergone very different selective pressures than fishes with more standard ecologies. Identifying some of the key aspects related to the ecology of cleaners provides a good example of how the evolution of a species can be affected by the demands of a highly social life. In this system, I argue that competition among service providers and conflicts of interests between cleaners and clients appear to be the major drivers of adaptation