Modelling how cleaner fish approach an ephemeral reward task demonstrates a role for ecologically tuned chunking in the evolution of advanced cognition

What makes cognition “advanced” is an open and not precisely defined question. One perspective involves increasing the complexity of associative learning, from conditioning to learning sequences of events (“chaining”) to representing various cue combinations as “chunks.” Here we develop a weighted graph model to study the mechanism enabling chunking ability and the conditions for its evolution and success, based on the ecology of the cleaner fish Labroides dimidiatus. In some environments, cleaners must learn to serve visitor clients before resident clients, because a visitor leaves if not attended while a resident waits for service. This challenge has been captured in various versions of the ephemeral reward task, which has been proven difficult for a range of cognitively capable species. We show that chaining is the minimal requirement for solving this task in its common simplified laboratory format that involves repeated simultaneous exposure to an ephemeral and permanent food source. Adding ephemeral–ephemeral and permanent–permanent combinations, as cleaners face in the wild, requires individuals to have chunking abilities to solve the task. Importantly, chunking parameters need to be calibrated to ecological conditions in order to produce adaptive decisions. Thus, it is the fine-tuning of this ability, which may be the major target of selection during the evolution of advanced associative learning.

Cleaner fish are sensitive to what their partners can and cannot see

Much of human experience is informed by our ability to attribute mental states to others, a capacity known as theory of mind. While evidence for theory of mind in animals to date has largely been restricted to primates and other large-brained species, the use of ecologically-valid competitive contexts hints that ecological pressures for strategic deception may give rise to components of theory of mind abilities in distantly-related taxonomic groups. In line with this hypothesis, we show that cleaner wrasse (Labroides dimidiatus) exhibit theory of mind capacities akin to those observed in primates in the context of their cooperative cleaning mutualism. These results suggest that ecological pressures for strategic deception can drive human-like cognitive abilities even in very distantly related species.

Sex differences in the cognitive abilities of a sex-changing fish species Labroides dimidiatus

Males and females of the same species are known to differ at least in some cognitive domains, but such differences are not systematic across species. As a consequence, it remains unclear whether reported differences generally reflect adaptive adjustments to diverging selective pressures, or whether differences are mere side products of physiological differences necessary for reproduction. Here, we show that sex differences in cognition occur even in a sex-changing species, a protogynous hermaphroditic species where all males have previously been females. We tested male and female cleaner fish Labroides dimidiatus in four cognitive tasks to evaluate their learning and inhibitory control abilities first in an abstract presentation of the tasks, then in more ecologically relevant contexts. The results showed that males were better learners than females in the two learning tasks (i.e. reversal learning as an abstract task and a food quantity assessment task as an ecologically relevant task). Conversely, females showed enhanced abilities compared with males in the abstract inhibitory control task (i.e. detour task); but both sexes performed equally in the ecologically relevant inhibitory control task (i.e. ‘audience effect’ task). Hence, sex-changing species may offer unique opportunities to study proximate and/or ultimate causes underlying sex differences in cognitive abilities.

A role for ecologically-tuned chunking in the evolution of advanced cognition demonstrated by modelling the cleaner fish market problem

What makes cognition 'advanced' is an open and not precisely defined question. One perspective involves increasing the complexity of associative learning, from conditioning to learning sequences of events ('chaining') to representing various cue combinations as 'chunks'. Here we develop a weighted-graph model to study the conditions for the evolution of chunking ability, based on the ecology of the cleaner fish Labroides dimidiatus. Cleaners must learn to serve visitor clients before resident clients, because a visitor leaves if not attended while a resident waits for service. This challenge has been captured in various versions of the ephemeral-reward task, which has been proven difficult for a range of cognitively capable species. We show that chaining is the minimal requirement for solving the laboratory task, that involves repeated simultaneous exposure to an ephemeral and permanent food source. Adding ephemeral-ephemeral and permanent-permanent combinations, as cleaners face in the wild, requires individuals to have chunking abilities to solve the task. Importantly, chunking parameters need to be calibrated to ecological conditions in order to produce adaptive decisions. Thus, it is the fine tuning of this ability which may be the major target of selection during the evolution of advanced associative learning.

Effects of Group Behavior in The Predatory Raid on Damselfish Nests By The False Cleanerfish Aspidontus Taeniatus

The benefits of group behavior have been reported in a variety of animals. The false cleanerfish Aspidontus taeniatus, which resembles the bluestreak cleaner wrasse Labroides dimidiatus, is the best-known example of mimicry in vertebrates. This mimicry system has been viewed as an aggressive mimicry to bite fish fins. However, recent field studies have reported that large individuals of the false cleanerfish form groups and jointly raid fish nests to eat eggs that are guarded by their parents. Since the cleaner wrasse does not form such groups or specialize in egg-eating, the feeding groups of the false cleanerfish is assumed to reduce the effectiveness of mimicry. Here, we conducted field observations to clarify the functions of group behavior in egg-eating in the false cleanerfish. The false cleanerfish formed groups of 2–12 individuals when they raided breeding nests of 13 damselfish (Pomacentridae) and one triggerfish (Balistidae). The results showed that the group behavior has two effects: a dilution effect, which reduces the risk of being attacked by egg-guarding fish, and an increase in foraging efficiency. We conclude that the false cleanerfish need to form cooperative foraging groups during egg-eating because the egg-guarding parents could see through the mimicry.

Effects of the cleaner fish Labroides dimidiatus on grazing fishes and coral reef benthos

Territorial and roving grazing fishes farm, and feed on, algae, sediment, or detritus, thus exerting different influences on benthic community structure, and are common clients of cleaner fish. Whether cleaners affect grazing-fish diversity and abundance, and indirectly the benthos, was tested using reefs maintained free of the bluestreak cleaner wrasse Labroides dimidiatus for 8.5 yr (removals) compared with controls. We quantified fish abundance per grazing functional group, foraging rates of roving grazers, cleaning rates of roving grazers by L. dimidiatus, reef benthos composition, and fouling material on settlement tiles. Abundances of ‘intensive’ and ‘extensive’ territorial farmers, non-farmers, parrotfishes and Acanthurus spp. were lower on removal than control reefs, but this was not the case for ‘indeterminate’ farmers and Ctenochaetus striatus. Foraging rates of Acanthurus spp. and C. striatus were unaffected by cleaner presence or cleaning duration. This suggests some robustness of the grazers’ foraging behaviour to loss of cleaners. Acanthurus spp. foraged predominantly on sediment and detritus, whereas C. striatus and parrotfishes grazed over algal turfs. Nevertheless, benthic community structure and amount of organic and inorganic material that accumulated over 3.5 mo on tiles were not affected by cleaner presence. Thus, despite greater abundances of many roving grazers, and consequently higher grazing rates being linked to the presence of cleaners, the benthos was not detectably affected by cleaners. This reveals that the positive effect of cleaners on fish abundance is not associated with a subsequent change in the benthos as predicted. Rather, it suggests a resilience of benthic community structure to cleaner-fish loss, possibly related to multiple antagonistic effects of different grazer functional groups. However, losing cleaners remains a problem for reefs, as the lack of cleaning has adverse consequences for fish physiology and populations.