CORDIS - EU research results

Efficiency and regulation of teaching under environmental constaints

Final Report Summary - TEACHING IN ANTS (Efficiency and regulation of teaching under environmental constaints)

In many animal species, individuals acquire information from others. Until recently there was no evidence for teaching in non-human animals. Now, teaching has been demonstrated in a small but diverse group of non-human animals, all being cooperative breeders. Because humans seem to have an additional sophistication that animals do not have, our focus on the underlying similarities among different taxa that achieve functionally similar outcomes, e.g. teaching, increases our understanding of the minimal criteria required for more complex behaviour.

Teaching in the tandem running ant, Temnothorax albipennis (Curtis), has been convincingly demonstrated by recent studies. The ant T. albipennis can forage for a resource such as food by two main strategies: (1) foraging alone or (2) teaching by tandem running, i.e. individual ants (leaders) teach naïve ants (followers). The goal was to examine whether a colony’s internal and external environment regulates the degree and success of teaching in the tandem running ant T. albipennis.

We have looked at the high level of communication between a teacher and her student. This allowed us to gain new insights about the efficiencies and degrees of teaching at the colony level and, on an extremely fine scale, at the individual level. We examined whether the foraging site (in terms of time, space, and quality) and internal state of the individual and of the group can regulate the degree (i.e. deployment) andefficiency of social information transfer. We used behavioral observations and mathematical modelling of the tandem running ant T. albipennis.

Teaching is expected to evolve where its costs are outweighed by the inclusive fitness benefits that result from the tutor’s relatives being more likely to acquire the valuable information. We found that unlike other theoretical studies. teaching is not only favoured where the pupil can’t easily acquire the information on its own but it highly depends on other factors such as the pupils’ and group’s internal state, e.g. starvation level. The perception of food starvation for the same time period of food deprivation can vary both between and within species. We have developed a simple method that can be applied to other social taxa, to estimate the perceived starvation level of the colony. The results of the study indicate that an increase in the perceived level of starvation elevates both individual information acquisition and teaching. We suggest that re-examination of the level of starvation, rather than the period of food deprivation, can resolve the inconsistencies among previous studies regarding whether low food availability increases or decreases the benefit of teaching (Gottlieb, et al. 2013).

Interestingly the starvation level of the individual and the colony did not only decrease the frequency of teaching the location of food but also was found to be a crucial factor determining the frequency of teaching the location of a new nest site. Both carrying behaviour, i.e. recruiting an individual without teaching and teaching via tandem runs are known to be an integral part of nest emigration in T. albipennis. The existence of carrying behaviour questions the role of teaching in the process of emigration. We found that although teaching occurs in the migration process, starvation can affect the quality of teaching; ants with a higher energy budget lead more tortuous tandem runs (Gottlieb, et al. in preparation). Furthermore, colonies that had to choose and emigrate to one of two new nests that differ in their quality showed that when not allowed to teach, the colony did complete the emigration process but had to compromise on the nest quality (Stuttard, et al. in preparation). Thus, suggesting that teaching has a crucial role in group’s decision making.

Increasing the difficulty of the teaching task did not affect the level of activity. Teaching in T. albipennis is a slow process. Thus, competition with an additional colony on the same food source may limit the time food is available. In order to increase the amount of food retrieved to the colony, teaching is expected to be less informative and more a means of rapid recruitment. The initial results suggest that the frequency of teaching does not differ between a competitive and non-competitive environment. However, competition does affect teaching characteristics, i.e. the communication between the follower and the leader. Given a fixed time/energy budget, every individual faces a fundamental trade-off between exploring for better resources and exploiting known resources to optimize overall performance under uncertainty. Although there has been much research on this trade-off, whether this trade-off affects the characteristics of teaching, has rarely been explored. The results indicate that teaching can handle the fundamental exploration-exploitation trade off successfully by implying an adequate design of the exploring option within the process of teaching.

Not all the conditions that affect the frequency of teaching behaviour also affect the dynamics between the leader and follower. Data on teaching characteristics was collected when knowledgeable ants taught the location of new source of food. Food was placed near and far from the ant’s colony. The probability of a naïve ant to find food located far from the nest is lower than to find food located near the nest. Nevertheless, there were no differences in frequency of teaching at both distances but we did find differences in teaching’s characteristics. We found that if food is located far from the colony the teacher will transfer the information more accurately to the naïve individual. The results further suggest that intentionally introducing error in the information transferred to the naïve individual can be adaptive and probably an explorative behaviour (Gottlieb et al., in press).

The ability to continuously explore novel domains of activity while teaching, most probably originated from a behaviour that is most commonly related to predators defined asincidental predation. Incidental predation occurs when secondary prey items are encountered and subsequently consumed, not through directed search for such prey but through their consequential encounter by a predator engaged in search for primary prey. This is suggested to be beneficial only when the abundance of prey is high. We suggest that the explorative behaviour might have probably evolved in an environment with low abundance of prey but high enough for small diversions from the direct path, i.e. exploration, to increase encounter rates.
We further suggest that although current evidence for explorative behaviour to a goal with known location is sparse, its rarity is more likely to reflect difficulty in data collection and interpretation. The same behaviour, if there is no good control, maybe often mistakenly interpreted as navigational constraint.

To broaden the understanding of the minimal criteria that are necessary for teaching to evolve and for teaching to be beneficial we are currently designing a mathematical model to see under what conditions teaching can evolve from “incidental predation” behaviour. Although there has been extensive research into related topics, such as the evolution of social learning, learned communication, and learned cooperation, currently there is no formal theory of the evolution of teaching. Thus, the results of this model and as such of this study, are expected to interest multiple academic disciplines, including evolutionary and behavioural biology, psychology, anthropology, archaeology, economics and education.

In summary, the program’s goal was to determine how, under different environmental conditions, simple rules can explain the evolution of teaching behaviour. To accomplish this goal, we used a multidisciplinary approach involving mathematical modelling and behavioural observation.
The topic of the study is a very recent and exciting idea that lies in what is today one of the most rapidly expanding areas of research – behavioural ecology and the evolution of human teaching. Thus, the results of this work (detailed above) contribute not only to the on-going debate as to whether teaching as expressed in humans exists in non-human animals, but also has a major impact in the field of behavioural ecology. In particular, the novelty and scientific quality of the research provides significant insights into the evolution of teaching and its implications on evolution adaptive decision-making. This study provides much data of general interest that will be published (six manuscripts in preparation) and is already published (Gottlieb et al., 2013) in top scientific journals.