On the interface of genetic, cultural and ecological dynamics: social learning and the emergence of foraging cultures

Prof. dr. Hogeweg

Results 2006

The fundamental research of this project focusses on interactions that arise when groups of individuals learn what to eat in a rich environment. The simulation model that we have developed over the past years has been devised in order to allow such interactions and the resulting complexity to arise. Our research fulfills a role in gaining a beter understanding of the complexity that arises when simple forms of learning occur in groups in complex environments.

Complex forms of social learning and large cultural variation are still thought to be one of the major differences between humans and non-humans. Our research shows that the most simple forms of learning (trial-and-error) already lead to culture in foraging groups. This suggests that cultural processes are likely to be prevalent in the foraging behaviour of many group foragers and questions the uniqueness of such processes to humans. By implementing rich complex environments in our simulations, we show that the environment plays a major role in shaping the outcomes of learning. In this way our results emphasise the importance of rich environments in the evolution of behaviour. This contrasts with the general focus on individual cognition by others.
The impact of the environment on the outcomes of learning is manifest from interactions that arise between learning, grouping and resource distributions. These can only be discovered if studied in an intergrated way as is done in our approach. Our counter-intuitive results arise easily and show the limited predictability of such complex systems. In addition, they demonstrate that social influences on learning and cultural transmission can be side-effects of grouping in certain environments. These insights are relevant for biology, evolutionary anthropology, cognitive psychology and artificial life.
From this point on, our research will proceed along the same lines. We have extended the model to include the adaptation of digestion capacities and observational learning. In this way we have expanded our research to gene-culture co-evolution and the role of different cognitive mechanisms.

Our results show how resource distributions can affect diet development via trial-and-error learning in group foragers. The main results are that homogeneously distributed resources lead to diet differentiation within groups due to local competition. In contrast, patchy environments allow individuals to share learning contexts and lead to group level diet similarity. Different groups develop different diets, however, because they follow different learning trajectories.
An important insight from these results is that trial-and-error learning is under social influence as soon as it occurs in groups and asocial learning no longer exists. Social learning is therefore merely a side-effect of grouping and does not require a selective advantage.
Taking the convergent social influence on learning that arises in patchy environments, we have studied whether the group level diets that arise can be inherited and lead to diet traditions. Our results show that indeed, trial-and-error learning in groups allows for diet traditions. Moreover, we find a cumulative cultural process that leads to an increase in diet quality. Hence we find to contrasting types of cultural phenomena, where traditional differences in diet arise mainly for low to intermediate levels of foraging selectivity, while cumulative change arises for high foraging selectivity. In the latter case a group-level selective process arises, which in positive feedback with individual-level selectivity, drives the cumulative process. We therewith present a highly parsimonous mechanism for the origin of cultural phenomena, namely that culture arises as a side-effect of grouping.
Given that we find such rich cultural behaviour already with trial-and-error learning, this begs the question of what the added cultural value of observational learning is. We have started to study this issue, together with Bas van Ursem, by adding observational learning in our model, but work is still in a very prelimenary stage. So far results suggest that even in the patchy environment, observational learning still allows for further convergence between individuals even though we had suggested it may be redundant in that environment.
We have also started work on the evolution on digestive systems with which we would like to study the coevolution between digestion and diet cultures. We have run preliminary simulations with an environment with a somewhat more natural structure in terms of digestion. We model fruits, seeds, leaves and insects according to their nutritional properties which then affect digestion. So far we have looked briefly at how digestive retention time adapts physiologically according to diets that individuals learn. Results seem to suggest that a given retention time, can stabilize diets through the way it affects resource preference development.

Our model has been formulated with primates in mind, but is sufficiently general to be relevant to other group foragers and should therefore have a wide appeal. Especially our findings on diet variation in different environments could provide search images for researchers in the field.
Moreover, in our study on diet traditions we find that traditional differences are found mainly on lower quality consumed resources. It is therefore possible that such a pattern of inter-group differences, but within-group convergence, on lower quality consumed resources could be a signature for diet traditions in natural group foragers.