Experimental approaches to the study of early hominin technology and cognition using great apes as behavioural models

Abstract

Humans are a ubiquitous species on the planet and our success in adapting and transforming the environments we inhabit is arguably the result of our ability to successively improve traits across generations to increase our fitness (Henrich, 2017). This process is known as cumulative culture (Boyd & Richerson, 1996; Tomasello, 1999) and involves the modification and transmission of detailed information via the direct observation (e.g. copying) of a model's actions or products. Despite the growing interest in human cumulative culture, its origins are still debated. Some authors have proposed that early stone tools represent the earliest signs of cumulative cultural evolution in our lineage. According to this hypothesis, early hominins could not have learnt how to produce or use early stone tools individually without copying a model. A more parsimonious hypothesis regarding the learning mechanisms underlying the production and use of early stone tools could be derived from great apes (Tennie et al., 2009), who learn most of their tool behaviors individually without the need of copying a model (Chapter 2; Tennie et al., 2009). According to this hypothesis, early hominins acquired early stone tool making and using abilities via individual learning catalyzed by non-copying social learning mechanisms. Given that it is not possible to test early hominins directly in behavioral experiments, I investigated if and how two species of great apes (chimpanzees and orangutans) could acquire early stone tool making and using abilities in a series of baselines and social learning experiments (Chapters 2 to 5). The main finding from these experiments is that not all behaviors involved in the production and use of early stone tools are learnt in the same way by naïve, unenculturated chimpanzees and orangutans. Orangutans, but not chimpanzees, spontaneously perform percussive actions in which an active element is used to strike a hard surface. In one occasion, this behavior led to the detachment of sharp-edged stones from a flint core. In addition, one orangutan spontaneously used a human-made flake as a cutting tool to severe the lid of a baited puzzle box. Finally, after seeing demonstrations performed by a human model of how to make and use flakes, one orangutan and two chimpanzees engaged in percussive actions using an artificial hammer to strike a core. The results from these experiments show that certain behaviors such as the unintentional production of sharp-edged stone tools and the use of readily-made flakes as cutting tools can be individually learnt by orangutans. By phylogenetic proxy, such findings suggest that our last common ancestor with orangutans might have already presented the cognitive and physical abilities necessary to perform these behaviors 13 Ma. However, some behaviors such as the intentional production of sharp-edged stone tools for their subsequent use as cutting tools, seem beyond the individual and social learning abilities of ecologicallyrepresentative (unenculurated) chimpanzees and orangutans. It is therefore possible, that the intentional production of sharp-edged stone tools only emerged in our lineage when certain abilities (such as copying social learning mechanisms) evolved in our lineage. If this were to be the case, the intentional production of sharp-edged stone tools would represent the starting or catalyzing point of cumulative culture in our lineage.