New open access paper out! @K_nelds, @miss_daffodil, @CTennie, Grant, Tomaselli, & Nielsen @royalsociety Open Science: "A cross-cultural investigation of young children’s spontaneous invention of tool use behaviors"! https://royalsocietypublishing.org/doi/10.1098/rsos.192240 1/25

We are interested in the origins of human tool use and cultures. Why? Because tools have played a key role in human evolution, having allowed us to produce the cumulative culture we are surrounded by today (e.g., machines, books, medical and musical instruments). 2/25

To find out why human tool use and cultures are so special, particularly when compared to our closest living relatives, the great apes, lots of research has focused on SOCIAL LEARNING. 3/25

It has shown that our capacity and motivation for high-fidelity social learning allows us to learn and improve upon tool behaviors which we couldn’t have invented on our own. And so, we have gradually ratcheted up the efficiency and complexity of our technological culture. 4/25

Yet, to reach a more complete understanding of human tool use, we also need to know which tool-use behaviors humans can invent through INDIVIDUAL LEARNING, i.e., on their own, without the need for social learning. In other words: What is the baseline of human tool use? 5/25

Testing this baseline is challenging, as we are embedded in culture from early on. Theoretically, one would need an “Island test”: letting some babies grow up on a lone island, far away from any cultural influence and then testing their tool-use abilities as adults. 6/25

As the Island test has to stay theoretical, the next best option is to study humans who have relatively less cultural experience than human adults....Pre-school children! Of course they are fully immersed into culture, but they still have less exposure than adults. 7/25

So, we want to study which tool-use behaviors humans can invent on their own, without social learning or cultural knowledge. To achieve this, we can 1) study pre-schoolers and 2) give those children tool-use problems that they are unlikely to have encountered before 8/25

With these goals in mind, an earlier paper by @CTennie, Sarah Beck, Ian Apperly and @miss_daffodil @royalsociety studied which tool-use behaviours 2.5- to 3-year-old children could invent on their own. https://royalsocietypublishing.org/doi/pdf/10.1098/rspb.2015.2402 9/25

Individual children from the UK and Germany were presented with a test battery of 12 tasks (Great Ape Tool Test Battery - GATTeB), each consisting of an apparatus and a tool. In each task, children had to use the tool to retrieve a target object from the apparatus. 10/25

Importantly, children were not told that they had to use the tool – they were merely asked to get the target object out. So children had to use their individual learning abilities to solve the tasks. 11/25

The tasks were based on tool-use behaviors observed in wild chimpanzees and orangutans. We deemed these behaviors by our ape relatives as good candidate behaviors that young children might be able to invent on their own, while at the same time being novel. So what happened? 12/25

Children spontaneously invented 11/12 of the tool behaviors, indicating that these behaviors could be part of the human tool use baseline. 13/25

HOWEVER, studies based on a single population (here from Westernized, educated, rich milieus) are hardly representative. So, we suggested bringing the GATTeB to San Bushmen communities in South Africa to see how children living outside of this context engage with tasks. 14/25

So in our new paper we examined the tool use invention abilities of Bushmen children aged 2 to 5 years old in the Northern region of South Africa and a new sample of Westernized children from a large metropolitan city in Australia, using the GATTeB. 15/25

Results: We replicated the findings of Reindl et al. in another Westernized culture as well as in a less Westernized culture: children from 2 years of age spontaneously invented the correct tool-using behaviors. We found that: 16/25

1) Within each culture, children of all ages were equally likely to pick up and use a tool. Yet, older children were more likely to actually succeed in retrieving the target objects. This is probably due to better fine motor abilities. 17/25

2) Comparing the Westernized and the less Westernized sample, we found that tool use and success rates were higher in the Westernized sample. Why? 18/25

Maybe it had something to do with the tasks: Westernized children could have been more familiar with the task materials, some of which were very novel to the less Westernized kids. Or some social/environmental factor disadvantaged children from the less Westernized group. 19/25

3) We also replicated the finding that the tasks based on great ape behaviors observed more frequently in the wild were also easier for the children to invent, while tasks based on less frequent behaviors were more difficult to invent. 20/25

This suggests that certain tool behaviors are easier to invent (for all great apes, including humans), probably because the affordances of the task are more apparent, and to execute, probably because the actions involved are simpler. 21/25

To sum up: kids from diverse cultural settings and economic circumstances can independently invent the same set of tool behaviors observed in wild great apes. We conclude that these behaviors are part of the “baseline” of human tool use, which emerges from 2 years of age. 22/25

We also suggest that humans and other great apes may share similar physical cognitive capacities that allow for tool use invention. Does that mean we have ruled out all alternative explanations to the baseline argument once and for all? Not at all. 23/25

Tool behaviours that seem to occur universally in children of different cultures can still be culture-dependent (i.e., children may indeed use their prior cultural knowledge). And non-universal behaviours can still be part of the human baseline. It is a tricky one… 24/25

We thank Annette Hohenberger and two anonymous reviewers for very helpful comments on the paper and the communities we worked with – the !Xun, the Khwe and the ≠Khomani, Caritas creches and kindergartens, the Queensland Museum and @UQ’s Early Cognitive Development Centre. 25/25

Here is the link again: https://royalsocietypublishing.org/doi/10.1098/rsos.192240