Periodic Reporting for period 4 - RATCHETCOG (The Cog in the Ratchet: Illuminating the Cognitive Mechanisms Generating Human Cumulative Culture) Reporting period: 2020-03-01 to 2021-02-28 Summary of the context and overall objectives of the project In human populations, skills and knowledge accumulate over generations, giving rise to behaviours and technologies far more complex than any single individual could achieve alone. This ratchet-like property of human culture appears absent in nonhuman species, as socially transmitted behaviours in animal populations are generally no more complex than those that can be acquired by trial and error. Although there were many high-profile speculative theories for the distinctiveness of this feature of human culture, these were based on very limited evidence. In the RATCHETCOG project, we implemented a comprehensive systematic investigation into cumulative cultural evolution, using experimental methods developed for the project which enabled data collection across three critical research domains: nonhuman primate behaviour; children’s development; and adult human learning. The methods that we devised within the project allowed us to both measure and manipulate the complexity of transmitted behaviours, and as such they provided a powerful tool for analysing the extent of ratcheting under different conditions and across different populations. Each of the three research strands has provided a vital source of evidence. Our studies of nonhuman primates have revealed that ratcheting is theoretically possible in these species, but also that under natural conditions the potential for ratcheting is likely severely constrained due to low-precision use of information, especially under conditions involving cognitive load. Our studies with children have highlighted the importance of the development of general-purpose cognitive capacities such as memory in determining whether solutions to certain problems can accumulate improvements with social transmission. These have also shown that some capacities that are likely to be unique to humans, such as explicit metacognitive understanding, play an important role in the effective use of social information under certain conditions. Finally, by comparing different learning conditions in adults, we have been able to elucidate contexts in which advanced cognitive capacities such as explicit metacognition and theory of mind are (and are not) likely to play a significant role in facilitating ratcheting. Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far We developed a novel method (WP1) for experimental testing of capacities for cumulative culture (described in Caldwell et al., 2020, WIRES Cogn Sci). This method used a stimulus selection paradigm which we first validated using straightforward comparisons of the use of information acquired vicariously, versus through one’s own activity. This included a cross-cultural study of human children, (WP2: Atkinson et al., 2020, J Exp Psychol Gen) and studies with nonhuman primates (WPs 6&7: Renner et al., 2019, PeerJ; Renner et al., 2021, Sci Rep). We further developed this task to investigate the developmental progression of children’s capacity for cumulative culture (WP2), finding that children as young as 3 years can learn from a social demonstration in a manner that would support the accumulation of beneficial modifications, as long as there was no memory load. However, when memory was required, only older children (5-6 years) derived greater benefit from high-scoring, compared with low-scoring, demonstrations. This showed that the expression of cumulative culture depended on both the cognitive capacities of the learners and the specific demands of the task (reported in Wilks et al., 2021, J Exp Child Psychol). In further studies of human children we investigated other cognitive demands likely to be implicated in many real world cases of cumulative culture (WP3). These included studies in which participants were required to: make inferences based on others’ reactions along with a knowledge of their goals (Blakey et al., revision under review, J Exp Child Psychol); understand others’ visual perspective (Blakey et al., revision under review, J Comp Psychol); and direct their learning towards those who possessed relevant information (Blakey et al., 2020, PsychArXiv). In all cases we found that these demands placed significant constraints on the use of social information by very young children (e.g. under 5 years). We used further variations of our stimulus selection task to study potential constraints on cultural transmission in adult human social learning (WPs 4&5). We found that under certain conditions, intentional knowledge transmission by cultural parents strongly facilitated the accumulation of beneficial information in chains of adult learners (Mackintosh et al., in prep). In contrast, we found little evidence to suggest that learners make inferences about others’ past behaviour, even when accurate inferences were theoretically possible and could boost performance (Atkinson et al., 2020, Cogn Sci). Finally, we found some suggestion of a possible role for “System 2” cognition in facilitating selective social learning in human adult participants (Dunstone et al., 2021, PloS One). In studies with nonhuman primates (WPs 6&7) we found that baboons and capuchin monkeys were, like very young children, in principle capable of learning from vicarious information in a manner which would lead to the accumulation of beneficial modifications over cultural transmission (Kean et al., in prep). However, it is likely that the conducive learning conditions required would rarely be encountered in the real world, and that the accumulation of benefits would only occur over a very limited number of generations of transmission. Overall we conclude from the findings of the project that there is no single factor precluding cumulative culture in nonhuman species. Rather, a wide variety of cognitive challenges are associated with the use of social information in most real-world cases of cumulative culture, and it is these (largely context-specific) challenges which constrain the expression of cumulative culture in other species. Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far) At the outset of the project, there were many different theories purporting to explain the distinctiveness of human cumulative culture, each focussed on relatively singular cognitive capacities or behavioural traits. However, there was little empirical evidence, largely as a result of the practical difficulties associated with collecting data on a large scale, with special populations, using the (limited) existing methods. The RATCHETCOG project aimed to develop a novel method which would significantly reduce the logistical obstacles associated with research on cumulative culture. We successfully achieved this, and as a result we were able to generate a rich body of evidence on a variety of factors promoting and inhibiting cumulative culture. As discussed above, our results suggest that there are numerous cognitive abilities which may be required in order for social information use to result in an accumulation of beneficial information. However, the specific abilities required will vary according to the situation. Our results with human children (described above) suggest that there are many challenges that may be encountered in real world social learning which mean that effective information use is outside of the capabilities of children under the age of five. These are also likely to be outside of the capabilities of animals. Thus the project has generated both a significant methodological advance, and – through the empirical data generated – a substantial theoretical shift, in the field of cultural evolution research. Experimental task used in Wilks et al., 2021, J Exp Child Psychol
