Denmark

**Author(s):** Tamariz, M.; Papa, A.; Cristea, M.; McGuigan, N. &lt;br&gt;&lt;br&gt;Introduction During our lifetime we typically learn several variants of a given cultural trait, e.g., from family, peers and experts. Social biases affecting which variants are adopted by learners are well-studied. Considerably less is known about influences on the selection of variants for onward transmission by transmitters. This study tests whether a social context congruence effect is in place, i.e., do we tend to transmit to friends behaviours that we learned from friends, to our children behaviours learned from our parents, and to novices behaviours learned from experts? Methods Each of 64 participants learned two cultural variants (methods to solve a puzzle): “E” from an Expert and “P” from a Peer. Then, we

Cultural Evolution Society 2022

Associative learning and cultural transmission: We teach to novices the variants we learned from experts, and to peers the variants we learned from other peers!

cultural transmission; transition biases; psychology; experiment

### Welcome to the fourth Cultural Evolution Society Conference, CES2022! 

As last year, Underline will support the online component of the conference. The ongoing COVID-19 pandemic is causing a great deal of uncertainty for all of us. Therefore, we seek to offer a hybrid format that will hopefully allow some folks to travel to Aarhus, others to meet up locally or regionally in CES hub-lets, and other yet to participate online. The field of cultural evolution is growing in size and scope; its theories, methods and applications - both academic and applied - are getting ever more robust. Do join us in what promises to be an exciting exchange of ideas and a celebration of our field.

##### ***For SoMe interaction and live tweeting please tag @CulturalEvolSoc and use #CESconf2022***


#### The CES 2022 local organising committee
* Luseadra Joy McKerracher
* Jesper Sørensen 
* Niels Nørkjær Johannsen 
* Benjamin Grant Purzycki 
* Mette Løvschal 
* Felix Riede
* Iza Romanowska 
* Karsten Olsen

Please click on the button bellow if you want to watch the videos

The field of cultural evolution is growing in size and scope; its theories, methods and applications - both academic and applied - are getting ever more robust. Do join us in what promises to be an exciting exchange of ideas and a celebration of our field.

**Author(s):** Tamariz, M.; Papa, A.; Cristea, M.; McGuigan, N. Introduction During our lifetime we typically learn several variants of a given cultural trait, e.g., from family, peers and experts. Social biases affecting which variants are adopted by learners are well-studied. Considerably less is known about influences on the selection of variants for onward transmission by transmitters. This study tests whether a social context congruence effect is in place, i.e., do we tend to transmit to friends behaviours that we learned from friends, to our children behaviours learned from our parents, and to novices behaviours learned from experts? Methods Each of 64 participants learned two cultural variants (methods to solve a puzzle): “E” from an Expert and “P” from a Peer. Then, we

technical paper

Almost all cultures around the world exploit language resources to construe experiences of phenomena in quantity terms (e.g., cardinals, grammatical number) yet the evolutionary processes underlying the development of grammatical (singular, dual, and trial) and lexical number in language is understudied. This study tests the hypothesis that grammatical number developed from cardinal number through a case study of dual marking in the Para Nyungan language family. Factors that motivate the development of dual grammatical number are examined. Number, as a category of grammar, describes the regular encodings of our experiences of phenomena in exact quantity terms (as singular, dual, trial, paucal and plural values) through morphological, syntactic, and/or phonetic (e.g., tones, clitics) configurations. Beyond grammatical number, the majority of the languages of the world exploit lexical number — such as cardinals (e.g., one, two, three, four) and ordinals (e.g., first, tenth) — for exact quantification. Some studies hypothesise that the dual, trial, and paucal grammatical number values evolve from the cardinal numbers two, three, and four (Corbett, 2000; Aikhenvald, 2018) with evidence from language families (Austronesian and Sino Tibetan) where the number features are often marked. Yet, the crucial role of shared descent in influencing such a development (Macklin & Round, 2022) is not central in these reports since examples are extracted from only a few of the languages that mark these grammatical number features. The hypothesis also assumes that lexical number (specifically, cardinals) evolves prior to grammatical number. The evidence that plurality (Grammatical number) is marked in the 907 sampled languages with lexical numbers (high limit of cardinals) but not the reverse seems to support the conclusion that lexical number precedes the development of grammatical number (Overman, 2021). The evolution of grammatical number and cardinals in languages can however be studied independently. And the results from such independent examination of evolution could shed light on the relation between these two ways of quantifying as well as socio-cultural factors underline their relation. The present study tackles the biases in sampling by focusing on language families where the grammatical number features of interest dominate. The data for the study has been taken from Grambank and Numeralbank, global databases respectively on the grammatical and numeral systems of languages worldwide. The Atlantic Congo dominates (with 222 (52%) of the 425 languages) families marking singular grammatical number on nouns (Grambank). Austronesian dominates (204 languages - 33%) the 616 languages that mark dual grammatical number on pronouns. Pama-Nyungan has the majority of languages marking dual on nouns (25% — 26 of 184 languages). Lastly, the trial grammatical number the least frequently marked grammatical number (8 out of 2,214 languages mark trial number) is dominant in Austronesian (with 4 languages). The study focuses examines the evolutionary time for cardinal numbers and dual marking in the Pama Nyungan language family. The numerals of these languages were extracted from numeralbank. Example 1: Development of dual marking in the Yanyuwa clade LANGUAGE DUAL MARKER CARDINAL TWO SOURCE OF INFLUENCE Yanyuwa _wiya, rri_ Kanymarda, Binary system: 3 = 2 +1, 4 = 2+2 Anyirarra Coordination: marda [Also, too, and] Dual relates to animacy Warluwarra _wiya kutya, guca Borrowed from Yanyuwa Barlarnu _wiya gutyiya, guciya Borrowed from Yanyuwa Yindjilandi _wiya ɡulharra ɡutharra Borrowed from Yanyuwa The results from the preliminary phylogentic reconstruction of cardinals and dual markers in Pama-Nyungan languages shows that (1) cardinals developed first and (2) dual grammatical number developed much latter. The cardinal number two usually grammaticalises as dual in the languages which mark dual number. Socio-cultural practices, including counting practices and objects counted, suggests that the grammaticalisation process is influenced by cultural factors. In other words, the findings shows that the development of dual grammatical number from the cardinal two is motivated by cultural practices. The example (1) shows that the independent development of dual in the Yanyuwa clade was motivated by the binary numeral system, the conception of addition ("Marda" [and] a clip from Kanymarda [two] used as coordinator) and the animate objects that were counted (animacy effect). The remaining languages in the clade borrowed the dual marker from Yanyuwa. The results raises questions about the possibilities for languages to mark grammatical number beyond five given similar sociocultural practices. While the presence of birth order naming seems a positive evidence of such grammatical number development, the feature is less popular in the family for any strong conclusions to be drawn. Moreover, the few instances of quantifying birth order on nouns of the languages, the birth order labels did not exceed the numeral limit of the language family. This preliminary results shows that grammatical number marking in Pama Nyungan develops from cardinal numbers and has implication for the evolution of grammatical number, the cognitive scientific theory of subitizing and the evolution of cardinals.

On the Development of Grammatical Number

INTRODUCTION: Phylogenetic methods are used for the reconstruction of historical relationships between cultural lineages. These methods, designed for biological evolution, assume traits can only be transferred via inheritance. Little work has been done to determine how the introduction of horizontal transmission, potentially ubiquitous in cultural evolution, affects their accuracy. OBJECTIVE: In this study we tested the robustness of phylogenetic methods when applied to a demographic process with horizontal transmission of traits between groups. APPROACH: We generated synthetic population histories represented as phylogenetic trees. We then simulated variant data on the trees, sampled the terminal nodes and used standard phylogenetic reconstruction methods to construct best fitting trees from the leaf data. We assessed the accuracy of the phylogenetic reconstruction as the distance between the best fitting tree and the ‘true’ tree representation of the demographic history. FINDINGS: Phylogenetic methods are reliable only when the borrowing is constrained to groups with a recent common ancestor. In addition, borrowing can be constrained by either shared ancestry or spatial proximity, leading to significantly divergent inferred trees. CONCLUSION: Preliminary results suggest that the introduction of demographic complexity, including extinction of groups and horizontal transmission events, significantly decreases the accuracy of phylogenetic inference.

Can cultural phylogenies accurately capture spatially explicit demographic scenarios?

**Author(s):** Kasser, S. M.; Blanco-Portillo, J.; Fortunato, L.; Laland, K. N. Objective: Gene-culture coevolution (GCC) theory routinely predicts interactions between cultural and genetic variation that should emerge through empirical analyses, yet evidence for such associations remains sparse. By adapting tools that quantify ecological drivers of population genetic structure ("landscape genomics"), we explore potential cultural drivers of human genetic variation across Oceania to evaluate the extent of GCC.

Methods: We will introduce landscape genomics as a method of quantifying gene-culture coevolution. In a short test case, we use high-density SNP data from Oceanian individuals across 17 island populations from the OGVP (a collaboration between Oxford and Stanford). We then analyse these frequencies using GESTE, a Bayesian procedure to estimate the influence of environmental factors on population genetic structure. For the present analysis, cultural and ecological factors relating to subsistence, kinship practices and island ecology are drawn from relevant cross-cultural datasets.

Results: We will present results on associations between cultural factors and genetic structure.

Discussion: We will discuss the results in the context of recent work on how culture may shape genetic variation beyond classic examples of GCC, and whether landscape genomics represents a useful analogy for quantifying GCC.

Unravelling gene-culture coevolution in the Pacific

**Author(s):** Matthews, Luke J.; Khan, Omair Introduction: Shared understandings of current events are crucial to any functional modern society but common understandings even about factual elements of “what happened” during an event are frequently and hotly contested. Methods: We applied cultural consensus analysis (CCA) and cluster analysis to hundreds of thematically coded journalistic articles from a 2019 air skirmish between India and Pakistan. We applied these methods longitudinally to test alternative hypotheses that either narrative consensus about factual elements increased or decreased over time. Results: Consensus about factual elements of the 2019 air skirmish decreased over a two month period as the narrative diverged into two increasingly thematically distinct clusters. Conclusions: At least regarding the 2019 India-Pakistan air skirmish, journalistic articles decreased in consensus about facts over time. If these results generalize, then media companies might improve shared understandings by repeating articles from closer in time to actual events, as these articles show greater potential for narratives reflecting factual consensus.

Rapid cultural evolution of current event narratives

**Author(s):** Bluet, A; Osiurak, F; Claidière, N; Reynaud, E. INTRODUCTION: Our technologies have never ceased to evolve, allowing our lineage to expand its habitat all over the Earth, and even to explore space. This phenomenon, called cumulative technological culture (CTC), has been studied extensively, notably using mathematical and computational models. One main factor that has received extensive attention is demography, and more precisely population size. However, the cognitive capacities needed for the emergence and maintenance of CTC remain largely unknown. In the literature, the focus is put on the distinctive ability of humans to imitate, with an emphasis on our unique social skills underlying it, namely theory of mind (ToM). A recent alternative view, called the technical-reasoning hypothesis, proposes that our unique ability to understand the physical world (i.e., technical reasoning; TR) might also play a critical role in CTC. Here, we propose a simple model that focuses on micro-societies experiments. Classically, we observe three forms of social learning in this literature: reverse-engineering (RE) where only the technology is present; observation (OBS) where a receiver witnesses a demonstration of the building of the technology by a sender; and communication (COM) with verbal interaction between a receiver and a sender. Thus, our model will implement these three social learning. Our model is composed of a simple environment with only one technology transmitted between generations of individuals. These individuals have two cognitive skills: ToM and TR, and can learn using the aforementioned different social-learning conditions. Their goal is to improve as much as possible the technology. Afterwards, we extend this model to include multiple chains of transmission in order to assess the effect of population size on CTC. OBJECTIVES: To model the transmission of technical information between two individuals in a micro-society. Mainly, we intend to investigate the role of ToM and TR in the social-learning conditions found in the micro-society literature. Another goal of this model is to replicate and solidify the results of said literature. Finally, we intend to test our model on a population scale. This allows us to test the different hypotheses of the impact of population size on CTC. APPROACH: In our model, we assume an environment composed of a single technology. We further assume that this technology can be decomposed into components, called traits. For example, if the technology represents a vehicle, traits represent the wheels, the seating, the steering wheel, etc. In each generation, an individual appears in the environment. This individual has a set of knowledge (named cog) for each technology’s trait, and two cognitive skills, namely TR and ToM. First, the individual learns from previous generations using one of three social-learning forms (RE, OBS, COM) to improve its cogs. This learning phase depends on its cognitive skills (TR and ToM; depending on the social-learning form). Afterwards, the individual tries to improve the technology by refining its traits. Refining the technology’s trait equals increasing its quality. Subsequently, the individual might have an opportunity to innovate, which is equivalent to adding a new trait to the technology in our model. This opportunity comes with a fixed probability, and the individual needs to cease it either by having a good enough technology or by being himself a quite advanced individual compared to its environment. Finally, the individual disappears from the environment and is replaced by a new one. This process is repeated for N generations (N = 1000 if not stated otherwise), thus creating transmission chains. In the extension of our model, we run multiple transmission chains at the same time. Each chain has its technology and individual, the latter of which can learn from conspecifics in other transmission chains. This emulates a fully connected population of individuals. It is important to note that we assume that innovations are transmission through social learning, meaning that an individual will bring back through learning the innovation from another transmission chain. RESULTS: In our model, all three social learning forms yield CTC in the form of exponential growth of technology’s quality. However, both RE and OBS produce lower technology quality than COM, the former both having very similar behaviours and results, which agrees with classical results of the literature on transmission chains, where RE has the lowest technology quality, closely followed by OBS and then well ahead COM. When comparing model of COM with and without ToM, we see little difference between the two, demonstrating a limited impact of ToM. Further analyses indicate that ToM has an impact only during the first generations, indicating that its effect may be to accelerate the acquisition of an optimal technology. When we add multiple transmission chains, we observe an effect of population size on CTC. This effect is weaker the greater the population size, hinting at a higher impact of adding individuals for a smaller population size. Further analysis showed that our model is driven by the frequency of innovation, thus explaining this result. Indeed, adding individuals is equal to adding potential innovators. However, we observe that once population size reaches a certain threshold, adding more individuals does not bring more innovation because there are already enough innovators. DISCUSSION: Our model results are in accordance with the transmission-chain literature, where communication conditions often outperform non-communication ones. This both strengthen the result of said literature and validate our model. Furthermore, we observe exponential growth of technology’s quality, something that is also observed in our cultural evolutionary history (Lehman, 1947). We posit that this exponential growth is due to a snowball effect: an increase in technology’s quality is accompanied by an increase in the understanding of this technology, thus increasing in return the capacity of the individual to increase the technology’s quality and so on. This snowball effect occurs because TR supports both the transmission of information and the modification of the technology. This effect of both the understanding and the technology’s quality improving together is something that we have already shown in a recent study using transmission chains, where an increase in the speed at which a wheel rolls down an axis is correlated to the increase in the understanding of how mass repartition affects this movement. Our results show that ToM has little to no impact on CTC. This seems surprising because ToM can only impact learning positively in our model. However, further analysis showed that ToM has an impact early on in our model, acting as a catalyst for the first burst of increases in technology’s quality. In this regard, we consider the impact of ToM as a launching pad for the snowball effect described earlier, being neither necessary nor sufficient for the emergence of a CTC. When applied to the scale of a population, our model shows interesting and well-known results. Indeed, a larger population size has been extensively linked to an increase in CTC and this effect has also been linked to the frequency of innovation. These are encouraging results, and in the future, we intend to further extend our model to implement other aspects of demography that have been shown to have an impact on CTC, such as population structure and dissemination of knowledge.

Impact of technical reasoning, theory of mind and population size on cumulative technological culture: insights from a model of micro-societies

**Author(s):** Fedorova, N.; McElreath, R. Introduction: Shelter is a key human technology, and varies extensively throughout the ethnographic record. Like other technologies, human investment in shelter is strategic and has consequences for the evolution of settlement landscapes. We present findings from a household study of the drivers of building behaviour in the ger districts of Ulaanbaatar, Mongolia. Methods: We formulate a Stochastic Dynamic Programming (SDP) optimality model to generate non-null predictions about building behaviour. With the use of Approximate Bayesian Computation we compare simulated model outputs with empirical data and select parameter combinations that can generate the observed data. Results: The SDP model allows us to explicitly interrogate trade-offs between saving and building through time. Preliminary results suggest that the dynamics of saving influence the time horizon and extent of building, through changes in family composition. Conclusion: Human decisions are always balancing trade-offs, which are difficult to access with traditional analysis methods. We hope that through the robust comparison of empirical work to simulated non-null predictions we take productive steps to understanding human landscape investments.

Strategic investments in housing in Ulaanbaatar, Mongolia

 As one of the two moderators, I will predominantly lead the theoretical discussion of what are social norms and how could they be identified in animals. I come from the perspective that rather than focusing on some cognitive mechanism that is taken to be necessary for having normativity, investigations into social norms should focus on community level patterns that are supported by social maintenance—positive and negative reinforcement by community members.

Theoretical angles on framing and identifying norms in nonhuman taxa

**Author(s):** Gunst, A; Nylund, E; Marklund A Introduction: Advantageous inequity aversion (i.e., the tendency to respond negatively to unfairness that benefits oneself) usually develops in 6–8-year-olds. However, little is known about the biological and cultural selection of this phenomenon. Materials and Methods: Using data collected from 120 4–8-year-old Finnish children, we tested two evolutionary explanations for the development of advantageous inequity aversion: reciprocal altruism (i.e., benefiting from sharing when the roles are likely reversed in the future) and inclusive fitness (i.e., benefiting from sharing with biological relatives that carry the same alleles). We asked children to distribute five erasers between themselves, a sibling, a friend, and a stranger. That is, an equal distribution was only possible if throwing away one eraser. Results: We found no support for advantageous inequity aversion being shaped by either inclusive fitness or reciprocal altruism. Participants did not show a preference for biological relatives or friends. Rather their behavior reflected a more general tendency for fairness. Discussion: It is possible that adherence to social norms to avoid negative consequences as explains the development of advantageous inequity aversion.

Norms of Fairness: Inclusive Fitness or Reciprocal Altruism Does not Explain Advantageous Inequity Aversion

**Author(s):** Schnell, E; Muthukrishna, M Individuals belong to many different cultural-groups which sometimes have competing interests. For example, one’s company may require frequent flights, which competes with environmental commitments at a larger-scale of cooperation. As such, individuals may be required to choose to cooperate with some groups at the expense of the others. We analytically model cooperation supported by indirect reciprocity with individuals who belong to both a small-scale, local group embedded within a larger-scale, global group. To model cooperation occurring simultaneously at different scales, we extend previous work (Panchanathan & Boyd, 2004) modelling cooperation in a mutual-aid game (MAG) conditional upon cooperation in a previous public goods game (PGG). In our extension, individuals can contribute to either a larger, global PGG or a smaller, local PGG with multipliers adjusted to create competing interests. Individuals then play a MAG amongst their local group. The model shows that by manipulating the multipliers of each PGG, these competing scales can undermine one another. This suggests that indirect reciprocity can maintain cooperation within one cultural-group, while leading to defection among groups with competing interests.

Competing Scales of Indirect Reciprocity

**Author(s):** Borgerhoff Mulder, M; Andrews, JB Introduction: The conditions responsible for the uptake, spread and decay of institutions lie at the heart of cultural evolutionary theorizing. Conservation interventions provide “natural experiments” to explore such dynamics. We use a failed Reduced Emissions from Deforestation & Land Degradation (REDD+) project in Zanzibar to explore two questions. How do institutions introduced by REDD+ continue to spread across the island, despite the absence of carbon payments? Second, has exposure to REDD+ influenced residents' willingness to participate in future payment for ecosystem schemes? Methods: We use interviews with community leader interviews and Willingness To Accept (WTA) data at a community and individual level to investigate how exposure to programming affects future willingness, and how various traits (age, gender, time preferences, wealth, conservation committee membership and forest dependence) moderate the effect. Results: We show that (1) externally-seeded institutional interventions can accrete their own momentum and spread, despite the absence of payments, and (2) exposure to even a failed REDD+ programme encourages future participation in conservation among some demographics. As a Conclusion, we highlight the relevance of cultural evolutionary theory for designing sustainable and endogenously-spreading development/conservation policy.

Premium content

Downloads

Next from Cultural Evolution Society 2022

On the Development of Grammatical Number

Similar lecture

Tip for Science: New app lets you study cultural transmission of continuous traits