Social and cognitive constraints on the evolution of culturally transmitted variants: models, experiments and theory
- Autores: José Segovia Martín
- Directores de la Tesis: Sergi Balari (dir. tes.), Mónica Tamariz (dir. tes.)
- Lectura: En la Universitat Autònoma de Barcelona ( España ) en 2020
- Idioma: español
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
The emergence of shared cultural conventions in a population is shaped by the interaction between individuals' cognition and the structure of the society. Humans, more than any other species in the animal kingdom, are able to learn and transmit vast amounts of information, through language and other cultural products. Individual cognitive constraints include cognitive biases, value systems and memory among others. Additionally, humans have an extraordinary capacity to construct social niches that can be modelled as complex systems. Societies are shaped by the structure of the social network and other high-level hierarchical entities that constitute integrated systems of rules that structure social interactions (e.g. institutions). In this thesis I formalise some of the relationships between these factors using a variety of approaches. In particular, I explore the following three main research questions: (1) How do the interactions between individual cognitive traits and the temporal dynamics of social network connectivity, i.e the order in which individuals in a population interact with each other, affect the spread of cultural variants? (2) How do the interactions between individual cognitive traits and institutions affect the evolution of cultural diversity and the emergence of cultural conventions? (3) How might current iterated learning models, niche construction and evolutionary developmental biology be synthesised into a compatible framework for language evolution? Ch. 1 contains a review of the literature and an introduction to the assumptions underlying the models presented in this thesis. In Ch. 2, I present an agent-based model manipulating specific network connectivity dynamics, cognitive biases and memory. I show that connectivity dynamics affect the time-course of variant spread, with lower connectivity slowing down convergence of the population onto a single cultural variant. I also show that, compared to a neutral evolutionary model, content bias (i.e. a preference for variants with high value) is the main driver of convergence and amplifies the effects of connectivity dynamics, whilst larger memory size and coordination bias, especially egocentric bias, slow down convergence. In Ch. 3, I report an experiment in the lab which has two main goals: First, to evaluate the effect of two connectivity dynamics (early and late) on the evolution of the convergence of micro-societies on shared communicative conventions under controlled conditions. Second, to compare the predictions of the agent-based model described in Ch. 2 against experimental data, and calibrate the model to find the best-fitting parameter setting. Results show that, as predicted by the model, an early connectivity dynamic increases convergence and a late connectivity dynamic slows down convergence. Expanding on the agent-based model, Chs. 4 and 5 explore the co-evolution of value systems and institutions by incorporating a comprehensive parameter combination of compliance, confirmation, content and frequency biases into the learning and production algorithm. Results show that, in general, institutional power facilitates the emergence of cultural conventions when compliance biases increase. In general, a compliance bias pushes diversity up when institutions are diverse, and pushes diversity down when institutions convey value systems with strong dominance of one or few cultural variants. In some regions of the parameter space, global conventions can also emerge in the absence of institutional power and therefore of institutions that are in place to guide convergence. In Ch. 6, I use the concept of niche construction to build bridges between eco-evo-devo accounts for cognitive capacities and cultural evolution guided by iterated learning processes. I propose a conceptual model that might be useful to act as a hypothesis-generating framework around which cognitive scientists can structure new triple-inheritance formal models.
