Vanessa Ferdinand
From Santa Fe Institute Events Wiki
Background: Anthropology, Linguistics, Cognitive Science
Currently 2nd-year PhD student in Language Evolution at the University of Edinburgh
Interests: I'm interested in mapping out the evolutionary dynamics involved in cultural transmission systems. I also wonder whether general evolutionary frameworks for culture are possible, or if different transmission systems (ex: social network typology) or the cultural information that's being passed on (ex: language, music, religious beliefs) are too idiosyncratic to be modeled generally.
Mainly, I research how the learning biases of individuals in a population map to the emergent structure of the language that share - and look at how transmission process itself modulates this. My master thesis addressed this problem with an iterated learning model of cultural transmission with Bayesian agents. Iterated learning models highlight language as a complex dynamic system that adapts to the environment which it propagates through; the environment here is the cognitive constraints of the humans (or agents) that learn and use the language. The main results we get for these types of models is that language becomes more learnable and structured via use.
I'd like to gain more tools to analyze these data sets with and as many outside perspectives on my work as possible.
Currently, I'm experimenting with human probability matching behavior and trying to formulate simple cognitive biases in terms of selection pressures in a culturally evolving systems. In particular, I'm searing for mechanisms and processes in social learning that can support the appearance of drift at the surface level of a culture: for example, counting up the number of different names that parents give their children each year is a surface level measure of cultural variation, where each variant is a baby name. And looking into the process by which parents actually choose their baby's name gives us a mechanistic understanding of how the variation we witness is generated.
Recent research is claiming that lots of cultural change over type just looks like drift. And generally, drift is something we try to formulate as a neutral model, which we would like to rule out to see if the change we witness is due to something more interesting then just sampling error - but to something like a selection pressure due to the functionality or fitness of one cultural variant over another. However, it is questionable whether true drift processes occur in cultural evolution: what could possibly be an "uninteresting" change that occurs for a system that must replicate by passing through human cognition? That's a great question...