Matthew A. Turner, Ph.D.


Code on GitHub


Hello and welcome! I am currently a postdoctoral researcher at Stanford University studying the evolution of social learning in uncertain environments, advised by James Holland Jones in the Earth System Science department.

Prior to this postdoctoral work I obtained my PhD from the University of California, Merced, advised by Paul Smaldino in the Cognitive and Information Sciences program. Before receiving my PhD, I worked professionally as a data engineer and research software developer after receiving a MS in applied physics from Rice University and a dual BS in mathematics and physics from Syracuse University.

I love to use computational modeling to develop and test theories of social behavior and to solve difficult practical problems. If you have a problem you think could be addressed through computational modeling, please consider contacting me to see if I could be of assistance developing models, performing analyses, or implementing your ideas in computer code via my consulting services.


How social structure emerges from cognition and social influence

How and why do social structures, such as extremism and polarization, emerge? What about population-level tendencies, such as how frequently individuals base their decisions on what others are doing, instead of relying on their own personal experience? Why is there a "replication crisis" and other cracks in the foundations of psychological and other science? These questions motivate my research.

In focusing on extremism and polarization, I have identified a number of root cognitive factors that lead to rising extremism and polarization, including initial extremism, communication noise, and random chance ("Paths to Polarization", Complexity, 2018). It seems that the empirical fact that extremists are more stubborn leads to rising extremism as social influence progresses, but empirical studies may need improved statistical methods to properly measure this ("Stubborn extremism as a potential pathway to group polarization", 42nd Annual Cog. Sci. Conference, 2020; follow-up journal articles in preparation). My work benefits from clear thinking about what models are and how we learn from them ("Mechanistic modeling for the masses", Comment on forthcoming Yarkoni, "The Genearlizability Crisis", in Behavioral and Brain Sciences).

More recently I have begun studying the fundamental cognitive, social, and evolutionary processes and factors that determine how frequently people learn from others. This work will be presented as a conference talk at the American Association of Biological Anthropologists in March, 2022, and will be presented in the form of conference and journal articles in the coming months.

For a full list of my research publications please see my CV, including the paper I co-authored with Paul Smaldino and Pablo Contreras Kallens on how "Open science and modified funding lotteries can impede the natural selection of bad science", Royal Society Open Science, 2019.

Practical application of computational modeling: consulting services

In addition to my academic research work, I provide programming and research consulting services. In one ongoing project, for example, I am contributing to the Berkeley National Lab's BEAM high-performance transportation modeling project.

If my expertise may be of service, please contact me to schedule a time to talk about your work and problems you face, and see how I may help.