Science has always produced knowledge. What it has never had, until now, is a scientist studying the process of knowledge production itself with the same rigor scientists apply to everything else. That is the work of James Evans.
As Max Palevsky Professor of Sociology and director of UChicago’s Knowledge Lab, Evans uses machine learning, large-scale data, and computational modeling to investigate how ideas emerge, spread, and calcify into consensus. His questions are deceptively fundamental: How do collectives think? What do they know? And crucially, what are they missing? Using decades of scientific literature as raw material, Evans has mapped the patterns that govern how researchers form hypotheses, which ideas attract attention, and where the blind spots of an entire field tend to cluster.
One of his most striking findings is that as science has become more connected, it has in some ways become narrower. Greater access to published research, he found, paradoxically tends to concentrate attention on already-popular ideas, crowding out the atypical combinations that most often lead to breakthroughs. The implication is pointed: The structure of knowledge itself can become an obstacle to discovery.
Evans builds tools designed to work around those obstacles, constructing algorithms that surface the connections human researchers are least likely to make on their own. His work sits at an unusual intersection of sociology, data science, and the philosophy of science, and has been published in Nature, Science, and the Proceedings of the National Academy of Sciences. He is also an external professor at the Santa Fe Institute.
The goal, as Evans frames it, is not simply to understand how knowledge works but to make it work better, for science, for society, and for the future of human and machine intelligence working together.