Jacob Foster

Jacob Gates Foster is a UCLA computational sociologist and founding co-Director of the Diverse Intelligences Summer Institute, which studies topics such as artificial intelligence, animal cognition, and the role of culture and language in shaping the human mind. Much of Foster's work involves machine learning and the development of formal models of the structure and dynamics of ideas and institutions. He is a member of the American Sociological Association's Sociological Methodology editorial board.

Early life

Foster grew up in White Post, Virginia and attended Wakefield Country Day School and Woodberry Forest School. He was recognized by the Virginia General Assembly for earning the highest possible score on 11 Advanced Placement exams, placing him number one in the world, and scoring a perfect 1600 on the Scholastic Aptitude Test.

He received his B.S. with distinction from Duke University, writing an honors thesis on "Physics with Two Time Dimensions" under the supervision of nuclear physicist Berndt Müller. He studied mathematical physics at Oxford University as a Rhodes Scholar, writing his transfer thesis under the supervision of philosopher of science Roger Penrose; and received his Ph.D. in Physics (with a specialty in Complexity Science) from the University of Calgary, writing his thesis under the supervision of Maya Paczuski and Peter Grassberger. Foster explains, "Like many physicists, I was drawn to the study of complex systems because it licensed me (after a fashion) to work on all sorts of systems that physicists aren't 'supposed' to—complex networks, evolutionary dynamics, etc."

Research and theories

Foster notes that computation can go beyond supplying new methods for studying culture, to change the way people think about culture. He argues that computational thinking, a theoretical stance encouraging detailed analysis of the “procedures for arriving at adaptive decisions based on approximate and noisy information, is badly needed. In Foster's view, "machine learning and computational cognitive science offer the best hope of formalization that remains faithful to the core insights of cultural sociology while offering a range of starting points," ranging from Bayesian models of cognition to deep learning.

Much of Foster's work pertains to the study of science itself. For example, he notes that the scientific literature reflects a growing focus on established knowledge at the expense of new opportunities. "High-risk" research that explores new chemical relationships is more likely to be ignored but also more likely to achieve high impact and recognition. Foster's study of 137 different prize winners in biomedicine and chemistry led him to conclude that given the professional demand for productivity this additional reward is insufficient to compensate for the additional risk; accordingly, he advocates for policy interventions to foster more innovative research.

In a 2011 Science article, Foster and University of Chicago sociologist James A. Evans note the potential for metaknowledge researchers to seek previously hidden connections in natural language processing. They cite the example of Google's identifying the emergence of influenza epidemics by tracking related Google searches. They point out also that metaknowledge can help uncover trends in which scientific ideas become entrenched as studies repeatedly find conclusions that support previous claims by well-known scholars and when students of distinguished researchers go on to do their own work that reinforces existing claims.

Some of Foster's theories concern the topic of wikis and their limitations. While noting that Wikipedia's success gives credence to the Web 2.0 vision of collective intelligence, he argues that the hive mind is ill-suited to the kinds of activities that public intellectuals typically perform, such as writing editorials that express the opinions of one writer in a timely way. Drawing on Timothy Messer-Kruse's account of his experiences attempting to correct Wikipedia's coverage of the Haymarket affair, Foster argues that Wikipedia emphasis on "reliable sources" rather than truth is a form of epistemic collectivism that can have corrosive effects on the crowd.

Accordingly, Foster laments the decline of the "middle" class of "not-quite-public intellectuals — the local or narrow expert whose insight may in fact outstrip the insights of his or her far more prestigious and well-attended-to 'betters' in the elite". As a remedy, he calls for democratizing social media through representative meritocracy: "Rather than a mob who can speak only through the occasional paroxysm of viral coordination or through involuntary algorithmic aggregation, we would have a digital republic — animated by devotion to individual creativity and the critical sense, able to draw effectively and efficiently on distributed expertise and knowledge."

During the 2020 coronavirus pandemic, Foster argued that the terminology "social distancing" should be replaced with the more precise term "physical distancing" since social connectivity is still needed even when public health requires that people maintain a physical distance from one another.