Background: People need knowledge in order to act. Sometimes people need to act collectively (e.g. in a business organization, or a democratic society), and then they need group knowledge. How can they get it – especially under adverse conditions? One way is for the individuals that constitute these groups to learn from their network of peers. Perhaps they can connect to epistemically prosper?
This talk reports the models, methods, and initial findings of PolyGraphs, an interdisciplinary project supported by the Royal Society and others. Using a model devised in economics, and adapted by philosophers, the PolyGraphs team runs computer simulations on graph datasets to investigate knowledge and ignorance in communities of inquiring agents.
Our presentation will touch on: the role of social network structure in producing knowledge and sustaining ignorance; homophily models of mistrust and the possibility of polarization; measures of group knowledge; and the effects of mis- and disinformation in trusting and cautious communities. A common theme will be the persistence of the trade-off between error and agnosticism, first discussed in the 19th century debate over ‘the ethics of belief’, now updated for the digital age. Link to data animations for Project 1 & Project 2.
Dr Brian Ball is Associate Professor of Philosophy at Northeastern University – London, and a member of the AI Ethics Advisory Board at NU’s Institute for Experiential AI. Born in Canada, he studied philosophy and linguistics (alongside anthropology, biology, computer science, psychology, and other subjects) at McGill University in Montreal, before moving to the UK for graduate study in philosophy at Oxford University, where he subsequently held lectureships at St Anne’s and Balliol Colleges. He moved to the capital in 2009, where his amateur volleyball team won the London Premier League before the pandemic. Now he uses his weekends to teach his kids to play table tennis.
Wednesday, 06 December 2023
11:00 - 11:45 GMT
Share this event on social media: