Professor Hegselmann’s lecture was entitled ‘Computational social epistemology: a case study on two-armed bandits versus Carnapian truth seekers and epistemic free riders’.
To alleviate less-philosophically informed audience members' puzzlement about the lecture’s daunting title, Professor Hegselmann began the lecture by explaining the basics of social epistemology, which revolves around social exchange of opinions under the influence of empirical 'signals’ that come in from the outer world. In a second part, he focused on the social exchange process and presented the so-called bounded confidence model that he developed together with his co-author Professor Ulrich Krause more than twenty years ago. Recently, the duo discovered that they had overlooked decisive features of their original model. That led them to revisit their seminal contribution all over again. This newly updated (and yet to be published) understanding of the bounded confidence model was being presented to the PPE community in the JSM Lecture.
Professor Hegselmann explained in a third part what Carnapian truth seekers, epistemic free riders and two-armed bandits actually are, and more importantly what their significance and use are to his research. Roughly, Professor Hegselmann developed a simulation model that allows to study how groups of people (divided in those doing thorough research to find the truth, Carnapian truth seekers, and those piggybacking off the research of the truth seekers, the epistemic free riders) can try to find the better option in a situation where a choice between two strategies that have unknown success probabilities is to be made.
By easing into the matter with the introductory social epistemological prelude, the model was clearly comprehensible for the audience. And through live simulated experiments projected on the lecture screen the workings of the model were made even clearer!
Fittingly, Professor Hegselmann reserved some of the last minutes of his lecture to expound on what John Stuart Mill would have thought about the model and its applications, concluding that according to his writings on ‘the method of difference‘ in his book A system of Logic, Mill would have found computational social epistemology extremely valuable. The method of difference is a logical principle designed to identify the cause of a phenomenon, but Mill thought it to be unusable as nature doesn’t provide suitable spontaneous experiments and what he called ‘methods of artificial experiment‘ weren’t available in his time. The computational approach however easily allows for this type of artificial experiment and thusly realise Mill's method of difference.
At the end of the lecture the PPE audience was well-acquainted with professor Hegselmann’s model, which quickly sparked a lively question and answering session lasting more than an hour. Questions on the specifics of the model, to questions on methodology, to questions on the applications of the model to economics (e.g. to understand stock buying behaviour) or politics (e.g. to understand and combat disinformation and polarisation) and even questions on how the model could be potentially used for a Bachelor’s thesis were all satisfyingly answered by professor Hegselmann. It is safe to say then, that this year the John Stuart Mill lecture once again proved a memorable end to the academic year at PPE!