FINDING TRUTH WHEN MOST PEOPLE ARE WRONG
Drazen Prelec, MIT Sloan School of Management
(joint work with H. Sebastian Seung, MIT Department of Brain and Cognitive Sciences)
Thursday November 30, 2006   4:00-5:30 pm
3 Cambridge Center, MIT Building NE20, Room 336 Conference Room

Abstract
Effective use of subjective judgment is essential in all fields of knowledge. We present a method for finding the true answer to a question from individual judgments. The method makes use of a mathematical instrument, called the Bayesian Truth Serum, which scores respondents for their judgments and their meta-knowledge of other people's judgments. In a probabilistic model of belief formation, respondents' scores converge to the subjective probabilities they implicitly assign to truth, even though subjective probabilities are not directly elicited and truth cannot be independently verified. An experiment demonstrates that the method outperforms majority rule in selecting truth and in identifying respondents with expertise.

Speaker bio
Drazen Prelec is the Digital Equipment LFM Professor of Management at the Sloan School, MIT, where he has taught since 1991. His interests include decision theory, marketing research, behavioral economics, and neuroeconomics. His work has been recognized by several fellowships and awards, including those by Harvard Society of Fellows and the John Simon Guggenheim Foundation.

 


Sign Up to receive email reminders of coming speakers, join the CCI Seminar  Series mailing list