Large Scale Experiments Webinar Series - Marina Agranov
Cambridge-INET, Prof. Sanjeev Goyal and Dr. Frederic Moisan are hosting a series of webinars on "Large Scale Experiments". The fifth webinar of this series will be given by Marina Agranov (Caltech) on her paper "Information Aggregation on Networks: an Experimental Study", joint with Ben Gillen (Claremont McKenna College) and Dotan Persitz (Tel Aviv University).
Abstract
Social scientists have been interested in learning and information transmission over networks for a long time. Most of the formal theoretical literature considers a general setting, suggested by DeGroot (1974), where a group of individuals is tied together by a network of social connections. These individuals are endowed with an idiosyncratic noisy piece of information regarding the true state of the world. In each period they make a choice that reveals their belief regarding this state and observes choices of her neighbors. As a result, she may or may not update her belief and her future choices.
While the theoretical literature provides fascinating long term analysis of various versions of this framework when the society tends to be of infinite size, it provides very few general insights into the dynamics of belief formation in finite, yet relatively large, groups. Thus, within this setting, our main objective is to use laboratory experiments to make the first steps into uncovering the effects of the network specifics on the process by which society members form their opinions, views, and beliefs when communicating over a finite, yet large, network.
We take the setting of DeGroot (1974) to the laboratory. Our goal is to characterize the convergence process of subjects’ beliefs and the conditions for successful knowledge accumulation. In particular, we are interested in structures and signals’ distributions that sustain heterogeneity of opinions. In addition, we will identify key locations for knowledge accumulation and transmission. The experiment is designed to enable the comparison of these positions not only with theoretical measures of centrality but also with key positions as perceived by subjects.
Event Date: Thursday 16th July 2020
Time: 04:00pm - 05:00pm
Event Contact: Marion Reusch - inet@econ.cam.ac.uk
Tags: