Authors:
Marwa Shekfeh
and
Ali A. Minai
Affiliation:
Department of Electrical Engineering and Computer Science, University of Cincinnati, Cincinnati, OH 45221-0030, U.S.A.
Keyword(s):
Associative Learning, Social Learning, Cognitive Agents, Misinformation.
Abstract:
A significant amount of information acquisition in human groups occurs through social learning, i.e., individuals learning through communication with their peers. Since people communicate what they know and their information is not completely accurate, such peer-to-peer learning can lead to the spread of both knowledge and misinformation over social networks. How much of each occurs depends on many factors, including the quality of knowledge in the group as a whole, its initial distribution over the network, and the learning styles of individuals. The number of configurations in which these factors can occur is infinite, but multi-agent network models provide a promising way to explore plausible scenarios. In this paper, we use such a model to consider the joint effect of two factors: 1) The proportion of initially well-informed and ill-informed agents in the population; and 2) The choice of each group to learn in one of two plausible ways. The simulations reported find that both fac
tors have a large effect.
(More)