Effects of agent information behavior on network performance

Abstract

In today's information rich society, individuals are often faced with the problem of making decisions in the presence of an overwhelming amount of information. Everyone takes on slightly different strategies. Should they make the decision as soon as possible, or wait until they have a great deal of evidence? Should they stop looking at new information once a decision has been made, or should they keep an open mind and continue searching? I propose a simulation model taking into account such individual differences in cognitive strategies. Five parameters are used to determine how an agent acts on available information: competence, decisiveness, corroboration factor, closed-mindedness, and engagement. The simulation provides evidence that agents with a propensity to cooperation are best able to handle the surplus. By treating the network as a sieve, agents with low closed-mindedness and high corroboration factor allow the most useful information to filter through to them. The engagement and decisiveness parameters, however, take on a much more open-ended role, and have an effect highly dependent on the problem and types of evidence at hand.