Incentives, Dynamics and Collective Wisdom

Scott Page

Short Abstract
The ability of a collection of individuals to make accurate forecasts depends on individual competence and collective diversity. In this talk, I will discuss the ability of various incentive systems produce those individual and collective characteristics. The model that I present will assume that individuals rely on models to make forecasts. With this model, I will study the population dynamics of these models. I will show that incentive systems that produce highly accurate individuals often produce too little diversity reducing the collective accuracy of the crowd. I will also show how different network assumptions maintain different levels of model diversity. The analysis will rely on both mathematical and computational methods.