An elementary humanomics approach to boundedly rational quadratic models

Abstract We take a refreshing new look at boundedly rational quadratic models in economics using some elementary modeling of the principles put forward book Humanomics by Vernon L. Smith and Bart J. Wilson. A simple model is introduced built on fundamental gratitude/resentment felt corresponding action responses reward/punishment form higher/lower payoff transfers. There are two timescales: one for strictly self-interested action, as economic equilibrium, another governed feelings gratitude/resentment. One three timescale scenarios investigated: where changes much more slowly than equilibrium (“quenched model”). Another model, which occurs over slower time evolution (“annealed” model) set up, but not investigated. The quenched with homogeneous interactions turns out to be non-frustrated spin-glass model. two-agent heterogeneous aligning (ferromagnetic) analyzed yields insights into critical probability p ( 1 − ) that represents empirical frequency opportunity agent i benefit (hurt) other invokes mutual gratitude (resentment). ∗ , = 2 exists each agent. When will choose their self-interest. > sensitive interpersonal thus reward/punish initiating benefit/hurt. find greater one-half, implies agents averse resentful behavior punishment. This was result its properties, consistent Axiom 4 about asymmetry resentment. Furthermore, who receives less behavior; i.e., has higher probability. For this particular Nash no predictive power properties since rewards same behavior, behavior. Accordingly, we see Gibbs does indeed lead richer properties.