Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
In an influential recent paper, Mailath–Samuelson formalize learning and reasoning through “model-based inference” and Bayesian updating. In this announcement, we substitute DeGroot’s heuristic for Bayesian updating by (i) furnishing a plausible interaction matrix that agents use to weigh each other’s beliefs, and by (ii) using this matrix to derive properties of the process for the DeGroot updating of beliefs by agents and oracles. The alternative argumentation that we provide facilitates bridging the literature on networks and that on model-based learning and inference; and it identifies productive and ongoing directions for further investigation.