Optimal prediction of decisions and model selection in social dilemmas using block models

Abstract

Advancing our understanding of human behavior hinges on the ability of theories to unveil the mechanisms underlying such behaviors. Measuring the ability of theories and models to predict unobserved behaviors provides a principled method to evaluate their merit and, thus, to help establish which mechanisms are most plausible. Here, we propose models and develop rigorous inference approaches to predict strategic decisions in dyadic social dilemmas. In particular, we use bipartite stochastic block models that incorporate information about the dilemmas faced by individuals. We show, combining these models with empirical data on strategic decisions in dyadic social dilemmas, that individual strategic decisions are to a large extent predictable, despite not being "rational. " The analysis of these models also allows us to conclude that: (i) individuals do not perceive games according their game-theoretical structure; (ii) individuals make decisions using combinations of multiple simple strategies, which our approach reveals naturally.

Complex decision-making strategies in a stock market experiment explained as the combination of few simple strategies

“…In this work we have shown the suitability of the MMSBM for the study of social systems, particularly for modeling, predicting and understanding human decision-making processes. MMSBMs not only provide more accurate predictions than other state-of-theart methods [14,17], but they are interpretable.…”

“…In the case of the SBM we use simulated annealing to find the set of model parameters (θ * , η * , p * ) that maximizes the posterior probability [14]. In the case of the MMSBM, we use the expectation maximization approach described in [14,27] (see Materials and Methods). We make our predictions about unobserved decisions using these maximum a posteriori parameters.…”

“…Next, we turn to the model parameters obtained for the most predictive representation of the data. In particular we focus on player groups, which we identify with guessing strategies [14]. As mentioned before, MMSBMs assume that players can mix several strategies and that contexts can also belong to different groups simultaneously.…”

Powered Dirichlet Process for Controlling the Importance of "Rich-Get-Richer" Prior Assumptions in Bayesian Clustering

“…The Bayesian clustering approach received broad attention in the last years. A non-exhaustive list of application includes medicine, [12], natural language processing [3,31], genetics [21,18,24], recommender systems [1,9,23], sociology [11,5], etc. The key idea is to simulate a corpus of independent observations by drawing them from a set of latent variables (clusters).…”

Community Detection in Bipartite Networks with Stochastic Blockmodels

“…We are not the first to show that advantages of a bipartite approach to understanding network structure. Recently, for instance, bipartite analyses have been used in game-theoretic analyses of human decisions and strategies [27] as well as efforts to create scalable predictions of user-item ratings [9], which naturally form a bipartite structure. Indeed, all 24 empirical networks which we analyze in later sections come from bipartite analyses (see Table ) across domains.…”