The Calhoun Experiment Study by Means of Agent-Based Modeling

TL;DR

This paper uses agent-based modeling to simulate Calhoun's mouse experiments, aiming to reproduce the observed social disintegration despite abundant resources.

Contribution

The paper proposes a set of parameters to qualitatively and partially quantitatively reproduce Calhoun's findings.

Findings

The proposed parameters allow for a qualitative reproduction of Calhoun's behavioral sink phenomenon.

The simulation shows some quantitative deviation from the original experiment's results.

Missing numerical data from the original study complicates full quantitative replication.

Abstract

More than 50 years ago John D. Calhoun conducted a series of experiments devoted to studying the social behavior of mice. The longest of them lasted more than four years and led to the creation of the concept of the so-called “behavioral sink”. The population of mice, which had all possible resources at their disposal but were located in a limited room, disintegrated after an initial phase of strong development. In our paper we are going to reproduce the effects of this experiment. The crucial problem in every simulation approach is to determine the set of the most important parameters which influence the global as well as local effects of the simulated process. In the studied case we have the problem that a lot of important information is missing. The author of the original work focused rather on social mechanisms, often omitting key numerical data related to the course of the experiment. In this paper we try to propose a certain set of parameters. By using them we can reproduce Calhoun’s results qualitatively and, with some deviation, quantitatively.

Figures