Human behaviour versus optimising agents and the resilience of farms - Insights from agent-based participatory experiments with FarmAgriPoliS

This paper aims to examine the extent to which human participants show higher resilience compared to computer agents in agent-based participatory experiments. We motivate and examine three types of resilient behaviour of farmers during a crisis or as response to competitive pressure: successful survival, loss-minimising farm exits, and path breaking respectively path creating growth strategies. Our experiments revealed that human decision makers recognised and exploited such resilient strategies in periods of crisis or under challenging circumstances in general better than myopic optimising agents, although they did not perform better on average. The reason can be seen in a substantial heterogeneity of human decision makers, for which we identified four categories: negligent gamblers, actors missing opportunities, solid farm managers and successful path breakers.