Simulating Heterogeneous Farmer Behaviors
increased messaging and the use of influential figures can lead to a net increase in the population ’ s behavioral intent .
A natural next step is to perform more rigorous validation experiments that are coupled with traditional survey techniques that have been modified to incorporate the specific parameters of our model . Another avenue for future work is to test the impact of counter-messaging . Corporations , political candidates , and various organizations often engage in persuasive messaging designed to drive people away from their opponents . In our model , the choice for each agent was either to act or to do nothing . An additional action that is diametrically opposed to the first could easily be added . Proponents of each action could then choose between positive messaging for their product , negative messaging against a competitor ’ s product , or a combination of the two .
One need only turn on the television or log on to social media to confirm that corporations and governments alike believe that persuasive messaging campaigns are an effective way to increase sales and participation . One website estimated that over $ 187 billion was spent on advertising in the United States in 2015 ( Goodman , 2016 ). And it is clear that celebrities and athletes supplement their income by acting as influential figures on behalf of organizations hoping to increase their sales . Given the enormous sums of money already being spent on persuasive messaging , it follows that at least some measurable benefit comes from the effort . But how many messages are lost among the rest , failing to stand out , and how many have the exact opposite reaction from the original intent ? In short , models can provide insights into this process , which in turn could potentially lead to organizations gaining a larger return on their investments in messaging campaigns .
However , those types of results will likely only be obtained with continued experimentation and empirical analysis , and synthesizing the results of both can be made easier with agent-based models of human behavior and decision-making .
Acknowledgments
We thank our colleagues at MITRE Jon Cline and Sasha Lubyansky for their invaluable assistance in completing this research . We also thank Gary L . Klein and Chris Glazner for their continued input , peer review , and technical insights .
References
Ajzen , I . ( 1991 ). The theory of planned behavior . Organizational Behavior and Human Decision Processes , 50 ( 2 ), 179 – 211 .
Axtell , R . ( 2000 ). Why agents ?: On the varied motivations for agent computing in the social sciences . Center on Social and Economic Dynamics . Retrieved from http :// www2 . econ . iastate . edu / tesfatsi / WhyAgents . RAxtell2000 . pdf
Carmack , C . C ., & Lewis-Moss , R . K . ( 2009 ). Examining the theory of planned behavior applied to condom use : The effect-indicator vs . causal-indicator models . The Journal of Primary Prevention , 30 ( 6 ), 659 – 676 .
216