Simulating Heterogeneous Farmer Behaviors
Table 3 . Motivation to comply selection
Table 4 . Varying the influential figure start and stop times interval
Since our focus has been to inform persuasive messaging strategies , our final scenario investigates how the timing of an influential figure speaking out relative to the timing of the messaging campaign impacts behavioral intent . Again , we operate from the base case where messaging starts at t = 200 and lasts until t = 300 . We include the influential figure with 90 % of the population connected to them and those connected all have the highest motivation to comply with the influential figure ( i . e ., + 1 ). However , this time we varied the start and stop time of the influential figure in five different ways , as described in Table 4 .
The results of this final scenario are shown in Figure 8 . All five experiments showed an increase in the average behavioral intent across the population of agents . However , we also noted significant variation in the increase depending on when the influential figure spoke out . Experiment 5 — where the influential figure was active from 100 steps before the messaging campaign started to 100 steps after it completed — showed the highest increase in behavioral intent . The smallest increase occurred in Experiment 1 when the influential figure
spoke out before the messaging campaign started and stopped speaking out once the messaging campaign commenced . All of these experiments showed statistically significant increases in behavioral intent relative to the base case where no influential figure was present . These results are important because of the vast sums of money spent on endorsement contracts . The model suggests that timing of an influential figure ’ s endorsement relative to a broader messaging campaign can have different levels of impact on the overall population ’ s behavioral intent .
Discussion
Empirical studies show that static regional- or demographic-based models of human behavior do a poor job of predicting the impact that persuasive messaging campaigns might have on a given population ( Gupta & Chintagunta , 1994 ; Wind , 1978 ). Social psychology indicates that , even within particular regions or demographics , populations are heterogeneous , and individuals formulate their behavioral intent based on a variety of factors . As
214