Journal on Policy and Complex Systems
Figure 4 . Vision / neighborhood size of agents depicted by the shaded area ; from left to right : v = 0 , v = 1 , v = 2 .
The vision of agents is a proxy for neighborhood or vicinity . As vision increases the size of the lattice that is considered an agent ’ s neighborhood increases . Recall from above that the number of agents within a neighborhood is how an agent ’ s utility is calculated . This means K C increases with v , while K S increases as v decreases . A simple run of 30 time steps of the model with all type C agents randomly distributed over the lattice and then another run of all S randomly distributed confirms this . The average cumulative score for clustering agents is 50 ( perfect ) when v = 2 and 47 for spreading agents when v = 0 . ( It is not 50 because when v = 0 agents are restricted in their movement , so the model is sensitive to initial conditions ).
To further illustrate , consider the lattices in Figure 5 , where v = 0 on the left and v = 2 on the right .
Figure 5 . Calculating utility depending on vision / neighborhood size . If agent X is a spreading agent , then K X
= 1 when v = 0 because it is the only agent in its relevant neighborhood . However , when v = 2 this same agent earns K X
= 0.97 , because now there are three other agents in this spreading agent ’ s neighborhood .
To create environmental turbulence , let vision change with some probability ( q ). At q = 1 vision changes at each time step ( most turbulent ) and at q = 0 vision never changes ( completely static ). In the simplest version of the model , all agents have the same vision .
124