Thresholds of Behavioral Flexibility in Turbulent Environments for Individual and Group Success
Behavioral Flexibility
The final major dynamic in this model is agent adaptability . One way to express this is to allow agents to change type between clustering and spreading . As we just saw above , if v = 2 and q = 0 then it is best to be a clustering agent . Allowing agents to change from spreading type to clustering type means that they can capitalize on — adapt to — this environmental condition and earn much higher utility than if they had to remain spreading types . The probability of changing type ( r ) can range from r = 1 , where agents always change type when a utility threshold is not met ( maximum flexibility ), to r = 0 , where agents never change type regardless of how poorly they are performing in a given environment ( minimum flexibility ).
One last variable that we can modify is the threshold ( θ ) below which agents are triggered to potentially change strategy ( depending on r ). Lower thresholds correspond with quicker satisfying . Recall that a perfect score for any agent over 50 runs is 50 . If θ = 1 this means that very early on we should expect to see agents stopping changing strategies . Even if later on the environment changes such that it would be advantageous to change strategies , these agents will not change . While this may seem like a disadvantage , as a preview of the results , and an illustration of the counterintuitive outcomes of even such simple agent-based models , we will see that extremely high thresholds are surprisingly actually associated with lower group utility .
A Single Time Step
To summarize , the events that take place during a single time step of the
model are as follows .
1 . Agent X is located somewhere on the lattice . The location is randomly assigned . The probability the agent is a clustering agent is given by ( p ). v is randomly assigned to either 0 , 1 , or 2 . θ between 0 and 1 is randomly assigned , as is the probability r .
2 . Agent X calculates the number of relevant neighbors , given by vision . 3 . The agent will then choose to move or stay :
a . If Agent X is a clustering type , it will move to a neighboring cell if the number of the agents ( n ) in the neighboring cell exceeds the number in Agent X ’ s current cell . If there are more than one neighboring cells with more agents , Agent X will move to the cell with the highest number . If there are no neighboring cells with n greater than the agent ’ s current cell , Agent X will stay in its current cell .
125