Economic Perspectives IV

Main Menu           Previous Topic                                                           Next Topic

Introducing "Absolute Incentives" Parameter

Note: in order to run the simulation referred to in this slide, go here, to the Java Applet version. You will be directed to download the latest version of the Java plug-in.

Now let us spice up the previous slide's simulation a little bit, by varying α - an "absolute incentives" parameter. α is a number between 0 and 1, and determines agent learning and sharing dynamics in accordance with this topic's hypothesis. That is, individually learned information goes up with increasing α, while the percentage of shared information goes down. For instance, when α = 0, each agent learns a small amount of information during every time step (say 0.01), and shares all of it with her partners. At α = 1, she learns a much larger amount (0.1), but keeps it all to herself.

In the simulation to the left, α increases by increments of 0.01 from 0 to 1 over the course of 100 time steps, its progress tracked by the blue line in the lower graph. At the same time, the amount of new information acquired by the organization at every time step is displayed as a green line in the graph above. This number is equivalent to the slope of the Total Information measurement from the previous slide's fixed-α simulation. We want it to be as high as possible, so that the organization as a whole produces information at maximum rate.

Note the inverted-U shape of the graph. It tells us that there is an optimal value of α somewhere between 0 and 1, corresponding to just the right mixture of absolute and relative incentives. It maximizes the organizational output by forcing the agents to make the best possible choices with respect to learning and sharing information. In the next slide, we will see how this optimal value may be affected by the network's connectivity.

                   Previous Slide                                                           Next Slide