Long Memory Properties and Complex Systems
capacity . Whenever a patch is exploited , the amount of sugar is decreased , but it has a grow back constant that allows restoring part of its sugar capacity .
These grid cells are randomly initialized in order to introduce spatial complexity in the simulation , as presented in Figure 18 .
Figure 18 . Sugar spatial distribution .
Hence , as can be seen in Figure 17 , the darker cells represent patches with higher sugar values , and the lighter ones patches with lower sugar values .
Moreover , individual agents that have different states initialized randomly according to a uniform distribution populate this grid :
• Amount of stocked sugar ( defined by a range of Minimum and Maximum Initial Sugar Stocks )
• Metabolism ( defined by a range of Minimum and Maximum Metabolism rates )
• Vision ( defined by a range of Minimum and Maximum Vision capability )
• Life Expectancy ( defined by a range of Minimum and Maximum Life Expectancy )
Hence , having these variables randomly initialized , the agent actions turn to be heterogeneous among them . Moreover , they are placed randomly within this grid , as is shown in Figure 19 .
Given that , agents can search and gather sugar from cells within their vision range , they consume sugar from their stock according to their metabolism and they die if they run out of stock or if they achieve their life expectancy — when they die , other agents with random initial states replace them .
Furthermore , they can explore only one cell grid at each tick and they
41