Collectives

Class #19 (Thu., Mar 22)

Reading:

  • Railsback & Grimm, Ch. 16.

Reading Notes:

Errata for Chapter 16:

  • Section 16.4.1, p. 213: The description of social status is not clear: it assumes ages are real numbers instead of integers. It should read:

    _The social status of a dog can be (a) “pup,” meaning its age is less than one;
    1. “yearling,” with age of 1 year; (c) “subordinate,” meaning age is 2 or greater but the dog is not an alpha; (d) “alpha,” the dominant individual of its sex in a pack; and (e) “disperser,” meaning the dog currently belongs to a disperser group, not a pack._
  • Section 16.4.1, p. 213: The first bullet of the description of pack formation should start “Determine how many times the disperser group meets another…”

  • Section 16.4.1, Fig. 16.1: Be aware that this figure does not depict the logistic function parameters in the Wild Dog model description. The figure shows a probability of 0.5 when the number of dogs in the population is 67% of carrying capacity (40 out of 60), while the model description says the probability is 0.5 when the population is 50% of carrying capacity. To match the model description, the figure’s curve should go through X = 30, Y = 0.5.

  • Section 16.4.2, p. 217: Setting the dog’s variable my-pack to the pack that created it (via the statement set my-pack myself) can cause a subtle error in the pack formation submodel. Disperser groups cannot form a pack with other disperser groups that originated in the same pack. In most people’s code, disperser groups use the value of my-pack from one of their dogs to determine which pack they came from. The error can occur when two disperser groups came from packs that have since died; when a disperser dog’s pack dies, NetLogo automatically changes its value of my-pack to the keyword nobody. Because disperser groups cannot form a new pack with another group that has the same value of my-pack, groups whose packs have died cannot combine with each other, even if they really came from two different packs. A solution is to set my-pack to the unique who number of the dog’s pack. Instead of set my-pack myself, use set my-pack [who] of myself. Use a similar statement when new dogs are created in the reproduction submodel. Then change the statements where a dog’s value of my-pack is compared to a pack, from (for example):

    set pack-members dogs with [my-pack = myself]

    to:

    set pack-members dogs with [my-pack = [who] of myself]

    And because my-pack now contains a who number instead of a pack, one of the last lines in the create-disperser-group-from code (page 220) needs to be changed, from:

    ask dogs-former-pack [set pack-members ... 

    to:

    ask pack dogs-former-pack [set pack-members ...

    This error appears to have little effect on model results.

  • Section 16.4.3, Figure 16.2: The figure should look like

    Corrected Fig. 16.2

    Corrected Fig. 16.2

arrow_back Prev: Class #18 "Stochasticity" Next: Class #20 "Patterns" arrow_forward