Collective Action Communities

Collective Action Communities preview image

1 collaborator

Default-person Jeremy Foote (Author)


community models 

Tagged by Evgeny Patarakin 6 months ago

Model group MAM-2016 | Visible to everyone | Changeable by group members (MAM-2016)
Model was written in NetLogo 6.0-M5 • Viewed 191 times • Downloaded 9 times • Run 0 times
Download the 'Collective Action Communities' modelDownload this modelEmbed this model

Do you have questions or comments about this model? Ask them here! (You'll first need to log in.)

Comments and Questions

Please start the discussion about this model! (You'll first need to log in.)

Click to Run Model

  mu ; Mean of the logisitic production function
  beta ; Steepness of the production function
  contribution-level ;; How much has been contributed
  progress ;; Ratio of project complete
  topics ;; Distribution of topics that a patch is "about"
  pindex ;; Index of the patch
globals [
  multiplier ; How much the parent patch multiplies the interest
  patch-gini ; Stores the gini of patch contributions
  turtle-gini ; Stores the gini of turtles
  patch-progress-gini ; Gini of patch progress
  interest ;; Baseline interest in projects
  resources ;; How much they have to contribute each tick
  contributions-made ;; How much they have contributed in total
  top-project ;; Current patch with highest utility
  topic-interests ;; Distribution of topic interests
  similarity-list ;; List of cosine similarity to each patch

to setup
  ;; Set the number of topics to use for topic-based
  let num-topics 20
  set-multiplier 1

  ;; Create the production functions for the patches
  ask patches[
    set mu random-normal global-mu 1
    set beta abs random-normal global-beta 1
  ;; If topic-based, then create topics
  if topic-based? [
    setup-patch-topics num-topics
  ;; Make the new turtles
  make-new-turtles num-turtles num-topics

to set-multiplier [x]
  set multiplier x

to make-new-turtles [n num-topics]
  repeat n [
    crt 1 [
      ;; Place turtle randomly
       setxy random-xcor random-ycor
       ;; Give it resources
       set resources abs random-normal .2 .05
       ;set resources random-float 3
       ;; And interest
       set interest random-float max-interest
       ;; Create topics, if necessary
       if topic-based? [
         set topic-interests make-topics num-topics

to go
  ask turtles[
    ; Find the best project
    ; Figure out whether to contribute to it
  ask patches[

; Creates topics for the patches, and creates an index to access the similarity matrix

to setup-patch-topics [num-topics]
  ask patches [
    ; Create random distribution of topics
    set topics make-topics num-topics
  ;; Create an index so that we can access the similarity list
  let n 0
  foreach sort patches [
    ask ? [
      set pindex n
      set n n + 1

; Turtle function; creates a turtle-patch similarity matrix

to make-similarity-list
  set similarity-list []
  foreach sort patches [
    set similarity-list lput cosine-similarity ? similarity-list

;; turtle function; reports the cosine similarity of a turtle and a patch's topics

to-report cosine-similarity [curr-patch]
  let patch-topics [topics] of curr-patch
  ; Dot product of topics
  let dot-product sum (map * topic-interests patch-topics)
  ; Then normalize by the magnitude of the vectors
  report dot-product / (make-magnitude topic-interests * make-magnitude patch-topics)

to-report make-magnitude [vector]
  ; Magnitude of a vector
  report sqrt sum map [? * ?] vector

;; Patch and turtle function

to-report make-topics [num-topics]
  let topic-list []
  repeat num-topics [
    ; For each topic, randomly choose how much of that topic
    set topic-list lput random 3 topic-list
  report topic-list

; turtle method

to find-top-project
  ifelse topic-based? [
    ; if it's topic-based, then look at all patches
    set top-project max-one-of patches [patch-utility]
  ; Otherwise, only look at neighbors
  set top-project max-one-of patches in-radius neighborhood-size [patch-utility]

to contribute-to-project
  ; Find the top utility of the top project
  let top-utility [patch-utility] of top-project
  ; If the benefit is greater than the cost, then contribute resources
  ifelse top-utility > resources [
    ask top-project[
      ; update the contribution level and progress of the project
      set contribution-level contribution-level + [resources] of myself
      set pcolor red
    ; Record that this turtle has made a contribution
    set contributions-made contributions-made + resources
    ; If they contributed, they are more likely to contribute in the future
    set interest interest * (1 + positive-reinforcement)
    ; But, they can't go above the max interest
    if interest > max-interest [
      set interest max-interest
    set color red
  ;; If they didn't contribute, then decrease their interest
  set interest interest * (1 - negative-reinforcement)
  set color blue

to update-patch-progress
  set progress get-patch-progress contribution-level

to-report patch-utility
  let x contribution-level + [resources] of myself ; What would the total contribution be?
  let weight [interest] of myself * multiplier / distance myself ; The utility is multiplied by the overall multiplier times my interest
  if topic-based? [
    ; Multiply the weight by how similar the patch topic distribution is to my interest
    set weight weight * item pindex [similarity-list] of myself
  let orig_utility weight * progress ; Current utility based on logisitic
  let total_utility weight * get-patch-progress x
  report total_utility - orig_utility ; Return the amount of utility added

;; Patch reporter; gives the progress at a given x

to-report get-patch-progress [x]
  let result 0
    ; Update the patch progress based on the current contribution level.
  set result 1 / (e ^ ((mu - x) / beta) + 1)
  ;If the exponential is too lrge, it means we're really close to 0
  set result 0
  report result

to recolor-patches
  set pcolor scale-color orange progress -1 2

; Turtle function

to ask-to-die
  if interest < .05 [

to update-ginis
  ; Code basically stolen from the wealth distribution model
  set patch-gini gini [contribution-level] of patches
  set turtle-gini gini [contributions-made] of turtles
  set patch-progress-gini gini [progress] of patches

;; this procedure recomputes the value of gini-index-reserve
;; and the points in lorenz-points for the Lorenz and Gini-Index plots

to-report gini [contributions]
  let sorted-contribs sort contributions
  let total-contribs sum sorted-contribs
  if total-contribs = 0 [
    report 0
  let contrib-sum-so-far 0
  let index 0
  let gini-index-reserve 0
  let lorenz-points []
  let num-contribs length contributions

  ;; now actually plot the Lorenz curve -- along the way, we also
  ;; calculate the Gini index.
  ;; (see the Info tab for a description of the curve and measure)
  repeat num-contribs [
    set contrib-sum-so-far (contrib-sum-so-far + item index sorted-contribs)
    set lorenz-points lput ((contrib-sum-so-far / total-contribs) * 100) lorenz-points
    set index (index + 1)
    set gini-index-reserve
    gini-index-reserve +
    (index / num-contribs) -
    (contrib-sum-so-far / total-contribs)
  report (gini-index-reserve / num-contribs) / 0.5

There are 6 versions of this model.

Uploaded by When Description Download
Jeremy Foote over 2 years ago Fixing documentation Download this version
Jeremy Foote over 2 years ago Bugfixes and new plots Download this version
Jeremy Foote over 2 years ago Added lots of plots; topic-based similarity Download this version
Jeremy Foote over 2 years ago Adding some features and info tab Download this version
Jeremy Foote over 2 years ago Working to add LevelSpace Download this version
Jeremy Foote over 2 years ago Initial upload Download this version

Attached files

File Type Description Last updated
Collective Action Communities.png preview Preview for 'Collective Action Communities' over 2 years ago, by Jeremy Foote Download
collective_action_abm.pdf pdf Design Document v1 over 2 years ago, by Jeremy Foote Download
JeremyFoote_May9.pdf pdf Progress Report - May 9 over 2 years ago, by Jeremy Foote Download

This model does not have any ancestors.

This model does not have any descendants.