1 / 26

The Emergence of Symbiotic Groups from skill-differentiation and tags

This paper explores how cooperative groups of individuals with complementary skills can emerge without elaborate mechanisms such as contracts or reputation systems. It proposes the use of tags as a recognition mechanism for group membership and cooperation. The model shows that when individuals with similar tags cooperate, they can form small cooperative groups that can grow and flourish over time.

julier
Download Presentation

The Emergence of Symbiotic Groups from skill-differentiation and tags

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. The Emergence of Symbiotic Groups from skill-differentiation and tags Bruce EdmondsCentre for Policy Modelling The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-1

  2. Part 1: The Issues Summary of problem and proposed solution The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-2

  3. The Problem I Am Focussing On • How can cooperative groups of individuals with complementary skills emerge… • in an evolutionary setting where it is always possible for parasitical individuals to join… • …without elaborate mechanisms such as: contracts; reputation by name; etc.? • In other words how might cooperative groups of collaborating individuals have initially developed as a foundation that might allow for more sophisticated social mechanisms to be developed later? The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-3

  4. Constraints upon possible solutions • Individuals must be able to be selfish so that this is undetectable by others in advance • There must be no direct accounting, reputation or punishment system • There should be no “magic” increase in value as a result of donation • Plausible that any such mechanism might be evolved (i.e. pretty simple) The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-4

  5. The Proposed Solution Condition • There might be a disparity of skills/abilities among the individuals due to one of: • Time/resources takes to develop skills (e.g. different trades in humans) • Necessary trade-offs between different abilities (e.g. heat retention due to size vs. speed) • Sheer evolutionary happenstance • So that it is advantageous for groups of individuals with complementary skills to form where group members share • I.e. symbiotic relationships The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-5

  6. Proposed Group Member Recognition Mechanism: tags • Tags are socially observable cues… …that can be used as a (fallible) guide as to group membership/whether to cooperate … …depending on how “close” they are to one’s own (set of) tags. I.e. the rule is: cooperate with those with similar tags • Can be: single- or multi-dimensional; continuous or discrete • Are not necessarily unique to an individual – they can be “forged” by others • Are not necessarily associated with any other characteristics of the individuals who have them The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-6

  7. A Brief History of Tags • Idea proposed by John Holland in 1993 • Developed by (among others): Rick Riolo (1997, 2001, etc.); and David Hales (2000, 2001, etc.) • Nature paper in 2001 by Riolo, Cohen and Axelrod exposes tags to wider audience • But this model is flawed (Roberts & Sherrat 2002, Edmonds and Hales 2003) • Further work fixes these flaws, explores conditions where tags work and works towards applications • This paper is part of this development Censored due to lack of time The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-7

  8. How tags work • By some process (e.g. chance) a small cooperative group with similar tags occurs • Due to benefits of cooperation those in the group reproduce more than others • Eventually a parasite appears in the group • The parasite (and its progeny) do even better than the cooperators in the group • Thus parasites reproduce more and come to dominate the group and cooperation ceases • Thus the benefit of the group (compared to others) disappears and the group dwindles away • But in the meantime other cooperative groups may have formed based round other tags etc. The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-8

  9. So… • For the tag mechanism to promote cooperation it is necessary that: • There is benefit to individuals to cooperate with similar others (this is easy to arrange) • When defectors arise they are self-defeating (e.g. they “kill” their own group) • New “seed” groups are always arising, so that when a dominant group dies others can grow • Thus there is a continual dynamic process of “tag groups” arising and falling allowing cooperation to flourish in new groups The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-9

  10. Part 2: The Model Design and Results The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-10

  11. Basic Design Ideas – static structure • Discrete time simulation • There are a variable number of individuals • There are n (necessary) food types • Each individual has: • A limited store for each food type (1 when new) • One skill, it can gather only one type of food • A tag value in [0, 1] • A tolerance value in [0, 1] • The tag and tolerance may be mutated during reproduction The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-11

  12. Each iteration individuals: • two new random individuals enter from outside • get some randomly distributed food depending on their skill • are randomly pairedp times • will donate share of some of any excess to those paired with if other’s tag is within its tolerance to its own tag (get 95% of value) • all stores taxed0.25; die if any 0, reproduce if all > 4 The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-12

  13. Tag value Tolerance value When donations occur between individuals Range of tag values The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-13

  14. The 3 colours indicate the 3 skills Individual with tag 0.79 and tolerance 0.4 etc. Age 1 Age 2 Age 3 Animation of example run The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-14

  15. Apparent phases in the dynamics of this model • No viable population • Growth phase of new seed symbiotic group • Resource competition between those within a group of symbiots • Predator-prey type dynamics between a parasite and collection of symbiots • Destruction of viable population The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-15

  16. Growth of seed group Growth of new seed group Parasitic Pred-Prey Co-existance Unviable population Unviable population Example run – subpopulations The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-16

  17. Population profile for a typical run Tag values Time The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-17

  18. Av. proportion of time population not viable against maximum toleranceand number of pairings: 2 Food types Proportion of time population not viable MaximumTolerance Number of pairings The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-18

  19. Av. proportion of time population not viable against maximum toleranceand number of pairings: 3 Food types Proportion of time population not viable MaximumTolerance Number of pairings The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-19

  20. Av. proportion of time population not viable against maximum toleranceand number of pairings: 4 Food types Proportion of time population not viable MaximumTolerance Number of pairings The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-20

  21. Viability and donation rate against size of reservoirs (averaged over 25 runs) Percent of time population is viable (with SD bars) Average donation rate The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-21

  22. What the model suggests to increase cooperation • A sufficient number of pairings (compared to population size) • A sufficient but low size of reservoir (i.e. resource is difficult to store) • A larger maximum tolerance (though this is a mixed effect) • A smaller number of necessary food types • Delay in deleterious effects of parasites are more likely to kill any symbiot groups • Mechanisms that facilitate the continual formation of new seed groups The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-22

  23. Part 3: Concluding Discussion Consequences, limitations, and future directions The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-23

  24. So far so good… • This seems to be a mechanism that allows the medium-term duration of symbiotic groups under a range of conditions …however, every now and then the whole population in this single niche collapses …thus it requires some continual “seeding” from outside (albeit at a very low level). • But it still might provide enough stability to allow further group mechanisms to develop. The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-24

  25. What would be nice is…. If I (or someone) built on this so that …new “seed” groups carried some of the characteristics of the “parent” groups …so that those groups with better defences against parasites would preferentially prosper compared to others and hence …themselves evolve! Also, if the assumption of: all individuals having a unique skill but insufficient for its own survival/fitness could be weakened …maybe by becoming a more realistic model. The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-25

  26. The End Bruce Edmonds bruce.edmonds.name Centre for Policy Modelling cfpm.org The Emergence of Symbiotic Groups, Bruce Edmonds, AISB Symp. on Socially Inspired Commuting, Univ. of Herts. 2005. http://cfpm.org/~bruce slide-26

More Related