Preference Co-operation in General Social Dilemmas: Complements

Consider a social dilemma where public-goods contributions are complements. This corresponds to Evolutionary Stability, Co-operation and Hamilton's Rule, Ingela Alger and Jörgen Weibull, section 4.4.2.

Model

Let player strategies x, y be strategy profiles in X, the set of strategies in a symmetric personal-fitness game G = (X, π) fully defined in sec. 2.1 of the paper. Let behaviour rules φ, ψ be functions of φ : X → X assigning a non-empty subset of the strategy space as a response to the other player's strategy. Then, given strategies

Equation 21,

a pair of real-valued traits α and β is mutually compatible when satisfying

Equation 25.

In the first equation, α is the first player's weight assigned to the other player's utility. In the second case, this is β. In words, a positive α or β expresses altruism; a negative value, spite; and a zero value, selfishness. Next, let x*(α, β) be the unique strategy of an α-altruist when matched to a β-altruist. The increase in fitness is then

Equation 26.

Finally, letting H be the gradient of evolutionary stability (as defined in section 2.2 of the paper), the evolution gradient over α, the incumbent degree of altruism, and α′, the mutant degree, where subscripts are derivatives, is defined as

Equation 28.

Gallery

Consider the the multiplicative contribution case,

Equation 39.

Stills

Fixing τ = 0.5 (strategic substitutes), relatedness r = 0.5, and cost c = 1, this produces the following static plots of function H, where v is linear, over intervals of α and β.

Full-size | eps
Full-size | eps
Full-size | eps

Animations

Fixing cost c = 1 and relatedness r = 0.5, fluctuate τ monotonically-increasing over [0, 1] with fixed τ = 1.0 (strategic independence).

Full-size
Full-size
Full-size

Fixing cost c = 1 and τ = 0.5 (strategic substitutes), fluctuate r montonically-increasing over [0, 1].

Full-size
Full-size
Full-size

← Back to index. Kristaps Dzonsons, $Date: 2011/02/02 16:45:43 $