Table 6

For nearly all combinations of parameter values, plasticity accelerates the discovery of a new optimal genotype network

N

α

ωnative

M

Sample size

Mean t*, control

Mean t*, plast

p-value


8

0.02

0.5

1000

498

160.3

55.1

< 2.2 × 10-16



0.08

498

160.3

53.98

< 2.2 × 10-16



0.01

498

160.3

53.49

< 2.2 × 10-16



0.05

200

498

482.27

129.2

< 2.2 × 10-16


10000

498

63.06

25.14

< 2.2 × 10-16


0.2

1000

498

53.57

29.74

< 2.2 × 10-16



0.8

498

235.86

70.17

< 2.2 × 10-16



0.95

498

251.77

79.7

< 2.2 × 10-16


16

0.01

0.5

1000

467

2476.75

1858.21

4.2 × 10-6



0.05

462

2437.04

1886.83

0.0001328



0.025

200

248

3381.98

3389.867

0.778


10000

500

548.65

392.58

1.5 × 10-14


0.2

1000

500

225.74

187.78

5.4 × 10-7



0.8

283a

4172.47

3494.55

0.00204


The number of generations that a population takes to 'discover' a circuit in a new genotype network is significantly lower when we allow plasticity (t*, plast <t*, control), according to a Wilcoxon signed-rank test.

The value of d is that of the old genotype network. We analyzed 500 pairs of evolving populations for each combination of N, c and d. We discarded population pairs in which any of the populations had not reached the new genotype network by the end of the simulation (t = 104). Thus, our actual sample size was lower than 500 populations. The remaining parameter values are as follows: d = 0.25; μ = 0.5. c ≈ 0.4 when N = 8 and it equals 0.2 when N = 16.

a We analyzed 1500 pairs of evolving populations for the simulations that we report in this row, but the populations found the new genotype network only in 283 cases.

Espinosa-Soto et al. BMC Evolutionary Biology 2011 11:5   doi:10.1186/1471-2148-11-5

Open Data