Research Article

Repeatability and Heritability of Behavioural Types in a Social Cichlid

Table 4

Heritabilities 2 of offspring behavioural type using the variance components approach (GLMM REML).

Random effect(s)n pairs or parent, n offspringVariance pair or parentVariance broodVariance errorAsymptotic covariance matrixa 2 ± SE
Within pair or parentWithin errorBetween pair or parent versus error

Pair74, 13270.117***0.653***0.0007870.000677 −0.000046 0 . 1 5 2 4 ± 0 . 0 3 1 6
Mother49, 13180.121***0.665***0.0010430.000697 −0.000035 0 . 3 0 8 2 ± 0 . 0 3 4 7
Father50, 12810.097***0.664***0.0007240.000716 −0.000036 0 . 2 5 6 0 ± 0 . 0 3 0 8
Pair + Brood within paira74, 13270.091**0.069**0.614***0.0009350.000641 −0.000002 0 . 1 1 8 2 ± 0 . 0 3 8 0
Mother + Brood within mothera49, 13180.096**0.074***0.616***0.0010990.000649 −0.000007 0 . 2 4 5 3 ± 0 . 0 4 0 5
Father + Brood within fathera50, 12810.076*0.085***0.605***0.0008710.000647 −0.0000002 0 . 1 9 8 0 ± 0 . 0 3 8 7

aFor brevity, covariances involving the random brood effects are not given ( 𝑛 = 1 6 2 broods for 74 pairs, 𝑛 = 1 6 0 broods for 49 mothers, 𝑛 = 1 5 6 broods for 50 fathers).
𝑃 < . 0 5 , 𝑃 < . 0 1 , 𝑃 < . 0 0 1 .
Heritability estimates are twice the intraclass correlations for mother and father effects, standard errors calculated according to [70].