Table 3 Effect of communal litter size and litter birth order on relative investment gained from partner mother.

From: Cryptic kin discrimination during communal lactation in mice favours cooperation between relatives

Effect

Fixed effect

χ2

df

pa

All pupsb

Communal litter size

−0.006 ± 0.003

5.83

1

0.016

Pup from 1st vs 2nd born litter

−0.035 ± 0.017

4.09

1

0.043

Partner food intakec

1.675 ± 0.110

68.24

1

<0.0001

Partner relatedness

−0.001 ± 0.016

0.01

1

0.94

4 way interaction

 

20.18

11

0.043

Relatedness × communal litter size × birth order

10.49

5

0.063

First-born littersd

Communal litter size

−0.005 ± 0.003

2.34

1

0.13

Partner food intake

1.696 ± 0.144

38.63

1

<0.0001

Partner relatedness

−0.018 ± 0.020

0.81

1

0.37

Second-born littersd

Communal litter size

−0.007 ± 0.004

3.61

1

0.057

Partner food intake

1.683 ± 0.160

31.06

1

<0.0001

Partner relatedness

0.020 ± 0.025

0.64

1

0.42

Relatedness × communal litter size × partner food intake

13.70

4

0.008

Relatedness × communal litter size

 

9.34

1

0.002

Second-born litters in sister nestsd

Communal litter size

 

0.14

1

0.71

Partner food intake

 

21.30

1

<0.0001

Second-born litters in unrelated nestsd

Communal litter size

 

12.84

1

0.0003

Partner food intake

 

11.68

1

0.0007

  1. aValues in bolded text are statistically significant (p < 0.05).
  2. bMixed-effects models of relative investment gained from partner mother (mother included as a random effect in each model; pair had zero variance and so was not included). Interactions that had significant effects are shown. Data for all pups from 16 communal nests (n = 44 first born and n = 33 second born pups in sister nests; n = 41 first born and n = 40 second born pups in unrelated female nests), shown in Fig. 4 and Supplementary Data 1.
  3. cProportion of food intake over the labelling period that was due to partner mother.
  4. dAnalysis broken down into four different subsets to interpret the significant 4 way interaction for all pups. Adjustment for multiple post hoc comparisons using a conservative Bonferroni correction has  a threshold for significance of p < 0.0125.