Table 1 For each period, the mean ± standard error of the features that describe the diurnal intake pattern of included growing-finishing pigs, obtained from the generalised additive models (GAMs).

From: Circadian rhythms and diurnal patterns in the feed intake behaviour of growing-finishing pigs

GAM feature

Period 1

Period 2

Period 3

Period 4

Period 5

P-value

ICC

Days 5–18

Days 19–32

Days 33–46

Days 47–60

Days 61–74

Pen

Pig

Timing of highest intake peak (Hours post-midnight)

16.4 ± 0.2a

14.6 ± 0.2b

15.7 ± 0.2c

16.0 ± 0.2ac

16.3 ± 0.2a

< 0.01

0.04

0.39

Highest intake peak (Detrended g/h)

195 ± 5a

213 ± 5b

228 ± 4c

236 ± 5c

245 ± 9c

< 0.01

0.02

0.10

Width of highest peak (Detrended g/h)

17 ± 2a

18 ± 1ab

21 ± 1b

23 ± 2b

24 ± 3ab

0.01

0.00

0.06

Lowest intake (Detrended g/h)

41 ± 2a

38 ± 2ab

35 ± 2bc

32 ± 2cd

29 ± 2d

< 0.01

0.06

0.49

Proportion of intake at night

0.17 ± 0.00

0.17 ± 0.00

0.18 ± 0.01

0.17 ± 0.01

0.17 ± 0.01

0.25

0.12

0.32

Lowest probability to start eating

0.18 ± 0.01a

0.17 ± 0.01bc

0.17 ± 0.01ab

0.15 ± 0.01cd

0.14 ± 0.01d

< 0.01

0.00

0.53

Highest probability to start eating

0.63 ± 0.02a

0.65 ± 0.01b

0.68 ± 0.01bc

0.69 ± 0.01cd

0.72 ± 0.01d

< 0.01

0.11

0.44

  1. The P-value reflects whether the feature differed between periods, and if so the superscripts (a, b, c and d) show which periods differed. For each feature, the intraclass correlation coefficient (ICC) is given for both pens and pigs, reflecting which part of the variation in the feature is explained by a pen resp. pig effect. Higher values for the ICC indicate more variation explained by the pen or pig effect, meaning that for low ICCs pens/pigs differed more from each other and were more consistent across time than for high ICCs. Note that for the width of the highest peak higher numbers reflect narrower peaks and vice versa.