Table 3. Principal component loadings of the behaviours.
| Behaviour | PC 1 | PC 2 |
|---|---|---|
| Contact-D | 0.76457 | 0.05006 |
| Contact-F | 0.83250 | 0.12250 |
| Contact-L | -0.89613 | -0.13959 |
| Inactivity-D | -0.41347 | -0.85549 |
| Exploration-D | 0.15037 | 0.82661 |
| Exploration-L | -0.19767 | -0.63679 |
| Grooming-D | -0.08038 | 0.42876 |
| Activity-D | 0.56716 | 0.61371 |
| Activity-L | -0.49598 | -0.21244 |
| Running-D | 0.47287 | 0.34835 |
| Vocalisation-F | 0.38216 | 0.07442 |
| Change of segment-F | 0.70109 | 0.51393 |
| Object segment -L | -0.87210 | -0.15841 |
| Object segment-D | 0.83888 | 0.12061 |
| Object neighbouring segment -L | -0.73723 | -0.22699 |
Loadings of the behaviours in principal component (PC) 1 and PC 2 gained from the principal component analysis of the novel-object test; loadings above 0.71 in bold type, loadings above 0.63 in italics (cf. [40] loadings above 0.71 rated “excellent”, loadings above 0.63 rated “very good”); D = duration (total time in s), F = frequency, L = latency (time in s until behaviour was first shown).