Table 2.
Hypotheses (A) and model selection results (B–D) for analyses examining sources of variation in circulating corticosterone in juvenile and adult Red Crossbills
| 2A Models and associated hypotheses | ||||
|---|---|---|---|---|
| Model | Hypothesis | |||
| Context + rTL | Circulating corticosterone depends on individual or partner telomere lengths, independent of food availability | |||
| Context + body condition | Circulating corticosterone depends on individual or partner body condition, independent of food availability | |||
| Context * rTL | The relationship between circulating corticosterone and individual or partner telomere lengths depends on food availability | |||
| Context * body condition | The relationship between circulating corticosterone and individual or partner body condition depends on food availability | |||
| 2B Juvenile baseline CORT, individual predictors (n = 47) | ||||
| Model | Params | ΔAICc | Weight | R 2 |
| Context * Body condition | 6 | 0.00 | 1.00 | 0.65/0.59 |
| Context * rTL | 6 | 17.75 | 0.00 | 0.32/0.31 |
| Context + Body condition | 5 | 18.43 | 0.00 | NA/0.33 |
| Context | 4 | 19.11 | 0.00 | NA/0.24 |
| Context + rTL | 5 | 20.15 | 0.00 | NA/0.26 |
| 2C Adult stress-induced CORT, individual predictors (n = 46) | ||||
| Model | Params | ΔAICc | Weight | R 2 |
| Context + rTL | 5 | 0.00 | 0.53 | 0.79/0.3 |
| Context * rTL | 6 | 0.41 | 0.43 | 0.73/0.19 |
| Context | 4 | 5.58 | 0.03 | 0.73/0.18 |
| Context + Body condition | 5 | 10.38 | 0.00 | 0.73/0.19 |
| Context * Body condition | 6 | 14.10 | 0.00 | 0.73/0.2 |
| 2D Juvenile baseline CORT, social partner predictors (n = 46) | ||||
| Model | Params | deltaAICc | Weight | R 2 |
| Context * Partner rTL | 6 | 0.00 | 0.74 | NA/0.34 |
| Context + Partner rTL | 5 | 3.42 | 0.13 | 0.37/0.28 |
| Context | 4 | 3.70 | 0.12 | NA/0.24 |
| Context + Partner body condition | 5 | 7.97 | 0.01 | NA/0.25 |
| Context * Partner body condition | 6 | 11.71 | 0.00 | NA/0.25 |
Note: Circulating corticosterone was log-transformed before analysis and predictor variables were either measured from focal individuals to examine intrinsic influences (B, C) or social partners (D). Model tables in which the top supported model was the null model are not shown. All AICc tables associated with individual and social partner predictor variables can be found in Supplementary Tables 6 and 8, respectively. R2 values represent conditional (variance explained by fixed and random effects) and marginal (variance explained by fixed effects) values respectively. Conditional R2 with NA values represent estimates wherein the variance explained by random effects equals zero.