Table 2. Details of the linear mixed models (round and size) for the search time analyses.
| Model parameters | Hypothesis testing | ||||
|---|---|---|---|---|---|
| Model: Round. Random term = (1|BeeID) | |||||
| Variables | Coefficients | SE | X2 | d.f. | P-value |
| Intercept | -0.65 | 0.09 | |||
| Colour | 0.82 | 0.11 | 205.18 | 1 | <0.0001 |
| OT | 0.09 | 0.11 | 0.02 | 1 | 0.88 |
| Round | -0.02 | 0.03 | 5.03 | 1 | 0.02 |
| Colour:OT | -0.03 | 0.11 | 0.07 | 1 | 0.79 |
| Colour:Round | -0.01 | 0.04 | 0.13 | 1 | 0.71 |
| OT:Round | -0.04 | 0.04 | 1.18 | 1 | 0.28 |
| Model: Size. Random term = (1|BeeID) | |||||
| Variables | Coefficients | SE | X2 | d.f. | P-value |
| Intercept | -0.59 | 0.08 | |||
| Colour | 0.64 | 0.09 | 42.30 | 1 | <0.0001 |
| OT | -0.06 | 0.09 | 0.38 | 1 | 0.54 |
| Size | -0.005 | 0.003 | 3.29 | 1 | 0.07 |
| Colour:OT | -0.03 | 0.11 | 0.07 | 1 | 0.79 |
| Colour:Size | 0.008 | 0.003 | 6.14 | 1 | 0.01 |
| OT:Size | 0.004 | 0.003 | 1.20 | 1 | 0.27 |
In parenthesis = the most parsimonious random term. OT = odour treatment.
Flower size itself did not affect search time, but its interaction with colour did (Table 2, P = 0.01). To study this interaction, we reanalysed colours independently. When bees were searching for red flowers, search time increased with size (slope = 0.005, SE = 0.002; X2 = 4.59, df = 1, P = 0.03). For blue flowers, in turn, the slope of the regression was slightly negative (slope = -0.002, SE = 0.001), although not statistically different from zero (X2 = 1.22, df = 1, P = 0.27).