Fig. 3

Relationships between learning performance and responsiveness to sucrose and light. Learning performance was assessed as the percentage of correct choices made in the non-reinforced test of each task. Sucrose responsiveness was quantified via the gustatory response score (GRS), which represents the number of proboscis extensions in response to antennal stimulation with water and six ascending sucrose concentrations. Responsiveness to light was quantified as the mean walking time (in seconds) to each of the six different relative light intensities. The composite phototaxis score was calculated from the six mean walking times via principal component analysis. No significant correlations were found between learning performance and the phototaxis score (n = 24): (A) 1^st phase of reversal learning (1^st phase RL): rho = 0.35, p = 0.09, R² = 0.12; (B) 2^nd phase of reversal learning (2^nd phase RL): rho = 0.22, p = 0.30, R² = 0.01; (C) Negative patterning (NP): rho = 0.35, p = 0.09, R² = 0.02. (D) No significant correlation was found between the GRS and the phototaxis score (rho = -0.09, p = 0.67, R² = 0.007). Each dot represents the data from a single bee. The regression red dashed lines represent nonsignificant correlations and are displayed alongside their 95% confidence intervals shown as gray dotted lines.