Figure 1.
Ethanol increases sting extension response (SER) threshold of honey bee foragers A) Mean voltage threshold for SER for bees in 0%, 2.5%, 5%, 10%, and 20% ethanol treatment groups.
Increased ethanol concentration treatment is associated with increased voltage threshold for SER. Linear regression of mean threshold response is statistically significant (p<0.05). B) Sting response before threshold. Proportion of individuals responding before threshold to the stimulus increased as ethanol concentration increased. Comparison between concentration groups using Pearson's X2 revealed significant differences across the groups (X2 = 4.75, df = 4, p<0.05). Numbers above bars are bees tested at each ethanol concentration.
Figure 2.
ESA conditioning in honey bee foragers using 12 volt, 50
Conditioned individuals (solid line) showed a greater increase in learning index over time compared to the control group (hedged line). Two-way ANOVA resulted in a significant time effect (F (9,153) = 4.05, p<0.01), significant group effect (F (1,17) = 21.02, p<0.01), and significant time and group interaction effect (F (9,153) = 5.74, p<0.01).
Figure 3.
Ethanol treatment affects ESA conditioning in honey bee foragers.
A) Learning index over time for 0%, 2.5%, 5%, 10%, and 20% ethanol treatment groups (see legend on plot). ANOVA analysis resulted in non-significant treatment effect (F (36,855) = 0.59, p = 0.97), significant time effect (F (9,855) = 4.60, p<0.001) and non-significant time and treatment interaction (F (4,95) = 1.51, p = 0.21). These results demonstrate that even with ethanol administration bees were able to learn to avoid color associated with 12 V punishment. However, early phase of training differs across ethanol treatment groups, at 5% or higher ethanol treatment longer training time is required for similar improvement in learning index (see Results). B) Mean and SE of time spent in shock side for 0%, 2.5%, 5%, 10%, and 20% ethanol treatment groups. Increased ethanol concentration treatment is associated with an increase in time spent in the shock side during the ESA conditioning, indicating lower learning performance. Linear regression of mean time spent in the shock side is statistically significant (p<0.05).