Figure 3. After conditioning Drosophila do not discriminate decaborane from β-mercaptoethanol. All graphs represent the mean relative distribution of flies in the arms of the maze ± SEM after the indicated treatment. Significant differences are calculated relative to the performance of naïve animals (open bars) by LSM contrast analysis and are indicated on the graphs with a single star. A, ANOVA indicated significant differences (F(2,59) = 8.5292, p = 0.0006). Flies shocked in the presence of β-mercaptoethanol augmented their avoidance toward it (light gray bar, p = 0.0005) and surprisingly a similar augmentation was shown by animals shocked in the presence of decaborane (dark bar, p = 0.0006). Indeed, the performances of the two differently trained groups were not significantly different (p = 0.5915). n ≥ 12 for all conditions, with a total of 900-2000 flies per condition. B, In the converse experiment ANOVA indicated significant differences (F(2,30) = 15.3343, p < 0.0001). Flies shocked in the presence of decaborane avoid it instead of being attracted to it as naïve animals do (light gray bar, p < 0.0001) and a similar response is observed in β-mercaptoethanol-trained animals (dark bar, p = 0.0006). The two trained groups do not show significantly different responses (p = 0.1150). n ≥ 8 for all conditions, with a total of 500–900 flies per condition. C, Animals shocked against benzaldehyde (10%) or (10%) hexanol do not change their spontaneous response to β-mercaptoethanol. ANOVA indicated significant differences (F(3,60) = 7.9061, p = 0.0002). Subsequent contrast analysis revealed significant differences in the performance of β-mercaptoethanol-trained animals (light gray bars) versus those trained with benzaldehyde (p = 0.0006) and hexanol (p < 0.0001). The responses of flies trained with benzaldehyde and hexanol (dark bars) were not significantly different (p = 0.3997 and p = 0.0813, respectively) from naïve animals (open bars). n ≥ 9 for all conditions, with a total of 700–2000 flies per condition. D, Animals shocked against benzaldehyde or hexanol do not change their spontaneous response to decaborane. ANOVA indicated significant differences (F(3,43) = 4.9331, p = 0.0052). Contrast analysis revealed significant differences in the performance of decaborane-trained animals (light gray bars) versus those trained with benzaldehyde (p = 0.0003) and hexanol (p = 0.0004). The responses of flies trained with benzaldehyde and hexanol (dark bars) were not significantly different (p = 0.2403 and p = 0.1561, respectively) from naïve animals (open bars). n ≥ 8 for all conditions, with a total of 500–900 flies per condition. E, ANOVA did not indicate significant differences (F(2,26) = 0.1105, p = 0.8959). Anosmic animals shocked in the presence of β-mercaptoethanol or in the presence of decaborane did not augment their response to either (light gray bar, p = 0.6792 and dark gray bar, p = 0.7059, respectively). Furthermore, the performances of the two differently trained groups of flies were not significantly different (p = 0.9624). n ≥ 8 for all conditions, with a total of 600-700 flies per condition. F, In the converse experiment, ANOVA also did not indicate significant differences (F(2,48) = 0.7551, p = 0.4757). Flies shocked in the presence of decaborane or β-mercaptoethanol did not change their irresponsiveness toward decaborane (light gray bar, p =0.24 and dark bar, p = 0.42, respectively). The responses of the two differently trained groups were not significantly different (p = 0.69). n ≥ 8 for all conditions, with a total of 800–1000 flies per condition.