This is a very timely new development in invertebrate neuroscience. Hot on the heels of the publications of the Honeybee genome, Vergoz et al. present a new classical (Pavlovian) conditioning procedure to complement the already classic "Proboscis Extension Response" (PER) conditioning. While both procedures are instances of olfactory conditioning, PER conditioning is appetitive and Vergoz et al.'s Stinger Extension Response (SER) conditioning is aversive.
With this development, the honeybee community ups the ante for the Drosophila community, which has been using complementary appetitive and aversive olfactory conditioning paradigms for a few years now. In Drosophila, the punishing electroshock seems to be mediated by dopaminergic neurons, while the sugar reward appears to be mediated by octopaminergic pathways. Vergoz et al. have used a pharmacological approach which suggests a similar dichotomy in honeybees as well. While it is reassuring that these results appear to confirm other insect studies, these experiments also contain the only weakness of the paper: the authors could have used the symmetric design of the two conditioning procedures (PER and SER) to counter the standard critique of pharmacological approaches (concerning either the specificity or the concentration of the drug) by inhibiting each pathway in each procedure. Symmetrical results with each drug blocking learning in only one of the two procedures but leaving the other unaffected would have been even more convincing than the current results already are. However, I expect this new development to yield much more interesting insights than merely confirming results from Drosophila, which alone warrants publication of the work as it is. I expect this paper to be seminal the the invertebrate learning and memory field.
Procedurally, the PER/SER pair has several advantages over the Drosophila experiments. For one, the behavioral output both during training and during test is a simple reflex in SER/PER. In Drosophila olfactory conditioning, there is no way to assess training performance and the test procedure involves an operant choice in a T-Maze. The Honeybee paradigms are single-animal preparations, while the fly experiments are mass assays. With the vast genetic toolkit in Drosophila and the advent of RNAi in Honeybees, both preparations will make for a great dynamic duo.