Variable rewards and discrimination ability in an insect herbivore: what andhow does a hungry locust learn?

With the exception of honeybees, there have been fewgood invertebrate models for associative learning. Grasshoppers and locusts (Orthoptera: Acrididae) possessa number of characteristics that make them excellentcandidates for such studies, and in this paper we present anovel protocol, based on a Y-maze, that is specificallydesigned for studying their learning and choice behaviour.Three separate experiments were conducted usingindividual gregarious forms of the desert locust,Schistocerca gregaria. In our first experiment, colouredarms of a two-sided Y-maze provided a large or smallamount of wheat for nine choice-trials. In the secondexperiment, locusts discriminated odours with wheatrewards for nine choice-trials. The odour–wheat rewardcombinations were then reversed for an additional ninechoice-trials. For the third experiment, the locusts againdiscriminated odours, but here we used artificial foods andthe rewards differed in their concentration of proteinand digestible carbohydrate. The results indicate that,in addition to showing good acquisition of choiceperformance, the locusts also took less time to reach thelarger-rewarded option. The data indicate that ourprotocol is highly sensitive for recording choice behaviourin acridids and reveals the potential they have foradvancing our current understanding of associativelearning and the field of learning in general.