In days past, we used to lure bees through a narrow entrance into a lab room for experiments. They got strong sugar water inside, which was what kept them coming back.
Nowadays, we test bees outdoors on a table top. Humans, with the help of a stop watch, replace video camera and computer in tabulating the areas in which the bees are searching.
But the tables and set up outdoors are similar to the times in the past when the bees were tested indoors (below).
Brightly coloured landmarks guide the bees to the target. The entire array of sugar dish and landmarks is moved about on the table from one visit to the next, so that the landmark constellation is the only reliable source for pinpointing the reward.
This story below was conducted in this setting, just outside the little house that is my office building. A picture of this ended up in an article in the Journal of Experimental Biology.
In one study, we (Cheng and Wignall, 2006, Animal Cognition, 9, 141-150) investigated the honeybee's memory when she was given two different tasks in succession. She was given Task 1, tested on Task 1 (Test 1), then given Task 2 at the same location, and tested on Task 1 again (Test 2). We found that performance on Test 2 was sometimes worse, a retroactive interference effect. This happened when the two tasks had different and incompatible response requirements. When we minimised this response competition, the retroactive interference effect was much reduced or eliminated altogether. We concluded that honeybees hold on to memories. The interference effect arises from uncertainty about which of the two memories to rely on for a test.
In another study (Cheng, 2005), the
retroactive interference effect was eliminated if Task 2 took place in a
different place. In the picture above, Task 1 would be at one of the tables,
and Task 2 at one of the other tables. Having a context change, a shift
of no more than a couple of meters, eliminated the retroactive interference
effect even when the two tasks had competing response requirements. Context
triggers the correct memory, a theme often found in studies on arthropod
learning and memory.
Nowadays, we test bees on open grass in an invertebrate compound:
Sometimes, we find enclosures for other arthropods, or potted trees for a Queensland fruit fly program run by Phil Taylor. In the back is a 'bee shed' that Andy Barron uses for his bee research.
Honeybees from a nearby hive come to an outdoors-style plastic table for experiments:
Catherine Prabhu used such set ups for the experiments in her thesis. These honeybees were trying to discriminate the smells coming out of the containers in the corners of each 'patch' to find the patch that contains sugar water rather than tap water.
Bees sometimes had to discriminate visual stimuli in the form of coloured cardboards:
The general theme of her research was to examine how honeybees dealt with conflicting evidence, contingencies that changed over time. In one study, bees were rewarded for visiting one colour (e.g., yellow) for 20 trials, and then conditions changed and they were rewarded for visiting another colour (e.g., green) for the next 10 trials. At different delays post training (and without further rewarded training trials), the bees were tested with tap water on all the cardboards. Their preferences showed a circadian pattern:
Task-1 target refers to the rewarding colour in the first phase. Searching at this target was compared with searching at the target colour for phase 2. (The never-rewarded third colour was hardly visited, and can be ignored.) We can see that immediately after training in phase 2, the bees preferred the target associated with phase 2. As time passed, the preference for the target associated with phase 1 increased, up to 22 h post training, which corresponded roughly with the start of training in phase 1 the day before. Two hours later, which corresponded to the time at which training in phase 2 ended, the preference had reversed again. For the full story, read the paper: Prabhu, C., & Cheng, K. (2008). One day is all it takes: circadian modulation of the retrieval of colour memories in honeybees. Behavioral Ecology and Sociobiology, 63, 11-22.