In my previous post I wrote about how my colleagues and I found that bumblebees could learn which flowers had pollen, and remember this long-term. However, most bumblebee individuals don’t only forage for nectar or only for pollen; they collect both. This begs the question: if a bee collects nectar from one type of flower and then collects pollen from another type of flower, can she remember which flower type had which reward? It might seem obvious that if a bee can remember which flowers had nectar and which had pollen separately, then she should be able to learn these both at once. However, it is possible that she just learns that a flower has a reward, and doesn’t differentiate between the two types of food (nectar and pollen).

To address this question, we carried out an experiment which was recently published in the journal Biology Letters. We let bees forage on artificial flowers of two types: one with a yellow corolla and one with a blue corolla. Either the yellow-corolla flowers had nectar (but no pollen) while the blue-corolla flowers had pollen (but no nectar) or the reverse.

To train bees by giving them a chance to get experience with both flowers, I let an individual bee into a room where she could fly around and collect nectar and pollen from the artificial flowers. She then returned to her colony, removed her pollen loads and regurgitated nectar into honeypots in the colony (making honey). After a few times of going between the foraging array and colony it seemed like bees had learned where to go to collect nectar and where to go to collect pollen. It was easy to tell whether the bee was trying to collect nectar or pollen from a given flower because bees collected pollen from the artificial anthers of the flowers and nectar from a well at the base of the anther (see video below). 

To test whether bees had learned where to go, I then gave each bee a ‘test’. In this test, I presented a bee with four artificial flowers. Two of these flowers were the ones she had previously encountered (blue and yellow) and two of these flowers were completely new (being orange and purple). In this test phase, none of the flowers had any nectar or pollen on them.

I found that if a bee had been trained to find pollen on yellow flowers and nectar on blue, then she searched for pollen on the anthers of the yellow flowers, and probed for nectar in the nectar wells of the blue flowers. This showed that bees remembered which flowers had previously contained which specific reward.

Interestingly, the bees also generalised what they had learned for yellow and blue flowers to the two novel colours (purple and orange). This meant that if a bee had learned that yellow flowers had pollen she also searched orange flowers for pollen (the colour most similar to yellow). Similarly, if she had learned that blue flowers had nectar she also searched purple flowers for nectar (the colour most similar to blue).

This finding that bees can simultaneously learn which flowers have nectar and which have pollen teaches us something new about bee behaviour. However, this discovery also means that we can now use bees to look at how animals more generally learn when they are dealing with different types of reward. As humans, we don’t just have one type of reward; our rewards vary both within a particular context (e.g. eating different types of food) but also we can be ‘rewarded’ across different contexts (e.g. the reward of eating versus the reward of sex). Learning about multiple types of reward is presumably more difficult and potentially confusing than learning about a single reward type. Bees now give us a way to investigate these questions in more detail.


Muth, F., Papaj, D. R., & Leonard, A. S. (2015). Colour learning when foraging for nectar and pollen: bees learn two colours at once. Biology Letters, 11(9), 20150628. doi:10.1098/rsbl.2015.0628