For the last 10 months or so, not a single day has gone by where I haven’t spent some amount of time thinking about honey bees, reading about honey bees, working with honey bees, or listing off cool honey bee facts to anyone willing to listen.  If you’ve been following the blog in recent months, then you may have already read a little bit about my missteps as a researcher along the way to my first full field season. If not, you can read those posts here and here. They’re largely considered modern classics in the behavioral blogging game.

Simply put, I study how consistent behavioral variation within individuals in a society affects how a group operates and performs. This idea of consistent individual variation in behavior is sometimes called “animal personality”—some individuals are consistently more bold, aggressive, or exploratory in comparison to others in the same population [1]. Theoretically, I could study how personalities affect group dynamics in many different animals. After all, many animals live in groups and we now know that animal personality seems to be the rule rather than the exception in organisms across the animal kingdom [2,3]. So then, why honey bees?

The Venn diagram above shows the intersection of my current research interests. On the left side is the animal personality stuff I mentioned before. On the right is the general notion that honey bees are an inherently intriguing animal society. They are complex. They are efficient. They are endlessly fascinating. More to the point, honey bees are a society mainly centered around work.  Honey bees are largely thought to be so successful in nature because of their division of labor—their ability to distribute tasks, such as offspring care and food collection, among many individuals and to have groups of workers specialize in various tasks [4].

Classic research in our field demonstrated that this allocation of tasks mostly works such that certain tasks are all performed by bees of the same age [5]. In fact, this type of system guides the allocation of work in many insect societies. However, we know that even individuals within these age-based castes can further specialize on certain tasks (i.e. they have a work-based “personality”)[6,7], but whether individual bees have larger preferences for certain kinds of tasks has yet to be explored. This is essentially the idea of a “behavioral syndrome”—when two or more personality traits or behaviors are consistently expressed together in the same individuals. Generally speaking, this could mean that individuals that are “bolder” are also more likely to be “aggressive” [1]. For bees working in a hive, this could mean that individuals that consistently prefer a certain task, such as building the nest, may also be more likely to perform other tasks, such as cleaning the nest [8].

More specifically, my research looks at bees that perform social immunity behaviors—behaviors that improve the health and disease resistance of the colony. To give a human example, think of washing your hands after using the bathroom: it’s a behavior that not only reduces your chance of contracting a disease but also your chance of spreading a disease to the people you interact with. Honey bees have many behavioral ways of fighting off disease, such as removing diseased larvae from the hive or grooming other bees [9], but no one has ever looked to see if these behaviors are usually performed by the same individuals.

A honey bee taking a diseased larvae (the white thing) out of the hive. This is a really important social immunity task for bees.

So that’s my research, more or less. But what does a normal day in the life of a behavioral bee scientist look like? Well, it’s highly variable. The bread and butter of behavioral research is the observation hive—a glass-walled hive that allows bees to behave as they would in a regular hive while also allowing an observer to watch individual bees as they perform their work. I set 4 of those up in April using some colonies I had out in the field as source colonies. I then gave them a month to get back up to full strength. This mostly means making sure they have additional food (sugar solution) to stimulate population growth and possibly adding in frames of emerging brood to help boost colony numbers. “Brood” is just another term for developing bees and “emerging brood” is a term used for bees that are at the last stage of development and are about to come out of their cell and get to work!

Since I’m interested in individual behavior, the next thing I needed to do for my actual experiment is mark bees so that they are individually identifiable. It takes anywhere from a few hours to about a full a workday to do anywhere between 300-500 bees. We use tags that are specifically made for bees, albeit for queen bees. From what I’ve been told, non-research beekeepers sometimes use them to keep track of sexually transmitted diseases in their hives. Putting them on worker bees essentially involves the same process as tagging a queen: we glue a numbered tag onto their thorax and then add an additional paint mark to their abdomen. It’s actually much easier than it looks and only takes a minute to learn.

A few answers to questions I usually get: 1) I don’t anesthetize the bees in any way. These bees are fully awake. 2) These bees are, however, only one day old. The day before I plan on tagging bees, I go out to my field colonies and look for frames of emerging brood. I then keep those frames in an incubator overnight so that I have a bunch of newly emerged bees the next morning. One day olds have very little venom, are reluctant to sting, and can’t fly, so they’re very easy to handle. 3) Yes! They can sting you still but it really doesn’t hurt. Usually, since bees die after they sting, it just sucks because you spent time tagging a perfectly good bee just to have them sting you and die.

After tagging, I introduce the bees into an observation hive and give them 13 days to develop to the age in which bees usually perform the behaviors I’m interested in.

On day 13, I then introduce frozen killed brood (FKB) to the hive, which is used to induce the larval removal behavior I mentioned earlier. These treated larvae mimic diseased brood and elicit similar social immunity behavior in worker bees. The gist of the experiment is to record the behavior of a whole group of bees and see if the ones performing the larval removal behavior are also more likely to perform other social immunity behaviors, such as grooming and removing dead adults from the hive.

So, for the next three days after introducing the FKB to the observation hive, I spend about 10 hours a day recording the behavior and location of every tagged bee in the hive every hour. These are long days and require a lot of focus. That being said, it’s pretty cushy “fieldwork” compared to what other researchers do. I get to spend the whole day indoors, drinking coffee and listening to music and podcasts while recording hour after hour of data.

This is one of the behaviors I’m interested in seeing. Two bees are getting groomed by other bees after the bee on the left specifically performs a dance inviting others to groom her [VC: Tez Stair]

As of writing this article, I’ve spent a little over 100 hours watching honey bees do their thing—only 9,900 more until I become a master if you believe Malcolm Gladwell. Luckily, the next step in my research is to automate my experiment a bit and introduce video recording into my project. And that’s essentially where I’m at now in my field season.

Although not of experimental interest, this is one of my favorite behaviors: the shaking signal. It tells other bees to get to work! [VC: Tez Stair]

Meanwhile, I’m also taking care of my field colonies in 90°+ Davis weather so keep me in your thoughts!

Literature Cited

1. Sih, A., Bell, A. and Johnson, J.C., 2004. Behavioral syndromes: an ecological and evolutionary overview. Trends in ecology & evolution, 19(7), pp.372-378.
2. Gosling, S.D., 2001. From mice to men: what can we learn about personality from animal research?. Psychological bulletin, 127(1), p.45.
3. Bell, A.M., Hankison, S.J. and Laskowski, K.L., 2009. The repeatability of behaviour: a meta-analysis. Animal Behaviour, 77(4), pp.771-783.
4. Oster, G.F. and Wilson, E.O., 1979. Caste and ecology in the social insects. Princeton University Press.
5. Seeley, T.D., 1982. Adaptive significance of the age polyethism schedule in honeybee colonies. Behavioral Ecology and Sociobiology, 11(4), pp.287-293.
6. Kolmes, S.A., 1989. Grooming specialists among worker honey bees, Apis mellifera.. Animal behaviour.
7. Visscher, P.K., 1983. The honey bee way of death: necrophoric behaviour in Apis mellifera colonies. Animal behaviour, 31(4), pp.1070-1076.
8. Walton, A. and Toth, A.L., 2016. Variation in individual worker honey bee behavior shows hallmarks of personality. Behavioral ecology and sociobiology, 70(7), pp.999-1010.

---

Adrian Perez is a 1st year PhD student in the Animal Behavior Graduate Group at UC Davis.  He studies honey bee division of labor.