Searching for blog posts tagged with 'forage'

Listening to the dance floor noise during waggle dance communication improves the mapping of foraging locations



An interview with the lead author, Dr Roger Schürch, Postdoctoral Researcher at the School of Life Sciences, of a new article which shows how the foraging locations of honey bees can be mapped more accurately.

Why was this piece of research necessary?

Honey bees use a unique, highly ritualised behaviour to communicate where they have foraged. During this “waggle dance” on an upright comb within the hive, they run in a straight line, circle back to where they started, and run the straight line again – they do this over and over again. Because they waggle their bellies back and forth in the straight run, we call this behaviour the waggle dance. Over the past century, researchers have been able to decode the waggle dance, so that we can now observe a dance and know where a bee has been to collect her food: the bees communicate a heading and a distance in the angle and the duration of the straight waggle run.

Understanding where bees have foraged is important if we want to be able to help them. There are two problems with the way this has been previously done. First, most of the studies studying where bees have foraged using the waggle dance have used a calibration curve published by Karl von Frisch in 1946[1]. We noticed that this calibration curve gave wrong results for the bees studied at the Laboratory of Apiculture and Social Insects (LASI). Secondly, the calibration curves used up to know were focusing exclusively on the averages of communicated angles and distances. In reality there is a considerable amount of noise in the location communicated by the bees. In other words, the bees are pretty imprecise in their communication. This means that when we observe a single dance, we cannot be sure where the bee has been. Previous studies have ignored this uncertainty.

[1] See von Frisch, K. 1946. Die Tänze der Bienen. Österr. zool. Z. 1:1-148

Honey bee feeding on a flower


What were the main results, and did these results turn out as expected?

First, we were able to show that our bees have a different calibration for distance and duration from those used by von Frisch, but that our calibration curve closely matches one published by Adrian Wenner in 1962[1]. This confirmed our initial suspicions that Karl von Frisch’s curve was not valid for our bees. Secondly, we have been able to quantify the precision with which bees are able to communicate visited locations to their fellow nestmates in both the directional and distance component. In the case of the angular precision, we have been the first to publish a concentration parameter capturing the angular variability among dances. For both vector components the imprecision that we found to be inherent in the waggle dance communication actually matches well the precision of recruitment found in previous studies[2].

[1] See Wenner, A. M. 1962. Sound production during the waggle dance of the honey bee. Anim. Behav. 10:79-95

[2] See Towne, W. F. & Gould, J. L. 1988. The spatial precision of the honey bees' dance communication. J. Insect Behav. 1:129-155.

Honey bee comb by Alex Wild

What has it added to our knowledge to the honey bee dance language?

The noise we found will enable us to make more honest assessments of where the bees have gone: instead of saying 25% of bees have visited that field, we say between 23-24% of bees have visited that field with 95% certainty. The latter statement is less precise, but more honest. But the real benefit lays in the ability to then test hypothesis like whether a colony has visited a particular field or crop more than you would expect by chance.

Does it change the validity of previous work on the waggle dance?

This is very hard to judge. Does the calibration curve of Karl von Frisch differ because he used a different bee, he was in a different location, or because he did not take random samples (see anecdote below)? At this point we simply don’t know, but based on our results we must warn against using Karl von Frisch’s calibration without validating it with the focal bee strain, and in the focal location.

Any fun anecdotes whilst collecting data?

For a very long time, we were brooding over one troubling feature of our data. Karl von Frisch’s data suggested a non-linear response in the distance-duration calibration, whereas ours could just not be accommodated with such a non-linear form. It was not until Lars Chittka came to give a seminar talk here in Sussex that we realised that we should be cautious with interpreting von Frisch’s data. Over lunch, he related to us a story Martin Lindauer, one of von Firsch’s students, had told him. When doing feeder experiments, Karl von Frisch often ordered “bad dancers” to be “executed”[1]. This creates a non-random sample, which made it probably easier for von Frisch to discover the general principle of the dance language. However, this non-random sampling is a bit like asking people who favour the colour red what their favourite colour is. Not surprisingly, the answer will be red.  Such a bias in the data makes it impossible to predict feeding locations from observed dances of a random bee, and presumably led to the non-linear curve von Frisch observed.

[1] This annecdote is also told in Chittka, L. & Dornhaus, A. 1999. Comparisons in physiology and evolution, and why bees can do the things they do. Ciencia al Dia International 2:1-17.



To read more about this research, see the full article in the Journal of Comparative Physiology A

More on the honey bee waggle dance, read Other Nations' blog post

Visit The Laboratory of Apiculture and Social Insects website 

photo credit: John Kimbler (top) and Alex Wild (