In addition, the researchers fitted five bees with transponders and tracked them with radar as they developed traplines. This revealed that flight paths between trapline segments were relatively straight and that between their first and last bouts, bees cut their total travel distance by 80% (from 1,953 to 458 m).
In contrast to computers, bees did not find the absolute shortest route of 312 m even in this simple experimental arrangement. But they came very close, especially considering that they explored only a small fraction of the possible routes, and established traplines relatively rapidly. This tradeoff between perfection and speed highlights the differences between mathematical and biological solutions to the travelling salesman problem.
How do bees develop such efficient routes so fast? The researchers assessed three possibilities — that bees optimize foraging routes by visiting flowers in the order of discovery, by shuffling them randomly, or by visiting those that are closest together — but found that the first two failed to fit their observations while the third did not fully explain them.
Rather, the researchers propose that bees optimize foraging routes through trial and error, combining exploration with learning from previous bouts to progressively adjust their routes as they find shorter paths. Based on the bees' movements during trapline establishment, the researchers developed a model linking experience to the likelihood of visiting particular flowers. Bees are well-known to be able to compute and memorize distances between locations, and the model assumes that they remember the length of the shortest route so far, compare it to the length of the current route, and then choose the shorter of the two.
Over time, choosing the more efficient route favors shorter segments over longer ones. The model is a good fit with the researchers' observations, predicting, for example, that bees will develop and stick to optimal routes in 20–25 bouts.
Besides shedding light on how bees develop traplines, this work suggests that small-brained animals can use simple methods to solve complex routing problems without the need for cognitive maps of spatial relationships, as has been suggested. It remains to be seen whether big-brained animals can also develop traplines with such elementary tools. But if so, that would free up their brain power for other tasks.
Authors: Lihoreau M, Raine NE, Reynolds AM, Stelzer RJ, Lim KS, et al. (2012) Radar Tracking and Motion-Sensitive Cameras on Flowers Reveal the Development of Pollinator Multi-Destination Routes over Large Spatial Scales.
Pages: 1 · 2
More Articles
- Ferida's Wolff's Backyard: The Snow Turned Into Snowy Rain As It Fell From The Trees. Still Lovely ...
- National Institutes of Health: Common Misconceptions About Vitamins and Minerals
- Oppenheimer: July 28 UC Berkeley Panel Discussion Focuses On The Man Behind The Movie
- Ferida Wolff's Backyard: Awesome Goldfinches, Part of the Incredible Possibilities That Nature Might Offer
- Ferida Wolff's Backyard: Nature’s Jigsaw Puzzles
- Julia Sneden Wrote: Love Your Library
- Rebecca Louise Law: Awakening on View at Honolulu Museum of Art
- Scientific Energy Breakeven: Advancements in National Defense and the Future of Clean Power
- Ferida Wolff's Backyard: Dandelion Landing
- Center for Strategic and International Studies: “The Future Outlook with Dr. Anthony Fauci”