Criminal bee-haviour
30 July 2008
At first glance, bumblebees and serial killers don't have much in common, but scientists at Queen Mary, University of London are using geographic profiling, a technique used to catch serial killers, to identify different types of foraging in bees.
It is hoped the research will help fine-tune geographic profiling, which is used across the world to predict where a serial killer lives from where the crimes are committed.
In the study, Dr Nigel Raine and Dr Steve Le Comber from Queen Mary, University of London worked with Professor Kim Rossmo, the former detective who invented geographic profiling. The technique has already been used with foraging data to predict where bats roost, but the current study is the first time geographic profiling has been tested in the laboratory.
In a real-life setting with a criminal, detectives may have information on what actually happened, but not on the other 'potential' crimes or choices available to the criminal. In the experimental set-up, the researchers were able to give the bees a fixed number of choices - in this case, of different flowers to visit.
The team observed how the real bees responded to the different scenarios. By comparing the bees' behaviour to different theoretical models of foraging using geographic profiling models, the researchers were able to fine-tune the technique and hopefully improve its accuracy when it is used to hunt criminals.
The results also showed that researchers should be able to predict the location of bee nests by applying geographic profiling to their foraging behaviour. Like criminals, bees seem to create a 'buffer zone' around where they live. Just as criminals tend not to commit crimes within a certain area around their home, bees appeared not to forage in the area directly surrounding their hive, probably to reduce the chance of parasites and predators locating it.
Image: Bumblebee (Bombus terrestris) visiting yellow artificial flower; Tom Ings
References
Raine NE et al. Geographic profiling applied to testing models of bumble-bee foraging. Journal of the Royal Society Interface 2008;doi:10.1098/rsif.2008.0242 [Epub ahead of print]