Bumble bee attention span
The EU-funded 'Visual search and attention in bumblebees' (VSIB2009) project investigated the mechanisms of visual search and attention processes in the bumble bee. This study has provided information on how the bee makes decisions and gives an insight into the minimal neuronal hardware required to make similar decisions. Researchers looked at whether bees can see at a glance. Humans can make a decision when presented with scenes for only 12 ms. Bees could distinguish between elemental targets such as colour in as little as 25 ms but were not able to handle a more complex discrimination task involving whole image analysis even when given 100 ms. This is significant as the bees failed to detect the cryptic yellow spider, a deadly predator. Bees could switch between multiple targets, such as two colours. Moreover, when posed with rewards such as high sucrose as well as saliency (prominence of colour, for example), both factors influenced search. The ultimate test for the bees was to compare multiple search tasks while simultaneously avoiding predators using different modalities – visual and olfactory. The scientists found that bees performed both searches when one was an olfactory task and the other was visual. However, they failed to perform searches when both tasks were visual. Project research showed that bees have evolved several adaptations to perform visual tasks with flexibility and efficiency. As such, the bumble bee could be an ideal model system to study mechanisms of attention compared with vertebrates. Using the relatively simple bee nervous system could pave the way for bio-inspired algorithms for robots or computers. The work also has great significance for agriculture as those plants with strong flower aroma could also help bees avoid predators visually. Not surprisingly, the work has appeared in Journal of Experimental Biology and more publications are expected. Results have also featured at national and international conferences.
Keywords
Bumble bee, attention, visual search, complex discrimination, multiple target, predator, algorithm, robot
