The modern supermarket offers a rainbow cornucopia of fruits and vegetables. Peppers, avocadoes, strawberries, cucumbers—they’re all made possible by bees. But “there just aren’t enough pollinators in the natural world” to take care of our global crop load, says Sarah Arnold, an ecologist at the University of Greenwich. So farmers release commercially reared bees by the thousands onto their fields, where the insects buzz along diligently and pollinate billions of dollars’ worth of crops every year. As bees dip into flowers to find food, their fuzzy little bodies pick up powdery pollen that gets spread when they visit the next flower, and the next, and the next. 

But commercial bees sometimes stray from farm fields to peruse nearby wildflowers. Now, scientists have found that—like for many humans—a jolt of caffeine helps bees stay on task and get the job done more efficiently. Arnold and her colleagues showed that feeding bumblebees caffeine while exposing them to a target floral scent encourages them to seek out that smell when they leave the nest. The caffeinated bees visit the target-scented flowers more quickly and often than those without that extra boost. The findings could be applied to industrial agriculture to train bees to stay more on track, the team reported Wednesday in Current Biology.

Pollinators had already been known to learn which flowers to visit by being exposed to scents inside the nest, says Jessamyn Manson, an ecologist at the University of Virginia who was not involved with the new research. And previous studies had shown that bees like to visit artificial flowers that produce caffeine, Arnold notes—but how the caffeine itself might impact bees’ actions was unclear. Other research shows that tethered honeybees exposed to a target scent while eating caffeine stick out their tongues in response for longer periods of time, but those bees were unable to freely choose which flowers to visit.

To investigate more deeply, Arnold and her team set up three groups of bumblebees. One got caffeinated sugar water and a blast of strawberry-flower odor. Another received plain sugar water and the odor, and yet another got just the plain sugar water. None of the bees had previously encountered any type of flower or floral scent. Each group was released from its hive and into a laboratory arena dotted with robotic flowers, some of which puffed out the same strawberry smell and others that released a completely different “distractor” floral scent. All of the fake flowers contained reservoirs of sugar water (without caffeine) for the bees to lap up upon selection.

The caffeinated bees showed a clear preference for the faux strawberry flowers, with 70.4 of them visiting the target blossoms right away. Just 60 percent of the noncaffeinated but odor-primed subjects made a beeline for the plastic strawberries first, and the bees that received neither caffeine nor the priming scent visited the strawberry flowers a little under half of the time, an expected result because they had never “learned” which plants to try in the first place.

Bees exposed to both caffeine and odor formed a “super strong association” between the two, Arnold says, suggesting that a bee might think: “When I had that odor in the past, I got this really nice [caffeinated] sugar and I remember that really clearly.” With each consecutive flower visit, these bees’ pace also increased faster than that of the noncaffeinated bees—indicating that caffeine might additionally enhance their motor skills.

Though the positive association was strong, it eventually wore off: After visiting dozens of flowers the caffeinated bees started investigating the distractor flowers too, and Arnold points to the laboratory setup as one cause. “Finding plastic flowers that are just a few inches apart from each other … it’s quite an easy task for the bees to solve,” she says. “The bees would sooner or later try out the distractor flowers and realize that they’re equally as rewarding.” But in a field of strawberry plants, real-life “distractor” flowers would be much farther away, and it might take the bees longer to stray from their task. In an agricultural setting, caffeine could be supplied alongside priming scents for specific plants in commercial hives, Arnold says. Farmers could place the caffeinated hives in their fields for the bees to pollinate more efficiently.

Manson says this strategy might be more applicable to farms in the United Kingdom than to those in the United States; U.K. farms tend to be smaller, and it is easier for the pollinators to wander off if untrained. U.S. crops pollinated by bees are often planted in huge fields that are harder to stray from, or grown in greenhouses from which bees cannot escape, she adds.

Whatever industrial application the new findings might lead to, Manson says these experiments’ use of caffeine as a priming stimulant is particularly revelatory. Humans actively seek out caffeine, “and I expect pollinators do, too,” she says. “It’s delicious and awesome.” But because this study had caffeine given in the nest rather than being doled out as a reward at the flower, she says, the experiment is a “strong demonstration” of how caffeine can help teach bees which plants to pollinate.