That's more or less the idea behind an approach to agricultural pest control proposed in a paper just released online at PNAS: if you saturate insect pests with a predator warning signal, they become used to the signal, and more vulnerable to predators [$a]. Aphids are the target pest—they form huge, clonal swarms to literally suck the life out of plants, as described very nicely in this BBC Nature video.
As the video notes, those clonal swarms are vulnerable to all sorts of predators, most famously ladybird beetles. So when attacked, the aphids emit an alarm pheromone to warn the rest of the clone. But it's possible to habituate aphids to the alarm pheromone—if they're surrounded by it long enough, they won't respond to it by running away. The new study's authors proposed genetically engineering crop plants to produce the alarm pheromone, to automatically produce that habituation.
To see if this would work, they raised aphids on a line of Arabidopsis thaliana (the white lab mouse of the plant world) that had been engineered to produce the alarm pheromone. And, indeed, habituated aphids were much less likely to be repelled by the alarm pheromone—and were even in some cases attracted to it. Perhaps the most telling test involved leaving habituated and non-habituated aphids on an experimental plant with ladybird beetles introduced—habituated aphids were less likely to survive 24 hours with the beetles.
This is a pretty clever approach to pest control, but there's an obvious caveat. I don't see any reason why aphids couldn't evolve a way around this attempt to swamp out their own alarm signals—the paper notes that different aphid species have different responses to the particular alarm pheromone tested, so engineering one pheromone into crop plants doesn't leave the aphids without evolutionary options. Unless it's very cleverly designed, any pest-control strategy creates strong natural selection—and the better the strategy is, the stronger the selection is—to evolve resistance. Alarm-pheromone-producing crops might be another tool for pest control, but they won't be the last one we need.
A ladybird beetle makes short work of some aphids. Photo by kenjonbro.
See also this press release from Cornell University, which discusses the paper's results.
Corrected, 17 Aug 2010, 2305h: Fixed the photo of the ladybird with aphids, which was meant to be full-width, and added a jump. Why do I keep forgetting those?
de Vos, M., Cheng, W., Summers, H., Raguso, R., & Jander, G. (2010). Alarm pheromone habituation in Myzus persicae has fitness consequences and causes extensive gene expression changes. Proc. Nat. Acad. Sci. USA DOI: 10.1073/pnas.1001539107