A terrifying robot can thwart invasive mosquito fish

Invasive mosquito fish are often fearless. 

Free from the predators of their native range, these mosquito fish run rampant, throwing naive ecosystems from Europe to Australia out of whack. To keep the problematic fish in check, scientists are trying to strike fear back into the hearts of these swimmers with a high-tech tool: robots.

In a laboratory experiment, a robotic fish designed to mimic one of mosquito fish’s natural predators increased fear and stress responses in mosquito fish, impairing their survival and reproduction, researchers report December 16 in iScience. 

While robofish won’t be deployed in the wild anytime soon, the research highlights that there are “more creative ways of preventing unwanted behavior from a species” than simply killing them, says Michael Culshaw-Maurer, an ecologist at the University of Arizona in Tucson who wasn’t involved in the study. “It’s just wonderful seeing work in this area.”

Native to parts of the western and southeastern United States, mosquito fish (Gambusia spp.) were let loose in freshwater ecosystems around the globe last century in a foolhardy effort to control malaria. But instead of eating malaria-transmitting mosquito larvae, the mosquito fish mostly gobble up the eggs and gnaw at the tails of native fish and amphibians, making them one of the world’s most destructive invasive species, according to the International Union for Conservation of Nature.

Efforts to combat mosquito fish, and many other introduced, invasive species, usually rely on mass killing with traps, poison or other blunt methods, says Giovanni Polverino, a behavioral ecologist at the University of Western Australia in Perth. “For most of the invasive species considered problematic, this doesn’t work,” he says, and can often harm native species too.

The problem isn’t necessarily the presence of mosquito fish in these ecosystems, Polverino says, but their wanton behavior enabled by a lack of predators. While predation prevents prey numbers from ballooning, merely the fear of predators can influence prey behavior in ways that ripple throughout an ecosystem (SN: 5/5/19). Polverino and his colleagues wanted to see if a robotic fish crafted to mimic one of mosquito fish’s natural nemeses, the largemouth bass (Micropterus salmoides), could be just as scary and take some of the bite out of mosquito fish’s negative impact.

In the lab, researchers set up 12 tanks that each housed six mosquito fish (G. holbrooki) with six native Australian tadpoles (Litoria moorei) that are commonly harassed by mosquito fish. After a week of acclimatization, the team transferred each group to an experimental tank for one hour twice a week for five weeks. There, half of the groups faced a robotic predator designed to recognize and lunge at mosquito fish when they got too close to the tadpoles.

Fear of the robot altered the behavior, shape and fertility of the mosquito fish, both during exposure and weeks later. Mosquito fish facing the robot tended to cluster together and not explore the tank, while the tadpoles, free of harassment, ventured farther out. Even in the safety of their home aquariums, fish exposed to the robots were less active and more anxious — exhibited by seconds-longer freeze responses — than mosquito fish that weren’t exposed. 

The cumulative stress taxed the fishes’ bodies too. Exposed fish lost energy reserves, becoming slightly smaller than nonexposed fish. Exposed males became more streamlined, potentially to quicken escape behaviors, the researchers say. And the sperm count of scared fish decreased by about half, on average.

“Instead of investing in reproduction, they’re investing in reshaping their body to escape better after only six weeks,” Polverino says. “Overall, they became less healthy and less fertile.”

The long-term impact that such robotic predators would have on wild mosquito fish and their neighbors remains unclear. That’s beside the point for Polverino, who says the main contribution of this study is showing that fear has significant consequences that may reduce the survival and reproduction of invasive species. 

“Our plan is not to release hundreds of thousands of these robots in the wild and pretend they will solve the problem,” Polverino says. But there may be more than one way to scare a mosquito fish. Giving the fish a whiff of their predator, for example, might induce similar changes.

“These are not invincible animals,” he says. “They have weaknesses that we can take advantage of that don’t involve killing animals one by one.”