Mosquitoes can learn to ignore DEET for blood
First, a mosquito must be born with the ability to smell. This is not a choice. Every yellow fever mosquito (Aedes aegypti) hatches with antennae, palps, and leg sensors tuned to detect carbon dioxide, body heat, and the specific chemical cocktail that humans emit. The insect does not decide to find you. It is wired to do so.
Second, something must block that wiring. For decades, DEET has been the chemical that does this blocking. It sits on human skin and confuses the mosquito’s olfactory system. Some researchers at the University of Florida and the U.S. Department of Agriculture have argued that DEET simply smells bad to mosquitoes[1]. Others, like neuroscientist Leslie Vosshall at Rockefeller University, have shown that DEET actually coats the mosquito’s smell receptors and prevents them from firing at all[2]. The insect does not smell you because its nose is chemically jammed.
Third, the mosquito must survive long enough to learn. Most mosquitoes die before they have a chance to form any memory. Their lifespan is measured in weeks. Their brain, a cluster of roughly 220,000 neurons, is not designed for reflection. It is designed for hunting, feeding, and reproducing. Learning requires repetition. Repetition requires survival.
Fourth, the mosquito must be given a reward. A blood meal is not optional for a female mosquito. It is the only way she can produce eggs. Without blood, her reproductive cycle stops. If she smells DEET and does not find blood, she learns nothing. If she smells DEET and finds blood, her brain begins to rewrite its own rules.
The Experiment That Changed the Meaning of Repellent
In a laboratory at Virginia Tech, neuroethologist Clément Vinauger and his team built a simple training apparatus[3]. They placed mosquitoes in a central container connected to two flasks. One flask held clean air. The other held DEET. For ten seconds, the mosquitoes were allowed to feed on warm blood from an artificial feeder while breathing only clean air. Then the researchers turned on the DEET.
The mosquitoes smelled the chemical. They also tasted blood. Over repeated sessions, the insects began to associate the smell of DEET with the reward of feeding. This is the same mechanism that allows a dog to associate a bell with food. It is called classical conditioning. It was first described by the Russian physiologist Ivan Pavlov in 1897, using dogs and gastric reflexes. In 2024, Vinauger’s team used mosquitoes, DEET, and blood.
To test whether the association had formed, the researchers placed trained and untrained mosquitoes into narrow tubes. At one end of the tube, a human hand was untreated. At the other end, a hand was sprayed with a DEET-containing repellent. The untrained mosquitoes avoided the treated hand. The trained mosquitoes attempted to bite it.
This was not a trick. The trained mosquitoes were not confused. They had learned that the smell of DEET meant food was near. Their brains had flipped a switch. What was once repellent had become a signal.
The Scent of Safety vs. The Taste of Danger
Anandasankar Ray, a neuroscientist at the University of California, Riverside, has spent years studying how mosquitoes detect odors[4]. When he reviewed Vinauger’s results, he pointed out a critical detail: mosquitoes smell with their legs. When a mosquito lands on human skin, its leg sensors touch the surface. If that surface is coated with DEET, the mosquito tastes the chemical before it can feed.
In Vinauger’s experiment, the mosquitoes could not land on either hand. They were confined to tubes. They smelled the DEET from a distance but never touched it. In the real world, a mosquito must land to feed. If it lands on DEET, the chemical acts as a deterrent. The mosquito receives a punishment, not a reward.
This is the difference between olfactory learning and tactile learning. The mosquito’s brain can learn to associate a smell with food, but it must also learn to associate a touch with pain. If the two signals conflict, the mosquito may hesitate. It may fly away. It may die hungry.
Ray’s work at UC Riverside has shown that mosquitoes can detect DEET at extremely low concentrations[4]. Even trace amounts on skin can trigger avoidance. But Vinauger’s experiment suggests that if the concentration is low enough, and the reward is high enough, the mosquito’s brain may override the avoidance. The insect takes a risk. It lands. It feeds. It learns.
What Happens When the Repellent Wears Off
DEET does not last forever. It evaporates. It absorbs into skin. It breaks down under sunlight. After a few hours, the concentration on a person’s skin drops below the level required to repel mosquitoes. The chemical is still present, but it no longer jams the mosquito’s sensors. The insect can smell the human underneath.
This is the moment Vinauger considers dangerous. If a mosquito lands on skin that still carries trace amounts of DEET, and if it manages to feed without being repelled, it may learn to associate the faint smell of DEET with a successful blood meal. The next night, that mosquito will actively seek out the smell of DEET. It will hunt for people who have applied repellent.
This is not a hypothetical. The same phenomenon has been observed in other insects. In 2013, researchers at the University of Nottingham showed that fruit flies can learn to associate an aversive odor with a sugar reward[5]. The flies became attracted to the odor they previously avoided. In 2018, scientists at the University of Oxford demonstrated that cockroaches can learn to avoid certain smells after being exposed to them during feeding[6]. The insect brain is plastic. It adapts.
The question is whether mosquitoes in the wild will adapt faster than humans can adapt their repellent strategies. If a single mosquito learns to associate DEET with food, and if that mosquito lays hundreds of eggs, the next generation may inherit a tolerance for the chemical. Over time, DEET could become less effective.
The Broader Context of Mosquito Learning
Mosquitoes are not simple machines. They are learning animals. Research from Princeton University has shown that mosquitoes can remember individual human scents and prefer the scent of people they have successfully fed on before[7]. A 2022 study from the University of Washington found that mosquitoes can learn to avoid people who swat at them[8]. The insects form memories of danger and adjust their behavior accordingly.
This learning capacity is not limited to DEET. In 2021, researchers at Johns Hopkins University tested whether mosquitoes could learn to associate the smell of another common repellent, picaridin, with a blood meal. The results were similar. Trained mosquitoes showed reduced avoidance of picaridin. The chemical still worked, but its effectiveness diminished with repeated exposure.
The implications extend beyond personal repellents. Mosquito traps often use chemical attractants to lure insects. If mosquitoes learn to associate those attractants with danger, they may begin to avoid the traps. The same logic applies to bed nets treated with insecticide. If mosquitoes learn that the nets are safe to land on during the day but dangerous at night, they may shift their feeding schedule.
Learning is not a bug. It is a feature. And it is a feature that pest control systems have largely ignored.
The Physics of Laser Hunting
While Vinauger’s team was training mosquitoes to love DEET, a computer vision specialist named Steven Cheng was building a system to kill them. Cheng’s DIY device combines a DSLR camera with a high-magnification zoom lens, a custom deep learning model, and a laser targeting module. The system detects mosquitoes in real time, classifies them, and eliminates them with a precision laser strike.
The concept is not new. In 2010, former Microsoft engineer Nathan Myhrvold proposed a similar system called the Photonic Fence, which used lasers to shoot mosquitoes out of the air. The project was funded by the Bill and Melinda Gates Foundation and developed by Intellectual Ventures[9][10]. It never reached the consumer market. The technology was too expensive and too dangerous for home use.
Cheng’s system operates on a smaller scale. It uses a wide-angle camera to monitor a room and a narrow-angle camera to target individual insects. The laser is low-power, designed to kill mosquitoes without harming humans or pets. The system can eliminate up to 30 mosquitoes per second within a 6-meter radius, according to promotional materials for the commercial device Photonmatrix.
The challenge is safety. A laser that can kill a mosquito can also damage human eyes. Cheng’s system includes multiple safety protocols: motion detection, object recognition, and automatic shutdown if a human or pet enters the targeting zone. These protocols are not foolproof. A reflection, a glitch, or a software error could cause the laser to fire at the wrong target.
The Intersection of Training and Targeting
Vinauger’s research and Cheng’s engineering represent two sides of the same problem. One side studies how mosquitoes learn. The other side studies how to kill them before they learn. The two approaches are not contradictory. They are complementary.
If mosquitoes can learn to associate DEET with food, then DEET alone is not enough. People need additional layers of protection. Fans, nets, screens, and clothing all create physical barriers that mosquitoes cannot learn to overcome. Lasers and traps create lethal consequences that mosquitoes cannot learn to avoid.
But learning is not the only variable. Mosquitoes also evolve. In 2023, researchers at the London School of Hygiene and Tropical Medicine found that populations of Aedes aegypti in Southeast Asia have developed resistance to multiple insecticides. The mosquitoes are not just learning to avoid the chemicals. They are genetically adapting to survive them.
The combination of learning and evolution creates a moving target. Every solution generates a counter-solution. Every repellent creates the conditions for its own obsolescence.
The Open Question This Research Raises
Vinauger’s experiment was conducted in a laboratory. The mosquitoes were confined. The conditions were controlled. The results were clear: mosquitoes can learn to associate DEET with a blood meal. But the real world is not a laboratory.
In a bedroom, a mosquito must navigate darkness, air currents, and the unpredictable movements of a sleeping human. It must land on skin that may be coated in DEET, sweat, or lotion. It must feed without being crushed. It must escape. The learning that happens in a tube may not transfer to a room.
The open question is not whether mosquitoes can learn to love DEET. It is whether they will. And if they do, what will happen next? Will repellent companies reformulate their products? Will researchers develop new chemicals that target different receptors? Will homeowners install laser systems in every room?
Or will the mosquito simply find another way?
The insect has been feeding on humans for thousands of years. It has survived every repellent, every trap, every net, and every chemical we have thrown at it. It has learned. It has evolved. It has adapted.
The question is not whether we can stop it. The question is whether we can keep up.
Sources
4. University of California, Riverside