Remote controlled cockroach


Our pals at Backyard Brains (makers of the terrific SpikerBox kit, which allows you to study the electrical impulses of insect neurons) are developing circuitry to control which direction a cockroach walks.

By modifying the HEXBug toy "Inchworm" circuitry to deliver pulses, we stimulated the antenna nerves of the discoid cockroach to "trick" the cockroach into turning upon command. Stay tuned! as we make the preparation easier, more reliable, and lighter!
Working RoboRoach Prototype (Via Make)


  1. Geez. Never thought I’d feel sorry for a cockroach.

    Can someone tell me what practical, world-bettering applications this might have? The linked page doesn’t have much more than, “We can drive a cockroach.”

    1. Haven’t you seen The Fifth Element? Remote-control cockroaches, with little cameras! You know…bugs.

    2. +1 This reminds me of the human “white room” experiments in THX-1108.

      This just seems ham-fisted and possibly inhumane. Sure, you’re turning the cockroach but may also by causing pain/damage to this poor critter.

      And don’t tell me “cockroaches don’t feel pain.” Even fishes feel pain, so I think there’s a good probability cockroaches do as well.

      1. I’m not shocking the cockroach. I am stimulating locomotion neural circuits in the antenna with tiny electrical pulses. The cockroaches only have the backpacks on for a couple minutes. The cockroaches are not killed. They are allowed to retire and make cockroach babies and live out the remainder of the cockroach lives eating organic lettuce and carrots and playing in small wooden jungle gyms. I’m serious.

        What’s the point of all this? It’s to teach high school students about principles of microstimulation of neurons, the proper stimulation frequencies, the proper circuit design, basic neurophysiology, etc. This type of research is typically only done at large research universities, and we are bringing such neurotechnology into high schools to accelerate public understanding of neuroscience. Similar neural implants are used to treat Parkinson’s disease, depression, and deafness. The technology is very crude. With more students working on the problems of nervous system manipulation and treatment, progress will undoubtedly be made. 1 in 4 people will be affected by a nervous system affliction in their lifetimes, and the current treatments available if you have a problem with neuronal function are very poor. Ask anyone who has a relative with mental illness, dementia, or stroke. Are you satisfied with the state of neuroscience treatment today?

        1. See? These guys have a high-minded purpose, and aren’t deliberately torturing arthropods because it’s “cool.”

          Don’t hate them. Hate me.

          Seriously, BackyardBrains, that’s some great and valuable work. I hope my endorsement does not taint your image.

    3. There isn’t enough processing power to form any sort of being in a cockroach. Inflicting needless pain on your quads while on the stepmaster would be a more legitimate ethical concern.

  2. When I was a kid we had to make do with pulling the wings off flies, or burning ants with a magnifying glass, or squeezing hamsters… Amazing how far technology has come.

  3. I don’t care how much you hate bugs, this is just disgusting and wrong. It’s just as wrong to do animal testing on ugly animals as it is on cute animals. This is not a wonderful thing.

    1. Um, where did they say that they “hate bugs”, or that they think they are “ugly”? They seem fascinated, not disgusted…project much?

  4. One could easily duplicate these results with two electric dog collars and a crawling baby.

    1. I don’t believe I am. I confess to having a Dr Frankenstein streak in me a mile wide. Plus, I have long been at war with Order Blattaria

      “The Lord God made them all,” but I still choose mine own enemies.

      You may now dismiss me as a valueless jerk, if my love of chocolate covered cherries wasn’t reason enough.

  5. “Torturing bugs?” “Feeling sorry for cockroaches?” Really?

    If memory serves, they don’t even have brains. They’ve got pathetic little bundles of crossed neurons that aren’t even capable of PERCEIVING pain, discomfort or annoyance, or anything except the basic instincts that come from being a cockroach. They’re essentially little organic robots, why not re-program them?

    1. For the record, insects do have a “brain.” It’s referred to as the central complex (CC), or sometimes the central body complex depending on the species. The CC has been a hot area of research in invertebrate neurobiology for the past ten years or so. It’s unfortunate that it’s been studied so little as, unlike mammalian brains, we have a real chance of understanding the complete circuitry at a very high level of detail. The CC is a bilateral structure, meaning that, like our brain (and unlike the other insect ganglia), it receives input from both sides of the body and contains many connections that cross over the centerline. The CC is thought to be involved in multi-sensory integration, sensorimotor integration, and decision-making, amongst other things. The CC shares a columnar structure that could be compared to the cortical columns of the primate visual cortex that organize visual information spatially in “maps.” Indeed, maps that organize polarized light angles “topographically” across the columns of the CC have been found in the locust. Surrounding the central complex are a host of other ganglia and neuropils that process sensory information (optic and antennal lobes), that are involved in learning and memory (the mushroom bodies), and that relay proprioceptive information about the state of the animals body.

      There may be fewer ethical qualms about working with insects for many and the U.S. government doesn’t consider them animals for the purpose of testing. Despite this, all good invertebrate neurobiology labs still treat their bugs as humanely as possible. This is done for several reasons. First, ripping the legs off of an unanesthetized cockroach (carbon dioxide or a brief chill in the freezer usually) tends to stress it out (surprise!), causing the animal be less responsive and screwing up your experiment. Second, labs are places of training for undergraduate and graduate students who may down the road end up working with animals for which there are strict protocols and greater ethical considerations. You don’t want to encourage young scientists to torture their research subjects, no matter how lowly. I hope that the Backyard Brains folks are considering this important aspect of scientific ethics in the material they’re putting together for kids.

      Finally, kudos to Backyard Brains for developing a clever and fun hack to teach schoolchildren about neuroscience and electronics. It should be noted that, as previously published work has shown, it’s very likely that the effect of the stimulation dissipates over time. The animals either seem to start ignoring the stimulation or their nervous system acclimates to it. To my knowledge, no one has developed a long lasting “neuroprosthesis” for an insect yet. That may be why DARPA had the crazy idea of combining electronics with the nervous system during the larval stage of insect, but that’s another story…

  6. Eww! Cool! Eww! Cool! Eww! Cool! Eww! Cool! Eww! Cool! Eww! Cool! Eww! Cool! Eww! Cool! Eww! Cool! Eww! Cool!

  7. perhaps, when the oil runs out, we can wire up all the cockroaches in the world and live off their psychic energy. they just keep on going and going …

  8. Not cool, BackyardBrains.

    You are sending direct electrical shocks to their nerves. Arguing that they’re “impulses,” not “shocks,” is a distinction without a difference. Unless you’re in a position to ask the cockroach what it feels like (and get an intelligible response), then you have no way of knowing whether it’s just moderately ticklish or excruciatingly painful.

    Jeremy Bentham had it right: It’s reasonable to assume that animals feel pain, and cruel to inflict needless pain. “The question is not, Can they reason? nor, Can they talk? but, Can they suffer?” When in doubt, err on the side of caution.

    I appreciate and admire your desire to teach your students. However, I don’t believe that showing arguable cruelty towards nonhuman animals is the appropriate way to do it. One can at least have a reasonable debate about causing possible suffering to animals in a laboratory environment “for the greater good,” but I can’t swallow that for merely demonstration purposes in a high school environment.

    As for Gainclone, until last century, the prevailing scientific view was that all animals were just mindless robots, since they “don’t have a soul,” and any expressions of “pain” were just neurophysical reflexes. Can you say affirmatively that a cockroach doesn’t have some capacity to feel pain? If not, is it OK to possibly cause pain, without knowing how they experience it?

    1. All notable points. Behaviorally, how can you know if an animal is in pain? especially a cockroach?

      1) Cockroaches poop and throw up (noxious chemical release) when they are attacked, threatened, or in pain. Next time you see a cockroach on the street try to catch it and pick it up. This behavior doesn’t happen with the RoboRoach stimulation.

      2) The stimulation no longer causes an effect after 6 or 7 stimulation trials (you can’t see it on the video). The neurons in the antennae adapt, which is what neurons do. After about 6 or 7 sequential stimulations the cockroach just ignores it and keeps walking in a straight line.

    2. “Can you say affirmatively that a cockroach doesn’t have some capacity to feel pain? If not, is it OK to possibly cause pain, without knowing how they experience it?”

      The ability to prove a negative (x doesn’t y) is very tricky. “Pain” is an unpleasant sensation that is produced by nociceptors in reaction to potentially damaging stimuli. Cockroaches do not have nociceptors. So they do not feel pain. They don’t have chlorophyl so they don’t photosynthesize.

      That’s an easy negative to prove.

      We don’t know what would feel good or feel bad to a roach, how good or bad it would feel (orgasm versus eating a tic tac) and ultimately how it would affect them on a long time scale (pain is important to humans because we remember what hurt and it leaves a permanent impression on us).

  9. That’s awesome, the should do it with bigger animals next.

    I wanna control a gorilla!

  10. Perhaps Anon would like to liberate the roaches from these experiments and bring them to live with him in his apartment.

    They’re BUGS. Their brains are so simple that before you ask if they can experience pain, you should be asking if there’s anyone in there conscious enough to experience it. Can you prove there is? Because the neurological evidence is against it.

    And for what it’s worth, there is a big difference between an “impulse” and a “shock.” For a demonstration, put your finger across the speaker wire of your stereo. That’s an impulse. Now stick it in a light socket. That’s a shock. Impulses are normal electrical activity for neurons, so there’s no justification for using a loaded word like “shock” to make it sound like someone just hooked Kitty up to the electrical mains.

    Unless, of course, that biased emotional spin is what you were aiming for.

    1. Anon has almost a half philosophical but not scientific question.

      Can anyone prove that atoms are not sentient and we’re torturing them by performing horrible chemistry experiments on them?

      Science is designed to explain the phenomenon we DO see, not the ones we don’t. If someone finds a mechanism for a pain analog in a roach, then that’d be good science. Something to work with.

      1. You, nor anyone else, knows what causes pain, nor what an insect or anything else experiences ‘inside’.

        How dead, insensate matter gives rise to subjective experience is not only a mystery at the present time, but is likely to remain a subject in philosophy forever beyond the ken of science. With our present level of knowledge, Occam’s Razor suggests that there is no point at which interior, subjective experience stops. If there were, you would need to posit a framework in which some things feel, and others do not. Please feel free to do so, because no one has been able to yet.

        Obviously, you cannot feel something that has no informational connection to you. If a cell in your finger dies, and no one is around to hear it as it were, you do not feel it. But that doesn’t mean the cell doesn’t feel it. If you examine your perceptions and thoughts closely, you will find that your stubborn sense that the rock you just kicked is not alive and didn’t feel anything is based entirely on your not feeling pain in response. That you feel pain when you injure a person, or animal you can emotionally map yourself to, is your mirror neurons creating real pain in you.

        In the absence of mirror neurons, one might imagine people would have a solipsistic philosophy that only we as individuals feel pain, and no other humans. It is telling that evolution has woven the truth into our brains.

        1. Oh no, all matter might be sentient, atoms could be alive, and Other people really could be Hell (Oh weep for all those billions of molecules forced to bend to the evil will of Mankind)

          This wretched thing you call Life is just a never-ending subatomic holocaust for those poor trapped electrons and protons in your body, so I suggest you immediately discorporate explosively, and set your oppressed matter minions free.

          1. Nice strawman, Anon.

            Yes, in fact I believe the entire universe is alive, and possesses subjective experience. No, that doesn’t mean I imagine a tiny person inside of everything.

            I notice you have not advanced a theory that grants people and cute fluffy animals sentience, but not atoms. Extraordinary claims require extraordinary evidence, anon. BTW, as many know, sentience =/= sapience.

            This puzzle is not going to remain idle philosophy for long. In our lifetimes we will be able to make machines that appear conscious in every respect. There will be considerable debate as to whether these entities are ‘real’, and whether they can suffer. I hope people like you are not in charge of their development, because they may decide to adopt your ignorance.

  11. I once experienced electrical muscle stimulation during a physical education class, and while it was weird to lose control of my hand and watch it clench, it wasn’t painful.

    In the cockroach experiments, it sounds like you’re really just messing with their perception. It reminds me of the cat tape experiments.

  12. “Cockroaches poop and throw up (noxious chemical release) when they are attacked, threatened, or in pain.”

    “Cockroaches do not have nociceptors. So they do not feel pain.”

    BackyardBrain + Blaine = rationalization lulz.


  13. I was “anon.” I didn’t realize that I wasn’t signed in. Sorry about that.

    I’m not an entomologist, a neuroscientist, or an insect. I’m not qualified to answer whether cockroaches can feel pain. However, many people who are far smarter than I am believe that there’s at least an open question on this issue, at least for some insects. One link here:

    (In case the link breaks, an article that discusses Hwang, R. Y., et al. (2007). Nociceptive Neurons Protect Drosophila Larvae from Parasitoid Wasps. Curr. Biol. doi: 10.1016/j.cub.2007.11.029)

    Simply saying that insects doesn’t have the same pain receptors that we have doesn’t prove anything.

    I’m not arguing that insects can feel pain. I do know that they are animals with a nervous system, and it’s possible that they feel pain (or whatever unpleasant sensation might correspond to pain for a cockroach). It’s hardly unreasonable — for essentially any animal to survive, it has to be able to respond favorably to positive stimuli and negatively to negative stimuli.

    As a matter of ethics, I believe that one should avoid situations that cause unnecessary suffering to other animals. I also believe that when in doubt, err on the side of caution.

    I’ll give credit to Backyard Brains for acknowledging the issue. For the others, do you believe that it’s OK to pull wings off of butterflies just for fun? If they can’t feel pain and they’re just mindless automatons, then there’s no harm, no foul.

    1. “Simply saying that insects doesn’t have the same pain receptors that we have doesn’t prove anything.”

      Actually it is a very important facet to the point you’re trying to make. The article you’ve linked talks about how a fruitfly has multidendritic neurons which may behave similarly to the pain receptors in humans. An unstated major premise of your argument is ‘It is wrong to inflict pain on cockroaches because they interpret pain as a vertebrate would’.

      I mean, even in the article you provided. That even fruitflies, which have been uniquely identified as having ‘pain sensors’:

      “…But even if the brain is involved, it is unlikely that fruit flies feel pain as we do, because human pain has emotional aspects which are processed in the association areas of the cerebral cortex. ”

      That being said, I’ve reached the end of being interested in this discussion because…

      ” For the others, do you believe that it’s OK to pull wings off of butterflies just for fun? If they can’t feel pain and they’re just mindless automatons, then there’s no harm, no foul.”

      Reductio ad absurdum rears it’s ugly head. Comparing a merit based scientific experiment to that is asinine.

      Particularly when you just posted a reference article where scientists inflict pain on poor helpless
      fruitflies. You didn’t seem to have any problems with them torturing those flies. Far from it, you used it to try and bolster your own argument. I guess you’re okay with pulling the wings off butterflies so long as it benefits you personally.

      Anyone can do reductio ad absurdum, it’s why it’s not a valid argument.

  14. I believe we’re at the top of the fucking food chain. Have at it before a meteor breaks up our self righteous little dirt ball. God, people are starving to death and dying of horrible fucking diseases. Screw a few million cockroaches if improves medical science for the rest of us.

  15. If you complainers aren’t all card-carrying vegans you need to shut your self-righteous murder-holes.

    And as to the uses – how about search and rescue? Not only would the roaches not need years of training like dogs, but they could enter the tiniest of spaces, crawl straight up smooth surfaces, etc.

    nehpetsE: Ye, exactly. That’s very insightful. We are, after all, only machines. And if you used your system, with a camera, to steer that baby away from a cliff you’d be a hero.

  16. crippling a butterfly out of cruelty is hardly comparable to giving a cockroach some direction for a while. stop being so dramatic and/or trolling. Kudos to your googling skills tho. and with this high ground you’re taking, are you saying you’ve never killed an insect, or let anyone in your presence do the same, ever? because they’re not even doing that. that mosquito on your cheek you reflexively smacked was just trying to feed it’s babies. besides, roaches will probably outlast humanity anyway, we should learn to control them before they learn to control us

  17. It’s hard to resist the urge to weigh in on the fascinating neuropsychology of misguided empathy. I do feel it… but I also understand it. I even got a little choked up when Wilson was lost in the movie Cast Away.

    Just cover the gadget with a little tiny fake saddle, with a cute little stormtrooper rider, and you’ll be good to go.

    Kudos on an astounding interdisciplinary teaching project that binds so many concepts into concrete demonstration.

  18. The main moral issue with this sort of bug remote control is to make sure that the bugs get recovered and the control circuitry properly recycled.

    Yeah, I don’t care much about the bugs.

  19. BackyardBrains, thanks for responding. It’s great to hear from the horse’s mouth on these things.

    It sounds to me like you’ve got good reasons for the experiments, and much as I dislike the idea of cockroach babies (I think only bedbugs & earwigs squick me out more), I’m glad the roaches live.

    Oddly, the thing that made me feel sorry for the roaches wasn’t the possible pain (which I tend to believe isn’t there) but the loss of autonomy. I think of those wasps that take roaches over, and it freaks me right out.

    I appreciate the info!

  20. It seems like a good time to share this link:

    While optical control of neural activity is the current cutting edge of neurophys/cybernetics research (think projecting holograms into brains to control behavior, or implanting fiber optic cables), what the backyard brains folks have done is less sophisticated, but not by much. It is remarkable that they have simplified these techniques for high school students, and they should be commended for creating better neuroscience education tools.

  21. I do hope that all of the querulous sanctimony is coming from avowed vegans. If you eat anything besides seafood, you’re contributing to the suffering and slaughter of animals that can very obviously feel pain, which have brains thousands of times larger than these insects.

  22. I’ve seen enough B movies to know for a sciencefact that messing around with living things can only end badly.

    Coming next Friday night on the SyFy channel: Roboroach Vs Cyberpede

  23. If it doesn’t “hurt” the roach, could it be trained to get a reward by submitting to the harmless stimulation. If they can I would tend to think that there is some pain involved.

    Can roaches be trained? Can they follow a maze that isn’t marked by their scent but rather other features of the environment?

    I know that roaches in my house are trained to run, but not to receive a reward. If you start training roaches mark them well so they don’t get killed as a runner.

    If it is just stimulation that doesn’t hurt, could all the great science students just experiment on themselves?

    Project proposal, ‘Made myself play the piano without lessens!’ No need to go to university if you pull that off, just find the best venture capital “reward.”

  24. If this has helped us learn about how to help people or animals with neurological problems there may have been some justification for doing it once.

    Repeating the experiment just for the sake of giving someone the experience of doing an experiment on an animal is going too far.

    Of course we humans are very proud our intelligence and some of use will use that most prized trait to justify doing whatever we want to species which have less intelligence as we measure it.

    1. Science is all about repetition and replication! Experiments and protocols build upon one another as we gain more knowledge. Promising results need to be confirmed by others to validate the science. In addition, one can’t learn neuroscience by reading about it in a book or even by reading journal articles. Students often replicate previous experiments in order to learn and refine techniques that are broadly applicable to research. No one is implying here, and no science is done just for the simple-minded sake of giving one “experience of doing an experiment on an animal” as you imply.

      1. When I studied about operant conditioning in high school psychology class I learned that rats deprived of water would eventually figure out that they could press a bar for water. I was able to learn that adequately without meddling with a living animal.

        Students in college level psychology classes relearned what was already known by repeating the experiment on live rats. It was more “hands on” than reading about it in a book but nothing new was learned by the exercise. Some rats were just pointlessly deprived of water and ultimately killed to give some college kids experience in performing useless redundant experiments.

        Nothing new was learned by this experiment to benefit humans or other animals. It was a needless repetition of an experiment with already known results carried out in a culture which regards animals as expendable.

  25. Wow, I blows my mind there are people who give a shit about the suffering of roaches.

    When I kill cockroaches I hope they experience fear and suffering and the survivors huddle in their dens trading stories about the terrible menace outside the walls who laughs as he deals misery and doom to their kind.

    1. I think most ethical objections are less interested in the specific suffering of roaches, and more to do with the speciesism of the experimenters (, regardless of their experimental subject. I do neurophysiological research on mice and this is a topic I struggle with, as do many others who are motivated to study biology. We, not surprisingly, often invoke pragmatic ethics to justify what we do, but it is impossible not to at least be aware of what philosophical lines we draw by our exploitation of other life forms.

    2. I blows my mind there are people who give a shit about the suffering of roaches.

      We call them ‘not sociopaths’.

  26. I killed some bacteria this morning. I imagine it’s death was slow and painful. I took amusement from it.

    -Destructo, destroyer of worlds!

  27. My qualm with the experiment is not the fact that a roach may or may not have been made uncomfortable. My qualm would be they are developing mechanical systems to control living creatures. We all know how this movie ends.

  28. Well, it’s going to be THEIR planet next so we might as well get our kicks in while we can!

  29. I find the idea of controlling any animal through electrical impulses and overriding its natural instincts an uncomfortable thought (but Backyard did say they were fine afterwards)

    On the other hand, I’ll quite happily kill a cockroach whenever I see one. They’re vermin. They are not endangered as a species, and benher’s quite right.

    On the other hand, this will help the general body of human knowledge, so that’s got to be OK under these circumstances. who knows, Backyard may inspire some scientist of the future to go and cure MS or something.

    On the other hand, do we have to do ‘live animal experiments’ repeatedly? Can’t we just rely on video records of what’s gone before rather than having to do it again?

    On the other hand, they are just cockroaches so who really cares?

  30. It should make for a better crunch when you step on it with all those crunchy chips on it’s back.

    Sorry, but I don’t feel much for roaches, thought they did act quite cool in Joe’s apartment…

Comments are closed.