Headless flies respond to light—Or: Why invertebrates are awesome


This is possibly the best opening line to a peer-reviewed research paper that I have ever read:

When I tell people I've been working on headless fruit flies' responses to light, they often look puzzled or laugh nervously. Allow me briefly to explain why I started cutting off flies' heads.

In this highly readable paper by Marc Egeth, we learn that flies continue to respond to light under conditions where they shouldn't be able to—namely, when their phyiscal movement is dulled by high doses of anesthesia, and (more astoundingly) when their heads have been severed from their bodies. This has some implications for the anesthesia—obviously, it doesn't completely restrict movement, so it would be interesting to know whether it's dulling pain as much as we think it is.

But it also raises some questions about what the heck is going on with the flies' sensory perception. Egeth has two theories. Prepare to get your mind blown a bit, on several subjects:

By what mechanism might headless flies detect light? Two possibilities are photodetection or thermodetection.

The light source itself was not hot because the light was channeled away from the halogen bulb by fiber optics (the instrument was a Schott ACE microscope light). In addition, I also found that headless anesthetized flies would move in response to a light from a 100-lumen LED flashlight, which does not get hot.

But, the flies might be responding to being heated by the infrared components of these lights. Or, the flies' extracephalic photoreceptors, which have previously been implicated in circadian entrainment, might also directly drive behavior. For readers of Perception, photoreception might be a more interesting alternative than movement in response to warmth, but either mechanism would be a novel phenomenon for the body of fly literature (Xu et al, 2006).

Headless flies are known to maintain posture, walk around, entrain to new circadian rhythms, engage in defensive behavior against conspecifics, and even learn to avoid a shock - and this even faster than flies with heads [emphasis mine](Booker & Quinn, 1981).

Butterflies have photoreceptors in their penises that guide mating behavior (leading to the phenomenon termed "hindsight"), and crayfish have "caudal photoreceptors" and thermoreceptors that can drive walking behavior without the involvement of the brain (Arikawa, Suyama & Fujii, 1997; Wilkens, 1988). After I submitted this paper, Xiang et. al. (2010) reported that light-responsive cells "enable [Drosophila] larvae to sense light exposure over their entire bodies and move out of danger." It seems plausible that similar cells are present in adult Drosophila bodies. These phenomena remind us that central human processes may be found distributed throughout the invertebrate body.

Naturally, the first thing this makes me think about is cephalopods. Like the flies, cephalopods are invertebrates. They also have a brain that distributes processing around the body. And, the like animals mentioned by Egeth, researchers have found evidence that cephalopods sense light in places other than their eyes. In fact, Roger Hanlon, a researcher at the Woods Hole Marine Biological Laboratory, thinks that octopuses might be color-blind in their eyes, but perceive color through light-sensitive parts of their skin. (You can find out more about this by fast forwarding to the 20-minute mark in the video I made for BoingBoing about cephalopod neurobiology.)

There's clearly some really interesting stuff going on in the world of the invertebrate senses. I, for one, cannot wait to see what we find out next.

Read Marc Egeth's paper on light-responsive headless flies

Watch a video Egeth made where you can see the flies respond as he turns a microscope light on and off.

Big thanks to Daniel Graham!

Image: Some rights reserved by Image Editor



  1. I’d like to see if humans have this. I sleep with a blanket over my face (hey I like being smothered and keeping the light out of my eyes) but if my arms or legs are exposed and the sun comes out, I wake up. No matter how tired I am. I feel like just having the sun on me makes me keeps me up.

    1. In humans, the blood is photosensitive. The backs of your knees and insides of your elbow are places that light can get to the vessels and wake you. If I wake up suddenly I usually have my arms or legs out in the open if there is ambient light present.

      When I was working nights I found that completely closing off the light sources to a sleeping room during the day can make a great difference on how well I slept off-cycle. Even a small leak made the sleep less thorough.

      As far as the flies, there is a human misconception that the center of intelligence is the brain. Since we have such a big one, you know, compared to anything else. We can’t quite fathom, from this vantage point, how an animal with a “small one” can “know” so much or function without one.

      The internet will slowly help us with that perception: We are increasingly letting the network store our info and devices be our memory, outsourcing our brain functions, and it will suddenly dawn on someone that maybe that’s how animals do it. Maybe birds use Google Maps For Animals to migrate.

      I know that sounds trite and dumb to the intellectual brain, but each moment brings a decision point. If we make the right decision at the right time, we are constantly in the right place. If you use a Tom Tom, you don’t ned to know the whole city, you just need to know where to turn.

      So all any creature needs is a way to divine what the next move should be, not store the whole process. It’s actually easier and more efficient to do so. Our neurotic and conflict-ridden stack of 7 evolutionary brain layers with new improved frontal lobes on top often just gets in the way. People spend years meditating to achieve what a bird does (living right without dwelling on it) every day.

      But science will continue to discount stuff which isn’t infinitely repeatable. After all, there’s no way that each moment is different and brings a new alignment of intent and possibility. A real scientist is someone who wins the lottery but won’t accept the money until they do it again in a blind test. :^)

  2. I have always wondered if humans, under anesthesia, still “feel” the trauma of surgery. More notably, an elderly family member underwent heart surgery, and experienced symptoms of PTSD, which I have dealt with for several years. The family member’s surgery was to improve quality of life; I recognize that his surgery would have left him sore, and medications can alter mood and functionality, but he experienced genuine panic, mood swings, nightmares, and a general feeling of complete psychological malaise.

    Wonderful article.

      1. “I’ve woken during surgery, so wouldn’t be totally surprised if your relative’s anesthesia was incomplete.”

        There’s a fallacy here somewhere…

        1. “There’s a fallacy here somewhere…”

          Not sure what you’re getting at, but I suppose I should make clear I was the patient, not the surgeon. ;-)

          “but.if.and.that in reply to jjsaul
          How horrible…”

          It wasn’t as bad as it sounds… I have no memory of any pain, just the very bright lights and movement. I would have assumed that I imagined it had the anesthesiologist not brought a group of med students in the next day to talk about it. Or I’d have recalled it later and be on the alien abductee circuit now.

          On the original topic, the idea of distributed processing in the fruit fly amazes me more than in cephalopods because they _fly_. It’s hard to imagine achieving the complex coordination involved in the absence of strict centralization.

      2. Oh, lovely. Thanks for the reminder to try to stay out of the hospital unless it’s a do-or-die situation. But seriously, thanks for the information. How horrible that you experienced “waking up” during surgery.

    1. Demand to have your surgery under local anesthesia. Then you can decide whether or not you’re getting adequate pain control because you’ll be awake. Plus you won’t feel gross for a week from the anesthetic.

      These days, it’s common to give patients Versed (midazolam), which is an amnesiac. It doesn’t do anything for pain control, which may or not be adequate; it just makes you forget so you won’t drive to your doctor’s house with a sawed off shotgun. The role of anesthesia in modern, cattle-call medicine is to shut you up.

      1. I’ve heard same about scopalomine in childbirth.
        And in colonoscopies.

        Haven’t had either, can’t say. Keep a notepad with you as you come out, else you forget, like a dream.

      2. I consider myself to be fairly well-read, but all of this is news to me. Thanks for giving me a new direction to explore (on paper, of course!). There is more than one nightmare in this story. I might, however, need some Versed for work.

      3. Midazolam also relaxes muscles (very very important during surgery) and lowers the amount of gas anesthetic required (which is, presumably, safer). Usually an additional analgesic is given. I wouldn’t be so hasty to sell amnesia short – wouldn’t you like to avoid remembering unavoidable, uncontrollable unpleasantness?

        Here is a quick and dirty list of the priorities of an anesthesiologist for the patient, in order:
        1. paralysis
        2. memory of event
        3. pain control
        4. unconsciousness
        The least important thing is that you’re ‘unconscious’, so if the other stuff is working don’t worry about it. But the more of these points you cover, the better the (anesthetic) outcome. (I study anesthesia professionally, but IANAA.)

        I’ve woken up early from anesthesiology a few times but I don’t remember the pain.

        1. Actually that sounds pretty terrible… So for some patients they’re paralyzed, aware of and feel that there being cut open and then just forget that it happened? Honestly I don’t want to have to experience being cut open at all even if somebody told me I wouldn’t remember it.

          1. “So for some patients they’re paralyzed, aware of and feel that there being cut open and then just forget that it happened?”


            The incidence of awareness may be as high as 1 or 2 for every 1000 patients

            But think of the profound implications that has. If you truly can’t remember anything does it matter?

            I feel better knowing that we usually get powerful pain relieving drugs and amnesiac drugs if the surgery is serious.

        2. “Here is a quick and dirty list of the priorities of an anesthesiologist for the patient, in order:
          1. paralysis
          2. memory of event
          3. pain control
          4. unconsciousness”

          Here’s a better one for general anaesthesia, and I do do it for a living.

          1. Keep your patient alive.
          2. Keep your patient unconscious for the appropriate duration.
          3. Provide the surgeon with conditions necessary for the procedure to take place.
          4. Reduce the physiological impact of the surgery as much as possible.
          5. Wake your patient up as comfortable as possible.

          1. Interesting. I was sort of taking your item #1 for granted. I’d like your opinion on – to focus on item #2 though (my item #4) – I submit that ‘consciousness’ is a really low priority. Do you really think being unconscious comes before controlling memory and sensation?

            I just study how information is encoded in the cortex, so I fixate on that aspect of anesthetics and have never knocked a patient out or anything (so I welcome other opinions/information). But the anesthesiologists I work with give me lists like mine. Maybe they want our lab to make a better anesthetic depth monitor (because they complain about the current devices).

          2. I guess I’d argue that pain and conscious perception are inter-related. Without consciousness it’s merely nociception, modulation of which will reduce the physiological stress response to the surgical stimulus.

            Unconsciousness however is really all most anaesthetic induction agents (like thiopental, propofol, etomidate) and volatile agents (ether, chloroform, isoflurane, halothane, sevoflurane etc) are good for, yet early anaesthetics which utilised agents like these alone were hailed as a great breakthrough because they allowed patients to avoid the mental trauma of remembering consciously experiencing being cut apart – even though they still woke up in pain. I know that memories stop being laid down before consciousness is lost and learning can still occur without explicit conscious recall. But I find the idea of someone being painfully *aware* of an operation yet merely unable to recall it later a little metaphysically disturbing. So I’d argue unconsciousness is more important than analgesia.

            If adequate analgesia and unconsciousness are present then formal paralysis using neuromuscular blockers is not necessarily required, it depends on the operating conditions the surgeon needs for the procedure.

            Your anaesthesiologists probably do want a good depth of anaesthesia monitor, we all do. Devices we have at present are either cumbersome, less than accurate or so wrapped up in copyrights and patents we aren’t permitted to know how they work. Make one that works, is convenient to use and publish exactly how it works so we can follow the maths and we’ll kiss your feet. :-)

      4. Many operations are extremely difficult or even impossible to do on spontaneously breathing patients under local anaesthesia. Part of the role of general anaesthesia is indeed to keep a patient still to allow the operation to happen — modern anaesthesia opened the chest, head and upper abdomen for surgeons to operate routinely. Anyone fancy trying to breath on one lung while a thoracic surgeon is delving around in the space where the other lung should be? No? Didn’t think so… ;-)

        Neither is it necessarily in your best interest to demand a particular anaesthetic technique. Local anaesthesia may well mean a change in surgical technique to accommodate as well, but you are generally better off getting what your anaesthetist and the rest of the surgical team are good at. What you want may be something they can’t do or haven’t done in years. There is safety in practise.

        That said, I quite like it when patients ask to be awake if I give that as an option. Not so keen if the opening gambit is “I want you to do it this way” before I have chance to make an assessment and know what’s feasible.

        1. I’ll take a general if I have open heart surgery, but for minor surgeries, just give me a little Fentanyl.

          I had a hard time not getting overmedicated for a simple colonoscopy. I had to reschedule the procedure so that I could see the one doctor who’ll allow patients to choose no sedation. Then I told the nurse that I couldn’t have benzodiazepines and she replied, “That’s okay honey, we’ll just give you a little Versed.”

          I am Jack’s complete lack of confidence.

          1. Sure, colonoscopy and gastroscopy are perfectly feasible with no sedation and a bit of titrated analgesia. A bit uncomfortable and you remember everything, but not an issue if you’re game for it.

            The nurse who didn’t know Versed is a benzo is why generic prescribing is a good thing…

  3. The reason that’s among the best openers to a peer-reviewed research paper is because almost no peer-reviewed research papers get to the point of publication using this kind of informal language. The whole thing is written as an extremely informal anecdotal retelling of the events that lead to this line of research, and includes the sentence “So, hmmm.”

    In other words, this may be published in a peer-reviewed journal, but this is not anything like an actual peer-reviewed scientific article. Not infrequently are articles published reporting on interesting discoveries like this in peer-reviewed journals, but they aren’t peer-reviewed articles in the normal sense.

    Looking at this guy’s site, he seems to publish a lot of this informal stuff. Not knocking that, or him, at all – I thoroughly enjoy this type of informal science, and understand that it will likely lead to more formal, rigorous studies later – but it’s a bit disingenuous to pass this off as rigorous peer-reviewed research!

  4. People’s reactions to anaesthesia vary immensely. I have a friend who can’t be anaesthetised to the point of unconsciousness without stopping her breathing so all operations are live. I, on the other hand, am a cheap date: one preliminary whiff of the ether and I won’t be back till tomorrow.

  5. Maggie et al. please look up chromophores in the invertebrate brain. Turns out there are photo-sensitive neurons in those critters separate from the eyes. This may be relatively recent (eg last year)
    research; I read Science on the loo. Interesting stuff, but not too suprising, the planet has been spinning at the same rate for a LONG time, why not exploit?

  6. In college I had a work-study job for genetics professors in which my job was to cut the wings off of fruit flies and mount them to be photographed. So I literally spent two years or so getting to pull the wings off flies.

  7. We just learned that butterflies have penis-eyes, and you guys are going on about anaesthesia? Can we not focus on what’s important here?

    1. Seriously! Can someone please point me to a good article on Butterfly – Eye – Genitals?!

      (A quick google search only turns up terrible tattoos.)

  8. I clicked on the link to read the PDF and the very first thing I saw was the following as the top left header:

    Running Head: HEADLESS FLIES

    I think he just ought to title the paper that.

  9. Can someone explain why the paper says “PLEASE DO NOT CITE” on the first page? Is it because it’s less a full paper and more just a report?

    1. Anon, sometimes journal publishers make a statement like this a requirement for self-archiving of preprints.

  10. in france they did experiment on ‘guillotined head’ with preliminary tune on the post code, essentially communicated with eyelid..check on your prefered motorhead search engine
    such guies with their headless flies & boingboing in général are always way behind true science..pity there is no youtube thing

  11. “100-lumen LED flashlight, which does not get hot” Maybe not hot, but all light sources emit heat. LEDs are not 100% efficient – I have some powerful ones that require a large heatsink. (Too hot to touch, and too bright to look at.)

  12. there’s been evidence for a few years that many organisms have light sensitivity beyond their eyes — I think that you can use the back of the knee in humans to entrain the circadian rhythm to light. crazy, but no longer new!


Comments are closed.