Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Medical Experiment to test variation in light perception (thoughts?)

  1. May 29, 2010 #1
    So I'm planning a senior honors thesis that'll start next summer. In my last post, I talked about testing magnetoreception, which is kind of risky. Another idea of mine is to take human subjects into a dark room, use filtered light to produce light at increments one 1 nm, and note the longest wavelength of light the individual can perceive. I'd then plot the distribution and study it. I'd love to find someone who was two or three standard deviations from the mean and could see light traditionally classified as IR! Of course, to find such a person, I'll need over a hundred human subjects.

    What do you guys think of this? It doesn't seem as well researched as one might expect. For instance, I can't even find an exact average for the longest perceivable wavelength. Most of us know it's about 700-750 nm, but I cannot find an exact wavelength anywhere, which leads me to believe there might be a lot of variation? I'd really hate to do this experiment and then find out that everyone has the same cutoff. But it's pretty safe to assume there'll be at least some variation, right? (Even my two eyes perceive color differently--my right definitely sees more blue).

    So I'd love any input you all have. Oh yeah, and I'm a neuroscience major, hoping to do a PhD after college.
     
  2. jcsd
  3. May 30, 2010 #2
    This sounds very, very interesting to me! I'd also like to see some subjective tests, e.g. "press the button when you consider the color 'orange'", or "press the button when you think the color is getting redder". I know that the photoreceptors in the eye can vary very much.
     
  4. May 30, 2010 #3
    there's a thread on the site somewhere about IR vision. apparently, some people's SOs can see the IR LEDs on their remote controls light up and see in the dark. female, IIRC. which is sort of interesting because some variability exists in cones and the wavelengths they are tuned to, and a few females exist that inherited genes for 4 different cones (this is not possible in males) giving them quad-phosphor vision and the ability to perceive colors that others do not. to further complicate matters, the sensitivity of the cones (or perhaps it was rods) can be influenced by diet, and there was some use of this fact during WWII to give Navy personnel enhanced night vision. but my searching on that was fruitless, not sure what they used.
     
  5. May 30, 2010 #4
    Carrots, I believe. But wasn't this a myth they spread out to disguise their use of radar?
     
  6. May 31, 2010 #5
    i don't know, didn't find anything. the only other things i think i know about it are lore from an old english teacher about not being able to smoke on deck at night because it could be seen from several miles away, some stuff about precision resistors being key in the early days of radar (stumbled across that when ordering some from Loeb Julie, once).

    there is some mention in Tom Clancy books about special forces guys eating vegetables to improve night vision, but you never know about that stuff.
     
  7. Jun 5, 2010 #6
    I haven't read the post in full, but enhancing color vision will not enhance night vision. Rods give you good night vision, not cones.
     
  8. Jun 6, 2010 #7
    Proton Soup, can you show me how to find that thread? I know there are rumors that some women have tetrachromatic color vision, but my understanding was that it hadn't been officially confirmed. Do you think these tetrachromatic women would be rarer than colorblind men, or about the same? I doubt I could find any tetrachromates in my sample size, unless they're somewhat common. If the experiment I described is done only among trichromates, do you all think it will still be fruitful?

    Thanks for the ideas, everyone!
     
  9. Jun 7, 2010 #8

    fuzzyfelt

    User Avatar
    Gold Member

    I believe the receptors of colour in humans are generally skewed compared with other trichromats, and that there are individual variations in this too, however, Ivo Kohler’s disputed work from the 1960s is interesting regarding perception.

    I don't have a direct link, but this link describes the experiment involving coloured goggles, begining towards the end of page 8, where perceptions seemed to change after some weeks of wearing goggles, and also took some weeks to revert once the goggles were removed.

    http://www.bristol.ac.uk/philosophy/hurley/images/papers/hunter_gatherers.pdf [Broken]
    (yet another link about synaesthesia)
     
    Last edited by a moderator: May 4, 2017
  10. Jun 7, 2010 #9
    the navy thing wasn't quite as i remembered, but it was a matter of color vision i think

    https://www.physicsforums.com/showthread.php?t=134077&highlight=infrared

    http://www.mindhacks.com/blog/2005/08/changing_diet_might_.html

    http://www.blogadilla.com/2008/06/08/are-you-a-tetrachromat/ [Broken]
     
    Last edited by a moderator: May 4, 2017
  11. Jun 10, 2010 #10
    Thanks for the links, Proton Soup! I am really encouraged by what you guys have found. It sounds like diet does effect visual perception, so I should have a good experiment.
     
  12. Jul 17, 2010 #11
    Wow, really cool idea.
    Any idea of how you will emit the light of different wave lengths?
    I imagine it would be quite hard to tell whether you are really seeing the near IR light, or you are imagining it, your eyes can see things in the dark that aren't there, and it will happen when people are trying to concentrate on the light .
    I imagine it would be easiest to show the light as a small dot and being up close to it.
     
  13. Jul 17, 2010 #12
    Hey Rathat,
    Yeah, I was asking one of my profs how to make light at specific wavelengths. He said I would need to use colored gels to filter the light, and then use a spectrometer to make sure it was the right wavelength. This is going to be a really hard experiment: I'll need lots and lots of colored gels, and I'll need lots and lots of participants if I want to find someone who is really abnormal, say two or three standard deviations from the mean.

    As for making sure the person is really seeing the light ... well, I'm not too worried about that. People do kind of see stuff in the dark ("Prisoner's Cinema," it's called), but I can always tell the difference between a real light and an imagined one. I'm more worried about people lying, saying they can see it when they don't. I'm going to do a few false ones to weed out any liars. That is, a few times, I'll have no light shining at all, and I'll ask the participant if they can see anything. Anyone who says yes is out!

    Thanks for the ideas, guys! Anyone else know how I can generate light of specific wavelengths at 1 nm increments??
     
  14. Jul 17, 2010 #13
    If you get a few friends in on it, you can make it double-blind, and then do a full round with placebo and light.

    Oh, and you can use lasers to generate light at specific wavelengths, but you would absolutely need to use an indirect method of exposing the eye to the laser source, to avoid damage.
     
  15. Jul 17, 2010 #14
    I asked my physics professor about this, and he said I would not be able to find lasers for every wavelength I need. Unless I can find lasers for [tex]\lambda[/tex]= 701 nm, 702 nm, ... 749 nm, 750 nm, I'll be out of luck. Plus, that many different lasers would be prohibitively expensive, I would think.

    Also, does anyone think I might have better luck testing the UV/near UV? My physics professor seemed to imply that would work better, though I can't remember why.
     
  16. Jul 18, 2010 #15
    Just found something incredible! Despite all conventional wisdom, humans can see UV light down to 300 nm! The lens of the eye blocks 300 - 400 nm, but after cataract surgery, people discover they can see UV light. It is not perceived as a new color, though :frown:. It appears whitish-blue or whitish-violet. Here are my sources:

    http://www.guardian.co.uk/science/2002/may/30/medicalscience.research

    http://neuronresearch.net/vision/files/tetracomparison.htm

    In light of that (no pun intended), I will probably test the near IR end of the spectrum in my experiment. I think I should use the three forms of vitamin A as variables. Vitamin A1 is the common one; anyone know how/where to get vitamin A2 and vitamin A3?
     
  17. Jul 26, 2010 #16

    Well, frequency is frame dependent, but you'd have to chunk that laser pretty fast though...

    Wow. I never snapped at how fast things must be moving away from us to produce such prominent redshifts.
     
  18. Jul 26, 2010 #17
    No reason that you should, given that we don't experience the contrast naturally in our daily lives in a way that we recognize. I kind of like the notion of hurling lasers by hand at friends while claiming that it's about frequency shifting... :evil:
     
  19. Jul 26, 2010 #18
    So I'll strap some lasers to a super rocket and ask my test subjects to look at it ... these answers are getting creative :biggrin:.

    Have you guys seen those bumper stickers that say, "If this bumper sticker looks blue, you're driving too fast?" :rofl:
     
  20. Jul 26, 2010 #19
    HA! I have to get one of those bumper stickers. How great would it be for a sprinter to have, "I'm not blue, I'm fast" on their shirt?

    Now, if you want to maximize the relative velocities of the laser and your... lets call the subjects "targets", lets take a lesson from particle colliders: You strap the laser to a railgun opposite a chair for your target, in a parallel and opposite railgun. Fire each at one another, with the laser angled to shine on the target at the moment when they pass each other.

    Now, you'll have a limited period of time to get info from your subjects before they die from trauma and hemorrhage, but you could weed some adrenaline junkies from the gene pool, AND study the human eye! :biggrin:
     
  21. Jul 26, 2010 #20

    DaveC426913

    User Avatar
    Gold Member

    My son was able to detect the IR light from our TV remote with 100% accuracy. I was very careful to try to fool him as best as possible, but he called it instantly every single time. No one else in our home was able to see it at all.

    Do you pay for flights? :biggrin:
     
  22. Jul 26, 2010 #21
    Hahaha, I'd pay for flights if it weren't for the all the colored light filters I'll have to buy and all the test subject I'll have to pay! But if you live within driving distance of Williamsburg, VA, feel free to drive down! Anyway, I tried something similar with my mom and dad, though maybe not as controlled and rigorous as what you did with your son. My mom could see the TV remote light, my dad couldn't. Hopefully I'll be getting a lot more and better data a year from now.
     
  23. Jul 26, 2010 #22

    DaveC426913

    User Avatar
    Gold Member

    Piffle. A mere 11 hour drive.

    Really? Huh. Blows my theory about it being highly correlated with youth.
     
  24. Jul 27, 2010 #23
    Unfortunately, if and when I do this experiment, I imagine I'll have to use all college students as subjects ... unless I can yank over a bunch of elementary school kids and senior citizens. I will look at sex as a variable, though, since color-vision is a sex-linked trait. Tetrachromacy is only supposed to exist in women, not that I actually expect to be lucky enough to find a tetrachromat. More likely it'll just be variation in the sensitivity of cone cells, and for that reason I'm still hoping to use different forms of vitamin A as a variable.

    One day, when I become a professor, I think I'll put up a big sign in my lecture hall that says, in letters only visible to those who can see infrared light, "If you can read this sign, see the professor immediately!"
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook