Very cool project. I am very skeptical of the article, but Nature doesn't publish foolishly and there are indeed plenty of "hacks" when it comes to color constancy or other aspects of color vision that always leave the door open for other counterintuitive phenomena. Looking forward to your DIY findings.
> but Nature doesn't publish foolishly yes they do. many meta-analyses have found the most prestigious journals publish the highest proportion of studies that don't replicate. the relationship continues down the prestigious spectrum. this is unsurprising because to get into nature you need to publish something amazing, and the most amazing things are things that are wrong. for example, magic tricks are amazing, but they are just clever deception
I think the bigger problem with the testing might be that RGB from a monitor simulating different wavelengths could give different results than you would get from comparing light that's actually different wavelengths. You're not actually testing your eyes' ability to recognize the difference in the colour of the light itself, but the colour your eyes perceive from changes in the balance of 3 other colours.
Thanks for the comment -- I don't think there's much of an issue, here. All of the colors in the chromaticity diagram shown in the video are in fact only obtainable as combinations of wavelengths. The only 'pure' colors shown in the diagram are those along the curved edge of the diagram (indicated with the wavelengths). So when the scientists did their testing along the protan and tritan axes, they were (necessarily) using combinations of wavelengths to produce those colors. While it's true that the particular spectra the scientists used will differ from whatever spectra my monitor is using, and I could conceive of that being an issue, it sounds like the way this method improves color vision is by general activation of the retinal cone cells overall, not by sensitizing them to particular wavelengths or changing their spectral response in a particular way. That may be wrong, of course, but that's my impression. So I don't think that using metameric colors (wiki: metamerism) should be a problem in the testing, unless something more subtle than expected is going on.
Agreed with Casey. While the Wavelength Discrimination Function is a sort of gold standard to judge an individual's color vision, it is usually just used academically, not clinically. The resolution along the blue-yellow (tritan) and red-green (e.g. Protan) axes, as being tested from the RGB monitor, is indeed proportional to the WDF. I am concerned however with using 8bit sRGB, since I don't think you'll quite have the precision to clearly detect even a 5% improvement to color vision.
Non-technical person here. Why does it have to be glasses? Could you create a web page that displays the right color so that when viewed full-screen all you can see is the color? (Maybe make some simple blinders?) But maybe monitors can't mimic precise wavelengths? And there's probably too much variation in monitors?
You're right that monitors can't reproduce precise wavelengths, but also the light intensity from a monitor is several orders of magnitude lower than that achievable with LEDs mounted directly before the eyes.
Strictly speaking, monitors can't reproduce pure 670nM light... however it's not clear to me exactly how strict the requirements are. The paper refers to previous research that explored (successfully, I take it) 650-900nM wavelengths. The sRGB red primary (i.e. what computers conceptualize as "pure red") is reasonably close to about 610nM, and the red LEDs in the pixels of computer monitors, at least in some cases, are going to be emitting some amount of 670nM light (see e.g. tables 7a and 8a here: clarkvision.com/articles/color-spaces/ ) ... given that the researchers are exploring lower-power lights with success (e.g. the above-mentioned paper), it seems plausible to me that a monitor could possibly have some effect, depending on what kind of LEDs it uses (one would need a spectrometer to be sure -- e.g. table 8b in the link above shows a monitor that does not emit that wavelength.) Maybe you would want to crank up the brightness and sit an inch from it, maybe you'd need a longer dose, etc. Given that I'm not a doctor I certainly can't make a recommendation. If this approach turns out to be useful and powerful I'd imagine that folks would be characterizing monitors by model and posting the data.
Thanks! I'm afraid I can't help you re: Europe (I got mine in the US) but in case it helps: the LEDs that I used were Lumileds L128-DRD1003500000 (in the Luxeon 2835 Color Line). These are surface-mount LEDs, though, so be aware that they are difficult to solder without damaging. Good luck!
Doesn't the light flux ultimately depend on the current you are driving them with? And with the potentiometer, how can you know how much current you are driving? In your schematic you mention a 100 or 250 Ohm pot + 4.7 per led, so assuming a switched off pot you are driving them with over 150mA... What I mean is that in the end it seems to me that either you don't put any pot and then you can calculate precisely the exposure, or with a pot you are going to just eyeball it (put not intended)... and yes I'm trying to build my own pair of glasses here :)
Hey Casey, man how smart you are! I wonder can you see out of these or is it just shinning a red light in your face. Impeding normal vision Also would a pair of glasses with a deep red lens cha get wavelengths of normal led lights say in the ceilings?
Hi - sorry, i thought i replied already: no you can not see out of these. In terms of using filters to get some benefit from ambient light, i don't think that would help: without the filters you still get the red light so adding the filters wouldn't change that... The other frequencies of light do not interfere with the effect (as far as i know). So you can just be in bright sunlight and probably get a similar effect, but these glasses just concentrate it and make it easy/fast to get a dose. In other words, the point of these glasses isn't to make you see only red light and no other light; the point is just to have some bright red light in your eyes.
@@caseyconnor2 got it ! So Ive read a study that red light glasses used on shift workers increased alertness and decreased stress. When we wear red lenses the ambient light is not converted to a red wave length of some sort? Because we see everything red
@@Nando_lifts2021 right, the filters do not convert the light to red, they just block the light that isn't red. If for example your surroundings are illuminated by a sodium-vapor light (which used to be common in streetlights and which has an extremely narrow frequency band of light that it emits) and you look through filters that filter out that frequency, you will see nothing at all. All that said, there is an effect where (in regular light, not sodium-vapor) you put filter glasses on and because the light is now dimmer your pupils dilate and let more light in, which does effectively "turn up" that frequency of light, so it does kind of work the way you imagined. (Perhaps this is how the study you described got a result.) Whether that effect is significant depends on the ambient light in question and how much more of it you want to get in to your eyes.
Check the numbers on-screen at 3:27 -- mine were spec'ed in radiant flux (lumens), but my estimation/calculation is that they are ~345 mcd at 75% of nominal current, so ~459 at nominal current. But again, that's based on a non-expert's math, so take it with a grain of salt. :-)
It's also important to consider that mcd is a measurement of the intensity of the light, and not the total light emitted. An LED with a narrower focus will have a higher intensity (mcd) than the same LED with a wider lens focus. In a case like this, where you're capturing most of the light, regardless of direction, lumens is a much more useful measurement.
You mean the LEDs used in Christmas lights? There's a chance they produce some of the necessary wavelengths, but it would depend on the particular lights. There is too much variation to make a broad statement.
