Smallest Difference in Wavelength the Human Eye Can Notice

[valdo333]

Hello - I'm having a difficult time finding this answer anywhere else on Physics Forums. I'm curious if anyone knows, or can figure out, the smallest difference in wavelength that the human eye can notice. For example: Can the human eye detect if a light changes from 568nm to 570nm?

 

[Drakkith]

There isn't really an answer for this. It depends strongly on the individual person. Some people can tell the difference between wavelengths very close together, while others, like my dad, are color blind to some extent and certain colors are much harder to tell apart than others. Even two "normal" people can have a large variance in their color perception.

 

[SteamKing]

The human eye can distinguish about 10 million different colors:

http://en.wikipedia.org/wiki/Color_vision

The perception of color is not confined solely to the eye: the brain plays a major role in determining what colors we 'see' in response to the stimulation of the retina.

 

[valdo333]

I figured there wouldn't be a definite answer. Thanks for the help!

 

[Vanadium 50]

Also, the intensity matters in whether or not two colors can be distinguished. You want both to be of similar, moderate intensity.

 

[sgb27]

You could pick two monochromatic colours a certain wavelength apart and calculate the ΔE (http://en.wikipedia.org/wiki/Color_difference) value for them, that would give you a good idea of how visible the difference between them would be.

From actual experience of red/orange colours around 600-650 nm, I'd estimate somewhere around 1-2 nm would be the smallest difference you would notice when viewed side-by-side (5 nm is easily visible). I have no idea for other colours, it could easily be very different.

 

[sophiecentaur]

If two areas of colour are touching then the discrimination is very good. 'Millions of colours' are needed in colour displays to prevent the eye from seeing contours in areas of subtle colour grading. That would suggest that the eye could, perhaps, discriminate 0.1% change in wavelength (very roughly).
But our discrimination in 'colour perception' varies quite a lot over the total gamut of our colour field. The most sensitive being in the region of 'skin tones'. We are not so sensitive to changes along the line of the spectral colours - because we evolved in an environment where there are very few actual (pure) spectral colours around.
It's worth looking at this link for an overview of our colour vision.