Hi Susan,

That's a great question and I suppose there are a couple of ways of answering it. Before we get into the heart of the matter, I think we should cover a few basics. Firstly, there are 2 different types of photoreceptors in the human eye: rods and cones (there are actually 3 different types of cones). The rods are mainly responsible for night vision whereas the cones are used for color/fine vision. One of my previous answers covers the basics of photoreceptors and the differences between the two, if you'd like a refresher (I also have some links there for more information and lots of pretty pictures).

These two kinds of photoreceptors respond to different intensities of light radiation (as you may know from physics, light can be thought of as either a wave of electromagnetic radiation or as photons; both forms carry energy). A dark-adapted rod photoreceptor can actually detect individual photons -- this is an absolutely spectacular feat of amplification within the cell!¹ Cones, on the other hand, amplify light signals to a much lesser extent than rods do: cone response amplitudes to single photons are only 5% of that of rods. ² The main reason why cones can't detect single photons is that this small detection event is buried in the high noise produced by biological activity in the cone cell (mainly due to thermal agitation of the visual pigment ³).

Although that is the response of individual photoreceptors, vision itself is mediated by complex visual pathways and these require the activation of more than just a single cell. If we look at just the rod visual pathways (e.g., night vision), we see that although these rod systems sacrifice high spatial and temporal resolution (not to mention colour vision), they gain the advantage of becoming extremely sensitive to very dim lights. Under optimum conditions, it has been found that fewer than 100 photons striking the eye (10 rods per 1/10 of a second) are enough for rod-mediated vision. ⁴ That means that the dimmest detectable visual stimulus is equivalent to the light from a candle placed 17 miles away!

But this is only half the story; what about colour vision? Dark-adapted rods are tremendously more sensitive than the colour-detecting cones and so it only makes sense that we need a greater stimulus to reach threshold in cone-mediated vision. The specific numbers for the red, green, and blue cones varies but, in general, for colour vision, the number of photons seems to be around 10,000 photons per second ⁵ (see also Adrian Popa's excellent MadSci Answer). I hope this helped answer your question. If you have any further questions, or find any errors, please feel free to drop me a line at rickys@sethi.org.

Regards,


Rick.

References

1  S. Hecht, S. Schlaer and M.H. Pirenne, Energy, Quanta
and vision  Journal of the Optical Society of America, 38, 196-208
(1942); 

Baylor, D.A., Lamb, T.D. & Yau, K.-W. (1979). Responses of
retinal rods to single photons Journal of Physiology 288, 613-634; 

Baylor, D.A., Nunn, B.J. & Schnapf, J.L. (1984). The photocurrent,
noise and spectral sensitivity of rods of the monkey Macaca
fascicularis Journal of Physiology 357, 575-607. 

2  Barlow & Mollon, 1982; Rodieck, 1998

3  Lamb, T.D. & Simon, E.J. (1977). Analysis of electrical
noise in turtle cones. Journal of Physiology 272, 435-468; 
Schnapf et al., 1990; Donner, 1992

4  David R. Copenhagen and Tom Reuter 

5  Hecht, Shlaer & Pirenne, 1942