Why is white on black considered higher contrast than black on white?

Question

While researching to answer Why are "Inverted Colors" considered an accessibility feature? I noticed the puzzling claim that "White text on a black background is a higher contrast to the opposite, so the letterforms need to be wider apart, lighter in weight and have more space between the lines." It's repeated a lot, particularly by design folks, but I couldn't find psychological research to support (or refute) it.

In terms of actual contrast ratio, inverting a black on white scheme won't change the contrast at all; the colors themselves are just as distant. Yet I see many claims that "inverted colors" is high contrast. The only real difference between the two I can think of is that the white background of black on white produces considerably more light than a black background. If this difference is enough to induce some level of ocular adaptation perhaps that could be a cause.

So is there a perceived difference in contrast between black on white vs white on black, assuming the same intensities for black/white? What causes it?

Answer 1

Your question is referring to display polarity. A positive polar display consists of dark letters on a light background, a negative polar display consists of light letters on a dark background. Polarity by itself is independent of text-to-background contrast, as you rightly state.

Generally, positive polarity facilitates performance (e.g. Buchner & Baumgartner, 2007). This effect is due to the higher overall display luminance (Buchner, Mayr & Brandt, 2009) possibly leading to a greater constriction of the pupil (Taptagaporn & Saito, 1990, 1993) and thus increasing the depth of field and decreasing spherical abberation.

However, this advantage of positive polar displays does not generalize to readers with visual impairment. People with low vision due to cataract (ocular clouding) perform better with negative polar displays (Legge et al., 1985, Sandberg & Gaudio, 2006). This effect has been discussed to be due to a scattering of light leading to a resulting veiling luminance and reduction in text-to-background contrast (Rubin & Legge, 1989).

In short: At the level of the display, a change of polarity has no effect on the text-to-background contrast. If the reader's eyes are cloudy, however, the contrasts at the retinal level is decreased for positive compared to negative polarity. This is why offering negative polarity is considered an accessibility feature.

As to why design folks claim negative polarity to be higher in contrast, I can only speculate on the basis of these findings. I suspect that it has to do with the adaptation of the eye to the lower overall luminance of the display. The intensity of the white letters may be perceived as more intense. Also the wider pupil dilatation would lead to a decreased depth of field. This in turn would require wider spacing and lighter weights of letters to achieve readability comparable to positive polarity. Unfortunately, I don't know any research to substantiate these speculations (which doesn't mean that it does not exist).