September 2005
Volume 5, Issue 8
Vision Sciences Society Annual Meeting Abstract  |   September 2005
Why is light text harder to read than dark text?
Author Affiliations
  • Lauren V. Scharff
    Stephen F. Austin State University
  • Albert J. Ahumada
    NASA-Ames Research Center
Journal of Vision September 2005, Vol.5, 812. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Lauren V. Scharff, Albert J. Ahumada; Why is light text harder to read than dark text?. Journal of Vision 2005;5(8):812.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Scharff and Ahumada (2002, 2003) measured text legibility for light text and dark text. For paragraph readability and letter identification, responses to light text were slower and less accurate for a given contrast. Was this polarity effect (1) an artifact of our apparatus, (2) a physiological difference in the separate pathways for positive and negative contrast or (3) the result of increased experience with dark text on light backgrounds? To rule out the apparatus-artifact hypothesis, all data were collected on one monitor. Its luminance was measured at all levels used, and the spatial effects of the monitor were reduced by pixel doubling and quadrupling (increasing the viewing distance to maintain constant angular size). Luminances of vertical and horizontal square-wave gratings were compared to assess display speed effects. They existed, even for 4-pixel-wide bars. Tests for polarity asymmetries in display speed were negative. Increased experience might develop full letter templates for dark text, while recognition of light letters is based on component features. Earlier, an observer ran all conditions at one polarity and then switched. If dark and light letters were intermixed, the observer might use component features on all trials and do worse on the dark letters, reducing the polarity effect. We varied polarity blocking (completely blocked, alternating smaller blocks, and intermixed blocks). Letter identification responses times showed polarity effects at all contrasts and display resolution levels. Observers were also more accurate with higher contrasts and more pixels per degree. Intermixed blocks increased the polarity effect by reducing performance on the light letters, but only if the randomized block occurred prior to the nonrandomized block. Perhaps observers tried to use poorly developed templates , or they did not work as hard on the more difficult items. The experience hypothesis and the physiological gain hypothesis remain viable explanations.

Scharff, L. V. Ahumada, A. J. (2005). Why is light text harder to read than dark text? [Abstract]. Journal of Vision, 5(8):812, 812a,, doi:10.1167/5.8.812. [CrossRef]
 NASA Airspace Systems Program

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.