June 2013
Volume 54, Issue 15
ARVO Annual Meeting Abstract  |   June 2013
Training for Identification of Colors Altered by Colored Filters
Author Affiliations & Notes
  • Thomas Kuyk
    TASC, Inc, Ft Sam Houston, TX
  • Alastair Smith
    TASC, Inc, Ft Sam Houston, TX
  • Semih Kumru
    Air Force Research Laboratory, Ft Sam Houston, TX
  • Footnotes
    Commercial Relationships Thomas Kuyk, None; Alastair Smith, None; Semih Kumru, None
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2013, Vol.54, 3019. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Thomas Kuyk, Alastair Smith, Semih Kumru; Training for Identification of Colors Altered by Colored Filters. Invest. Ophthalmol. Vis. Sci. 2013;54(15):3019.

      Download citation file:

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

  • Supplements

Purpose: Colored filters (CF) can alter the appearance of colored stimuli because they selectively absorb visible light. Last year we reported subjects can learn to rename narrow band (NB) hues (generated on a CRT) that were altered in appearance by a CF. We repeated last year’s experiment but with broad band (BB) reflective hues and also determined if learning to identify altered BB hues had any effect on identifying NB hues.

Methods: Six subjects were tested on a color naming (CN) task wearing a CF that blocked short wavelengths. The CN stimulus set consisted of 8 circular BB hues printed on a white background. One hue was presented at a time using a color wheel (CW) format. Each CW had 8 hues that differed in the set and order. One CW set had colors labeled by name and another set had no labels. After familiarization with the color set using labeled CWs, CN without a CF was tested with the unlabeled CWs. Each color was presented 8 times in random order and subjects had a choice of 8 CNs. The test was repeated with the CF. Then the CN test with NB hues was given without and with the CF. Next a block of training trials with the CF on was given using the labeled CWs. Subjects examined 64 colors (8 for each color in the set) at their own pace and were instructed to learn to associate the CN with hues as they appeared through the CF. No training with the NB hues was given. Immediately after the training block, the BB CN test was re-administered. Subjects were instructed to use what they had learned in training to name the colors. Re-testing without further training occurred again 2 and 7 days later. Also on day 7, the CN test with the NB hues was re-done. A control group of 5 completed a similar sequence of CN testing with BB hues but were never given any CN training. Data were CN error rates.

Results: CN without a CF was near 100% correct. With the CF, but before training, errors increased to 33% incorrect. After training errors decreased to 14% and stayed there at 2 days post training. At 7 days the error rate rose slightly to 20%. The initial error rate for the NB colors with the CF was 71% and was unchanged at day 7. CN performance for the control group showed little improvement with repeated testing.

Conclusions: Subjects were able to learn to rename a set of BB hues altered in appearance by a CF and retain that ability over a few days. There was no evidence what was learned for BB hues transferred to CN of NB hues.

Keywords: 469 color appearance/constancy • 471 color vision • 579 learning  

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.