Friday, April 20, 2007

Mini review. Psychophysics of Reading in Normal and Low Vision

In February I wrote about Siliva Zuffi and Carla Brambilla's work on the readability of colored text on a colored background (permalink). In the second paragraph I mentioned Gordon Legge's work. Legge has revisited his work and compiled it in a must-have book for anybody doing graphical user interfaces or working on digital publishing.

In Europe, and more so in Japan, there is great concern about low vision, because the population is rapidly aging and concomitantly there is more reliance on electronic displays. The eye has not evolved for the longevity of today's homo sapiens, and an aging population means that an increasing percentage has to cope with diseases like glaucoma and macular degeneration.

At the same time, we no longer make a phone call by turning a rotary dial but by selecting a number from a phone list displayed on a tiny and dim LCD screen, we no longer stay in line at a government office to speak to a representative but fill out a Web form, we no longer have instruments or warning lights in our cars but read the car status and our location on a navigation panel, etc.

In the United States the problem is not yet as grave—according to Legge's book there are only an estimated four million idividuals with low vision. However, the federal government has regulations for the accessibility of federal documents. It is not just a question of being considerate to our elders, it has become a requirement for doing business with the government.

My first encounter with Legge's work goes back to about 1987 or so, when I was working for Gary Starkweather, who asked me to look into the trade-off of resolution vs. gray levels on the new ionographic printers that started showing up on our desks. At that time, I had the chance to meet Legge at an OSA vision meeting in San Francisco and ask him for advice on setting up a readability experiment. He did not tell me to read a paper or a book—he listened patiently at my problem description and then he sketched out how he would design the experiment if he would be at my place.

For the historian of science it is interesting to read Legge's papers and learn how his knowledge evolved and expanded. However, his book is what is valuable for us doing science, because in it Legge revisits all his research with what he knows now. This is why I think it is a must-have book. Here is the catalog information:

Gordon E. Legge
Psychophysics of Reading in Normal and Low Vision
Lawrence Erlbaum Associates, Mahwah, 2007
ISBN: 0-8058-4328-0

At $110.00 the book is quite expensive. The price includes a CD-ROM with full reprints of the twenty original articles in the Psychophysics of Reading series. The text of the articles has been type set in a uniform format, and the figures have been reprinted from the originals. The CD-ROM also contains a cumulative reference list with all the citations from the book and the twenty articles, unfortunately only in PDF, not in BibTeX format. Supporting material for the MNREAD test is also included—a score sheet which can be printed for use, and computer source code in C, MATLAB, and Perl for estimating MNREAD parameters from test data.