This html article is produced from an uncorrected text file through optical character recognition. Prior to 1940 articles all text has been corrected, but from 1940 to the present most still remain uncorrected. Artifacts of the scans are misspellings, out-of-context footnotes and sidebars, and other inconsistencies. Adjacent to each text file is a PDF of the article, which accurately and fully conveys the content as it appeared in the issue. The uncorrected text files have been included to enhance the searchability of our content, on our site and in search engines, for our membership, the research community and media organizations. We are working now to provide clean text files for the entire collection.
ofle
►How many colors can be used at
►How does the nature of the display an
the operator’s task affect how the disp1 should be colored?
How well do Navy sonar operators detect targets in their displays? We know that at times they do not report them as soon as possible; they may prefer to wait until the signal is stronger and they are more certain about their judgment. But even taking such psychological factors into account, it is clear that their performance falls short of what it might be.
Signal detection theory measures how well cathode ray tube (CRT) operators can detect targets under given conditions in relation to the theoretically ideal performance under those conditions. Human operators fall far short of the theoretical optimal performance. After eliminating the psychological variables, the discrepancy that remains must be a result of the way the visual system works. To improve performance, operators’ tasks must be matched to the ways human sensory systems take in and deal with information.
In the search for ways to improve performance, one long-standing suggestion has been to add color to displays. Indeed, color-coded CRT displays will soon appear on board U. S. submarines. Objects may stand out against a background because of their color or their brightness; a great contrast in brightness generally makes objects more visible than color contrast. When many objects are in the field of view, however, the brightness cue breaks down, and color coding tends
to become more effective. .
While adding color can often n12 . display interpretation easier, the incoff2 use of color can make it harder. The > lowing questions, therefore, must addressed: .fl
► What colors should be used and 1 what way?
time without confusing the viewer: ^
nlay
The best way to color a display ^ pends on the type of display and what operator is trying to determine from Thus, display types must be studied seP‘
98
Proceedings / February
^teIy, and the speed and accuracy of eir respective operators measured as the c°lors and the way in which the displays are used are changed.
. *n measuring the benefits of color cod- ln8> We first determined the number of *ors that a video display terminal °T) operator could handle. A sonar aterfall display, for example, shows the j*unds detected in the water through a of bearings over a period of time. Qe s°unds are converted to spots of light n the CRT screen. Noise produces a ran- 0rn display of lights, but each clear tar- jp Produces a line of lights along the earing at vv^ieh it is situated. Critical Ormation that has been determined °ut each target, such as its course, peed, and bearing, may be printed ound the margins of the display.
We
color
^herent color, and the alphanumeric in- ir^tion about each target would appear 0!”e same color as the target track. The the at°r wou^ have to match the color of dit'traC^ t0 corresponding data. Tra- ,)r'0nally, no more than about seven col- can be used without confusing the j}- rat0r- We studied sets of as many as Seecol°rs. Sometimes the subjects could tj ah the colors in the set, and some- ge(es they had to memorize them. A tar- rWas presented in one of the colors at as"1 °m’ anc* the subject had to match it if hUlCkly as Poss'hle to a color in the set col 6 Cou^d see the set, or identify the the r ^ ke had memorized the set. In ei- With CaSe’ Practiced subjects could deal ai) far more than seven colors without rors ndUe increase in response time or er- tors’ k*®Urc ' shows that while the opera- berresPonse time increased as the num- jnCr °f colors they must deal with (j0e®ased from one to seven, their speed Crnot get much worse with further in- sUb'SeS 'n tke mtmher of colors. It took tvitjWs several hours to memorize a set •Uo ^ colors, but sonarmen have far 'he t exPerience with their displays than 'he 6St a"owed- In short, it appears that tjVeinurnher of colors that can be effec- least^ Pfccessed is surprisingly large—at •j,,. > if properly chosen.
Cqi01s ma'ching method allows different fici 1 Sets t0 evaluated quickly and ef- fUseat|y- If two colors in the set are con- One^r,.11 ’s immediately apparent, and
sets
assumed that if such a display were coded, each target would be made a
0 'he colors can be eliminated. Some are significantly better than others. gr0llat color should the display back- VDT^ * **■ ’s Wpically black, but many gr0 tisers believe that a blue back- Testj makes the displays more legible. deterevealed no differences in target- c l0n against the two backgrounds,
■edings / February 1988
99