[comp.cog-eng] Color in Screen Displays

nib@hound.UUCP (N.BENIMOFF) (12/03/88)

Does anyone know where I can find some good empirical data relating to
the following question:  

Are there colors that should be avoided in constructing a user interface
for presentation on a CRT due to fatigue factors?  I.E., consider the
following scenario.  We have a user who sits in front of a color CRT
looking at text and graphics for the better part of an 8-hour shift.
Are there colors that, more than others, might cause problems of a 
fatigue or perceptual nature?

Thanks for any help.

Nick Benimoff
AT&T Bell Laboratories
att!hound!nib

chrise@bcsaic.UUCP (Chris Esposito) (12/06/88)

In article <2789@hound.UUCP> nib@hound.UUCP (N.BENIMOFF) writes:
>Does anyone know where I can find some good empirical data relating to
>the following question:  
>
>Are there colors that should be avoided in constructing a user interface
>for presentation on a CRT due to fatigue factors?  I.E., consider the
>following scenario.  We have a user who sits in front of a color CRT
>looking at text and graphics for the better part of an 8-hour shift.
>Are there colors that, more than others, might cause problems of a 
>fatigue or perceptual nature?
>
>Thanks for any help.
>
>Nick Benimoff
>AT&T Bell Laboratories
>att!hound!nib

One reference is:

    `ACE: A Color Expert System for User Interface Design'
    Barbara J. Meier
    Comp. Sci. Dept
    Brown University
    Providence, RI 02912

Found in the Proceedings of the ACM SIGGRAPH Symposium on User Interface 
    Software
Banff, Alberta, Canada
October, 1988



-- 
------------------------------------------------------------------------
 "what if we can't tell the difference between AI and artificial stupidity?
Chris Esposito                      | CSNET: chrise@boeing.com
Boeing Advanced Technology Center   | uucp: ...!uw-beaver!ssc-vax!bcsaic!chrise

andrea@hp-sdd.HP.COM (Andrea K. Frankel) (12/06/88)

In article <2789@hound.UUCP> nib@hound.UUCP (N.BENIMOFF) writes:
>Does anyone know where I can find some good empirical data relating to
>the following question:  
>
>Are there colors that should be avoided in constructing a user interface
>for presentation on a CRT due to fatigue factors?  I.E., consider the
>following scenario.  We have a user who sits in front of a color CRT
>looking at text and graphics for the better part of an 8-hour shift.
>Are there colors that, more than others, might cause problems of a 
>fatigue or perceptual nature?

You could check out the experiments on the physiological effects of
colour:  red (especially bright) raises the pulse and blood pressure,
tending to produce short-term alertness (and, after an extended
exposure, increased irritability!).  Yellow and orange tend to make
people do whatever they're doing faster (which is why most fast-food
emporiums decorate in same); blues and greens have a calming effect.
There is a certain shade of pink which has been experimented with in
correctional and mental facilities, which (if memory serves me) has
a strong calming effect at first, but after 30 minutes has a rebound
effect in terms of triggering violence.

Personally, I find that colour fatigue is best avoided by providing
user control of the palette (a la Microsoft Windows or X Windows
interfaces).  Once I find a set of colours that works well for me
(balancing the "focus" problem with distinguishability and fatigue
factors), I can work much longer without fatigue or vision problems.
However, the palette that works for me produces rather interesting
reactions from my coworkers!  For X, I tend to favor soft
lavender-on-black for windows I'm writing in for extended periods, and
a bright cyan on black for windows where I need to quickly spot
details; the cursor, border, and mouse sprite for those take on the third of
that trio.  Miscellaneous details in dark Violet, and an occasional-use
window in white-on-deep-lavender.  Add a black-and-white bitmap
background (currently set to some charming parrots) and some classic
yellow-on-red-plaid postits, and no decorator would touch it with a ten
foot pole.  In MS Windows, I use a black-on-light-cyan window, and set
the borders such that the active window has a red border and the
inactive ones go grey.  Few people can stand working at my terminal or
PC very long, but hey, it works for me!  And the default colour
combinations make me feel ill or sleepy in very short order.

Subversive question du jour:  now why is it that the "plebians" who are
"merely" users of VDT applications aren't normally given access to the
colour palette?


Andrea Frankel, Hewlett-Packard (San Diego Division) (619) 592-4664
                "...I brought you a paddle for your favorite canoe."
______________________________________________________________________________
UUCP     : {hplabs|nosc|hpfcla|ucsd}!hp-sdd!andrea 
Internet : andrea%hp-sdd@hp-sde.sde.hp.com (or @nosc.mil, @ucsd.edu)
CSNET    : andrea%hp-sdd@hplabs.csnet
USnail   : 16399 W. Bernardo Drive, San Diego CA 92127-1899 USA

trejo@nprdc.arpa (Leonard J. Trejo) (12/06/88)

In article <2789@hound.UUCP> nib@hound.UUCP (N.BENIMOFF) writes:
>Does anyone know where I can find some good empirical data relating to
>the following question:  
>
>Are there colors that should be avoided in constructing a user interface
>for presentation on a CRT due to fatigue factors?  I.E., consider the
>following scenario.  We have a user who sits in front of a color CRT
>looking at text and graphics for the better part of an 8-hour shift.
>Are there colors that, more than others, might cause problems of a 
>fatigue or perceptual nature?
>
>Thanks for any help.
>
>Nick Benimoff
>AT&T Bell Laboratories
>att!hound!nib

The question you ask does not yet have a simple answer.  It is the
subject of much current research.  Sometime next year, Plenum will
publish a book, entitled "Color in electronic displays," in which I and
Harry Snyder will have a chapter on measurement of human performance
and David Meister will have a chapter on human performance issues.
A NATO working group (RSG-13) is sponsoring the book.  Until the book
appears, attached are some references that may be helpful to you.
DOD-sponsored technical notes & reports can be obtained from
DTIC (Defense Technical Information Center), Cameron Station,
Alexandria, VA 22304-6145.

Off the cuff, there are two critical factors to be aware of:

1. Do not use color contrasts that primarily produce blue-cone
difference signals (e.g. blue text on yellow background, etc.) for
small symbols.  This will provide poor symbol edge definition, 
reduced legibility, and inadequate cues for accomodation--all of 
which will produce eye strain.  Red- or green cone contrast is 
necessary for small symbols, and the addition of some luminance 
contrast is best.  

2. Use a light background color (e.g. gray) rather than black.  This 
will allow for better perceived levels of colored symbol saturation 
and better color discrimination across varying conditions of
ambient illumination.

			REFERENCES

Boynton, R. M. (1979). Human Color Vision.  New York:  Holt,
                       _____ _____ ______
    Rinehart, & Winston.
Burnette, C. A. (1985) Uses of Color in Complex  Information
                       ____ __ _____ __ _______  ___________
    Displays  (HFOSL  Technical  Note 71-85-04).  San Diego:
    ________
    Navy Personnel Research and Development Center.
Hardesty, G. K. C.  &  Projector,  T.  H.  (1973).  NAVSHIPS
                                                    ________
    Display Illumination Design Guide.  Section I: Introduc-
    _______ ____________ ______ _____   _______ _  _________
    tion to Light and Color.  Annapolis: Naval Ship Research
    ____ __ _____ ___ _____
    and Development Center.
Heglin, H. J. (1973). NAVSHIPS Display  Illumination  Design
                      ________ _______  ____________  ______
    Guide.   Section  II:  Human Factors (Technical Document
    _____    _______  __   _____ _______
    223).  San Diego: Naval Electronics Laboratory.
Meister, D. (1984). Human Engineering Data Base  for  Design
                    _____ ___________ ____ ____  ___  ______
    and Selection of Cathode Ray Tube and Other Display Sys-
    ___ _________ __ _______ ___ ____ ___ _____ _______ ____
    tems (NPRDC Technical Report 84-51).   San  Diego:  Navy
    ____
    Personnel Research and Development Center.
Merrifield,  R.  M.  and  Silverstein,  L.  D.  (1986).  The
                                                         ___
    development and evaluation of color systems for airborne
    ___________ ___ __________ __ _____ _______ ___ ________
    applications:  Fundamental   visual,   perceptual,   and
    ____________   ___________   ______    __________    ___
    display  systems  considerations (NADC Report No. 86011-
    _______  _______  ______________
    60).  Warminster: Naval Air Development Center.
MIL-STD-1472C (1981). Human Engineering Design Criteria  for
                      _____ ___________ ______ ________  ___
    Military Systems, Equipment and Facilities.  Washington,
    ________ _______  _________ ___ __________
    D. C.:  U. S. Department of Defense.
Trejo, L. J., and Lewis, G. W. (1987).  Brain mechanisms for
    human  color  vision.   In  A.  M.  Crawford  and  M. H.
    Metcalfe (Eds.)  FY86  Independent  Research/Independent
                     ____  ___________  ________ ___________
    Exploratory  Development  (NPRDC  Special  Report 87-1),
    ___________  ___________
    January, San Diego: Navy Personnel Research and Develop-
    ment Center.
Trejo, L. J., & Lewis, G. W. (In press).  Sensitivity to hue
    differences  measured  by  visual evoked potentials.  In
    Proceedings    of    the    First    Navy    Independent
    ___________    __    ___    _____    ____    ___________
    Research/Independent  Exploratory Development Symposium.
    ________ ___________  ___________ ___________ _________
    Laurel, MD: Chemical Propulsion Information Agency.
Wagner, D. W. (1977). Color Coding: An annotated  bibliogra-
                      _____ ______  __ _________  __________
    phy (NWC Technical Publication 5922).  China Lake: Naval
    ___  ___ _________ ___________ ____
    Weapons Center.
Wyszecki, G. & Stiles, W. S. (1982). Color Science: Concepts
                                     _____ _______  ________
    and  Methods, Quantitative Data and Formulae.  New York:
    ___  _______  ____________ ____ ___ ________
    John Wiley & Sons.

norman@sdics.ucsd.EDU (Donald A Norman-UCSD Cog Sci Dept) (12/06/88)

     In article <1724@hp-sdd.HP.COM> andrea@hp-sdd.UUCP (Andrea K. Frankel)
     writes:
     ...  red (especially bright) raises the pulse and blood pressure,
     tending to produce short-term alertness (and, after an extended
     exposure, increased irritability!).  Yellow and orange tend to make
     people do  ...

Umm, gulp, the professional scientist in me, coupled with being a
professor, makes me say --- wait a moment!  There is no evidence for
these effects -- no scientific evidence, that is.  And, moreover, it
would be very difficult to get such evidence, because of the very
large numbers of variables that would need to be controlled in such
experiments.  Yes, I know all about the popular literature on color.
But just because it is popular doesn't mean it is correct.

Color clearly has effects on mood and motivation, but the effects are
apt to be subtle, to vary with multiple factors, and to differ across
people (and probaboly differ for the same person at different times).

There ARE standards for appropriate colors to be used: there will be a
panel at the CHI 89 meetings on this topic.   To the writer of the
article who was at HP -- contact wanda Smith at HP who is expert on
these matters (and who will be on the panel).

don norman

Donald A. Norman	[ danorman@ucsd.edu   BITNET: danorman@ucsd ]
Department of Cognitive Science C-015
University of California, San Diego
La Jolla, California 92093 USA

UNIX:  {gatech,rutgers,ucbvax,uunet}!ucsd!danorman
[e-mail paths often fail: please give postal address and all e-mail addresses.]

andrea@hp-sdd.HP.COM (Andrea K. Frankel) (12/07/88)

In article <658@sdics.ucsd.EDU> norman@sdics.UUCP (Donald A Norman-UCSD Cog Sci Dept) writes:
>
>     In article <1724@hp-sdd.HP.COM> andrea@hp-sdd.UUCP (Andrea K. Frankel)
>     writes:
>     ...  red (especially bright) raises the pulse and blood pressure,
>     tending to produce short-term alertness (and, after an extended
>     exposure, increased irritability!).  Yellow and orange tend to make
>     people do  ...
>
>Umm, gulp, the professional scientist in me, coupled with being a
>professor, makes me say --- wait a moment!  There is no evidence for
>these effects -- no scientific evidence, that is.  And, moreover, it
>would be very difficult to get such evidence, because of the very
>large numbers of variables that would need to be controlled in such
>experiments.  Yes, I know all about the popular literature on color.

You may notice (or maybe you didn't), that I said nothing about the
spirituality of purple, the effect of yellow on the third chakra,
assumptions about the psychology of the viewer, or anything in that
vein.

The description of the effects for red is the one that I have seen in
reports from the medical world, and not the popular press.  The most
recent report concerned the time it took for wounds to clot, and showed
a clear effect.  I believe it *is* well documented, but I don't happen to
have the references handy.

>But just because it is popular doesn't mean it is correct.
>
>Color clearly has effects on mood and motivation, but the effects are
>apt to be subtle, to vary with multiple factors, and to differ across
>people (and probaboly differ for the same person at different times).

I agree wholeheartedly.  However, there is evidence for some effects
that do not.

To elaborate on the case of "red", the popular press reports that red =
anger, and/or that red = sexual excitement.  However, in the oriental
cultures, red = joy.  Similarly, in the west we associate white with
purity, cleanliness, holiness; in the oriental cultures, white is
associated with death.  (We dress our brides and altarboys in white;
they reserve it for shrouds.  They dress their brides in red, which
some westerners correlate with ladies of the night.)  In my previous
posting, I was very deliberately NOT getting into the cultural
interpretations, but sticking with the underlying physiological
phenomena - obviously, if red is raising the pulse and blood pressure,
that could be interpreted as either joy or anger depending on the
context.

>There ARE standards for appropriate colors to be used: there will be a
>panel at the CHI 89 meetings on this topic.   

[flame on]
I hope they look past the ends of their own noses.  WHY, oh why, do
people refuse to look into the results obtained in other fields before
concocting their own theories and rules?  Simply waving it all off as
popular press and therefore wrong isn't good enough.  The putative
experts aren't doing a great job, judging from the gawdawful user
interfaces I keep getting subjected to.  If you ever want to get past
the stage of arguing about personal aesthetic preferences, you need to
look at the research that has been done rather than assuming it can't
exist!
[flame off]

And now I'll leave all you "experts" and go back to work...


Andrea Frankel, Hewlett-Packard (San Diego Division) (619) 592-4664
                "...I brought you a paddle for your favorite canoe."
______________________________________________________________________________
UUCP     : {hplabs|nosc|hpfcla|ucsd}!hp-sdd!andrea 
Internet : andrea%hp-sdd@hp-sde.sde.hp.com (or @nosc.mil, @ucsd.edu)
CSNET    : andrea%hp-sdd@hplabs.csnet
USnail   : 16399 W. Bernardo Drive, San Diego CA 92127-1899 USA

croot@lynpac.oz (Admin) (12/09/88)

In article <1724@hp-sdd.HP.COM> andrea@hp-sdd.UUCP (Andrea K. Frankel) writes:
>In article <2789@hound.UUCP> nib@hound.UUCP (N.BENIMOFF) writes:
>>Does anyone know where I can find some good empirical data relating to
>>the following question:  
>>
>>Are there colors that should be avoided in constructing a user interface
>>for presentation on a CRT due to fatigue factors?

From some work long ago, all I CAN remember is that you should never use
red-on-blue or vice-versa (eg. red text on blue background). Apparently
the eye adjusts its focus slightly for some colours, red and blue being
the worst two (This sounds like crap - and probably is - but the original
assertion is true), therefore the eye *cannot* focus properly since one of
the red/blue foreground/background will be slightly out of focus. I know
that on my terminal, I find it very hard to read. Matt.

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"life? don't talk to me about life!"          Matt Atterbury. :-) :-( :-) :-(
"klaatu nikto barada"     Lynwood Pacific, Sydney, Australia. :-( :-) :-( :-)
matt@lynpac.oz.au (Europe/Internet) || uunet!munnari!lynpac.oz.au!matt (UUCP)
-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"life? don't talk to me about life!"          Matt Atterbury. :-) :-( :-) :-(
"klaatu nikto barada"     Lynwood Pacific, Sydney, Australia. :-( :-) :-( :-)
matt@lynpac.oz.au (Europe/Internet) || uunet!munnari!lynpac.oz.au!matt (UUCP)

bph@buengc.BU.EDU (Blair P. Houghton) (12/14/88)

In article <234@lynpac.oz> croot@lynpac.oz (0000-Admin) writes:
>In article <1724@hp-sdd.HP.COM> andrea@hp-sdd.UUCP (Andrea K. Frankel) writes:
>>In article <2789@hound.UUCP> nib@hound.UUCP (N.BENIMOFF) writes:
>
>From some work long ago, all I CAN remember is that you should never use
>red-on-blue or vice-versa (eg. red text on blue background). Apparently
>the eye adjusts its focus slightly for some colours, red and blue being
>the worst two (This sounds like crap - and probably is - but the original
>assertion is true), therefore the eye *cannot* focus properly since one of
>the red/blue foreground/background will be slightly out of focus. I know
>that on my terminal, I find it very hard to read. Matt.

It's physics, and due to the achromaticity of the lens.  Red and blue,
being at opposite ends of the spectrum (ROYGBIV and all that) refract
to very different angles in the eye.

This explains the fact that they are focussed differently.  I say this
because many such things (e.g. misdetermination of intensity due to
differences in contrasting surroundings) are a result of retinal processing
and Higher Neural Function.  This one just happens to have a purely
physical basis.  (I'd expect that the visual system was capable
of eliminating the problem through some sort of processing, but it
doesn't.)

The really interesting thing is that they appear (it's a Famous Optical
Illusion) to be at different distances from the eye, when clearly they
are not, considering that it is very difficult to determine distance
from focus alone, and parallax is nonexistent if the two colors are in
the same plane.  Perhaps that slight misfocus is perceived as a separate
object.  Perhaps not.  It's easy to get fooled by these things.

				--Blair
				  "White on DarkOliveGreen."

u-adomma%ug.utah.edu@wasatch.UUCP (Alex Dommasch) (12/14/88)

In article <234@lynpac.oz> croot@lynpac.oz (0000-Admin) writes:
>
>From some work long ago, all I CAN remember is that you should never use
>red-on-blue or vice-versa (eg. red text on blue background). Apparently
>the eye adjusts its focus slightly for some colours, red and blue being
>the worst two (This sounds like crap - and probably is - but the original
>assertion is true), therefore the eye *cannot* focus properly since one of
>the red/blue foreground/background will be slightly out of focus. I know
>that on my terminal, I find it very hard to read. Matt.
>

I recently read an article in Scientific American where they discussed
just this phenomenon.  It turns out that the boundary recognition cells in
the retina (or whereever) only operate on luminance information, not color
information.  So with red on blue, or any other pair of colors with similar
luminances, the eye really sees no boundary at all, and the characters
wander all over the place, or don't form at all.

Alex.

================
Alex Dommasch		u-adomma@ug.utah.edu
Salt Lake City, Utah