Dear Colleagues,

Some of you have been wondering what was the upshot of my recent query about
looking at computer screens in near darkness. I have taken a little while to obtain permission to quote you but most people were happy to permit this.By way of thanking you all publicly here is the question and a summary of the best replies;

The Question:

Dear all,
I recently had an inquiry from a computer firm where some of the staff
apparently prefer to work in rooms that are so dark they can only just see
the keys on the keyboard. I was asked if this could cause eyestrain. I did a
quick search and found no evidence one way or the other. I suspect that
slowness of dark accommodation coupled with the need to see other things as
well as the screen might be the reason for any eyestrain. I remember half a
century ago  (in the fifties) when television was still a novelty in England
(and I was still a child) I was told not to watch the TV in a dark room as
it "draw my eyes", that is to say, tire them needlessly. Is there a simple
answer to this question?
Regards, David McFarlane.

The Answers:

The most helpful answer I received was from Owen Evans. He pointed me to the reference in Grandjean. Figure 161 on page 272 shows that a background 5 times darker than the centre of the visual field reduces contrast sensitivity to 77% of the value found in a no-contrast situation. This is based on research by S. Guth,  (1958),  "Light and comfort", Industrial Medicine and Surgery, 27, pages 570 -574.

Owen commented that contrast sensitivity is a measure of our ability to distinguish small objects with varying contrast between the object and the background is a topic that is well-covered by Salvendy "Handbook of Human Factors and Ergonomics".
He also provided a "Rule of thumb": if it is screen intensive work, buy the biggest and sharpest screen that you can afford, and update them frequently.

        *       Thanks, Owen! I had always thought - incorrectly - that this contrast problem only arises with glare problems where the brightest light was peripheral to main visual task.

Jenny Long said;

The first question you should be asking is "what is the polarity of the screen?". If the workers are using a light coloured screen (e.g. black characters on a light background) and they are working in the dark, then there is a large amount of contrast between what they are viewing and their surrounds. This can cause a problem with dark adaptation (especially seeing reference material or their keyboard after viewing the screen) which could result in eyestrain symptoms. It also means more work for the muscles in the iris controlling the pupil size, as the pupil would need to change size frequently to cope with the bright screen and the dark surrounds. (By the way, this is the basis for the instruction you received as a child, David, about not watching the TV in the dark.)
If, however, these workers are using a dark screen with light coloured characters, then this might explain why they prefer to work in the dark. As a rule of thumb, it is best to keep the ambient illumination levels similar to the brightness of the screen.



 One person commented that;

If the screens are a bit old, or the users are trying to use them at maximum resolution, there may well be a focus issue - as you turn up the brightness, the pixels defocus a bit, sometimes a lot. If the screen is old, the phosphor may not be as efficient as it once was, requiring higher brightness settings than usual. In order to get the sharpest image that they can to work with, the users may be turning the intensity down, and then need lower ambient light levels in order to minimise reflections and achieve adequate contrast.

Terry Simpson commented;

"I have had some experience with this scenario. Users sometimes say that it they prefer it darker. Often this is a dislike of the current (poorly specified and maintained) lighting rather than a preference for low light. It also sometimes indicates poor display equipment. Low light levels will open the iris and put a greater demand on visual correction. This might result in symptoms of eyestrain."

Denny Ankrum commented;

"Theory and the research would suggest that they would have a problem. The trouble with that is that the research, etc., does not conform to what happens. Hmmm. I guess reality must be wrong. I've run into situations similar to what you describe. My advice would be to let them continue. Suggest that if they do feel discomfort that they might want to raise the ambient light levels a bit. In any event, they are not going to do permanent damage to their eyes."


Tom Mayfield commented;

"Reading by torch light under the sheets was always my mother's cue for saying I'd harm my eyesight.  I also remember the old TV's and that observation.  John is right about reducing contrast generally for ease of readability and I seem to remember some info in the old VDT Manual.

There is the issue of your peripheral vision being more sensitive in darkeness, so there could be a constant switching of attention to peripheral movement (if there is any) and the consequent shutting and opening of the iris as the eye moves from the lighted screen to the dark room.  This would also happen if the user is a "hunt and peck" typist.

It is an interesting question, as I have tried working with my lap top in the dark, and the constant accommodation changes that Peter Howarth mentioned become unbearable for me after a very short time (I'm a "hunt and  peck" typist!).  It would be interesting to look at your programmers' error rates compared to working in lit conditions.

Peter Howarth commented;

"Normally one thinks of 'eyestrain' as being caused by either the accommodation sytem or the oculomotor system.  The symptoms of presbyopia, for example, are brought about by the increase in the proportion of ones amplitude of accommodation needed to focus a near object.

This increase occurs when the total amount of accommodation available declines (with age). Our bodies are often pretty lazy, and the eye uses just enough accommodation to bring an object into 'good enough' focus, rather than precise focus.  When focus is not precise, a point source will be imaged on the retina as a 'blur circle' .

The circle has the shape of the eye's pupil, and its dimensions are determined by 1) the amount of defocus and 2)the pupil size.  In a bright environment the pupil size decreases - which is why uncorrected presbyopes can perform near tasks so much better under higher light levels.  In a dark environment the pupil will enlarge, and so for a given amount of blur the amount of accommodation needed will be greater - with the potential for more discomfort.  The closer the visual tasks, the greater will be the problem. If the workers prefer to work under near-dark conditions, then there is something wrong with the lamps.

Many thanks for all your contributions!
Regards, David.