Human factors - lighting Visual distress in the office environment Arnold Wilkins zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA More than half the headaches and eye-strain suffered by office workers can be avoided by the appropriate choice of office environment. This perhaps surprising conclusion can be drawn from the results of a recent study undertaken jointly by the Medical Research Council,theBuilding Research Establishment and Thorn Lighting. The study compared two types of fluorescent lighting: one was conventional, exhibiting the 40 per cent fluctuations in intensity that occur 100 times/second; the second was a new, steady form of fluorescent lighting. The two forms of lighting were of identical appearance and differed only in the circuitry hidden in the lamp housing. Volunteers from a large government legal office were asked to keep weekly diaries of their headaches and eye-strain for a year. Half way through the winter months the lighting in some offices was swapped over without the occupants being aware of the change. The incidence of headaches and eye-strain among those exposed to the steady lighting was less than half that among those exposed to the conventional fluctuating light. Other characteristics of the lighting, such as the colour of the tubes and the speed with which they ignited did not seem to be important. Since the lighting was allocated at random and participants did not know what sort of lighting they were exposed to, it is likely that it was the lighting rather than any other uncontrolled factor that affected the incidence of headaches. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Unnatural brain excitation The difference between the groups exposed to the two types of fluorescent lighting suggests that light from conventional fluorescent fittings can cause headaches and eye-strain and that these complaints arise from the imperceptible fluctuations in light output. Although this fluctuation is too rapid to be seen, it affects the firing of cells in the brain. Nerve cells fire in response to each flash from a fluorescent tube and this unnatural form of nervous excitation may be the cause of the problem. It should be noted that the fluctuation is due to the circuitry controlling the tubes and not to the tubes themselves. Computer displays The use of computer display terminals has been associated with complaints of headaches and eye-strain. The study may help to explain why, because it suggests that fluctuating light can provoke headaches even when the fluctuation is too rapid to be seen. Computer displays typically use cathode ray tubes and the picture is created by a flying spot that scans the screen, illuminating it intermittently. The fluctuation in brightness is greater than that from fluorescent lighting and it is also slower. It may, therefore, have a greater effect. Possible solutions Several practical steps can be taken to reduce the incidence of headaches and eye-strain. One obvious solution is to change the circuitry controlling thefluorescentlighting. The new form of circuitry known as high-frequency solid-state ballast is cheaper to run - it consumes 40 per cent less power and emits only very slightly less light - and can be obtained from major lighting companies. One complication is that because people prefer this form of lighting they tend to leave it switched on for longer. Participants in the lighting study were able to switch the lights on from wall switches and from pull cords attached to the luminaires. Clocks inserted in a sample of luminaires showed that the new form of lighting was switched on for 30 per cent longer on average. In the short term it may not be possible to change the lighting circuitry, but some help can nevertheless be given to those individuals who suffer frequent headaches and eye-strain. These people are more likely to be young women, but all categories of staff may be affected. The study showed that seating people near a window where daylight can dilute thefluctuationcan reduce the incidence of headaches. It was also found that headaches and eye-strain decreased with the height of the office above the ground: the higher the office the more daylight it received, because the windows overlooked buildings of similar height. The increase in daylight with office height may, therefore, have been responsible for the decrease in the incidence of headaches. If daylight is not available, then the fluorescent tubes above the individual's place of work can be removed and tungsten- halogen uplighters installed. These give a steady natural light and are sufficiently bright to make up for any loss of light occasioned by the removal offluorescenttubes. Workers complaining of headaches and eye-strain from VDU use may benefit from the use of a liquid-crystal display (LCD). These displays provide a rather low-contrast image but differ from conventional displays in that they do not flicker appreciably (unless they are back lit). PC-compatible LCD displays with the resolution of the IBM colour graphics display (200 x 640 pixels) are now available as add-ons. Problems from patterns So far we have considered only the problems posed by fluctuating light. But steady light, if it is patterned, can also induce headaches and eye-strain and even epileptic seizures. As an extreme example, look at the pattern of stripes in Figure 1. Many people will find this pattern objectionable to look at and it will provoke anomalous visual effects including illusions of colour, shape and motion. The patterns that can induce discomfort and anomalous visual effects have very specific properties. The worst patterns are stripes, and the worst stripes are those that have even width and spacing. As the width and spacing become uneven the effects decrease, althought the width of the stripes needs to be less than about 5 per cent of the distance between them before the effects can be ignored. The greater the difference in brightness between neighbouring stripes the worse their effects, although even faint stripes can be aversive. The effects are negligible when the contrast (defined as the difference in luminance divided by the sum) is less than 0.1, but they get progressively worse as the contrast approaches 1.0. Whether stripes with the appropriate shape and contrast will cause a problem depends on their size and the distance from Facilities Vol 6 / No 4 / April 1988 9