Condign volumes 1 to 3

Read from the scan by GLM-OCR; expect the odd misread word.

# UNCLASSIFIED

UK RESTRICTED

AFTER-IMAGES AS A RESULT OF FLASHES OF LIGHT

## BACKGROUND

1. Illusions observed in darkness or low light conditions, after exposure of one or more eyes to a bright light are known as 'positive after-images' and are often similar in colour to the inducing light. Those illusions seen in moderate illumination are called 'negative after images' and are often of the approximation to complementary colours. In practice the actual appearance of after-images is complex and likely to depend on many factors. However, adaptation from dark to light takes much less time than from light to dark - hence there is an awareness of the appearance of a light in dark conditions much more quickly than the appearance of a shadow in lighter conditions. Excessive eye stimulation, as is well known, produces blindness in the limit.

2. Within the eye the sequence of changes which occur in reception after light is absorbed differs between rods and cones, classes of cones, and suffers further adaptation by the nerve cells of the retina and of the brain. However, the result is the persistence of the sensation of light after the stimulus has been removed.

3. Primary Sensation Human eye response is from $ \lambda=0. 4 $ to $ \lambda=0. 7 \mu \mathrm{m} $ with an unaided peak at $ 0. 5 5 \mu \mathrm{m} $ ; and a detectability from $ 1 0 ^ {- 6} $ cd.m $ ^{-2} $ to 5000 cd.m $ ^{-2} $ . There is a range of conditions over which the primary sensation produced by a flash of light depends on its luminance-time product (i.e. in general, the integration of luminance with respect to time). This is independent of temporal distribution. This was proved experimentally by regularly repeated flashes as long as 150 years ago and confirmed this century using extremely short flash durations now possible. The practice holds for flash values down to $ \sim 4 \times 1 0 ^ {- 7} \mathrm {s e c} $ and is considered valid for high intensities. For longer duration flashes the measurements are less researched. Hence, the integration of flashes is less reliable as a guide

at an illumination intensity of $ 0. 5 \mathrm{c d}. \mathrm{m}^{-2} $ at which pulse durations of 27 millisec may be the eye integration limit $ (3^{\circ} $ fovea fov). The time duration may be greater than 27 millisec for dimmer flashes and shorter in period for those which are brighter.

4. Results of Experiments Preliminary experiments [1] to investigate after-image conditions for eye stimuli of different lengths produced the following results:

- Stimuli ~2 sec maximum. The whole course of the (positive or negative) after image, excluding its first 15 sec, was dependent on total light in the stimulus and not on the light's intensity distribution in time.

- Stimuli 2-5 sec. The result only differed slightly from the 2 sec stimuli length above, but with some time distribution differences. In particular, there were some differences between positive (dark background) and negative (bright background) after-images.

- In the after-image observation time t=10 to t=30 it is believed that the retinal illumination decays exponentially (from white light stimuli), but the after-images caused by other colours may decay at other rates. The actual detection threshold is inversely proportional to flash duration.

5. The human eye can nevertheless distinguish between extremely short spaces between flashes (pulses of light). Experiments have shown that 4 millisec intervals can be distinguished from intervals as little as 0.28 millisec. Hence, in the context of UAP observations, pulsed (i.e. modulated) lights should be distinguishable by observers, even with very short durations - and should not get mis-reported as steady lights. This is