The Pulfrich Illusion

More Java applets here.
If your browser supports Java, you should see an animated image above (generated by a Java applet embedded on the page). Otherwise, you are wasting your time here.

The animated image is simply a white ball moving back and forth sinusoidally about 45 cycles per minute (four seconds for 3 cycles). On my monitor, the horizontal range of motion is about 2.6 inches (6.5 cm).

Viewing the Illusion
View using a dark filter over one eye (and no filter over the other eye). Best results are achieved using a very dark filter. This generally necessitates using several lighter filters in combination. Filter suggestions: I get good results by looking through the blue filters of 3 pairs of 3D glasses.

An ideal filter can be made from the two "lenses" of a pair of polaroid sunglasses. This is the best option because you can vary the darkness by rotating one of the "lenses" relative to the other. I broke apart a pair of clip-on polaroid sunglasses. If you have two pairs, you can use them together without having to break them.

Note submitted April 10, 2002 by Andi Karrer:

With an flat panel LCD or TFT display, this is even simpler. You need only one polaroid sunglass lens because the display is already polarized. Just hold one "lens" of a pair of polaroid sunglasses in front of one eye and rotate until the ball becomes appropriately dark.

The following (static) image illustrates a degree of darkening that works well for me:

The effect seems to increase as you view it longer. Moving back away from the screen may also be helpful.

Free Pulfrich Glasses
Available from Rainbow Symphony
Ask for the "TV glasses".
What You Should See
When viewed with a dark filter in front of one eye, the animated ball at the top of the page will appear to be moving in and out of the screen as it cycles back and forth. Instead of moving back and forth in a straight line, the ball appears to be travelling in an elliptical path (a flattened circle) in a horizontal plane. The direction of apparent motion around the ellipse depends on which eye has the filter. If the filter is on the right eye, the rotation will seem to be counter-clockwise (from an overhead sense). If the filter is on the left eye, the rotation will seem to be clockwise. Increasing the darkness of the filter increases the effect (up to the point where the filter is so dark that one eye can no longer see the image).

For me, the ball appears to move in and out by a distance of about .75 inch (2 cm) from the screen.

Your perception of the darker image (viewed by the eye with the filter) is delayed by several hundredths of a second relative to the brighter image. This causes the perceived position of the darker image to lag behind that of the brighter image. The amount of lag varys with the speed of the ball. When your brain integrates the two images, which are perceived to be at two different horizontal positions, the result is a stereoscopic 3D effect.

With the filter over your left eye and the ball moving to the right, the two lines of sight converge at a point behind the screen. When the ball is moving to the left, the two lines of sight converge at a point in front of the screen.

Another Illusion: Pulfrich Television Snow
Now that you have gone to the trouble of acquiring a dark filter, try viewing "television snow" with it. Tune a TV set to an unused channel, or disconnect the video input, so that the television screen shows a constantly changing pattern of random dots ("snow"). When viewed with a dark filter over one eye, the snow pattern will appear to break up into two sheets at different depths, one sheet moving to the left and the other moving to the right.
The "Pulfrich Effect" was first explained in 1922 by the German physicist Carl Pulfrich (who could not see it, due to blindness in one eye). I learned about it from the "Amateur Scientist" article by Jearl Walker in the March 1978 issue of Scientific American magazine (Walker01). The "Television Snow" illusion was discussed in a Scientific American article: "The Resources of Binocular Perception" (Ross01). Various "Television Snow" Illusions were discussed by Jearl Walker in the April and May 1980 "Amateur Scientist" articles (Walker02) (Walker03).

Copyright (c) 1997-2014 by Mark Newbold, all rights reserved.
<Dogfeathers Home Page>   <Mark's Home Page>   <Mark's Java Stuff>
Email: Mark Newbold
This page URL: