The Spinning Woman

I received an email link to an animated spinning woman (new window) in the Daily Telegraph. Supposedly, if you see the woman spinning clockwise you are right-brained and vice versa. For me, she spins in both directions and changes directions seemingly randomly. I can sometimes get her to switch directions, but I don't know how to do it at will. I guess according to this test I am "ambi-brained". I sent the link to some others and a few reported back that she spins one way or the other. There were some people, like me, for whom she switches directions somewhat randomly. Here is the original spinning woman from the Daily Telegraph:
A "deconstruction" (new window) posted on a blog claims that the shadows of her feet are the culprit for the illusion. I didn't buy that explanation — truth be told, I didn't even understand the explanation. So, I wondered if the illusion is the same if the shadows of the feet are cropped out:
Someone was interested in viewing the individual frames. So, here are all thirty-four frames (new window) from the original animation. When she randomly switches directions for me it is quite often when my eyes are not focused on the image and I notice the switch in my peripheral vision. But it seems that when she switches direction, her extended leg is usually fully extended to the right or left. So I had this theory that the direction she appears to be spinning is influenced by the motion of her foot from when it is fully extended to one side or the other. If the animation loop begins with her foot fully extended to the right, she has a tendency to appear (to me) to be spinning clockwise. I created two additional animations to test this theory. In the example below I shifted the animation so that the first frame in the loop has her leg fully extended to the right. I also modified the animation to stop after two revolutions. When this animation begins, she has a tendency to appear (to me) to be spinning clockwise. Click on the image below to open in a separate browser window so you can refresh just the image, not the entire page. When you use F5 to refresh the image, the browser will run the animation loop again.
Conversely, in the example below, I shifted the animation so that the first frame in the loop has her leg fully extended to the left. When this animation begins, she has a tendency to appear (to me) to be spinning counter-clockwise. Again, click on the image to open in a separate browser window and use F5 to restart the animation.
The results seem inconclusive to me. She easily spins in both directions for me with the two tests shown above. However, for me there is a tendency for her to spin clockwise more often with the first test and counter-clockwise with the second test. Lastly, I was curious if there would be any difference if I reversed the sequence of the frames in the animation. Would she tend to spin in the opposite direction. Below on the left is the original animation. On the right is the reversed sequence.
In the two examples above, she still changes directions randomly. There are times when the two animations, side-by-side, are spinning in opposite directions. It seems that the original animation is more often spinning clockwise and the reversed animation is more often spinning counter-clockwise.