[LINK] optical illusion not a psych test

[Robin Whittle]

Stephen Loosely wrote:

http://www.news.com.au/common/imagedata/0,,5675247,00.gif

    wait till this loads .. then ..

Is your dancer turning clockwise .. or anticlockwise .. look again
..   .. and, reported here as an optical illusion ..

http://scienceline.org/2007/10/29/ask-hsu-spinning-girl-right-left-brain-hemispheres/


This animation:

  http://www.news.com.au/common/imagedata/0,,5675247,00.gif

is contradictory.  When the outstretched foot goes right to left, it
is lower in the image than when it goes left to right.  This
unambiguously shows the rotation is clockwise, looking from the top.

However, the shadow only appears for the right to left motion, which
would only be the case if the rotation was anticlockwise.

There are modified images at:

  http://ofb.net/~whuang/imgs/spin/

as pointed to by:

http://scienceline.org/2007/10/29/ask-hsu-spinning-girl-right-left-brain-hemispheres/

with added outlines which give clear indication of both clockwise
and anticlockwise, but the contradiction between the shadow and the
foot height remains.

I don't believe a contradictory image such as this would be much use
as a psych test because the final perception probably depends on
which of the two cues the viewer perceives first.

There is an article on optical illusions:

   http://www.sciam.com/article.cfm?id=the-neuroscience-of-illusion

Image 3 in the slideshow:

  http://www.sciam.com/slideshow.cfm?id=the-neuroscience-of-illusion

is an extraordinary demonstration of the brain adjusting perceived
brightness in shadow situations.  This is the Checker Shadow
Illusion by Edward H. Adelson:

  http://web.mit.edu/persci/people/adelson/index.html

The raw image is here - large and medium sizes:

http://web.mit.edu/persci/people/adelson/images/checkershadow/checkershadow_illusion4full.jpg
http://web.mit.edu/persci/people/adelson/images/checkershadow/checkershadow_illusion4med.jpg

The background of squares A and B are actually the same brightness,
but it sure doesn't appear this way to my brain!

My father Jeff has a simple proof - cut out two holes in some opaque
cardboard to obscure the entire image apart from squares A and B.

My approach to proving it was to edit the image and drag a patch of
one square over the other.

Edward H. Adelson's proof is here:

http://web.mit.edu/persci/people/adelson/images/checkershadow/checkershadow_proof4full.jpg
http://web.mit.edu/persci/people/adelson/images/checkershadow/checkershadow_proof4med.jpg

but I think my Dad's approach is most satisfying.

There is a whole slew of illusions which generate perceived motion
in our peripheral vision, from Prof. Akiyoshi Kitaoka:

  http://www.ritsumei.ac.jp/~akitaoka/index-e.html

  http://www.ritsumei.ac.jp/~akitaoka/rotrays.gif

  http://www.psy.ritsumei.ac.jp/~akitaoka/saishin31.html
  http://www.psy.ritsumei.ac.jp/~akitaoka/rotearths2.jpg
  http://www.psy.ritsumei.ac.jp/~akitaoka/rotsakura2007.jpg
  http://www.psy.ritsumei.ac.jp/~akitaoka/rotladybugsnight.jpg
  http://www.psy.ritsumei.ac.jp/~akitaoka/murasakiimojpg.jpg
  http://www.psy.ritsumei.ac.jp/~akitaoka/exp4typhoonss.jpg
  http://www.psy.ritsumei.ac.jp/~akitaoka/expsnakes.jpg

I had fun playing with these images with Photoshop's Image > Adjust
> Hue/Saturation > Hue.  Also Image > Adjust > Replace Colour.  This
can be used to turn the yellow-green into a deep red, which seems to
stop the movement.  Colour is not vital to the process.  Turning
these images into greyscale still produces the movement for me, but
inverting the greyscale (Control I in Photoshop) stops it.

Also, get your rulers out to check the twisted and bulging grids in
these:

  http://www.ritsumei.ac.jp/~akitaoka/SwampA2.jpg
  http://www.ritsumei.ac.jp/~akitaoka/Bulge02.jpg

 - Robin