P3D RE: P3D The double-nail illusion

[jeremy.hinton@xxxxxxxxx]

Thanks to Abram Klooswyk, p3d 3784, 11 Mar 2000, following Gabriel 
Jacob, p3d 3773, 07 Mar 2000.

I find the 'montage' a very powerful demonstration. I am not certain 
that what I can see in the figure is quite what Abram is describing. 
Fusing Abrams figure (below), I see the ambiguous dot/s (the lowest of 
the three) aligned in depth _whichever_ of the _other_ (upper and 
middle) dots I am fixating. Only if my convergence drifts even 
slightly, then the lower dots revert to the window depth as shown in 
Abrams Figure B.  If you fixate alternately on the upper dot, middle 
dot and window depth, then you might see what I mean - level 3 (lower 
dot/s) suddenly turns from a pair of dots at window depth into a single 
dot at the fixation depth, with monocular 'flankers' which appear to be 
at the depth of the non-fixated dot.  

Is that the percept you are talking about?


> Because I ran out of virtual mounts, I have put 
> them together in one mount: 
  ^I love that^ !


Abrams version:
>               _____________________ _____________________ 
>              |                     |                     |
> 1            |         o           |           o         | 
> 2            |           o         |         o           |
> 3            |         o o         |         o o         |
>              |_____________________|_____________________|
> 

Because of the relatively large disparity in this figure, I am unable 
to fuse any two levels (window, front and back) simultaneously. So I 
have modified the figure to reduce the disparity:

Reduced disparity version (front and back dots):
>               _____________________ _____________________ 
>              |                     |                     |
> 1            |          o          |           o         | 
> 2            |          o          |         o           |
> 3            |         o o         |         o o         |
>              |_____________________|_____________________|
I can't get an alternative percept from this one. (well not easily 
anyway)

 Reduced disparity version (all 3):
>               _____________________ _____________________ 
>              |                     |                     |
> 1            |         o           |          o          | 
> 2            |          o          |         o           |
> 3            |         oo          |         oo          |
>              |_____________________|_____________________|

I get the same effect as for Abram's original figure when fixating on 
the near dot.
But I cannot duplicate it for the rear dot. I think that this is also 
interesting as it suggests that the occlusion cue overrides a possible 
stereo percept in the periphery... 

> 

> An interesting real objects-demonstration does the same as the 
> stereograms above, it is the double-nail illusion, first 
> published by Jodi D Krol and Wim A van de Grind, Perception 
> 1980 (9), 651-669. After that paper many variations were tried, 
> leading to a Phd thesis by Jodi Krol in 1982 (I was present 
> when he defended this thesis at the University of Amsterdam).
I will try to find these at lunchtime. Meantime thanks for getting me 
thinking!


Jeremy 

(PS I am using wall-eye viewing, but I don't see why cross-eye 
shouldn't work just as well. You'll just see a different dot as being 
in front of or behind the window.