[vslist] Re: stereoanomaly test

[Raymond van Ee]

Hi Alex,
In answer to your questions below, attached is Vision Shell c-code 
that is based on the code I used to test subjects for stereoanomaly.

A key feature is that the stimuli need to be flashed so that 
observers can not make eye movements. Therefore your Julesz 
stereogram is unnecessarily more complicated than a stimulus 
consisting of just two disparate vertical bars.

The attached code (and a ready to run project version) is quickly 
hacked just to give you an idea of the test. I realize that the code 
is not optimized and, anyway, I am not a good programmer. Whenever 
you think that the code might serve your purposes, feel free to ask 
for more information. The (identical) test and the data provided by 
the test are described in a recent publication (which is also 
attached) [R. van Ee & W. Richards (2002). A planar and a volumetric 
test for stereoanomaly. Perception, 31, 51-64].

If you plan to conduct an experiment in which the stimulus 
presentation duration is long and in which subjects are free to make 
eye movements, then don't worry about stereoanomaly. In such an 
experiment subjects will adopt eye movement strategies that overcome 
their stereoanomaly. In experiments in which subjects are not able to 
overcome the stereoanomaly handicap the anomaly is indeed a concern 
as I have now shown striking correlation between the outcomes of the 
stereoanomaly test and a variety of tasks that require stereo vision. 
(Just like there is a correlation between coloranomaly tests and 
results in color tasks).

kind regards,
Raymond van Ee
----------------------------------------
Asst. Professor, Physics
Utrecht University
Helmholtz Institute
PrincetonPlein 5
3584 CC Utrecht
The Netherlands
http://www.phys.uu.nl/~vanee


>Has anyone programmed a simple visionshell program to screen people to
>ensure that they are not stereoblind?  I envision a program which
>would perhaps present two halves of Julesz stereograms, one on each side
>of the screen (I am using a mirror stereoscope)
>
>I recently read a paper claiming that 30% of undergraduates are at least
>partially stereoblind, so I feel the need to screen people even for my
>very simplistic experiment..
>
>thanks
>--
>   Alex Holcombe
>
>` aholcombe at psy.ucsd.edu
>   H:(619)861-6565 W:(858)534-3924
>   http://psy.ucsd.edu/~aholcombe
>  `--------------------------------- - -+
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: VanEeStereoAnomalyTest
Type: application/mac-binhex40
Size: 349537 bytes
Desc: not available
Url : http://visionscience.com/pipermail/visionlist/attachments/20020524/157dde0b/VanEeStereoAnomalyTest.hqx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SmallScreenPlanarTest.c
Type: application/mac-binhex40
Size: 20024 bytes
Desc: not available
Url : http://visionscience.com/pipermail/visionlist/attachments/20020524/157dde0b/SmallScreenPlanarTest.hqx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vanEe&Richards.pdf
Type: application/pdf
Size: 220743 bytes
Desc: not available
Url : http://visionscience.com/pipermail/visionlist/attachments/20020524/157dde0b/vanEeRichards.pdf