UNSW Perception Project
The first version of the UNSW perception project is complete! It took about 50 hours with all the fiddly little changes, I am now working on a second version. The value of the experience I got from this project is immeasurable. Let me break it down for you:
I was working for Dr. Tom Beesley, making a psychological experiment to test perception and memory. For a detailed explanation about what was being tested, written by someone who actually knows what they're talking about, click here to go to Dr. Tom Beesley's website.
The program is an Oculus Rift-enabled room with several variously-coloured shapes floating in the subjects's field of vision, and it looks like this, but now with fewer objects:
Pretty colours oooooh
The idea is that the test subject needs to locate, amongst all those cylinders, the one differently-shaped object, a capsule (A cylinder but with rounded ends). They look around for the object with the Oculus Rift, and once they locate it, center it in their vision and hit the space bar. This is repeated and repeated. Unbeknownst to the subject, the same patterns keep popping up again every now and then. The test is to see if people subconsciously notice and recognise that repetition, and therefore find the correct object faster. The program records all sorts of data about this process, such as time taken for each object, number of missed tries, head tracking data, etc. Each pattern is randomly generated at run-time, within several constrictions; no correct answers way off at the edges, none smack bang in the middle, not too many or too few of one colour, all sorts of stuff.
Once that phase of the test is complete, the subject runs a different, similar test where the capsules are invisible entirely, and the subject must guess where the capsule should be. If the experiment is successful, they'll subconsciously remember the repeated patterns and accurately(ish) guess where the capsule was meant to be.
You can download a short version of the experiment here.
If you'd like the full version of the experiment program, please contact Dr. Beesley here.