A new article in Nature describes current research into
motion-induced blindness. MIB is believed to
occur because the right hemisphere of our brain appears to see the world
exactly as our raw sensory data shows it. But the left hemisphere tries
to suppress sensory information that conflicts with our internal model of
how the world should appear. The Nature article includes a simple
animation, called Bonneh's Illusion, that demonstrates the effect. A
more detailed discussion and
further examples of the phenomenon can be found on Yoram Bonneh's MIB
page and Jack
Pettigrew's web page. A better understanding of MIB and right-left
brain interactions could have applications to binocular or stereoscopic
vision systems in robots.
I don't see why interactions between raw and interpreted sensor data or
MIB are more related to binocular vision systems rather than to
monocular ones. The MIB is observed even with a single eye, I bet.
Yep, I just tried looking at the Bonneh's Illusion animation with one
eye shut and still experience MIB. Oddly, MID occurs (for me at least)
very frequently using only my right eye but almost not at all using only
my left eye. I wonder if that's related to a right/left brain thing or
just because of differences in my eyes.
Well, although I wasn't able to reproduce this difference of MIB for
left and right eye (it turned out to be difficult for me to observe MIB
at all), one could speculate about the nature of the disbalance. Since
the visual cortex corresponding to the right eye is in the left
hemisphere, which is incidentally blamed to be the one responsible for
supressing inconsistent data, there could be better chances to supress
the signal originated from the left visual cortex before it gets
propagated anywhere else.