Bill, I'm intrigued with what has been done on this. Has anyone ventured
to guess at what the bandwidth looks like for tactile stimulation? And,
how does this bandwidth get reduced as sensors are more closely spaced.
If you knew the answers to this, even a ballpark idea, you could easily
show that the technique is limited to simple figures. I'm curious to
know what the numbers are.
On Thu, 17 Aug 1995, Bill Gerrey wrote:
>
> Yeah Robert!
> As one of the original researchers on "the Tactile Vision Substitution System" (TVSS) studied at Smith-Kettlewell, you have it ever-so mostly right.
>
> The argument over stereo presentation is the weakest one; it is possible, given "size constancy" and other monoccular cues, to ferrit out distance information and so forth.
>
> Where such a system really falls down is the loss of internal detail as soon as adjacent nerves are stimulated. Our studies found that turning off every other stimulator in an array made no difference in performance in object-recognition tasks.
>
> Observation showed that the best subjects, Bill Gerrey being the star, identifies objects most successfully by keeping most of the image out of the field, at
> attending to detail on the edges simplified the task.
>
> The moral of the story was and is: display to the skin must be simple, not complex. High resolution won't make a difference. Research on this goes back to the early 1960's. Let us not schmoose with those who haven't done their homework.
>
> Bill Gerrey
>
This archive was generated by hypermail 2b29 : Sun Dec 02 2012 - 01:30:03 PST