[redsniper]

http://www.youtube.com/watch?v=piyY-UtyDZw&feature=player_embedded

Sensors in cat's brain show what the cat is seeing on a computer. Seriously. I just.. I don't even.. my mind...

[Flipside]

Amazing stuff, though the comment at the end that what the cat saw looked a bit 'catty' was more a product of human phsychology than cat

[PsychoLandlord]

Well ****. This should have some interesting applications in the future.

[Mikes]

Well ****. This should have some interesting applications in the future.

Anyone else thinking... Spy Cat!!!!?

[achtung]

Your pet is actually a security camera.

[Rodo]

that's creppy.

[Mars]

I do wonder what a human produced image would look like, with the amount of dedicated face recognition we have.

[watsisname]

http://news.bbc.co.uk/2/hi/science/nature/471786.stm
"Monday, October 11, 1999 Published at 19:10 GMT 20:10 UK"

:I

(That is really cool though.  Wonder why there isn't anything more recent about it?)

[Kosh]

http://www.youtube.com/watch?v=piyY-UtyDZw&feature=player_embedded

Sensors in cat's brain show what the cat is seeing on a computer. Seriously. I just.. I don't even.. my mind...

This is old news. Watch the first part of this, it's on google video. Made back in 2006.

[Ghostavo]

I do wonder what a non-human produced image would look like, with the amount of dedicated face recognition we have.

Fixed, since our condition limits us.

This makes me wonder how much of what we perceive actually is that way.

[Mars]

I have no interest in a non-human image. I am incapable of processing any other way.

[headdie]

I have no interest in a non-human image. I am incapable of processing any other way.

Being able to process a human's visual data would have it's uses, take training as an example, with this kind of technology a member of training staff can run through a training exercise and trainees can see where they should be focusing their attention, likewise turn it around and fit the equipment to a student they can then run through the exercise and the trainer/examiner can see exactly what the student sees and where their attention is focused. 

Also how about being able to drive your PC purely by being able to visually focus on an icon?

[Herra Tohtori]

I have no interest in a non-human image. I am incapable of processing any other way.

The image from retinal nerves is inlaid to visual cortex in a matrix shape. Thus the image's properties are mostly preserved and independent from which animal's brain the data is plucked out.  This, of course, assuming a visual cortex that does preserve the matrix of image data; I wouldn't bet too much success reading the image data off an insect, frog or fish brain this way.


Nevertheless, the cat brain's species-specific abstract processing of the image is one way further still. The only things that are varied on the image are quality, field of view, colour depth and spectrum, but that is the same with any kind of camera equipment and we can adapt to it fairly well.


However, cool as this is, it is of fairly limited use; what I'm interested in is the inverse process - ability to insert image data into visual cortex. That would have an immense amount of applications, ranging from artificial visual perception to recreational virtual reality or enhanced reality purposes.

Technically if they are able to read the data from the visual nerve as it's presented at the visual cortex, they should be able to insert a stimulating matrix of electrodes to insert image data instead. It would of course be somewhat different than the actual human eye experience, but the brain is very adaptable - with sufficiently high-resolution matrix and ocular equipment, I'm quite sure that especially young children would get used to the artificial sight.

And think of what gadgets you could include in the ocular implement... zoom in, cable connectivity to different image data - hey, would you feel like plugging your brain to a ten metre space telescope?

[General Battuta]

I have no interest in a non-human image. I am incapable of processing any other way.

The image from retinal nerves is inlaid to visual cortex in a matrix shape. Thus the image's properties are mostly preserved and independent from which animal's brain the data is plucked out.  This, of course, assuming a visual cortex that does preserve the matrix of image data; I wouldn't bet too much success reading the image data off an insect, frog or fish brain this way.

Nevertheless, the cat brain's species-specific abstract processing of the image is one way further still. The only things that are varied on the image are quality, field of view, colour depth and spectrum, but that is the same with any kind of camera equipment and we can adapt to it fairly well.

I'm having trouble parsing your posts but I think you are being way too generous about how much information is actually present in the raw input to the brain. Most of vision is a mirage introduced by neural processing using evolved rules. Pulling the raw information a cat sees isn't going to tell us much about what a cat sees at all.

[Herra Tohtori]

I'm having trouble parsing your posts but I think you are being way too generous about how much information is actually present in the raw input to the brain. Most of vision is a mirage introduced by neural processing using evolved rules. Pulling the raw information a cat sees isn't going to tell us much about what a cat sees at all.

That's very true, and I don't know the specifics of how this operation actually works.

I do know I've read from some source that I can't remember now that the retinal data is directly projected on an area of visual cortex, but obviously due to relative crudeness of eyes as an observation instrument (lots of pixels in the yellow spot, little pixels on the outside, colour data only on the center, etc.), the quality of the image projected and captured on the brain would suffer accordingly, which was demonstrated on the video.

Tapping into the higher brain processes that do the visual filtering and abstract handling of information would be even more advanced than hooking up to the direct datastream in the form it arrives to the visual cortex, so I'm assuming that's what they're doing. They probably have filters to handle saccades and to reconstruct the image from the data to a remarkable extent, and clearly it is working... somewhat.

That doesn't mean it's "what the cat sees"; more like "what the cat is watching", filtered through the eye, ocular nerve, signal capture and handling and possible post-processing.

[MP-Ryan]

I'm having trouble parsing your posts but I think you are being way too generous about how much information is actually present in the raw input to the brain. Most of vision is a mirage introduced by neural processing using evolved rules. Pulling the raw information a cat sees isn't going to tell us much about what a cat sees at all.

That's very true, and I don't know the specifics of how this operation actually works.

I do know I've read from some source that I can't remember now that the retinal data is directly projected on an area of visual cortex, but obviously due to relative crudeness of eyes as an observation instrument (lots of pixels in the yellow spot, little pixels on the outside, colour data only on the center, etc.), the quality of the image projected and captured on the brain would suffer accordingly, which was demonstrated on the video.

Tapping into the higher brain processes that do the visual filtering and abstract handling of information would be even more advanced than hooking up to the direct datastream in the form it arrives to the visual cortex, so I'm assuming that's what they're doing. They probably have filters to handle saccades and to reconstruct the image from the data to a remarkable extent, and clearly it is working... somewhat.

That doesn't mean it's "what the cat sees"; more like "what the cat is watching", filtered through the eye, ocular nerve, signal capture and handling and possible post-processing.

I'm digging back a few years of memory to neuropsych lectures (and my memory may be slightly faulty), but as I recall the majority of our sight actually occurs in the brain as a result of image interpretation, rather than what is directly received into the eye (and try to disprove that tautological hypothesis, I dare you... ).  It's true that the primary visual cortex appears to receive a lot of direct data from the receptor cells in our retinae, and areas of the retinae superficially map to the surface layers of the cortex, but that isn't what we actually "see."  So back to your original post, it would be unnecessarily cumbersome to try to insert an image into the "input" end of the visual cortex (and as that part of the brain engages in parallel processing instead of directional flow of information, impossible).  It would make much more sense to try the insertion at the point where the assembled visual information is being interpreted.  Otherwise, our built-in processing rules would disrupt what you want to display.

Like Battuta, deciphering your original post is proving somewhat challenging for me so maybe that's what you were saying to begin with.  Image insertion at output of visual cortex = practical.

Sight is mostly perception and interpretation, not a true reflection of what exists in our surroundings.  (Who needs illicit drugs when you have neuropsych to do the mindfrak for you )

[Bobboau]

So back to your original post, it would be unnecessarily cumbersome to try to insert an image into the "input" end of the visual cortex (and as that part of the brain engages in parallel processing instead of directional flow of information, impossible).  It would make much more sense to try the insertion at the point where the assembled visual information is being interpreted.  Otherwise, our built-in processing rules would disrupt what you want to display.

I disagree with you.
Fortunately this is science and there is one way we can settle this.
TO THE THUNDER DOME lab!

[Herra Tohtori]

I'm digging back a few years of memory to neuropsych lectures (and my memory may be slightly faulty), but as I recall the majority of our sight actually occurs in the brain as a result of image interpretation, rather than what is directly received into the eye (and try to disprove that tautological hypothesis, I dare you... ).  It's true that the primary visual cortex appears to receive a lot of direct data from the receptor cells in our retinae, and areas of the retinae superficially map to the surface layers of the cortex, but that isn't what we actually "see."  So back to your original post, it would be unnecessarily cumbersome to try to insert an image into the "input" end of the visual cortex (and as that part of the brain engages in parallel processing instead of directional flow of information, impossible).  It would make much more sense to try the insertion at the point where the assembled visual information is being interpreted.  Otherwise, our built-in processing rules would disrupt what you want to display.

Like Battuta, deciphering your original post is proving somewhat challenging for me so maybe that's what you were saying to begin with.  Image insertion at output of visual cortex = practical.

Sight is mostly perception and interpretation, not a true reflection of what exists in our surroundings.  (Who needs illicit drugs when you have neuropsych to do the mindfrak for you )

Well, I just can't personally see how it could be possible to make a generalized signal format that would actually be a match with the signals that visual cortex outputs as the WYSIWYG data. It would, by my understanding, have a lot more variation between different people - correct me if I'm wrong, but do we even have a clear concept where the visual cortex dumps the complete "visual perception image stream" that it reconstructs from the fragmented datastream from the retinas? What happens next and what format is the image data at this point?

I think visual cortex would have easier time to adapt into processing raw video signal, mimicking the ocular nerve inputs, than us figuring out what sort of info the visual cortex reconstructs from the sensory inputs from the eyes. It just feels likely to me that at this level, the data would be highly specialized and abstract rather than anything resembling raw image data, so even if you managed to patch together some software that mimics the functionality of visual cortex in reconstructing the ocular signals into a brain-comprehensible visual perception data, you'd likely start having latency problems without very powerful image processing units as well as the probable problems trying to adapt the system for general purpose rather than individually tailored for use.

It just feels like it would be a simpler solution to let the brain do the heavy lifting: Try to calibrate the input signal to visual cortex to match as closely as possible the qualities of a biological eyes' retinal inputs, and let the subject's brain adapt to it and learn to process and reconstruct a visual perception from that data. The brain is so far uniquely powerful and adaptable marvel of signal processing, so I bet it'd do the job better and faster than trying to bypass the image reconstruction phase done by the visual cortex. Of course, it would not be quite the same as the "natural" vision due to differences between signals from natural eyes and artificial ocular implants, but especially if the implantation surgery was done at young age, I think it would have spectacular resources... ignoring the current practical difficulties of hooking up raw image stream to the brain at sufficient and meaningful resolution.

[General Battuta]

Well, I just can't personally see how it could be possible to make a generalized signal format that would actually be a match with the signals that visual cortex outputs as the WYSIWYG data. It would, by my understanding, have a lot more variation between different people - correct me if I'm wrong, but do we even have a clear concept where the visual cortex dumps the complete "visual perception image stream" that it reconstructs from the fragmented datastream from the retinas?

Not reconstructs, really. More or less just...constructs.

[Bobboau]

well, my money is on RLE encoding with 7/5/5 HSL color space, I'm only putting down 5 bucks but if I win the odds will make me a billionaire.