Extracting video images from cat brains

[Royston]

I found this yestarday. Although the research is quite old now they've successfully reconstructed spatiotemporal scenes from the responses of visual neurons.

 

In essence they have recorded the visual input of a cat and rendered the data so it can be viewed. The implications are that when the research is ready for human subjects it will be possible to record and playback our dreams!

 

What would this look like though ? If you think about the images in your head when you think or dream, to me they are an incomplete mish mash of mental pictures and scenes, where 'you' are the only one that can decipher them into something of meaning.

 

It would be really interesting how a computer would interpret such images...could be quite scary as well.

 

http://www.mindpixel.com/chris/2005/06/extracting-video-from-cat-brains.html

[YT2095]

the only thing I don`t understand so far is that when we dream, why would any of our visual neurons need to be stimulated? surely all it`ll need is a selection of memories and nothing to do with the eye is needed to dream? (ignoring REM).

 

it just seem to be the long way around to picture a mental image if the brain has to stimulate Visual neurons to do it.

 

just a Though

[Royston]

Very good point. I'll have to have a browse to see what's responsible in the brain for producing mental images as opposed to visual input from the eyes...I guess it's a combination of lot's of neural responses, as you said like memory et.c

 

In the mean time check this out...slightly off topic, but still in the same vein when it comes to transposing thoughts to an external source. The research is in it's early days but very interesting, and also raises questions about the ethical side of such technology. Tell me what you think.

 

http://news.bbc.co.uk/1/hi/health/4472355.stm

[Morpheus]

Can someone find me the original .pdf file that was located in MindPixel's blog?

[badchad]

I believe I have located the original publication here:

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10479703&query_hl=3

 

I'm sure someone has institutional access on here and can grab it for ya. If not PM me with an email and I can send it to you (although I don't lurk around these boards often, so you'll have to be patient).

[GeminiinimeG]

Ok so if your imahining somehting your going into your brain and bringing up pictures from your neurons, meaning if we could access this we might be able to read peoples minds because wed be able to see every songle bit of information thats stored withing our memory, this is cool now spies dont have to be tortured, well just open up their brains and see whats under the hood

[Skye]

There's a link to a free version on that ncbi page:

 

Click here.

[ashennell]

the only thing I don`t understand so far is that when we dream, why would any of our visual neurons need to be stimulated? surely all it`ll need is a selection of memories and nothing to do with the eye is needed to dream? (ignoring REM).

 

Whenever you simulate anything in your head, like a percept or even an action you use parts of the cortex that actually represent or produce real percepts and real actions.

 

A simple way to understand this is by considering a population of neurons that respond when you see a certain face. They do not code the physical properties of the face (well they may partly) but the meaning of the face - the identity of the person , how it's appearance in your visual field will effect your behaviour, current goals etc, etc. The perceptual system's role could be viewed as transforming inputs that are physically similar into outputs that are semantically similar. So in simulation, activation of the visual areas dosn't just produce a phyical 'image' but also contibutes towards defining a meaning of the stimuli. (Nueroscientists: I know this is a big simplification)

 

I just had an intersting idea while typing this message actually. I would say that if you were asked to visualise an apple that the internal 'picture' would be a lot less vivid than when you have a dream or even a daydream. Maybe this is because an apple (or any other object), on it's own, has little semantics attached whereas a story has more meaning. Of course, when you are asleep or daydreaming sensory input is somewhat attenuated. Perhaps a fairer comparison would be between an apple and imagining buying a newspaper in a shop , or something else with narative, both while fully awake. I, myself, find it alot easier to get a vivid simulation of the more complex situation.

[ashennell]

The implications are that when the research is ready for human subjects it will be possible to record and playback our dreams!

 

Maybe one day. I think that coding of information in the brain is individual specific. The paper you refer to actually records from the visual thalamus. This is before the signals reaches the cortex. At this stage decoding is probably quite simple as the organisation is retinotopic and the flow of information is on way. However, once the information gets to the cortex things get a lot more complex. Activity here is subject to mass top-down influcence and is likely to depend on personal experience, interpretation, covert attention, etc,etc. Unforetunely we would need to decode this mess to be able to record our dreams. It would be cool though, we apparent never remember most of our dreams - it would be quite strange to wake up and watch what we created overnight for the first time not as a dream.

[bascule]

I think that coding of information in the brain is individual specific.

 

If you're referring to what Dennett calls a "phenom", an atomic unit of mindstuff, then yes, you're probably correct, but I'd say that the reasoning is because it maps against your own internal ontological structure, and when you abstract that away, you will be left with a coding which is common to everyone, at least for humans.

 

The paper you refer to actually records from the visual thalamus.

 

Yep, this is not an electronic gateway into the "mind's eye"