>>7824444 Nice quads, checked. Also as I said, I have no access to the original paper so i can't verify it. Maybe an anon with more google skills can find a digital copy and grant us some insight. The video is legit though, the comments are added by some anon making the webm out of a documentary on machine-brain-connections.
The input and output video resolution is low res on purpose to minimize the neurons which have to be monitored and simpler neuron output that can be distinguished from random noise. Wether or not the cat sees a cat face in the human face I couldn't make out of the summaries of the paper. The output face does have a more triangle shape though.
>>7824518 animals don't have intelligence. they might have awareness, but that isn't the same thing. you're not intelligence for recognizing human faces because that's a subconscious process in your neural networking. what makes you intelligence is forming a conversation, solving tasks, and so on that are outside your instincts (fixing a computer algorithm, designing an IC/engine). This cat isn't smart for having a section of its brain dedicated to processing faces as cat-like, as almost every living thing has that for a coping mechanism.
>>7825715 Nah I've wrote all explanations and provided scientific papers on /pol/ when I made the thread - cats do perceive us as bigger cats but with different behavior than their.
Everything they do to us is what they would do to other cats - it's because they're still slightly wild and not domestic - they never been trained for a specific purpose - they're just wild animals that tolerate us.
It's really no big deal - and it doesn't change anything - for long time I thought it was common sense as I first read about it in national geographic magazine as a kid.
>>7825246 >forming a conversation, solving tasks, and so on that are outside your instincts None of those things are outside of your instincts, though, abstract logic is a result of conditioning and your subconscious brain chemistry.
>>7825629 >What if it's us that sees cats as tiny humans when they're in fact demonic killing machines? We actually do, the baby faced kittens, and for that matter most mammal babies, are so cute to us because they have the same facial pattern properties as human babies, which we are hardwired to response to. And the cat does the same thing to us with their patterns. We see them as tiny humans and interpret human emotions and motives in them, they see big cats and think we are starving because we are not hunting mice and such.
If you think about it it also applies to dogs, why would that specific animal be so well integrated into human society? My take is that their social hierarchy system is similar to the human one, we too have a lot of different natural and cultural ranking systems in place. So it doesn't even occur to the dog that we are a totally different species, the patterns just snap well into each other.
>>7824501 Amazing how much focus is given to the eyes of the face. Makes sense too. Basic facial recognition starts at the eyes, and it is so easy to make a face from the most basic figures because of the strong response to this pattern. ^^ :( =) oo ._. ** >>
>>7825918 >>7825932 >Ill believe it when I see the papers so you mean the paper that was already linked half a dozend times already?
>>7825925 imagine one the technology gets advanced enough to monitor more or even all optical nerves. you could implant a chip to record anything you see. or alter what you see. would be some ghost in the shell tier stuff. fund it.
1999? why haven't there been repeat tests with better input/output different animals humans for comparison fucking makes my blood boil. How are we to take neuroscientists seriously when they can't even do their job right?
>>7825987 Weren't there some blind people who got augmented and had visual information from a camera induced into their optic nerves so they could see primitive shapes? AFAIK the body started to build up scar tissue around the transmitters which dampened and finally stopped the signals reaching the neurons.
> the experiment is reading the OPTIC NERVE but thats wrong. the paper explicitly states they are reading from the brain. "lateral geniculate nucleus - a part of the brain's thalamus that processes visual signals from the eye" but I guess ignoring the paper and just shouting your opinion from first glances is easier, right?
>>7828118 You seem quite confused, the conversation you jumped in the middle of was talking about the input being random noise, maybe you should follow reply chains before you reply so you understand the context next time.
I wonder if this is specific to the cat or is a common trait in domesticated animals by humans. If it's the latter then you would be able to extrapolate that over to dogs as well. I wonder how we would appear to them if it is that case since they've been domesticated for a decent amount of time longer.
>>7828141 Do you actually have a point or do you just repeat what you said before? At no point in the experiment there is a pure random noise signal that you could interpret freely, the neuronal response has some noise in it due to spontaneous discharge which is normal for all kinds of brain cells. While it would be technically possible to construct any image output from any input in any experiment by altering the signal enough in a prefixed manner, that would defeat the whole purpose of monitoring something wouldn't it and wasn't done here.
>>7825987 Totally agree with this. It is disgusting to mutilate a living creature like that without a clear set of goals to achieve. That video contributes NOTHING to the sum total of science. Everything you said about different animals and humans in comparison, is correct, if your gonna do it, do it right and go all the way.
Dogs hate me, I have an affinity to cats, do dogs see me as a cat? if so is it smell that triggers images? Leave experimentation on animals alone until you know where you are going - pricks!
>>7824501 >>7824438 If I'm not mistaken, output is calibrated by showing the subject many different images. The output is nothing but many of those images stacked on one another. If they show the subject many images of cats then output images will look like cats.
>>7828294 You are mistaken. The output is not a mix of different images, it is the added up signals of exactly the same video over and over. Also if you had bothered to even glance over the summary of the paper or read the thread you would know that they showed the cat portions of the movie Indiana Jones, not other cats.
>>7828272 How do you know where you can go with any technology until you do some basic research first? The real applications of this tech is developing Brain-Machine-Interfaces. The most obvious would be making blind people see again which was done in a primitive manner already. Then you have all the things that we currently consider still a science fiction like cyborgisation.
>>7828553 >If they show the subject many images of cats then output images will look like cats. Do you see cats in the input video? No other cats were in the input video. Does Indiana Jones even feature any cats? It does not feature any cats.
>"If I'm not mistaken, output is calibrated by showing the subject many different images. The output is nothing but many of those images stacked on one another. If they show the subject many images of cats then output images will look like cats"
>>output is calibrated by showing the subject many different images
>>output is nothing but many of those images stacked on one another
>>7828559 I didn't watch the video. I just know how the experiment works. I read about this experiment years ago.
Assuming that looking at humans consistently produces feline features in the output image relative to what a human would see in the same experiment (which isn't so clear since the output image is too fuzzy) then I suppose cats preferentially pick up on feline features. It reminds me of the Mass Effect 3 scene where the human, the salarian, and the turian all say that asari look like their own respective species. The human focusses on the face, the salarian on the skin, and the turian on the cartilaginous cranial ridge things.
>>7824430 I remember they did this with MRI data of humans as well, where they used machine learning to match patterns of fMRI activity to hundreds or thousands of videos during the training phase and then recreated the image presented to the subject, which would then look as a composite image of the various training videos presented.
This seems very different, though. I'd be interested in knowing where the electrodes are positioned. If it's in the visual cortex and these are truly just action potentials the image should be far more distorted.
Electrodes are in the lateral geniculate nucleus, a part of the brain's thalamus.
>match patterns of fMRI activity There is no pattern matching done by the science team, they are not comparing the unknown response to a known pattern. They are just adding up the discharges of the neurons in the thalamus and see what images emerge. This is a very different, more crude but also more impressive process.
>>7828682 what for, the optic nerve would show a distorted input image without much processing done on it. it is basically just a cable from the retina to the brain. the interesting work starts when the signals reach the pattern recognition parts of the brain.
>>7831452 googled a bit, there have been several similar experiments. the scarring seems to be the big problem of applying this for prolonged times. seems like pharmaceutical research has to be done first to progress this further.
>>7832591 >before pausing it >it is a webm of a the source, both of which anybody can pause at any time >I don't or don't want to understand a thing therefore I make up excuses how it fits into my narrow world view agreed anon and the moon landings were fake because there are no stars on the pictures, right.
>>7832830 what kind of argument is that, do you mistrust your doctor when he gives you an electrocardiography too? thats a computer generated graph based on the interpretation of signals from even less neurons. your "argument" could be used to doubt 99% of neurobiology and neurology.
>>7832833 >electrocardiography That doesn't target individual neurons, it gets an average electrical activity over a certain area and there isn't a bunch of additional processing and filtering algorithms to reconstruct a visual picture, its just signal amplitudes that give an estimate of the heartbeat that is much better than if he just felt for it himself.
I am not doubting the science, I am doubting your interpretation of what the picture is because if if was the real neural response it would show 170 graphs of signal pulses, not a reconstructed video.
>>7832849 >That doesn't target individual neurons, it gets an average electrical activity over a certain area and there isn't a bunch of additional processing and filtering algorithms to reconstruct a visual picture, its just signal amplitudes that give an estimate of the heartbeat that is much better than if he just felt for it himself. and what ultimately causes the electrical activity that the computer captures to give you graph? the sinoatrial node of the heart which is just a tiny lump of neurons. from there the signal is progressed, relayed and shaped even more until it over a lot of steps causes the electrical differences in the skin as a side effect that are measured by the device and show you the graphs.
>if was the real neural response it would show 170 graphs of signal pulses, not a reconstructed video. an image is the visual representation, not interpretation, of the data, so what would change if instead of giving you a video they would show you the bits the gathered data consists of? are you reading this text from pixels on your screen or is your computer reading out 0 and 1 to you, and would that make things more real?
>>7832870 >and what ultimately causes the electrical activity that the computer captures to give you graph? The design of the sensors provided the signal, the ECG does not measure the sinoatrial node, it has 12 leads spread across the limbs and body that is filtered and graphed as a simple electrical signal, the sinoatrial node consists of entire cells, nerves, and tissue and is much more than just neurons. If my doctor handed me a 3D image of my heart and said it the real signal straight from the ECG, I would be skeptical about his competence.
>would that make things more real? Actually observing what you have to say instead of interpreting a transmitted virtualization.
>>7832885 would you believe the 3D image if it was measured by an ultrasound device instead? that too is nothing else than the visual representation of data which started as a lot of raw signals of the device reading-head which were interpreted in the device microprocessor. the right tool for the right job.
>>7832910 I would believe it was ultrasonic reflections they measured which actually does include depth information, but I wouldn't call it a real picture of my actual insides, for that you would need an endoscope.
>>7832918 >wouldn't call it a real picture of my actual insides, for that you would need an endoscope. in a strict sense this is not the real Saturn's north pole but an image of the real Saturn's north pole. so you are arguing the semantics of the word "real"?
>>7832925 Most of the images NASA releases are heavily modified and aren't real, they shift other wavelengths of energy into various colors on the visual spectrum to make it appealing rather than accurate, such that it is an artistic interpretation rather than a direct real transcription.
>>7832934 >Most of the images NASA releases are heavily modified agreed. >aren't real semantics, as they show real objects.
let's take a step back, how can you be sure what you see with your own eyes is real? after all your brain is merely interpreting electric signals from your optical nerves. can you say for sure that anything you see is real? could it be possible that you are a brain in a jar hooked to a computer?
>>7832938 You are the one making statements about what is absolutely real and how that cat face is absolutely really a direct measurement just to scare some random anxious guy, so the burden of proof is on you, pal, you can't just go to your next jarred brain scare tactic when you fail to conclusive demonstrate your way more believable, but still fatally flawed cat face fear mongering.
>>7832949 >making statements about what is absolutely real you started with arguing that, not me. >just to scare some random anxious guy. that is a weird accusation. >burden of proof the burden of proof is on whoever wants to come to a scientific conclusion, if you want to ignore things based on whether they spook you or not that's your loss. >fail to conclusive demonstrate your way more believable in your view maybe, i think i presented my points well and you haven't really answered any of them to a conclusion either other than to say no. >fatally flawed cat face fear mongering made me chuckle, nice one.
i see we have reached an impasse. have a good day.
This. A lot of the interesting to look at "huh science" pictures would probably be virtually incomprehensible without a heavy dose of signal processing. Saying that these images are less "real" is about as insightful as saying that post processed music isn't "real".
>>7832999 Nice trips, checked. The brain is not really a stand-alone organism, there are a gorillion nerves and stem parts throughout the body, mostly in the spine, that not just transfer but also process information. For example the famous knee-jerk reflex is not caused in the brain. Separating and transplanting this mess is probably a lot harder than having interfaces directly implanted and connected.
>>7837617 >also active when expert birdwatcher are looking at birds as well as normal people. elaborate, because I understood that as the region is always active when matching patterns, how do you distinguish between born and learned activity?
Wouldn't you be perceiving the output on the screen through your own human perception, and thus it's probably not what the cat is actually perceiving? We perceive our brains (and their cat brains) through our own brains, after all. But it's not really that simple though, as what we call "consciousness" may have a far more complicated answer than simply "muh brain".
Also, iirc they have already done this thing with humans, and the traduction looks just as shitty.
>>7839248 >they have already done this thing with humans they stuck 170 metal needles into somebodies brain which causes scarring and irreparable damage? please do give us a link or name of the paper, that is badass.
>>7838857 are you so high you forgot the original source mentions the human face and even if you are the creator of the webm you added nothing new? 420 lmao dude weed, right.
>>7839479 in humans they usually use MRI and similar non-intrusive scanning methods to build a database of how certain patterns trigger different responses in the brain. then when the subject is shown a new video, the machine can guess what the object shown is by comparing the response to its database. you can't reconstruct the exact same picture out of this, only a guess on what the human is seeing.
>>7839456 Original source does not state that, no where the scientists said anything about it, no where in the scientific paper it is mentioned about the cat - it's just the journalist for the documentary opinion about it - just like other people in this thread think they saw a cat face there but it's totally not indicated by the scientific observation and analysis of many tapes.
Just pointing out that people arguing on the idea of the thread are arguing on popular opinion among unscientific people and not based on any scientific observation.
>>7839771 >Original source i ment the source of the webm, the documentary, which mentions the human face. i showed the muted youtube video to some people, they saw the human face too without anybody telling them about it beforehand.
>Just pointing out that people arguing on the idea of the thread are arguing on popular opinion among unscientific people and not based on any scientific observation.
yes that is what /sci/ about, and for that matter most of internet discussions. we are not a research group peer reviewing a paper, we are anons posting our opinions.
>>7825629 >What if us human really look immensely evil and ugly without neuronal filtering? We do, this is why you shouldn't look into the mirror when doing LSD. The cognitive dissonance lets you see your face without the myriads of associated subconscious reactions. Basically your sober brain knows it is your face, just as your brain knows it is your hand when you try to tickle yourself. Cut that middleman out and shit gets weird and spooky fast.
>>7846371 you take your unproven position, claim it is the standard/basic despite the absence of any data and absence of research pointing in any direction. then you somehow assume it to be simpler, despite proving either position would have extreme consequences.
a classic example of https://en.wikipedia.org/wiki/Begging_the_question
that's why I told you that is not how debate is done.
>>7846415 there is no data indicating a god existing. there is no data disproving a god. therefore any assumption on whether or not god exists are just that, assumptions, never proofs. did you understand it now?
All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.
This is a 4chan archive - all of the shown content originated from that site. This means that 4Archive shows their content, archived. If you need information for a Poster - contact them.
If a post contains personal/copyrighted/illegal content, then use the post's [Report] link! If a post is not removed within 24h contact me at email@example.com with the post's information.