I watched the video and I want to believe, but... I don't know. I have cats, I know cats are intelligent, cats can be trained to understand single words, etc. But I'm just not convinced any cat can string together English sentences like this. Also, the concept of time might be a bit abstract for a cat who is always living in the present moment. And then there's the most viewed video on that channel[1] where he presses the "cuddle" button, and then... doesn't cuddle. He seems like he knows pressing a button get a reaction, but he doesn't know which button is which. Good morning human, time to play with the disembodied voice buttons again.
Like I said though, I want to believe. Can you think of any longer videos that might convince a skeptic?
I think that video is terribly cute and the cat is adorable, but I also remain unconvinced that a cat can understand, "morning now, before night". That is too abstract. For context, it took my (then) 4 year old human child many months to understand "before" and "after". He would use them indistinctly to mean "not now".
I've watched a few in the past and I'm not convinced the more complex interactions are "real" language use instead of operant conditioning combined with confirmation bias. Too many of the videos will have button sequences that seem random get assigned a meaning, and even non-request button presses will get attention.
Buttons for "food", "water", "pet me", "play with me", "outside" almost definitely work. Before I lived with a cat again, I was skeptical of those as "well, duh, any of those buttons get a reward, they don't care what it is", but our cats ask for those specific things without buttons.
That is more compelling, but I'm still not convinced. Out of all the videos I've watched it's the only one that comes close. Smells like broken clock theory to me.
"No" is another higher-level abstract concept that I wouldn't expect anyone to master before more basic concepts. If the cat could say "ears, no" or "tummy, no" with similar confidence, then I would be convinced it understands combining the concept of "no" with another word. And those seem easy to test too. Cats don't like you messing with those areas.
Maybe there's more in videos I didn't watch, but I watched 5-6 videos and the cat seems to be pressing buttons randomly. It doesn't seem to correspond to anything it does.
This one [1] isn't very long, but I think it shows that the cat (Billi) does associate the buttons with a meaning.
The owner is playing with Billi with a toy on a string, then drops it on her back. She then quickly (within 10 seconds) presses "No", then "Back", to tell her owner to get the toy off of her back. If she was just pressing buttons to get a reaction, this would be no more likely than "Morning" "Love you", or "Want" "Hello", or any other arbitrary pair of buttons.
AFIACT, the videos are a kind of progress log for demonstrating the cat's learning in different concepts. The older videos focused on [demonstration of] learning of more "core" concepts ("food", "water", "pets", "outside", etc.); but since these concepts are now fully "taught", the owner seems to feel no more need to film the cat communicating these simpler core messages, except when they form an interesting sequence or larger idea.
Instead, the newer videos focus on learning of more fringe concepts — ideas that "come up" less often — and so progress is slower in these.
(Note that, watching the videos chronologically, the number of buttons the cat has available grows quite a bit over time. I think the owner thinks that the cat now understands the general concept of associating things-to-be-communicated with presses of word-buttons, and so feels that they can now feed them new concepts faster.)
I'm trying, but I don't see it at all with these examples.
1) just seemed like random pressing until the dog pressed "paw", then the owner repeated loudly "something in your paw?" The dog presented its paw, then the owner decided "hurt" "stranger" "paw" was some sort of splinter she found there. The dog wasn't even limping.
2) I didn't get any sense of the presses relating to anything the dog was doing, and since the owner was repeating loudly the thing she wanted the dog to find, I was a bit surprised. Then the dog presses "sound," the owner connects this with a sound I can't hear, then they go outside to look for something I can't see.
Billie the Cat: I simply saw no connection between the button presses and anything the cat did. The cat pressed "outside" but didn't want to go outside. The cat presses "ouch noise" and the owner asks if a sound I didn't hear hurt her ears. Then the cat presses "pets" and the owner asks if the cat wants a pet? The cat presses "noise" and the owner continues the monologue apologizing for the painful noise and offering to buy her cat a pet. Sorry to recount most of the thing, but I don't get it at all.
-----
Not trying to debunk talking pets, but I'm not seeing anything here. I at least expected the dog to be smart enough to press particular buttons for particular things, but I suspect the buttons are too close together for it to reliably distinguish them from each other. I'd be pretty easy to convince that you could teach a dog to press a button to go outside, a different button when they wanted a treat, and a different button when they wanted their belly rubbed. In fact I'd be tough to convince that you couldn't teach a dog to do that. Whatever's being claimed here, however, I'm not seeing.
-----
edit: to add a little more, I'm not even sure that *I* could reliably do what they're claiming the dog is doing. To remember which button is which without being able to read is like touch typing, but worse because the buttons seem to be mounted on re-arrangeable puzzle pieces. Maybe I could associate the color of those pieces with words, but that would only cover the center button on each piece.
If a dog were using specific buttons for language (or if I were doing the same thing) I'd expect the dog to press a lot of buttons, until he heard the sound he was looking for, then to press that button over and over. Not to just walk straight to a button and press.
I think the cat just presses the buttons when it wants the owner to come, and presses them again when the owner says something high pitched at the end and looks at it in expectation.
I've seen a few videos on YouTube of cats trained on these push-down language buttons, mounted in a hexagonal pattern, on a mat laying on the floor. The cats press a button and the button plays back a word which the owner has recorded.
I'd like to see more testing out of it but the cats seem to be rather expressive. They pick the same button, so at least they are consistent in their wants. They have a few verbs in the mix, it seems, not just nouns.
You should check out the YouTube of the cat that was trained to use buttons to speak. It sometimes says things that are surprising which it wasn't trained to say directly - like complaining that its owners music is bad. I realize that this isn't contradicting your thesis of language models not being conscious - but I just generally think people should check out things like that cat. So I'm not trying to disagree with you here. I just think the cat is cool.
I'm also a fan of ants. They have agriculture. They farm in their underground cities! They keep livestock too; insect livestock, but still! They give them medicines when they get sick, specially cultivated fungus for example. They are so cool.
The video where the cat begs (through the buttons) to go outside in the rain, takes one step outside, comes back in, and immediately complains (through the buttons) about the rain, is pretty convincing to me.
Also interesting is its use of the "angry" button in response to not getting its way, or having something irritating it, or when it's clearly frustrated by not having its needs understood.
In order to translate language, the animal has to be speaking language. That is, they have to be putting some sort of emotion or indicator into their vocalizations that would be able to be decoded in the first place.
There's a channel on YouTube called BilliSpeaks where someone trained their cat to be able to press buttons to speak. The cat can answer questions, communicate feelings and even have conversations with the help of their translator, but only because they were trained on how to use it.
I am one of those people who had one view before a youtube video and another view after.
We had cats when I was a kid, and I would observe that cats are very good at tracking prey, at opening doors or at getting humans to open them, at figuring out who is the boss, and at bearing very, very long term grudges, but otherwise they operate in a system where the world revolves around them (apparently quite literally in terms of how they map the world).
This utility-focussed view of the world means they always get fed, but it also leads them to get stuck when exploring -- never paying attention to the fact that the neighbour's garage door does not just open, it also closes, for example!
So I have tended to see them as well-optimised, intelligent, but not necessarily "bright".
But the cat-mirror-ears video -- where a cat sees a reflection in a mirror, apparently understands that it is its own reflection and then... realises it has ears... that changed my mind.
There is another "cat theory of mind" video where a mother cat purposefully and deliberately retrieves an object its kitten wants to play with, that scientists have talked about on twitter, but I can't find it.
A friend of mine has a cat that learned to play fetch as a kitten and never stopped playing. So unusual.
Like I said though, I want to believe. Can you think of any longer videos that might convince a skeptic?
[1]: https://www.youtube.com/watch?v=pvgfI9P377U
reply