Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Certain dogs are capable of learning the names for more than 100 different toys

scientificamerican.com

Post reply on HN
Print view View original

121–130 of 218 posts

Re: Certain dogs are capable of learning the names for more than 100 different toys

#121

Post by b112 »

ta8645 wrote: post #100
knodi123 wrote: post #25

Earlier quoted context omitted.

My dog recognizes different cars by their noise profile. So he'll wake up from a nap and get excited whenever my wife pulls onto our street, but stay down for any other car. It probably helps that we're the only prius on the block, but still impressive!

I heard about a test someone did, after wondering how their dog was always standing in the window at the end of each workday, anticipating their owners arrival home. After casting about, a theory was developed that the dog "smelled time". That is, as the day wore on, the scent of their owner, decreased in the air; at a low-enough concentration, the dog knew it was about time for the front door to open and their owner…

Certainly interesting enough to do another test, but instead without another human entering the house.

Maybe a sealed box that opens to vent the scent. You'd need to have the same box do the same, with no scent too.

Re: Certain dogs are capable of learning the names for more than 100 different toys

#122

Post by AeroNotix »

br3d wrote: post #83
grimgrin wrote: post #80

Earlier quoted context omitted.

I did not know what class inclusion was, but now I'm thinking it's more complex than that? After a little bit of reading. "all daisies are flowers" and not "all flowers are daisies", this example seems more like the "solid grasp" you're referring to. And not basic categorization that a 3 year old might have: "foods", "toys"

Hmm, I'm now struggling to remember basic developmental psych, but there's definitely a phase at which linguistic children struggle with things having two names (it can't be both "dog" and "Rex") but I think you're right - this phenomenon is subtly different to class inclusion. But either way, dogs can do something with language comprehension that speaking children can't, which is the bit I find really interesting

I need to read up more on this because in my extremely small sample size this kind of dual naming understanding came in really early with my daughter.

I feel like it's a linguistic subtlety that us adults are struggling with conveying the exact concept.

Re: Certain dogs are capable of learning the names for more than 100 different toys

#123

Post by doubled112 »

motohagiography wrote: post #117

If dogs are smarter than we knew, what are some additional tasks we could give them? I have manuals for disability support dog training (opening doors, getting medications, helping to dress, etc) but maybe we could teach them to: - clean or arrange rooms or outdoor spaces - find and pile firewood - collect litter - convey messages to named people What else?

There's something very human about "hey, this thing is smart, better put it to work for us"

Re: Certain dogs are capable of learning the names for more than 100 different toys

#124

Post by cogman10 »

qup wrote: post #107
legohead wrote: post #88

Earlier quoted context omitted.

I had a dumb labrador that was hard to train. Now I have a kokoni who knows 20 commands & words, and she is super easy to train. We haven't tried to test the limit of her abilities, we just ran out of things to teach her (that we cared about, at least). She never cared for toys though. I noticed while teaching her to "speak" that she would sneeze a lot, I guess it's kind of a reflex for her when she tries to speak. S…

She sounds like my dog, who has sneezing fits a lot. The shelter said she was a shepherd/boxer mix when I adopted her, but I believe she's a mountain cur. Either way, she's wonderful, and learned every trick I taught her in the first couple of months. She still knows them all 9 years later. She's learned a lot about me without my instruction, we just fit like a glove. I don't have an exact count, but she knows at lea…

> who has sneezing fits a lot

Sneezing is a way dogs say they want to play. That's adorable that she loves sneezing for you. [1]

[1] https://www.thekennelclub.org.uk/health-and-dog-care/health/...

Re: Certain dogs are capable of learning the names for more than 100 different toys

#125

Post by fnordpiglet »

ghshephard wrote: post #109
fnordpiglet wrote: post #54

Earlier quoted context omitted.

Why do we have K9 search and rescue dogs and drug sniffers instead of just having police officers sniffing everyone’s luggage? The answer from the article is, essentially, because dogs are much better at odor detection than us. It weirdly seems to say because we have the ability to sense odors we are better than we think we are, which implies we think we can’t smell at all or something. But dogs are much much better…

The article made a quite different argument than "dogs are much better at odor detection than us" - that is contrary to what the research shows. Humans have much better ability to smell some things than dogs do, but, dogs are (as you noted) capable of smelling some things better than humans.

They make some squinty suppositions from some limited studies that suggest dogs (which can smell cannabis flowers) can’t smell flowers better humans etc. But then they conclude with this statement that sort of refutes the entire premise:

“”” Besides having more olfactory receptor cells than humans, dogs also boast a specialized snout adapted to methods of breathing that deliver a steadier stream of information-rich scent. Dogs and some other animals even experience scent differently. Their olfactory system allows them to smell liquid phase chemicals that aren't airborne—think of layers of urine and other liquids on your neighborhood fire hydrant—by working like a pump to deliver them to a specialized nasal organ. “””

Which is pretty indicative that as everyone knows from daily experience with dogs, dogs are better at smelling things than humans. There may be a range over which that superiority falters at some chemicals but I saw nothing other than headlines and pop science supposition that indicates humans are superior at smelling than dogs.

Re: Certain dogs are capable of learning the names for more than 100 different toys

#126

Post by bentcorner »

tlavoie wrote: post #45
_whiteCaps_ wrote: post #20

I surprised that poodles were called out as a less obvious smart dog. My dog learned on her own that me holding keys + wearing glasses = car ride, while keys + no glasses = walk. It took me a while to learn how she knew what we were going to do.

Our dogs are also also very good at picking up other cues. Not only can we not say "walk," but the "well..." as in, "well, time to take the dogs" gets the same reaction now. Our SPCA-special (shepherd / husky / other?) learned after _one_ event that the tub running with the bathroom door open means that it's bath-time, and time to head out through the pet flap to the safety of the outdoors.

This is not a difficult task but my dog absolutely recognizes the various ways I end meetings during work or discord chats while gaming and always starts bothering me for pets when I'm done.

Re: Certain dogs are capable of learning the names for more than 100 different toys

#127

Post by dotnet00 »

AeroNotix wrote: post #120
wosk wrote: post #86

Earlier quoted context omitted.

I did not read the paper, so I cannot comment on the "solid grasp of class inclusion", but regarding the capacity that you described in your comment, I have a 2-year-old and it's been a long while since she has mastered this (book vs this book, toy vs toy, fruit vs an apple and so on). As far as I know, most two year old have already acquired this concept. (EDIT I see the other comment says something similar and you…

I didn't even know "class inclusion" was a thing really. Though obviously the concept makes sense. My daughter had a solid grasp of it definitely around 16-18 months. She could easily talk about books or toys, cars, food, drinks etc. Not sure if this is unusual but 7-8 as the other poster mentioned sounds crazy late for that kind of conceptual understanding to appear.

Yeah, it's not really convincing that it can typically take up to 7-8. By that age kids are already able to read, write and do basic math, which of course requires them to understand "classes" like numbers and letters, such that they could handle both "write a number" and "write 23".

Re: Certain dogs are capable of learning the names for more than 100 different toys

#128

Post by Extropy_ »

cush wrote: post #104
bongodongobob wrote: post #24

I adopted an Australian Shepherd from a friend a year ago and I'm absolutely blown away by how smart he is. He even knows the neighbor dogs' names. I can say "go get spot!" and he'll run after the right dog. He knows all my family members names. He knows that "you've arrived" on Google maps means we are there. He knows that "power on" coming from my Bluetooth headphones means we are going for a walk. He gets the conc…

I did this with my Aussie as well. I'd hide all of her toys throughout the house then instruct her "go get your hamburger!" and in a few minutes, she'd come back with her hamburger. It was wild that she'd remember what toy she was looking for during that time. More impressive still, was she would be able to learn new toy names through inference. I'd come home with a new toy turtle, hide all her toys, say "go get your…

I have a 3 year old pug. He'll run through the house searching for the right family member if you count down from 10 and say "Go find X! Where is he?!"

Re: Certain dogs are capable of learning the names for more than 100 different toys

#129

Post by dboreham »

kemayo wrote: post #77
airbreather wrote: post #3

I have owned six dogs in my life, most recently getting a 7 week old pup staffy x ridgeback, now 16 weeks old. He goes with our 9 year old girl staffy, who can follow conversations, eg planning to go to the park or beach without directly mentioning it in any way. All my dogs have understood a lot of words, some more than others, but all way more than 20. The aspects that promote this, I believe, are this: 1. They are…

The argument I've seen made for crating from an evolutionary-history perspective is that dogs are den animals -- they actively like having a small nook that they can feel safe in, where they're comfortable and nothing can sneak up on them. I have a crate for my dog. The door stays open all the time, it has a comfy mat on its floor, and sometimes he goes there to take a nap. He's only ever shut in it when there's a re…

Cats like boxes.

Re: Certain dogs are capable of learning the names for more than 100 different toys

#130

Post by dotnet00 »

doubled112 wrote: post #123
motohagiography wrote: post #117

If dogs are smarter than we knew, what are some additional tasks we could give them? I have manuals for disability support dog training (opening doors, getting medications, helping to dress, etc) but maybe we could teach them to: - clean or arrange rooms or outdoor spaces - find and pile firewood - collect litter - convey messages to named people What else?

There's something very human about "hey, this thing is smart, better put it to work for us"

To be fair, we wouldn't have domesticated dogs if they weren't smart enough to work for us as hunting/herding assistants.
Post reply on HN