4 ms·

Dogs Forget Events Two Minutes Later

3 points by Red_Tarsius 12y ago

anigbrowl 12y ago
Mmm....seems a bit dubious. My dog likes me to hide his rubber ball so he can go look for it, and he's so good at remembering places I've hidden it before that finding new places to make him really look for it is a problem (he's not very scent-motivated). Now of course he could be relying on associative memory because I praise him when he does a good 'find' but their test here looks overly reductionist to me. For sure, if I'm doing something else for a while and then ask him where his ball is, he may well have forgotten where he left it and looks in random places. So his memory is not that great; on the other hand, when I see that he has multiple balls (which he enjoys hoarding) in one spot I'll call him to bring one over, then wait a bit before telling him to go get another one, in which case he knows exactly where to go. I'll try timing the interval to see how long before he forgets his previous position. The other thing that occurs to me is some dogs' habit of burying bones and so on that they may want to store for later. Sure, they are using their sense of smell there, which is many times better than ours, but given that dogs don't have mediocre vision to start with (by comparison to ours), I wonder again about the quality of the experimental design. If you asked people to remember something using one of their inferior senses - say, a series of fabric textures that are only touched rather than seen - I suspect they would turn in mediocre results. It seems to me that the stimuli used in the experiment were too abstract to be sufficiently engaging to the animals, for the same reason that we don't have a very extensive tactile vocabulary, notwithstanding the importance of touch as a sense.
- dalke 12y ago
  I'll be generous and suggest that NatGeo didn't translate the research fully correctly. The quote from the researcher is "animals have no long-term memory of arbitrary events" though there may be "specialized memory systems hardwired to remember certain "biologically relevant information" (such as where to find food)." This would mean that it's much harder to have a discussion abou the topic without a detailed understanding of what "associative memory", etc. means - knowledge I don't have.

3 points by Red_Tarsius 12y ago

Mmm....seems a bit dubious. My dog likes me to hide his rubber ball so he can go look for it, and he's so good at remembering places I've hidden it before that finding new places to make him really look for it is a problem (he's not very scent-motivated). Now of course he could be relying on associative memory because I praise him when he does a good 'find' but their test here looks overly reductionist to me. For sure, if I'm doing something else for a while and then ask him where his ball is, he may well have forgotten where he left it and looks in random places. So his memory is not that great; on the other hand, when I see that he has multiple balls (which he enjoys hoarding) in one spot I'll call him to bring one over, then wait a bit before telling him to go get another one, in which case he knows exactly where to go. I'll try timing the interval to see how long before he forgets his previous position. The other thing that occurs to me is some dogs' habit of burying bones and so on that they may want to store for later. Sure, they are using their sense of smell there, which is many times better than ours, but given that dogs don't have mediocre vision to start with (by comparison to ours), I wonder again about the quality of the experimental design. If you asked people to remember something using one of their inferior senses - say, a series of fabric textures that are only touched rather than seen - I suspect they would turn in mediocre results. It seems to me that the stimuli used in the experiment were too abstract to be sufficiently engaging to the animals, for the same reason that we don't have a very extensive tactile vocabulary, notwithstanding the importance of touch as a sense.

dalke 12y ago
I'll be generous and suggest that NatGeo didn't translate the research fully correctly. The quote from the researcher is "animals have no long-term memory of arbitrary events" though there may be "specialized memory systems hardwired to remember certain "biologically relevant information" (such as where to find food)." This would mean that it's much harder to have a discussion abou the topic without a detailed understanding of what "associative memory", etc. means - knowledge I don't have.