Hacker News new | comments | show | ask | jobs | submit login
Ravens attribute visual access to unseen competitors (nature.com)
91 points by Hooke 9 months ago | hide | past | web | favorite | 11 comments

I'm impressed by the lengths to which the authors must go and have gone to refute the skeptics' counter-explanations. For example, they simulated the presence of a potentially observing competitor using an audio recording of a competitor that was not actually observing the raven caching its food. Indeed otherwise a skeptic might claim that the caching raven was acting as "observed" in response to the different sound, rather than (as the authors want to show) because it inferred that it can be seen when the peephole is open.

I shall be extremely surprised if this does not turn out to be the case for all sorts of other animals too.

I have seen dogs, cats, and rats (yes, rats) perform complex, planned actions clearly relying on good models of who were now and would later be looking at what.

Rats are very very smart, smarter than most dogs. If they had language and slightly less fear they would be super dangerous.

Yup. We had pet rats for more than ten years. The smartest among them were frighteningly smart. In the general scheme of things, it doesn't take all that much to lay a plan and outwit me, but when a small rodent does it, and does it again, I'm impressed. Like removing the least favorite pellets of food from the bowl, hiding them, and then calling my attention: The bowl is empty, if you please!. Had me fooled for weeks. Or, as that same individual did, wondering whether the cage would have an entrance on the back side, in the same spot where it had one on the front, hauling a huge piece of bread out of sight of the others, stashing it away, launching an expedition to the far side, discovering no entrance there, returning, picking up bread, and blustering her way past the others through the front door. Yes, that kind of thing is beyond many - probably most - dogs.

The house had mice for a while. I set down live-catch traps, which are a metal box with two levered entrances that trapped the mouse.

After a couple of catches the remaining mice blocked the entrances by packing them with junk they found on the floor.

I always wondered if that was random behaviour or deliberate intent.

You really have to wonder just how smart animals would become if they could speak and train subsequent generations like humans do - never mind if they somehow invented reading and writing.

Yeh we had a nice plague and the mice even the very young would very quickly learn that they must jump over the traps and just ignore them, when they were placed in a mouse thoroughfare

Same thing we do every night, Pinky.

Yeah the dog would just eat the bread immediately.

That, and an opposable thumb would put them at the top of the food chain.

An interesting experiment, but it all comes down to an effect at "p < 0.05." That's potentially a 1-in-20 chance of it being a meaningless fluke. Which given the very relaxed statistical standards in most behavioral studies isn't too surprising, but I thought in the "replication-crisis" era Nature at least would have more stringent requirements.

That's not a 5% chance it's a fluke. It means you'd only see this result 5% of the time if it was a fluke.

It's a big difference.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact