this is fun to watch and listen to. I liked the sawing mimicry.

Fascinating!

Interesting new research.

New Caledonian crows were presented with Bird and Emery's (2009a) Aesop's fable paradigm, which requires stones to be dropped into a water-filled tube to bring floating food within reach. The crows did not spontaneously use stones as tools, but quickly learned to do so, and to choose objects and materials with functional properties. Some crows discarded both inefficient and non-functional objects before observing their effects on the water level. Interestingly, the crows did not learn to discriminate between functional and non-functional objects and materials when there was an arbitrary, rather than causal, link between object and reward. This finding suggests that the crows' performances were not based on associative learning alone. That is, learning was not guided solely by the covariation rate between stimuli and outcomes or the conditioned reinforcement properties acquired by functional objects. Our results, therefore, show that New Caledonian crows can process causal information not only when it is linked to sticks and stick-like tools but also when it concerns the functional properties of novel types of tool.

http://www.plosone.org/article/info%3Ad ... ne.0026887

Samsa might have something to say about some of the inferences they made.

natselrox wrote:Interesting new research.

New Caledonian crows were presented with Bird and Emery's (2009a) Aesop's fable paradigm, which requires stones to be dropped into a water-filled tube to bring floating food within reach. The crows did not spontaneously use stones as tools, but quickly learned to do so, and to choose objects and materials with functional properties. Some crows discarded both inefficient and non-functional objects before observing their effects on the water level. Interestingly, the crows did not learn to discriminate between functional and non-functional objects and materials when there was an arbitrary, rather than causal, link between object and reward. This finding suggests that the crows' performances were not based on associative learning alone. That is, learning was not guided solely by the covariation rate between stimuli and outcomes or the conditioned reinforcement properties acquired by functional objects. Our results, therefore, show that New Caledonian crows can process causal information not only when it is linked to sticks and stick-like tools but also when it concerns the functional properties of novel types of tool.

http://www.plosone.org/article/info%3Ad ... ne.0026887

Samsa might have something to say about some of the inferences they made.

To be honest, after reading through the article a couple of times I was perplexed. I couldn't figure out what exactly they were trying to demonstrate by disproving an "associative learning" explanation, as I wouldn't have assumed that the crows were learning due to associative learning anyway.

I had then planned on writing up a post discussing the problems I thought existed in the study, until I quickly thought I should check the authors - only to realise I knew a couple of them. This left me in the position where I know that they know more than me, but I disagreed with their arguments.

However, just now I realised that they were disproving the 'associative learning' account in order to support a 'causal learning' account - in other words, classical conditioning isn't enough to explain the results, but it also requires operant conditioning (which I would have assumed to be the case anyway). Their general conclusion is what I would have expected; that the crows have no innate or specific ability to process information of stick tools, or the effect of stones on the environment, but rather they generate this understanding through experience with the environment.

I got thrown by the atypical terminology, but it otherwise looks like a good article to me. Thanks for bringing it up, Nats.

Mr.Samsa wrote:To be honest, after reading through the article a couple of times I was perplexed. I couldn't figure out what exactly they were trying to demonstrate by disproving an "associative learning" explanation, as I wouldn't have assumed that the crows were learning due to associative learning anyway.

I was thrown off guard by that as well. That qualifying statement seemed somewhat unwarranted (at least in the abstract).

I had then planned on writing up a post discussing the problems I thought existed in the study, until I quickly thought I should check the authors - only to realise I knew a couple of them. This left me in the position where I know that they know more than me, but I disagreed with their arguments.

However, just now I realised that they were disproving the 'associative learning' account in order to support a 'causal learning' account - in other words, classical conditioning isn't enough to explain the results, but it also requires operant conditioning (which I would have assumed to be the case anyway). Their general conclusion is what I would have expected; that the crows have no innate or specific ability to process information of stick tools, or the effect of stones on the environment, but rather they generate this understanding through experience with the environment.

Now it makes sense to me! I didn't assume an innate component to this behaviour but the dismissal of associative learning sort of startled me. Now I realise that it was only to include another form of learning. Thanks for the explanation, mate!

I got thrown by the atypical terminology, but it otherwise looks like a good article to me. Thanks for bringing it up, Nats.

natselrox wrote:Now it makes sense to me! I didn't assume an innate component to this behaviour but the dismissal of associative learning sort of startled me. Now I realise that it was only to include another form of learning. Thanks for the explanation, mate!

Yeah I'm so used to reading articles using "associative learning" as an description of "learning" in general so I was immediately caught off guard. I was doubly confused because I knew Alex Taylor was attempting to figure out the learning mechanisms behind certain intelligent behaviors of the crows..

Oh, and for those interested, the study included videos:

http://www.plosone.org/article/info%3Ad ... 0026887#s5

SUPPORTING INFORMATION:

Movie S1.

Discrimination between sand and water (Experiment 4). Fourth trial of Laura when faced with a tube filled with water and one filled with sand.

(MP4)

Movie S2.

Discrimination between heavy and light objects (Experiment 6). First trial of Mimic when faced with light and heavy objects of the same size and colour.

(The links are direct downloads of MP4 files).

Nora_Leonard wrote:

The_Piper wrote:What is it sliding on? Maybe it's trying to work at the item in some way? Maybe it's just sledding.

I really liked this clip. Although from the way the crow is working to separate his 'sled' into constituent pieces I do think the sledding is inadvertent, as much as it has entertained us all. And for all we know it might have gone on to make something really clever out of the parts.

For me it looks like the crow is pecking on the side to get it moving to, as onyx8 said, catch some gnarly rides.


But I know that is wishfull thinking.


But maybe...

I watched two crows "torment" ? a squirrel on the ground earlier today. Unsure if squirrel was startled from a fall or too young or inexperienced to run away quickly up the next tree. It just sat there for about a minute and the crows took turns and flew up close and scared the poor thing several times, before squirrel (a grey one) opted to get going and away from them.

I saw a couple of crows dropping some nuts or something (something roundish and black, nutsized) on the road right outside of where I live. One of the crows wasn't happy where its "nut" had rolled so it was picked up again, and put in a better position on the road. I didn't stay to see if they were successful or not.



http://www.bbc.co.uk/nature/18025926

Crows recognise familiar human voices and the calls of familiar birds from other species, say researchers.

The ability could help the intelligent birds to thrive in urban environments; using vocal cues from their human and avian neighbours to find food or be alerted to potential threats.

The team used recordings of human voices and jackdaw calls to test the birds' responses.

Going back to your BBC link from last year, akigr8:
http://www.bbc.co.uk/nature/14897544

The zoologists suggested that larger groups were more likely to contain a "skilled individual" that could operate the device for the benefit of the whole group.

I'm confused. Are they saying that this one bird stands at the machine, operating it just so that all the other birds can pick up food? Or does it simply show other birds how to operate it by (undeliberate) example, then let them get on with it? If it is the former, the only explanation I can think of is that it is playing with the machine, and doesn't mind the others benefitting.

PS, thanks for the link.

DavidMcC wrote:Going back to your BBC link from last year, akigr8:
http://www.bbc.co.uk/nature/14897544

The zoologists suggested that larger groups were more likely to contain a "skilled individual" that could operate the device for the benefit of the whole group.

I'm confused. Are they saying that this one bird stands at the machine, operating it just so that all the other birds can pick up food? Or does it simply show other birds how to operate it by (undeliberate) example, then let them get on with it? If it is the former, the only explanation I can think of is that it is playing with the machine, and doesn't mind the others benefitting.

PS, thanks for the link.

It's simply a statistical effect. All they are claiming is that an advantage of living in social groups is that as the number of individuals increase, the probability of an individual in that group being able to solve a specific task also increases. This is the idea of the "pool of competence" that the authors mention. I don't think it necessarily means that only one individual can perform the task, but presumably the number will tend to be limited to a select few (but this case is not focused on teaching or social transmission, so I'm not sure if they use that skill for this task). For those that do it, it's hard to tell whether it's for the benefit of the group or simply their own amusement - further testing would need to be done to be sure, but when the authors mention the "benefit of the group", the "benefit" is in having an individual who can perform a task that produces positive outcomes (e.g. access to food for everyone else), and not necessarily a claim that they are doing so with the welfare of others in mind.

Mr.Samsa wrote:For those that do it, it's hard to tell whether it's for the benefit of the group or simply their own amusement.

So, the answer, then, to my question is "We don't know".
If it is for the benefit of the group, there must be specialisations within the group, as in meerkat groups, where one or two will look after the babies, and another will climb a tree to act as lookout, while the others are foraging in the knowledge that the lookout will warn them. This involves a kind of rota, because it isn't always the same individuals every day doing these chores, of course (otherwise they would complain and/or rebel).

Has the kind of multi-year, detailed study of crows comparable to Cambridge U.'s study of meerkats in South Africa been carried out, or planned? If that was done, they could make a TV series out of it, maybe. It would be better than some awful fictional film "Crows", portraying crows as the worst threat since AQ!

... I suppose the big problem with studying crows is that they won't stay on the damned ground!