Animal Intelligence - information, images, and videos.

[brotherbear]

http://arstechnica.com/science/2016/01/s...rain-size/ 
 
SCIENTIFIC METHOD —
Social carnivores aren’t smarter—it’s all in the relative brain size
Despite their bulk, bears are the champion puzzle solvers.

JOHN TIMMER - 1/28/2016, 12:07 PM

Animal intelligence varies widely. Some have cognitive abilities that were once thought to be limited to humans, while others seem to act purely on instinct. It's not simply a matter of having large brains; birds don't have especially large ones, but they can master complicated problems or learn the solution from others in their social network.

So what can explain animal intelligence? One general trend that has been noted is that the size of the brain relative to the rest of the body seems to matter. Birds may not have big brains on an absolute scale, but their brains are relatively large compared to their body mass. Others have also noted that lots of the animals we consider smart seem to operate in social groups. These include birds, primates, elephants, and dolphins.

A new study looks at problem-solving across a wide range of carnivores and finds mixed support for these ideas. Belonging to a social group didn't seem to make a difference, but having a large brain to body ratio did. The surprising (or perhaps worrying) thing is that the brain to body ratio was high in some of the biggest carnivores tested: bears. 

The approach used here was simple, if limited. Researchers had devised a metal puzzle box that they could put food into. They took versions of this box, scaled for animal size, to zoos, where they handed it over to 140 different carnivores, which collectively represented 39 different species from nine families. They then tracked how long the individuals took to solve the problem (if they did at all), along with a variety of other measurements. Over all, 23 of the species represented managed to solve the problem at least once, with 35 percent of the individuals being successful.

Within that average, however, were some clear winners and losers. Befitting their reputation for having a "will do anything to get food" attitude, 70 percent of the bears managed to solve the problem, an average brought down by the southeast Asia's sun bear, the smallest of the lot. Raccoons and their kin also performed well, with half of them successful; weasels of various sorts managed just under half. When looking for smarts, however, you can skip the mongooses, which completely failed to get at the food even once.

(Enigmatically, nine animals managed to open the box but didn't bother to take the food inside it. Eliminating these cases did not change the results.)

One of the striking things about that is that the mongooses were a social species. Other social species (like wolves) did reasonably well, but in the end, it was a wash: social species had no advantage in this test. Rating the species for manual dexterity, which might make the task easier, showed no correlation.
 
What did matter was the brain size relative to body mass. You'd think something as large as a bear might need an enormous brain to compensate, but that's apparently not the case. The authors tested the volumes of some individual brain regions but found none of them showed a statistically significant connection to the success of the animal. And the weak correlation between absolute brain size and success wasn't significant when relative brain size was taken into account.



These findings are in keeping with a number of past studies outside the carnivores. It also makes sense in terms of a general trend among the carnivores: there's no relationship between relative brain size and a tendency toward forming social groups among these animals.



There are a number of limitations to this study. As the authors note, they're testing a small number of animals and subjecting them to only one test—they're not capturing all the dimensions of intelligence. But within the limits, the experiments were remarkably thorough. Zoos can also have very different environments and enrichment programs, which could affect these outcomes, but the authors included a variable that tracked the zoo in their statistical model and found it had no predictive value.



Still, the results raise some questions about the relationship between intelligence and social interactions. And some of the researchers are testing the puzzle against carnivores in the wild, which could help avoid some of these limits.



PNAS, 2015. DOI: 10.1073/pnas.1505913113 (About DOIs).



Promoted Comments

Schizoid

Ars Praefectus

et Subscriptor

JUMP TO POST

MisterAlex wrote:
Did they record/publish any footage of the animals at work? I'd love to watch them "figuring out" the puzzle box. I wonder how complex it was. Reply