Friday, June 5, 2009

Rooks use tools in the lab!

An Economist article reported this week on research published in Proceedings of the National Academy of Sciences showing that rooks, a member of the Corvidae family of birds, can use simple tools in a lab setting. They’ve never been observed doing so in the wild.

Researchers at Cambridge University placed food (a waxworm) on an out-of-reach platform. The birds could access the food by dropping a stone into a vertical tube, which provided enough force to “collapse” the platform, thus bringing it into reach (confusing, I know…see diagram).

The subset of successful rook-participants (four birds out of the twelve who were given an initial try) were then given more challenging tasks, such as figuring out how to drop a stick into a tube.

When researchers decreased the tube diameter, the birds figured out that they needed to select a smaller stone or stick in order to make the device work.

They were also able to select the right tool for a particular job: When given a stone that was too big to fit down a tube, for example, they selected a stick instead.

In later, more complex tests, rooks were able to use a “hook tool” to grab a bucket containing a waxworm. They were able to distinguish between different hook-like shapes and select the appropriate one for a job. Finally, they were tasked with creating their own hooks with which to extract the bucket. All four of the original rooks (Connelly, Monroe, Fry, and Cook) were successful.

According to the researchers, these later tests provide evidence of “insight,” or the “sudden production of new adaptive responses not arrived at by trial behavior,” because the rooks were able to transfer understanding of an original task to novel situations.

Much research has suggested that tool use is the main driver for the evolution of advanced physical intelligence. This study, claims its authors, contradicts that hypothesis. Rooks don’t use tools in the wild, yet they are able to use them when they need to, suggesting that tool use is simply a by-product of a more general “cognitive tool kit.” It is something that can be tapped into, depending on context.

Rooks’ “cognitive tool kit" may have developed because they are innovative in other ways in the wild—they’ve evolved the same general capacity as tool-users through other means.

To me, these two hypotheses seem more additive than contradictory. It makes sense that tool use has pushed the advancement of greater complexity in many animal species. But similarly, other complex, non-tool related actions—such as those performed by rooks in the wild—might push the same abilities. If different activities “push” animals in similar ways, it makes sense that they may develop similar types of complexity. I don’t think there’s a need for a war over tool use as a “driver” versus a more general “cognitive tool kit.” Many different activities, including tool use, probably drive changes in intelligence.

Displays of tool use in an increasingly broad range of animals should make us think about our current conceptions of intelligence. Multiple species of birds, for example, now appear to have out-performed chimps in a variety of tool-use tasks.

I think these experiments should also cause us to question our definition of “tool use,” and of tools as a marker of intelligence. As the Economist author notes, finding examples of animal tool use has become something of a “cottage industry” in the media. I’m not sure the connections we draw between different species' tool use acts are always as sound as we think. Are a rook that can create a hook out of wire, and a chimpanzee who can use a stick to procure food, really doing the same thing? To us, they are both "using tools," but cognitively, can we lump them together? I’m not sure. Perhaps fMRIs or other brain imaging techniques can or are exploring this question.

It seems like the next step for these researchers might be to change wild rooks' natural environments in an attempt to provoke tool use. (What would it take? How "close" are they?) This might provide more insight into the most unique aspect of this study--the fact that rooks are only tool-users in captivity.


No comments: