When we look for examples of intelligent animals, certain species always leap to mind. Ourselves of course, and our close relatives the chimpanzees and other primates. Perhaps the cunning corvids – crows and scrub jays – with their prodigious memories and talent for deception. Dolphins and whales are pretty bright. Many would even agree that there is a sort of intelligence governing the behaviour of social insects like ants. But sheep? Sheep are just thick. Except that they aren’t. Over the past few decades, evidence has quietly built up that sheep are anything but stupid. It now turns out that the humble domestic sheep can pass a psychological test that monkeys struggle with, and which is so sensitive it is used to look for neurological decline in human patients. Laura Avanzo and Jennifer Morton of the University of Cambridge were interested in a new kind of genetically modified sheep. These animals carry a defective gene that in humans causes Huntington’s disease, an inherited disorder that leads to nerve damage and dementia. The hope is that the Huntington’s sheep could be a testing ground for possible treatments.
For that to work, they reasoned, researchers will have to be able to track changes in the cognitive abilities of the Huntington’s sheep. So they decided to find out whether normal sheep could pass some of the challenging tests given to people with Huntington’s. If the sheep passed, that would mean that the Huntington’s sheep could be seen losing the ability as their disease progressed – and maybe regaining it if any treatments worked. So Avanzo and Morton put seven female sheep through a series of increasingly tricky challenges. In one test the sheep walked into a pen that contained two buckets, one blue and the other yellow, with some food in the blue one. Over the course of a few trials they learned what was going on and always went to the blue bucket. When the researchers put the food in the yellow bucket instead, the sheep changed their behaviour accordingly. They also mastered a subtler game in which the food was still in one of the buckets but the clue to its location was the colour of a cone placed nearby, not the colour of the bucket itself.
Next Avanzo and Morton stepped up the intellectual pressure, trying the sheep on intra-dimensional and extra-dimensional set-shifting. These tested the animals’ ability to shift their attention, something that requires a high level of mental control. In intra-dimensional set-shifting, the sheep still had to choose a bucket based on colour, but the set of colours was different: instead of blue and yellow, the choice was purple and green. Humans find this pretty easy. Extra-dimensional shifting is harder, as the sheep had to ignore the colour of the objects and instead focus on their shapes. In a touching piece of scientific understatement, Avanzo and Morton note that their decision to do these tests “was driven more by curiosity than expectation”. Humans and other primates can do set-shifting, but other large animals struggle with it – although researchers have persuaded mice and rats to do it. The task relies on the prefrontal cortex, a part of the brain that is much bigger in humans than other animals.
Impressively, the sheep passed the tests, learning to attend either to different pairs of colours or to the objects’ shapes as necessary. As well as being good news for the study of Huntington’s disease, it’s one more step towards rehabilitating sheep’s reputation. It really is about time we stopped making fun of sheep. They can not only recognise each other’s faces, especially sheep they are socially close to – they can remember significant others for at least two years. They can also discriminate breeds, preferring to look at their own. What’s more, there is evidence that they can group plants by family and memorise the correct route through a maze. They have sophisticated social lives too: rams become long-term buddies and stick up for each other in fights. There are even claims that sheep in the UK have learned to cross cattle grids by rolling across them, but further research may be needed on that point.
Source: PLoS One
February 22, 2011
Original web page at New Scientist