"Anyone who has spent time working with sheep will know that they are intelligent, individual animals who are able to recognise their handlers", said Prof Jenny Morton, who led the study. Later, they were able to recognize the images for which they had been rewarded.
Every time an animal picked the celebrity face instead of a different image on a second screen, it would get a food reward. During one set of trials, they were made to discern between their handler (whose photo they'd never seen before) and a new face. Researchers from Cambridge's Department of Physiology, Development and Neuroscience trained eight sheep to recognise the faces of four celebrities from photographic portraits displayed on computer screens.
A sheep "model" of Huntington's disease has been bred, displaying similar brain and social changes as witnessed in human patients.
"This ability has previously been shown only in humans", the scientists write.
Sen. Rand Paul Recovering from Brutal Attack Outside His Kentucky Home
Stafford said Paul had three displaced fractures, which can lead to life-threatening injuries, as well as lung contusions. Rand Paul said Monday that attacking the Kentucky Republican had "absolutely nothing" to do with politics.
Over a series of four training sessions, the sheep's ability to choose a familiar face, represented by one of the four celebrities, over a completely unfamiliar face improved. Celebrity profile photos were randomly paired with images of one of 62 objects, all head-sized but lacking faces. "Either the human face is similar enough to the sheep face that [it] activates the sheep face-processing system, or human-face recognition relies on more general-purpose recognition systems". After training, the sheep were shown two photographs - the celebrity's face and another face.
They then challenged the animals again, this time by showing them a picture of the same celebrity, but using a new photo of their face tilted at an angle.
As expected, the sheep's performance dropped, but only by about 15 per cent - a figure comparable to that seen when humans perform the task. "I guess they have extended our work to show that sheep generalize viewpoints of the faces, which does require a rich representation of the identity".