Well into the last century, comparative psychologists had animals perform arbitrary laboratory tasks unrelated to the problems they face in their natural environments. This theory-free 'behaviourism' never advanced our understanding of cognition to the degree that Darwinism has.
Evolutionary theory predicts cognitive similarities based on the relations between species and their habitats. It also tells us that if closely related species, be they octopus and squid or human and chimpanzee, show similar responses under similar circumstances, the most parsimonious interpretation is that the cognition involved is similar too. Humans and their closest relatives diverged so recently, in evolutionary terms, that it is hardly anthropomorphic to assume that shared ancestry suggests shared cognition.
He gives several examples, including the ability to recognize faces (he does not mention sheep - but as I recall they can recognize up to 50 other sheep faces):
Sold. I actually like his books - so I may be a bit biased towards his views already. Plus, he gave an excellent talk at the Darwin conference at Hofstra University last March. Read the full essay here (you will need subscription to view it). At some point we plan to invite him for our science & religion lecture series, for his work on the evolutionary origins of altruism.
A wealth of recent evidence supports this assertion, most of it discovered precisely because investigators have taken human capacities as their starting point. Only humans were thought capable of recognizing faces from the arrangement of the nose, eyes, mouth and so on. But other primates have this ability, and the same neural substrate seems to be involved3. Similarly, bonobos, golden monkeys and a variety of social mammals kiss, embrace, groom or mount their opponents after a fight. Calling this 'reconciliation', a term derived from human interaction, has proven appropriate given that these reunions alleviate stress and repair social ties4. In contrast, efforts to single out distinctly human capacities have rarely held up to scientific scrutiny for more than a decade, such as claims about culture, imitation, planning and the ability to adopt another's point of view.
Even distantly related species, such as elephants, dolphins, primates and birds, share an evolutionary history that may explain cognitive similarities, much as deep homologies in genetic instruction underlie the eyes and limbs of both flies and rodents. For example, neuroscientists first discovered mirror neurons in macaques, but have since found them in swamp sparrows, suggesting that they occurred in the common ancestor of birds and mammals. These neurons fire both when an animal performs an action and when it sees or hears another perform that action, and are thought to facilitate human imitation and empathy.