Not that there is anything wrong with that…
The researchers did lots of test modifications to try and rule out any source of bias in the tests.
The main test was the test monkey picking one of two tokens. One would let the test monkey receive a piece of food. The other would not only let the test monkey receive a piece of food but also another monkey would too.
Rather than being self-absorbed and just picking any old token to get itself some chow – the test monekys showed bias for picking the “pro-social” token that lets both monkeys get food. No factors appeared to significantly affect this – except hiding the other monkey from the test monkey. This sort of reinforces the social basis for the behaviour. The monkey is hardly going to try and give food to a monkey it doesn’t know is there.
Monkeys also had some sense of stranger danger too – they weren’t likely to be socially helpful to a complete stranger monkey. Showing a similar “circle” style of relationships (family>friends>others) that is used to describe human social networks.
Go read Ed’s post for further analysis on the greater impact on animal and human behavioural sciences
(yes I’m being a little lazy, I just unpacked into my new flat and am trying to offload numerous web item thingys)