Day Fifty Five

Your whole life shows in your face - be proud of that.

Lauren Bacall

Humans and animals are programmed from birth to look at the human face and deduce moods from what they see. 

 The eyes have it.   If I cover my eyes, I can get much closer to the birds I’m watching than if I advance, however stealthily with my face uncovered.

Birds look for the ‘eagle eyes’ or ‘hawk eyes’ of predators to reduce the risk of becoming prey.  If they can’t see the eyes, they don’t see the risk - although of course even if you wear a paper bag over your head, even a flock of bird-brains will take to the sky if you move suddenly in their direction.

At Cambridge University, a group of researchers trained sheep to recognise President Obama and actors Emma Watson and Jake Gylenhaal. 

Don’t ask me why. 

Jake Gylenhaal

Perhaps they wanted to discover why facial recognition software, first developed in the 60’s by Woody Bledsoe, Helen Chan and Charles Bisson, has taken over 50 years to develop - and still isn’t perfected, according to the BBC’s technology correspondent Rory Cellan-Jones. 

‘The more images you ask computers to process, the less accurate they tend to be.’ he says.

So the answer is clear: instead of dodgy software in top security buildings, just install a sheep.

The human brain is so attuned to facial patterns that we even see them in things which don’t actually have faces.  This is called “pareidolia”.

Despite the posh word, we’re not that good at identifying unfamiliar faces. 

When researchers presented guinea pigs (the human kind not the little squeaking creatures although  maybe the humans were squeaky creatures too) with two photographs and then asked them to differentiate on second viewing, 20% of people could not.

I’m hearted by this as it proves I am not alone.

I can’t tell the difference between Natalie Portman and Keira Knightley, however many times I see them and as every male actor between 20 and 35 has  a 4-day growth of beard, they all look alike to me.

I find it much easier to recognise a smiling handbag.

So who’s better at this face recognition thing, humans or computers?

“It depends on the task,” says Rob Jenkins, psychologist from York University.

“If it’s a human who knows the face they’re looking at, bet on the human every time. 

If it's a face that’s unknown, the machine has the edge.”

Okaay, let's recap:  humans who already know the face they are looking at, can recognise it?  

Did we really need research to tell us that?

And none of us can beat the sheep.