Forget about them crossing the road, this is the big one, folks!

A semi-serious anecdote from a former student of poultry behaviour…

The rise of facial recognition technology

Today, we live in a world dominated by technology, one particular part of which is facial recognition. Whether we are passing through an airport, walking down a city street or simply unlocking our mobile phones, our faces are being captured and used to identify us.

Facial recognition technology is based on using the geometry of the face, i.e. the relationship between features such as the distance between your eyes or where your nose sits in relation to your ears, to build a unique map, which can be used to identify you.

But it wasn’t always like this. Back in the late 1990s, facial recognition, the way we know it today, was a dream. Pure science fiction. In fact, it wasn’t until 1999 that a commercially viable product using iris recognition was available on the market.

Now, turn your attention to a young scientist, working towards his doctoral degree at Oxford University at around the same time. Poor fellow. He was investigating social discrimination in laying hens, which in plain English means: how the heck do chickens recognise each other? This might seem like something of an unusual pastime, not to mention use of taxpayer’s money, but it did have a serious point…


Cover of my doctoral thesis on chickens - Rob Gregory Author

My doctoral thesis. Yes, I did complete it and am still rather proud of it.


Facial recognition and the pecking order

Chickens have long been believed to have a strong pecking order. Indeed, the very phrase ‘peck order’ comes from work done by a Norwegian scientist, Thorleif Schjelderup-Ebbe, back at the turn of the twentieth century. I have my own thoughts on the validity of a strongly linear pecking order, i.e. one where the top dog dominates all those below it, the second in command dominates all but the top dog and so on, as indeed did Schjelderup-Ebbe, but I’ll save that for a future blog.

Suffice to say that scientists have long believed that the way chickens recognise their place in the pecking order is through the comb on their heads and to a lesser degree, the wattles that hang from their throats. The question I was grappling with, because it was indeed me who was that poor young scientist, was whether chickens used individual recognition or what was known as a ‘badge of status’. Think about a sergeant’s stripes and you’ll get the idea. You don’t need to know a sergeant individually, to know how to behave around them!


Head of a chicken showing the comb and wattles - Rob Gregory Author

Traditional wisdom says that the comb and wattles are what chickens use to recognise each other.


Where it gets serious, at least from an animal welfare point of view, is when you look at commercial flocks of hens, which can number in the tens of thousands. If chickens have to rely on individual recognition, then the cost of trying to remember several thousand one-on-one relationships just isn’t worth it and is likely to be highly stressful, which is not good for poor old Chicken-Licken. On the other hand, if they can switch to badges of status, then everything should be hunky-dory, and we can all sleep soundly at night, thank you very much.

So, there you have it. Took a long time to explain, but at least you now know why I was spending an unhealthy amount of time in the company of poultry. And as an enquiring young researcher, I had a question and it annoyed me…


How do chickens recognise faces?

At the time that the peck order is being formed, the comb and wattles are in a juvenile stage, i.e. they are still growing. And growing rapidly. The effect is like trying to remember your place when everyone around you, including yourself, has a new haircut each day. Of course, you could argue that on a day-to-day basis, the change is gradual and so the chickens don’t notice it, but as you can see from the images below, over a seven-week period, when all of the peck order action is happening, the change is really quite substantial.


Picture of a chickens head at different ages - Rob Gregory Author

The same chicken before and after puberty. Look how the comb changes but the other main features do not.


An alternative idea is that the size of the comb and wattles reflects the amount of testosterone the chicken has and so the bird ‘knows’ how strong it is without ever having to see its own features. But that doesn’t really work because there are diseases that cause the comb and wattles to enlarge and change colour, even when the amount of circulating testosterone is low.

And that’s another thing, the placement of the comb and wattles means that a chicken can almost certainly never see its own! It’s like that card game where you slap a card on your forehead and then try to guess what it is, based on what the other players tell you. I’m sorry, but there has to be a better explanation than: chickens use something that they can’t see, and which is changing rapidly, as the basis for settling all of their future arguments about food, water, access to mates etc. After all, if God doesn’t play dice with the universe, why should chickens play cards with their social status?

So, I set out to investigate, taking lots of photographs of chickens as they progressed through puberty and measuring various facial features with a ruler. No particle accelerators or whizz-bang tomfoolery here… just belt and braces British science at its best. And what did I find?


Schematic of a chickens head showing measurements made - Rob Gregory Author

Back then I was a brilliant artist as well!


Chickens have individual and distinct faces

Lo and behold, chickens have faces! Things like the diameter of the eye, the length of the nostril, the length of the ear feathers and beak length changed, on average, less than a millimetre during puberty, in contrast to the comb and wattles, which increased by several orders of magnitude during the same period. Furthermore, after puberty, the same facial features that I had measured essentially stayed the same, giving the animal a consistent set of cues with which to recognise other individuals as they progressed through life.

While this in itself, is not absolute proof that chickens are using those features to recognise other individuals, or indeed, even using individual recognition, it certainly called into question the use of the comb and wattles alone for recognition, which was good enough for me.

And of course, I took it one step further, which is where the link to modern-day facial recognition technology comes in. Consider the following paragraph lifted from my thesis:

Excerpt of text from the doctoral thesis of Rob Gregory Author

Perhaps I was ahead of my time for once? This was written in late 1999.


Unfortunately, I never did that particular piece of follow up research. If I had, then possibly I wouldn’t be sitting here today, writing this blog for you. But nonetheless, I can still get some small satisfaction out of the knowledge that my hypothesis, made on the basis of looking at the humble chicken, has since proven to be validated on a far more complex animal – us!




If you enjoyed this blog, please feel free to share it. And while you’re here, why not have a look at some of my other blogs and books?


Thank you!