One of the biggest evolutionary changes in our species took place under the skin. Our immune system has rapidly evolved to deal with the changing conditions we met as we migrated out of Africa. We even pinched some genes from the Neanderthals to help us fight disease. New research suggests we might have also evolved another trick to stop illness: our sense of smell.

Sniffing out disease

The research in question was relatively straightforward. Healthy volunteers were injected with a solution. Half were given an inert saline solution, whilst the other half received endoxin. The latter causes the body to mimic the symptoms of disease.  Photographs and odour samples were taken of each group and then shown to participants.

These participants were sitting in an MRI at the time, allowing the scientists to track how their brain responded to sniffing and looking at sick people. Curiously, it responded more to seeing/sniffing the sick people. This was promptly followed by the participants ranking those ill-looking people as less attractive than their “healthy” counterparts.

The results seem fairly clear. People can sniff out and (subconciously) identify those riddled with disease. And are driven to avoid them. It seems like it would be a fairly good evolutionary advantage for our ancestors to develop. Although, it won’t stop my girlfriend trying to snuggle up with me whenever she’s ill.

The results of trying to sniff out disease
The results of trying to sniff out disease

Of course, it’s never quite that simple. I think this study isn’t quite strong enough to start drawing far-reaching conclusions from it. Notably, the study only involved 30 people, 15 in each sample group. Plus the volunteers were from a Western university population, an issue I’ve complained about before.

But then, I wouldn’t throw out this research either. Neither researchers nor participants knew which samples were ill and which weren’t. This helped ensure it was free from bias. Plus, the use of MRI scans helps confirm there’s an actual effect here. All in all, I think this is a great first study; but more work is needed.

Evolving to be sick

So, preliminary work suggests that humans evolved to sniff out disease. But what if it was the other way around? Some have claimed that humans might have evolved exaggerated or easily detectable symptoms of disease. A big piece of evidence for this idea is that symptoms often appear exaggerated in social settings. This is taken to indicate that they serve some sort of signalling function, perhaps encouraging others to care for the sick person.

Whilst this view has been bubbling below the surface for a while, it’s never found wide acceptance. I suspect because it’s a bit too hyper-adaptationist for people’s liking. It’s easy to come up with just-so stories about the great benefits a trait offers. But that doesn’t make it true. See the space ape hypothesis for a great example of this. 

And yet it doesn’t quite die off either. Perhaps this research might be the final nail in the coffin. After all, if peopl detecting symptoms causes them to find the diseased person unattractive and shun them it wouldn’t be that beneficial for the ill person to show symptoms. In fact, it might even lead to a sort of evolutionary arms race. People evolve to mask their symptoms and simultaneously detect disease. It’s bacteria versus immune system in a battle of the ages!

Of course, the real upshot here is that there are alternatives to the idea that humans evolve to sniff out disease. Thus, we should be careful we don’t fall into a hyper-adaptationist view in that regard either. Yes, it makes sense we would have developed that ability. But that alone is not enough. Neither, sadly is the current research on the subject. More work is needed.

As long as I’m not the one that has to sniff diseased samples.


Regenbogen, C., Axelsson, J., Lasselin, J., Porada, D.K., Sundelin, T., Peter, M.G., Lekander, M., Lundström, J.N. and Olsson, M.J., 2017. Behavioral and neural correlates to multisensory detection of sick humans. Proceedings of the National Academy of Sciences, p.201617357.

Tiokhin, L., 2016. Do symptoms of illness serve signaling functions?(HINT: YES). The Quarterly Review of Biology, 91(2), pp.177-195.

Related posts

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.