One Trillion Smells...

For decades scientists have claimed that humans can discriminate only 10,000 different smells. But a new paper has shown that to be a myth.
24 March 2014

Interview with 

Leslie Vosshall, Rockefeller University


A man's nose.


For decades scientists have claimed that humans can discriminate only 10,000 different smells. Now a new paper published this week in the journal Science suggests that this is a myth and the number should be not 10,000 but at least a trillion. Leslie Vosshall, from Rockefeller University in New York, spoke to Chris Smith...

Chris - So, where did we get this number of 10,000 from in the first place?

Leslie - It came from an influential paper from 1927 where some psychologists were sort of mulling what the number should be and they came up with a theoretical number of a little over 6,000. And subsequently, that was just rounded to 10,000. It's had this pervasive pall over human confidence about how good our sense of smell is. So we've, for almost 100 years just felt that we're terrible smellers.

Chris - But some people as we know we smell great. No, I'm just kidding. So, when we're actually talking about smelling, what's actually going on when we smell something?

Leslie - So, it starts with inhaling. So, you inhale a bit of air that has smells in it - good smells, bad smells, food smells, flower smells, excrement smells and all of those smells are odour molecules and they land on a little strip of tissue in the top of the nose - the roof of a nose. There, they interact with receptors. Humans have about 400 different receptors and so, that's why the number of 10,000 never made much sense because we can see millions of colours with just 3 receptors in our eyes - in our retina. And so, if we have 400 receptors, right so a 100 times more than the eye, we should be able to smell much more than 10,000 smells.

Chris - So, when you say receptors, these are chemical docking stations, sitting there, looking for the molecules that we're sniffing in and interpreting them and then sending those signals on to the brain.

Leslie - Correct.

Chris - So, how did you then say, "Well, this must be wrong. Let's see how many smells a person really can discriminate"?

Leslie - So, the study was designed by a senior scientist in my lab, Andreas Keller. And so, he was playing with this idea, "Well, we can't bring humans in and have them smell 100,000 or a million or a trillion odours, but we can sample a subset of all possible odours that exist." And so, he started mixing odours together in mixes of 10, 20 or 30. When you do that, you get strange esoteric smells that don't particularly smell like anything, but you can have people tell them apart. So, does strange mix A smelled different from strange mix B. And people were remarkably good at this. So, a rather large proportion of people could tell many of the mixtures apart. And so, we could then extrapolate to what is the theoretical minimum of smells that humans can discriminate. We come up with a conservative estimate of 1 trillion.

Chris - How did you account for some people having a slightly more sensitive or acute sense of smell compared with others.

Leslie - That's a great question. So, we only enrolled people on the study who had at least the basics of a sense of smell. We exclude people who are terrible or unable to smell. But then we enrol a good cross-section of metropolitan New York city. So, we have smellers who are - our least good subject could only discriminate we think, 70 million smells. And then our best smeller was many orders of magnitude better, 1028 smells. And so, the discrimination of our subjects is across a broad spectrum, but the average for this human population is where we come to the trillion.

Chris - Did you have representation of different ethnic groups because is there evidence that in the same way that we all look different that different ethnic groups may be tuned or genetically programmed to be able to sense certain smells better than others?

Leslie - There is a lot of variability in humans and that cuts across whether you smoke, what your gender is. There's a little influence of race or ethnicity. Because we're in New York city which is a big multi-cultural place, we always strive to enrol the most diverse possible group. So, that's what we've done here - men, women, people of all ages and races.

Chris - Any professional wine tasters on the panel?

Leslie - We hope not. These are untrained members of the public. Presumably, the experts would do even better than the non-experts.

Chris - So, what are the implications of this then apart from the fact that all of the textbooks are wrong, all of the scientific research that quote this "10,000" number to date is wrong and you've got to put that right? What is the implication?

Leslie - We're really excited and I think everybody who read the story is excited that we've corrected this longstanding fact that was a made-up number. I think the implications are that it brings the sense of smell up to the level of the other senses. We don't need to be terribly insecure that we're terrible smellers relative to our house pets. I think it should make people more conscious that they can do much better than they think they could. I think expectations play a huge role in people's lives and I hope that people will, starting today or whenever they heard about the story, really engage more with the olfactory world.

Chris - It's ironic isn't it that if you ask most people, if you had to surrender a sense, which would it be? Most people say that they would give away their sense of smell, judging it to be far less important than it really is.

Leslie - They always do but if they did that experiment, they would end up - it has huge effects on mood and how you eat, and people get depressed if they can't smell. So, I think it's the least loved and least appreciated sense, but we think it's actually - it's importance is huge.

Chris - Now, what about other animals though because we're accustomed to the fact that dogs we know are very good at finding things - If you throw a ball for a dog, it can find it even in longer hours and even in the dark, despite the fact that it clearly can't see it. Dogs must have much better noses than us. So, what implication does this study have for them? Are we catching them up?

Leslie - I think we are catching them up. So, we talked about the receptors in the nose. So, dogs don't have a thousand times more receptors. They have probably at most, 4 times as many receptors. I think the difference between dogs and us is that they really care about their sense of smell. They use it. Noam Sobel did an experiment where he put undergraduates on the ground and have them search for chocolate and they were very good at it. So, I think it's mostly just - we are not using our sense of smell to its fullest capacity.


Add a comment