Smelling like a dog

Pretty much everyone (excluding you and I, of course) agrees that, for humans, smell and taste are the minor senses. Read a psychology or physiology text and you’ll see what I mean – huge attention paid to vision, then hearing, then touch and then …. ok, I suppose we should tack on half a page about smell and a few lines on taste, the latter sometimes accompanied by the now discredited tongue map of taste perceptions.

There’s no doubt that the human chemosenses are minor if judged by the amount of attention we have paid to them in the past. How else could we possibly explain that something so apparently fundamental as which parts of the tongue were sensitive to which tastes was transmitted through generations in error? And as recently as twenty years ago, quite a few taste scientists firmly believed that a total of four primary tastes was it. Now we can barely open a journal without falling over evidence for some new basic quality – fat, starch, umami, calcium, rubber. Ok, only my own data, rather hastily collected late one night point to the last one.

There is really nothing very unique about the chemosenses – accepted ideas are overturned in science all the time. But given the relative paucity of research in this area compared to other sensory systems, it would seem that there is a greater risk that chemosensory scientists will spend their time looking for new receptors or ion channels or orbito-frontal Christmas lights rather than going over old ground. Not many of us want to spend our time navel-gazing in front of the fire, sherry in hand, wondering what we actually know after all. Surely that’s the job of those Cartesians in the philosophy department on the floor below.

Which makes it all the more unusual and refreshing when someone does do this. At the recent
Association for Chemoreception Sciences meeting (, Matthias Laska of Linköping University in Sweden was motivated to question something that we all know to be ‘true’ [1]. This is the almost self-evident fact that many mammals – the dog is the foremost example – are much more sensitive to smells than are humans. You believe this, don’t you? And the reason is ….. ? If you’d have asked me a month ago, I would have agreed with you, and I would no doubt have said that I’ll be able to dig up the data – just give me a day or two.

Laska did do the necessary olfactory archaeology and the evidence is, in fact, …. missing. In other words, he did not find even moderately strong evidence for human inferiority in olfactory sensitivity. Laska examined published detection threshold values (that is, the lowest concentration of a compound that can be detected) for humans and compared them with those from a variety of other mammals. The list included dogs, rats, mice, bats, pigs, primates, seals and otters – so at least some mammals who seem to spend their days snuffling along the ground or twitching their noses. Because methods often vary, he took a quite conservative approach to the comparison, and only contrasted mean human threshold values (which would contain high as well as low values) with the lowest individual values reported for each other species.

The interest began at the onset of this research. While he estimated that thresholds for around 3300 odourants had been determined for humans, this number quickly fell for other mammals: a couple of hundred for different primates, 72 in the mouse, 45 in the rat, and a grand total of 15 odourants tested with dogs! In the comparison of actual threshold values with rats, bats, mice, monkey and otters, humans come out very well on the majority of odourants on which they and we have both been tested. And dogs? Well, they beat us on 10 of the 15 odourants for which there are data in common. They are mostly better than us at detecting carboxylic acids, for example, but much worse for alcohols and slightly worse for the few acetic esters on which we have been compared.

But the key issue here is not which compounds are higher or lower, but rather that the total number of odourants tested on dogs is so low, that even the slight advantage that dogs have in number of odourants for which they have tested as more sensitive is insufficient to be the basis of a supposed fact regarding dog superiority.

So, from where does this notion arise? Clearly, not from the scientific literature. My guess is from watching dog behavior. When your nose is that close to the ground, there’s an awful lot of information that you can gather that never reaches human noses, held aloft as they are. And let’s face it, dogs seem so
very interested in all sorts of smells. As well, there are scientific data that, if not exactly proof, very strongly suggest this interpretation. As reported in an earlier blog (, a clever experiment by Porter and colleagues [2] showed in 2007 that humans, like dogs, could successfully track a scent in an open-air environment once their noses were down amongst the grass.

All that appears to be left to our canine companions is their much better ability to recognize odours, including those of other dogs and multiple humans, and also their skills at being great detectors of drugs, explosives and so on. But even shortly after birth, babies can recognize their mother’s underarm odour, distinguish it from that of other mothers of newborns [3], and the rest of us are generally able to distinguish scents from older and younger people [4].

In contrast, though, humans are notoriously poor at odour identification. We can’t reliably even label foods that we eat regularly purely by their smell. Unless, of course, we are trained. There is no possibility that you or I, if we aren’t trained, can identify by smell the origin of a wine, coffee, or tea, but there are those who have trained for extensive periods that can do just that. Moreover, I can smell two wines and know that they differ but not know in what way. In contrast, my wine enthusiast friends or trained wine judges can immediately spot the “sweaty saddle” note in one and the “stewed prunes” in another, and this ability also serves to help them recognize each wine next time they encounter it. Indeed, there are few odour discrimination, recognition or identification tasks that humans do not seem capable of performing given sufficient training. It’s possible that if we spent the same amount of time as dogs sniffing each other’s ….. you get the picture.

At this same recent
AChemS, there was another intriguing presentation that gave us further reasons to be proud of our own sense of smell. The reasons for having two eyes or two ears are obvious, namely stereoscopic vision and binaural hearing, both of which help to locate the origins of shapes and sounds in out environment. But two nostrils? Data presented by Wang and Chen from Baylor College of Medicine in the USA [5] show that sniffing a pair of odourants with both nostrils enhances detectability of an odourant, relative to sniffing the same odourant with both nostrils. Hence, pulling in different smell information with each nostril stimulates a comparison process that “sharpens” the information available. Not such a minor sense after all.

1. Laska, M., Busting a myth: Humans are not generally less sensitive to odors than nonhuman mammals, in Association of Chemoreception Sciences. 2015: Bonita Springs, Fl, USA.
2. Porter, J., et al.,
Mechanisms of scent-tracking in humans. Nature Neurosci., 2007. 10(1): p. 27-29.
3. Cernoch, J.M. and R.H. Porter,
Recognition of maternal axillary odors by infants. Child Dev, 1985. 56: p. 1593-1598.
4. Mitro, S., et al.,
The smell of age: perception and discrimination of body odors of different ages. PLoS One, 2012. 7(5): p. e38110.
5. Wang, J. and D. Chen,
Stereo olfaction sharpens sense of smell in Association for Chemoreception Sciences. 2015: Bonita Springs, Fl, USA.