You smell fine

A few weeks ago, I came across a fascinating review article in Science with the instantly compelling title, Poor human olfaction is a 19th-century myth [1]. Like most people, I’ve been told that compared to many other animals, humans aren’t very good at smelling. Apparently this isn’t true, and stems more from 19th century notions of what human brains “should” be doing than from actual data.

The myth dates to the work of the pioneering and hugely influential anatomist and anthropologist Paul Broca, who concluded that the relatively small olfactory bulbs in humans and other primates enabled the occupancy of this space by facilities of intelligence and even free will: “…it is no longer the sense of smell that guides the animal: it’s intelligence enlightened by all the senses… This is the cause of the atrophy of the olfactory apparatus of primates” [2]. This presumed “atrophy” and poor sense of smell however, were not based on any actual data on primates’ sense of smell! The myth propagated and expanded in the 20th century, becoming dogma. Back in the 19th, even Freud weighed in, noting that the lack of a strong sense of smell contributed, in humans compared to other animals, to sexual repression and mental disorders.

The article is great, and has gotten quite a bit of popular press (e.g. NY Times, The Guardian). I’ll comment here on some neat experiments on humans’ sense of smell, the scaling of sense organs with body size, and general questions (that I don’t know the answer to) about fundamental limits on the senses.

Experiments: On the trail of the elusive banana

Apparently, actual studies of humans’ sense of smell reveal impressive abilities. We can track scent trails through grassy fields quite well, especially with training — the topic of a 2007 paper in Nature Neuroscience [3], with wonderful images of a person crawling through the grass, blindfolded and with ears covered, following a trail of chocolate essential oil:

In fact, with a nasal “prism” that mixes or merges input from each nostril, the authors showed that people use inter-nostril comparisons to aid their scent-based direction finding.

Why chocolate? A key point seems to be that animals have different sensitivities for different chemical cues. While dogs really do have excellent senses of smell, with a much higher density of olfactory receptors than people, people are better at discerning the scent of a banana. This complicates crude inter-species comparisons.

I find the recent date of the scent-tracking and inter-nostril comparison paper remarkable. It’s always a thrill to see experiments that could have been done centuries ago, but weren’t! If you’re wondering whether the experimental subjects get some actual chocolate at the end: I don’t know.

Smell and scaling

As Broca noted, the relative size of the olfactory bulb in primates is much lower than in many other animals. Does this matter? Its absolute size is large, many times larger in humans than in mice, for example. Is relative or absolute size important? The answer isn’t clear without a better understanding of what the olfactory bulb is doing, but there’s no compelling reason to think that relative size should be a determinant of sensitivity to smells.

Interestingly, the number of neurons in the olfactory bulb is remarkably similar across mammals: from humans to mice to to star-nosed moles, each contains about 10 million neurons:

How “should” olfactory sensing scale with animal size? Thinking about scaling helps us make sense of all sorts of aspects of animal structure and function, from the necessity of lungs for large animals (since surface area needs to scale with body mass) to the ability of small animals to walk on water (since surface tension forces scale with length). For smell, however, it’s not clear that there should be any pan-organismal scaling — it’s hard to see how the need to process or collect more olfactory data should be correlated with size, or other general features.

All this is a bit reminiscent of eyes: should eye diameter scale with animal size? There’s no obvious need for this; a big animal with a 2 inch eye collects light just as well as small animal with a 2 inch eye. Ever-larger eyes can gather more light, of course, and so give greater sensitivity, but presumably focusing and aberrations become harder to deal with. What do we find in nature? Larger animals do in general have bigger eyes than smaller animals, but eye size doesn’t grow proportionately with the rest of the body. As we’ve all seen, elephants have tiny eyes compared to their massive frames; whales, too. As far as I know, there’s no good explanation for the non-isometric, but not flat, scaling of eye size.

You look fine, too

Speaking of eyes: Vision is another sense for which the common wisdom is that humans are vastly outperformed by many other animals. At least in terms of sensitivity, this isn’t true. The human eye is sensitive to single photons, and the whole visual system triggers with just a few photons hitting the retina. This was discovered in beautiful experiments many decades ago that I start off my graduate biophysics course with a description of. With few-photon sensitivity, you’d think there isn’t a lot of room for improvement, and you’d be right. Animals such as cats have a reflective layer at the back of the eye that gives light two chances to be absorbed, and hence a factor of about 2 increase in sensitivity. The nighttime sensitivity of the tawny owl is about 2.5x that of humans [link] — even a stunning nocturnal predator is only 2.5x better (in terms of photon capture by photoreceptor cells) than you and me!

For vision, unlike smell, it’s pretty straightforward to trace the connections between the fundamental physics and limits on biological sensitivity — we can ask about the likelihood of detection of single photon events, for example. For smell, it’s not clear how to do this. Of course, there are “photons” of substances to smell (single molecules), but their detection probabilities must also be functions of the existence of similar molecules, diffusion, etc. Perhaps it’s possible to make decent estimates of the fundamental limits of olfaction, but I haven’t done it, and I haven’t seen it done.

One can ask similar questions of inter-species comparisons and fundamental limits for other senses. For hearing, sensitivity is frequency-dependent. I haven’t found a good graph of hearing thresholds for a lot of different species, but this one (marine mammals + humans, cats, and elephants) suggests that humans, at our optimal frequency, hear about as well as these other animals. Intriguingly, it also suggests that the minimum detectable sound should itself be a function of frequency, perhaps implying a fairly constant measurement time across species such that higher frequency notes are sampled for more cycles. It also suggests that it should be fairly easy to sneak up on a California Sea Lion.

In any case, it seems like hearing should be amenable to a deep biophysical analysis. I wouldn’t be surprised if this has already been done, but I also wouldn’t be surprised if it hasn’t!

Today’s illustration…

A peregrine falcon, based on a photo by Brett Cole. Staring at the photo, there’s more color in a falcon’s beak than I would have thought, though my painting is nonetheless too garish. Peregrine falcons, by the way, have interesting nostrils.

References

[1] J. P. McGann, Poor human olfaction is a 19th-century myth. Science. 356, 7263 (2017). http://science.sciencemag.org/content/356/6338/eaam7263/

[2] Note (3) in Ref. [1], from M. P. Broca, Recherches sur les centres olfactifs. Revue D’Anthropologie 2, 385 (1879).

[3] J. Porter, B. Craven, R. M. Khan, S.-J. Chang, I. Kang, B. Judkewitz, J. Volpe, G. Settles, N. Sobel, Mechanisms of scent-tracking in humans. Nature Neuroscience. 10, 27–29 (2007). https://www.nature.com/neuro/journal/v10/n1/full/nn1819.html