Notes on smell (Gilbert) [research notes]

I’m trying to learn more about smell.

I’ve written about olfaction a bit before, with a focus on how people use language to describe olfactory experiences. To the extent that I conduct my own research on smell, I assume that it will focus on the intersection between smell and language. But I think it will also be important––and just interesting––to develop a firmer background in how the olfactory system works.

To that end, I recently read What the nose knows, a book by Avery Gilbert. Yet the problem with just reading a book––even if one takes some notes––is that the information inevitably slips away. So I think there’s considerable value in writing a summary of some of the key points I’d like to remember, because:

  • The act of writing the summary helps consolidate some of those memories.
  • The summary itself serves as a usual artifact, which I can consult later to cue those memories.

So here are some of the things I found most interesting from the book.

It’s hard to figure out the natural “categories” of smell-space

People have been interested in taxonomizing smell-space for quite some time.

Linnaeus, that lover of categories, attempted an initial classification of smell. Others followed, like Hans Henning, who created the “odor prism”––representing the six “primary” dimensions of olfactory experience––which was appealing conceptually but (apparently) underwhelming from an empirical perspective. Later still, Ann Noble created the “Aroma Wheel” for wine, which apparently is still used by wine experts.

Yet the challenge is that––with some exceptions, like compounds containing sulfur––the relationship between the structure of an odor molecule and the olfactory experiences it elicits is hard to predict. Again, this is not to say there’s no systematicity whatsoever. But at least as presented in the book, it seems that––compared with domains like color or sound––the fundamental “dimensions” of olfactory experience are harder to categorize.

People are actually pretty good at smell

There’s a common belief that humans are bad at smelling things. And it’s true that we have fewer olfactory receptor genes than mammals like rats. But humans display some surprisingly good olfactory abilities.

  1. Parents can apparently pick out their infants’ diaper in a line-up (and they also find it less unpleasant to smell).
  2. Humans can also train themselves to recognize the volatile compounds in certain drugs (a task usually relegated to detection dogs, e.g., at the airport).
  3. Humans can also improve at smell; further, smell professionals (like sommeliers and perfume experts) display systematically different brain activity in an fMRI scanner than novices during a smell task––experts see more blood flow to the orbitofrontal cortex, which is sometimes associated with decision-making.

Humans have two olfactory pathways

There are two distinct ways of smelling: orthonasal (where odor molecules enter through the nose) and retronasal (where odor molecules are sent back up through the nasal pathway during the act of chewing and swallowing).

This latter pathway] is apparently particularly developed in humans, and plays a big role in our experience of flavor (as contrasted with the simpler experience of taste).

There was a brief surge of interest in smell at the cinemas

Gilbert also writes about the development of Smell-O-Vision (along with its competitor, “AromaRama”), a system that would pipe odors into a movie theater to accompany a film.

There was considerable buzz around the release of this system in late 1959, and also substantial financial investments. But sadly for its inventors, it just never really took off; this may in part be to a more disastrous premiere by AromaRama, which soured the public on the idea of smell as a cinematic experience. I found this part of the book quite sad: Hans Laube, the creator of Smell-o-Vision, struggled to recover financially and professionally from the failure of his invention.

Writers typically write more about vision than smell

Perhaps unsurprisingly, attempts to characterize literary works in terms of sensory characteristics suggest that writing is dominated by vision. Some writers, like William Faulker and Emily Dickinson, are well-known for their use of olfactory and gustatory language, though as far as I can tell, no one has conducted a serious empirical investigation of this question. This is something I’m hoping to do.

Concluding thoughts

Overall, I enjoyed and recommend the book. Gilbert obviously covers other topics as well, and the writing is (usually) entertaining. My next read is Smell: A very short introduction.

Written on June 18, 2022