When Life Hands You (S)-(–)-limonene . . . You Can’t Make Lemonade

Hello dear readers. Are you still browsing FirstNerve? All my new stuff, including this post, is available over at my Avery Gilbert Substack. Give it a look!

A Skull Full of Mush

 

There’s an interesting paper to be presented later this month at an Association for Computing Machinery virtual conference on the theme of Interactive Media Experiences.

The paper, by three engineers and a psychologist from the University of Liverpool, is titled “Predicting the colour associated with odours using an electronic nose.” The cross-modal associations between scent and color have been of interest for ages; my colleagues and I helped put them on a quantitative basis with a couple of papers in the late 1990s. The Liverpool group’s new twist is getting an e-nose to predict the color associations.

That’s all fine. It’s of a piece with attempts to predict a molecule’s odor character from physical parameters such as molecular weight, shape, charge, etc. The authors of the new study claim to find a 70 to 81% “machine-human similarity rating.” Whatever. I’m not that interested in the details of their e-nose or experimental protocols, but I’m fascinated by their conceptual point of view as expressed in this sentence fragment in the conclusions section:

Thus, highlighting the possibility of using e-noses to predict human olfactory perception and implying that the colour associated with odours are partly written into the molecular properties of the stimulus [5].

Reference 5 is to the color-odor study—using human noses—that I published with Sarah E. Kemp. In it, we discussed the psychological associations that emerged between the test odors and measurable parameters of color. Nowhere in that paper did we attribute those association to the “molecular properties” of the odor stimuli. It is a mistake for the Liverpool researchers to cite that paper in support of their claim.

The bigger problem is with their philosophical view, namely that an odor’s color associations “are partly written into the molecular properties of the stimulus.” This is a dumb but persistent conceptualization that’s popped up before. Here’s what I said about it in 2009 while commenting on a study by Mandairon et al. called “Humans and mice express similar olfactory preferences.”

What I can’t understand is Mandairon’s endorsement of a mathematical model that predicts odor pleasantness. The idea is that odor pleasantness is “partially dependent on the odorants’ physicohemical properties.” Of course this has to be true at some level: different molecules produce different smells because they have different structures. But Mandairon et al. go beyond tautology; the shared response of mice and men suggests

that olfactory preferences are indeed partially engraved in the structure of the odorant molecule

and

there is an initial part of the [odor] percept which is innate and engraved in the odorant structure.

Perceptions engraved on the molecule? This is simply a bizarre way to think. What else is engraved on a molecule of phenylethanol: Visual associations to red roses? The name of my florist? An olfactory memory of my dead grandmother?

Perceptions happen in the central nervous system of an organism. To talk as if odor pleasantness resides in the structural features of a molecule is animistic thinking, pure and simple.

My objection to Mandairon, et al. applies equally to Ward, et al. No matter how cool the math and the engineering, the idea that odor-linked human perceptions are “written” or “engraved” in the structural features of a molecule is rubbish and unworthy of a place in scientific discourse.

The studies discussed here are “Predicting the colour of odours using an electronic nose,” by Ryan J. Ward, Shammi Rahman, Sophie Wuerger and Alan Marshall, published in SensoryX ‘21: Workshop on Multisensory Experiences, together with IMX 2021: ACM International Conference on Interactive Media Experiences, June 21-23, 2021, and “Humans and mice express similarolfactory preferences,” by Nathalie Mandairon, Johan Poncelet, Mousafa Bensafi and Anne Didier, published in PLoS One 4:e4209, 2009.

The Memory Meme that Refuses to Die

 

A reader’s letter to the Financial Times relates a charming story about her son, who as a two-year-old blurted out that a lady in a store smelled like his godmother. Lo and behold, both women wore L’Heure Bleue by Guerlain.

Thirty-one years later, he pulls the same stunt at a restaurant: he correctly identifies his godmother’s signature scent on a lady sitting at the next table.

The letter writer thought she was illustrating a point made in a previous edition of the paper, namely “wouldn’t it be a fine thing if there was a good smell that reminded people of you?” In fact, she did so quite effectively. We can also applaud her son for having a good nose, a keen awareness of scent, and a long memory.

But why did the FT have to spoil it all by slugging the letter “The power of scent is like a Proustian madeleine”? There is nothing Proustian about this anecdote. The first incident shows a toddler with a precocious awareness of smell. The second shows that as an adult he retains his smell identification ability, along with a good memory.

In neither instance was he transported to a full-blown evocation of an earlier time and place—whether by the instantaneous, effortless process described by many pre-Proustian writers, or by the slow, labored process described by Proust in Swann’s Way.

My guess is that some junior editorial assistant probably read French Literature at Cambridge and simply couldn’t resist justifying all that tuition money by working in a Proust reference.

Basta. In the interest of accuracy and history, isn't it time we begin to forget Proust’s madeleine?

Yeah, About That TPS Report Cover Sheet . . .

 

Unlike ex-Initech employee Peter Gibbons, I’m a fan of TPS reports. That is to say, studies of terpene synthases (TPS) in cannabis. These are the plant enzymes that turn precursor molecules into the various fragrant terpenes behind each strain’s aroma profile. Some TPS enzymes convert geranyl diphosphate into a slew of monoterpenes, while others convert farnesyl diphosphate to a bunch of sesquiterpenes.

Identification of cannabis terpene synthases and the genes that produce them is a relatively new field. Canadian researchers led by Judith K. Booth at UBC in Vancouver were early leaders in this effort and last fall they published another study. This one, in Plant Physiology, was the subject of a “news and views” piece in the same issue, written by Marc-Sven Roell at the Heinrich-Heine University in Düsseldorf.

Like Bill Lumbergh, I have some issues with Roell’s TPS report cover sheet. Specifically, with his claims about what is required for future cannabis breeding efforts aimed at fragrance and flavor.

To predict and design cannabis smell and taste to meet consumer demands, two milestones have to be reached. First, a comprehensive understanding of terpene composition is required, which can be achieved by using quantitative terpene profiling in existing cultivars. Second the underlying molecular and biochemical mechanisms leading to these distinct profiles need to be understood.

Roell insists that we must first know everything about terpenes in all cultivars—their complete chemical composition plus their physiological means of production—before we can get around to breeding hybrids with specific consumer appeal.

Notice anything missing? How about sensory evaluation of the smell and taste of the existing cultivars? And how about relating perceived aroma to differences in terpene composition?

Roell no doubt expects to find cultivar to cultivar differences in terpene composition, which he assumes will equate to differences in aroma and flavor. But differences in chemical composition don’t necessarily translate into perceptible differences in aroma, much less differences that are meaningful to cannabis consumers. So his strategy of starting with comprehensive knowledge of terpene composition will be enormously inefficient.

This objection also cuts in the opposite direction. As a plant physiologist, Roell ought to know that even within genetically identical clones, terpene composition can vary with growing conditions. Thus, it is possible that within-clone variation could result in perceptible aroma differences.

So either way you look at it, Roell’s insistence that baseline chemical data is an absolute prerequisite for breeding better smelling cannabis is pretty weak, especially when he totally neglects sensory measurement.

In fact, I’d argue that the best way to assist breeding efforts is begin with quantitative sensory evaluation of a range of existing cultivars. Next, cross-tabulate aroma profiles with sales and consumer preference data to obtain a ranking of most-valued sensory traits. Go back to your list of cultivars and start hybridizing for likely winners. It’s the purely phenotypic selection method that worked very well for Luther Burbank in the days before gene sequencing and gas chromatography. This approach would get the program going a lot faster than Roell’s “study the hell out of everything before making a move” strategy. I think he has it totally backwards—sensory analysis should precede chemical analysis.

“Hello Marc-Sven, what’s happening?

We have sort of a problem here. Yeah, you apparently didn’t include sensory analysis in your new cover sheet on the TPS report.

Did you see the memo about this?

Yeah, if you could just go ahead and make sure you do that from now on that would be great. 

And I’ll go ahead and make sure you get another copy of that memo.”

The study discussed here is “Terpenes in Cannabis: Solving the puzzle of how to predict taste and smell,” by Marc-Sven Roell, published in Plant Physiology 184:8-9, 2020.

Smells from Beyond

 

In a series of pre-Halloween posts a decade ago, I examined ghostly smell stories and poems including examples from the Victorian English novelist Wilkie Collins, the 19^th century American writer Bret Harte, the Irish poet W.B. Yeats, and an English novel (“The Uninvited”) that was made into a movie starring Ray Milland.

I was reminded of them this week when I came across a new study: “Perceptual phenomena associated with spontaneous experiences of after-death communication: Analysis of visual, tactile, auditory and olfactory sensations.”

Have I started dabbling in occultism? Not exactly. The paper appears in Explore: The Journal of Science & Healing, published by Elsevier, which claims to address “the scientific principles behind, and applications of, evidence-based healing practices from a wide variety of sources, including conventional, alternative, and cross-cultural medicine.” Further, “It is an interdisciplinary journal that explores the healing arts, consciousness, spirituality, eco-environmental issues, and basic science as all these fields relate to health.”

OK, then!

So who conducted the study? Here’s the batting order:

Marjorie Woollacott, an emeritus professor of physiology whose focus is motor control.

Chris A. Roe, a professor of psychology at the University of Northampton in the UK who is President of the Society for Psychical Research and whose most recent publication is a chapter titled “Clinical parapsychology: The interface between anomalous experiences and psychological wellbeing.”

Callum E. Cooper, a senior lecturer in psychology at Northampton, who holds two doctorates and whose most recent publication is the entry on “Anal Intercourse” for the Encyclopedia of Sex and Sexuality.

David Lorimer, Programme Director of something called the Scientific and Medical Network, and an editor and author of many books including Survival? Death as a Transition; he is interviewed in a recent youtube video.

Evelyn Elsaesser, an independent researcher in Switzerland and author of, among other books, On the other side of life: Exploring the phenomenon of the near-death experience.

So what have they delivered?

The study itself consists of a lengthy 194-item online questionnaire that was filled out by 991 people, mostly women (85%) with a median age of 51 years. The olfactory data appear in Table 6. In response to the question “Did you smell a fragrance characteristic of the deceased which made you think he/she was present?” 28% of respondents said yes. Of these, 60% also said yes to the next question: “Did you feel that the deceased was conveying a message to you by way of this fragrance?”

The authors give us an idea of what smells were reported:

“The fragrances typically included aftershave lotion, a typical body scent, perfume or soap, but many odors were noted in the descriptions, including tobacco, food and flowers.”

Smells were the least common sensory impression (28%) reported by people who experienced ADC. The most frequent were tactile (48%), visual (46%), and auditory (44%). While noting that there is considerable variability in the representation of the individual senses in previous studies of ADC, the authors don’t have much to say about the proportions they found.

For what it’s worth, I would point out that the substantially lower incidence of smells compared to sights and sounds is consistent with what’s been reported for sensory impressions in dreams and hallucinations.

The paper is mildly interesting for the cross-sensory tabulations insofar as they relate to smells in phantosmia, dreams, and hallucinations. Whether there is anything else worth pursuing on the topic of after-death communication is, in my opinion, dubious. Interestingly, the funders of the study are publicity-shy:

Funding: This work was supported by a foundation that wishes to remain anonymous. The funding organization had no influence on the final research design, data collection, analysis, interpretation of data, the writing of the report or the decision to submit the article for publication.

UPDATE May 7, 2021

And right on cue—spooky new Netflix movie Things Heard & Seen features among other things a smelly ghost.

The study discussed here is “Perceptual phenomena associatedwith spontaneous experiences of after-death communication: Analysis of visual,tactile, auditory and olfactory sensations,” by Marjorie Woollacott, Chris A. Roe, Callum E. Cooper, David Lorimer, and Evelyn Elsaesser, published online in EXPLORE, February 23, 2021.

Keeping Up with the News

I’ve tracked new scientific findings about the human sense of smell for a long time. Rather than just file them away for my own interest, I posted one-sentence summaries of clinically relevant findings to the Disorders of Odor Perception website at hyposmia.com. I was doing so on a more or less weekly basis until about a year ago when covid hit the fan and disrupted my routine along with everyone else’s.

Over the course of the year, “regular” publications on smell science were overwhelmed by a flood of papers on covid-related smell loss. I tried to keep up with these for while, but it was simply too much to handle and eventually I stopped updating the site.

Recently, when I decided to resume posting new material to Disorders of Odor Perception, it occurred to me that the web design was hopelessly out of date: the site looked OK on laptops, but was barely legible on phone or tablet. So I redesigned the site to make it “responsive”, i.e., it adapts to the device you view it on and is (hopefully!) much easier to read on mobile devices.

As before, new content appears in three categories: Smell Loss (hyposmia and anosmia); Smell Distortion (parosmia, phantosmia, etc.); and News & Reviews.

I also updated the format of the “new content” emails. They now contain direct links to the original source items—no need to go back to website to find them. This ought to make the emails much more user friendly.

If you are a clinician or patient or just interested in the topic of smell disorders, I encourage you to sign up for the email alerts. It’s a painless way to keep up with new work in the field.

You Better Listen to the Radio

 

I’ve known Saskia Wilson-Brown since she first began organizing the Institute for Art and Olfaction in Los Angeles almost eight years ago. The IAO is now a thriving center for perfumery training, workshops, seminars, and scent-related events. Saskia’s latest innovation is Perfume on the Radio, a twice-monthly broadcast that is also available in podcast here and from the usual sources (Spotify, iHeartRadio, etc.).

I am one of the guests on Episode 5, dubbed “The Vice Show.” Saskia and I talk about the aromas of cannabis, how strain names get in the way of consumer experience, and the developing parallels between weed and aromatherapy. My segment begins at around minute 16. It’s free and it’s about 10 minutes long.

The Tao of Terpenes

 

When I moved to Colorado a few years ago I got interested in the aroma of marijuana. I founded a start-up called Headspace Sensory LLC and began a series of consumer research sniff studies. Three of these have been published so far, and they constitute the nucleus of the brand new field of cannabis psychophysics. You can find them here, here, and here. 

Along the way I’ve met many people and read a lot about the subject. My impression of the scientific and commercial “cannabis space” is that it is a mixed bag. There are some very talented, very focused people doing excellent work in areas like chemical analysis, plant genetics, and product development. On the other hand, there are lots of people whose hot air to solid content ratio tilts rather heavily toward HA. 

One thing that gets the alarm on the FirstNerve bogosity meter bleeping the loudest is the topic of terpenes. Terpenes are a chemical class of volatile molecules and they are responsible in large part for making weed smelly weedy. That much is true. But one doesn’t have to follow the online terpene trail very far to find perfumey punditry of the most egregious kind. [Take pity and refrain from naming names.—Ed.] [Okay, if you insist.]

So I was pleased when the editor of Terpenes and Testing magazine invited me to write a short article on the chemistry and perception of cannabis terpenes. It gave me the chance to describe some fundamental principles of odor perception and how they apply to the bouquet of terpenes found in cannabis flower. One important lesson: the sheer abundance of a given terpene doesn’t tell us much about its contribution to a flower’s overall aroma. 

In the article I also describe my dream of creating a cannabis aroma map on which individual cultivars are arranged according to smell similarity, and my belief that sensory evaluation—sniff testing—has a major role to play in segmenting the cannabis consumer market. 

If you would like to read it, you can download at this link.

Did Nicolas Guéguen Really Replicate the Cinnabon Effect?

Some years ago I blogged about a couple of scent-related field studies conducted by French academic researcher Nicolas Guéguen. In 2009, I wrote about a paper of his I had just come across, published in Psychological Reports back in 2001. In “Effect of a perfume on prosocial behavior of pedestrians,” Guéguen found that people were more helpful to a young woman who “accidently” dropped an item when she was wearing perfume, compared to when she was not wearing scent. I thought the paper was worth mentioning because it described a theory-free, empirical approach that could shed light on the natural history of perfume use.

Three years later, in a post called “The Cinnabon Effect Confirmed,” I wrote about another of Guéguen’s studies. This one, published in the Journal of Social Psychology, was titled “The sweet smell of . . . implicit helping: effects of pleasant ambient fragrance on spontaneous help in shopping malls.” The original Cinnabon Effect study was discovered in 1997 by social psychologist Robert Baron who found that shoppers in a mall were more helpful to a stranger when approached next to a pleasant-smelling retail shop (e.g., Cinnabon) rather than an unscented one (e.g., Banana Republic). In his new paper, Guéguen found similar results in a French mall, and extended the paradigm to include male as well as female “strangers.” I wrote approvingly that “Guéguen has confirmed the Cinnabon Effect and extended it to spontaneous acts of kindness.”

I recently became aware that Guéguen’s “sweet smell of” study has been the object of some impressive data sleuthing by Nicholas Brown and James Heathers. A new post on the excellent Retraction Watch website updates the story. In a nutshell, the pair claims that a bunch of papers by Guéguen don’t pass the smell test. The reasons for their skepticism are procedural and statistical. Guéguen is a prolific author yet one with no collaborators, and he may, according to Brown and Heathers, have skirted ethical requirements. Disturbingly, statistical patterns in his underlying data seem consistently fishy. The original B&H posts can be found here and here.

It’s important to note that Brown and Heathers do not claim outright that Guéguen’s studies are fraudulent.

We have not made, and do not make, any specific allegations of fraud, nor are any implied. The initial document that we released is entitled “A commentary on some articles by Dr. Nicolas Guéguen” and details a long series of inconsistencies in research methods, procedures, and data. The words “fraud” and “misconduct” do not appear in this document, nor in any of our communications with the people who helped with the investigation. We restrict ourselves to pointing out that results are “implausible” (p. 2) or that scenarios are “unlikely [to] be enacted practice” (p. 31).

You can download their detailed fifty-two page critique here and judge it for yourself. If you’re less hardcore, this Ars Technica piece covers the story nicely.

I think the points raised by Brown and Heathers are compelling and cast serious doubt on some, if not all, of Guéguen’s work. Accordingly, I’m putting a metaphorical asterisk on the two studies I cited, and flagging this action on my previous blog posts. Data integrity and transparency are core to the scientific endeavor, and when serious doubts are raised they need to be acknowledged.

Greener Grass Podcast: Where I Talk About the Smell of Cannabis

 A couple of weeks ago Lex Pelger, host of the Greener Grass podcast, dropped by my sensory lab in Fort Collins to tape an interview. His first questions were about my career in the science of smell and how smell works. Midway through (around the 16:45 mark) he turned to my latest research interest: creating quantitative aroma profile for cannabis strains. We discuss how this kind of sensory research can be put to use in the rapidly evolving cannabis industry. You can find the podcast here.

Frontier Psychophysics: The Prequel

Image ©Avery Gilbert

 When I moved to Colorado in mid-2015, it occurred to me that there were two local olfactory opportunities that might be commercialized. One was cow manure, the other was pot.

When the wind is right, the unmistakable scent of manure wafts over Fort Collins from large-scale cattle feedlots many miles to the east and south of town. Having grown up in California’s Central Valley, the smell of manure is, to me, familiar and not particularly disagreeable. But immediately downwind of feedlots it can be pretty intense, and the cattle industry is interested in cost-effective ways to reduce it. I did a literature search and concluded that the problem has not been adequately attacked from my point of view, i.e., as a the fragrance industry would tackle a new malodor issue. I think a combined chemical and psychophysical approach could yield commercial innovations.

But the more alluring opportunity proved to be pot, newly legal in Colorado for recreational use and an industry that is growing at an exponential rate. I sniffed my way through the wares of local dispensaries and was amazed at the enormous olfactory range on display. Some strains reeked of grapefruit, others were dense and funky. Some were almost cheesy, others mildly hay-like or even piney. From a fragrance standpoint, there is at least as much going on in weed as there is in wine, beer, coffee, or tea.

But there’s a big difference. Those other agricultural products have a long history of aesthetic appreciation. They have fully developed olfactory lexicons (think, for example, of the U.C. Davis Wine Aroma Wheel). And they have chemical standards and training programs.

Cannabis, on the other hand, maintains an outdated, underground feel. Sure, there are aficionados who can be eloquent in describing the aroma of a given strain. But there is no organized, empirically-based, agreed-upon basis for describing the scent of weed. For example, in the various “Cannabis Cup” competitions, smell is one of many dimensions on which a strain is judged, but as far as I can tell there are no specific criteria for evaluating a sample’s smell. If sheep or chickens were judged that subjectively at the county fair, there would be a riot in the livestock pavilion.

I made some inquiries at the local campus, Colorado State University. I had excellent discussions about possible collaborations with faculty in the psychology and chemistry departments. The conclusion was always the same: cannabis research was impossible. Or to be more precise, a university that takes even one dollar of federal funding is obliged to play by federal rules. That means that in order to do cannabis research, one needs a DEA license to hold and handle the drug. One can use only federal money (NIH, NSF, etc.) to carry out the research—no private or corporate funding. And finally, and most disheartening, one can only use cannabis grown at the officially approved federal plantation in Mississippi. (This product is frequently lampooned as being a throwback to 1970s era weed—the low potency, twiggy, seedy, stuff that one “cleaned” on an LP cover before rolling into a doobie.)

By this point I had landed on a simple idea—to treat locally available cannabis strains as a just another consumer product. I would do routine sensory research to characterize the relevant odor descriptors, and see whether consumers could tell one strain from another based on smell.

That was my original idea. To make it happen, I had to blaze my own trail. I found an angel investor and created a new company: Headspace Sensory LLC. I wrote an entire research plan and got it approved by an independent institutional review board. I found an excellent collaborator in Joseph DiVerdi, a Colorado State chemistry professor with an interest in cannabis and his own private lab. With his help I found a genial landlord willing to rent me a couple of rooms perfect for sniff testing.

All the elements were in place. But I still needed to find volunteer sniffers. That would prove harder than I had imagined.

“A Very Distinct Smell”

“Marijawana’s bad, and it also has a very distinct smell, okay? I’m gonna pass around just a little tiny bit. Now, I want you all to take a smell, so you know when someone is smoking marijawana near you.”

Mr. Mackey
South Park S2E3

I published a new smell study today in the scientific journal PLOS ONE: Consumer perceptions of strain differences in Cannabis aroma. You can download a free copy here. Sure, there’s some statistics and chemistry, but anyone interested in smell—or weed—can read it and understand the results.

In the spirit of Hollywood prequels, I will describe in a subsequent post how the study came about.

Spinning One's Wheels

Given my recent work on cannabis aroma for Headspace Sensory, LLC, I’ve been thinking a lot about sensory description. A short post at the Scent Culture Institute on “smell wheels” caught my eye. In it, Claus Noppeny (and/or one or more of his sidekicks at the SCI) looks at visual representations of smell, such as Michael Edwards’ Fragrance Wheel and Mandy Aftel’s Natural Perfume Wheel. (Both of these, I should note, were preceded by Ann Noble’s 1987 Wine Aroma Wheel, which really launched the modern enthusiasm for wheels.)

SCI asserts that the wheel format for displaying fragrance families (or aromatic notes in wine) has limitations, the primary one being that the circle implies completeness while allowing for no further additions (e.g., new fragrance notes). SCI immediately undercuts its latter charge by noting that Mandy is currently revising her wheel. However, the observation that wheels project an all-you-need-to-know—nothing else matters!—completeness is, I think, correct. Self-enclosed circles are a poor way to encourage new ways of thinking and perceiving.

But epistemological isolation is the least of the aroma wheel’s problems. For me, more serious problems arise from the format’s practical implications. Let’s consider three such problems.

Problem 1: Size of Slice

In a typical wheel layout, one moves radially outward from categories (“fruity”) through subcategories (“citrus”) to specifics (“grapefruit”). This creates a pizza-slice shape for each category. Some slices are a lot wider than others. Why? Because they have a lot of specifics (the Wine Aroma Wheel “fruity” category has 19), while others have a few (“woody” has just 7). Assigning a large fraction of the wheel to “fruity” implies that it is a dominant odor category, and that “woody” is relatively subordinate.

But from a sensory evaluation perspective, this is not necessarily true: a given wine may display several fruity notes, but it’s extremely unlikely that any wine will display all 19 fruity notes. Furthermore, even when several fruity notes are present, it doesn’t mean that they dominant that wine’s aroma. A Chardonnay’s fruit notes may be overpowered by oakiness.

Problem 2: Arrangement of Slices

My impression is that categories on most, if not all, aroma wheels are arranged based on intuition. That is to say, their ordering around the wheel is not based on data regarding relative similarity. More likely, it is based on the subjective impression that floral and fruity are more closely related smellwise than floral and woody.

Problem 3: Making Ends Meet

Even if there are data confirming the similarity sequence floral : fruity : woody, what justifies us linking floral to woody in order to make a circle? Would it not be more honest to portray the relationship as a linear smell spectrum? Or maybe a triangle? (But it wouldn’t look as cool.—Ed.)

These intellectual objections cannot offset the appealing simplicity and visual allure of the wheel format. But in an age of innovative data visualization, there are surely better ways to display sensory descriptors and their interrelationships.

Something in the Air

Via Gizmodo

My recent inquiries from online journalists are beginning to form a pattern and it’s not good.

First came an email from Eric Spitznagel at Vice Tonic asking about the science behind the age-old saying “he who smelt it dealt it.” His starting point was the idea that laws of gas diffusion and the concentration gradient of odor dispersion would invariably indict the smeller as the dealer. (From the published piece it appears he got this working hypothesis from an engineering professor at the University of Colorado.) My response was to distinguish between models that describe the behavior of ideal gases, and the more complicated turbulent currents and plumes found in real life. The non-ideal distribution of scented air streams is the basis for the “casting” behavior which many animals species use to localize the source of a smell (they zig-zag back and forth through the odor plume in ever-shorter tacks until they reach it). Given these atmospheric vagaries it is entirely possible that an emission from the guilty party might curl up an innocent person’s nostrils first.

Next heard from was Daniel Kolitz at Gizmodo who was putting together a “GIZ Asks” installment on the legitimate if somewhat feculent question “Why does dog poop smell bad to us but good to dogs?” Kolitz collates answers from a crack team of dog specialists and smell researchers including, beside yours truly, Alexandra Horowitz, Don Wilson, Peter Hepper, Cat Warren, and Charles “I’m publishing as fast as I can” Spence (I kid, I kid). What’s interesting is that several of the experts blithely assume that all human odor responses are cultural, while other take the (correct) view that certain smells or categories of smell are inherently (biologically) offensive. Click over to read the whole thing, but here here’s the pungent part of my answer:

Dogs don’t approach shit as an aesthetic experience—they treat it as a source of social information, like an olfactory Instagram. It answers a lot of questions: Who left it? How recently? Is the pooper healthy? We are able to extract similar information. The lingering cloud in the office restroom tells you who had lunch at P.F. Chang’s. Plumbing and ventilation rob us of the social signals in feces and leave us with mere disgust.

So where is this latest journo-trend heading? What follows farts and dog poop? I could make an educated guess, but I’ll take the lazy way out and just wait for the next email from an inquiring mind.

Going Fictional: Introducing the Nick Zollicker Stories

I’ve always been fascinated by the portrayal of smell in fiction. I made a brief study of it in What the Nose Knows, and here on FN I’ve taken a close look at various authors and how they weave scent into their novels. Among the writers I’ve discussed or quoted are E.L. James, Tom Robbins, Vladimir Nabokov, Toni Morrison, and P.G. Wodehouse. (You can find these posts via the FN Review, FN Retrospective, and Olfactory Art tags.)

In most of these works scent is mentioned in passing. It functions to characterize a person or place or, especially in the case of Nabokov, to provide a sepia-toned sense of nostalgia. Smell is rarely a central theme in fiction. There are exceptions, like M.J. Rose’s The book of lost fragrances, Süskind’s Perfume: They story of a murderer, and Roald Dahl’s brilliant short story Bitch.

One would think the world of contemporary commercial perfumery is an ideal setting for fiction. The business is an unstable blend of creative fashion and technical chemistry. It straddles the magical and the mundane. It simultaneously touts its innovation and its longstanding traditions. It is filled with characters of Dickensian dimensions.

Another setting ripe for fiction is the world of academic smell research. University labs are stocked with weirdos and drones as well as the rare brilliant scientist. Scientists share a lifestyle—monk-like yet dissipated—that is remote from the experience of most readers. Their scientific obsessions are remarkable if not often bizarre. It’s all fertile ground for fiction.

Well, somebody needed to step up to the plate, and it appears that the someone is me. I have created a contemporary smell expert named Nick Zollicker. He lives in Berkeley, California where he runs a secretive private olfactory research institute. He has wide experience in the hard-edged world of commercial perfumery yet thrills to pushing the boundaries of olfactory science.

My first two Nick Zollicker stores An Imperfect Mimic and Smothering the Savage. They are now available digitally on Amazon. (You can read them on your Kindle or download the Kindle app onto whatever device you prefer.) I hope you enjoy them.

Can We Predict a Molecule’s Smell from Its Physical Characteristics?

An extract of Yuanfang Guan’s winning code for odor prediction

A paper in this week’s edition of Science claims that computer models can predict the smell of a molecule. The paper describes the organization and outcome of an IBM Dream Challenge in which multiple laboratories competed to see whose model best predicts sensory characteristics from chemical parameters.

This crowd-sourced effort began with an olfactory dataset collected and published in 2016 by Andreas Keller and Leslie Vosshall. (Full disclosure: I previously collaborated with Keller and Vosshall on a different smell study.) They had 49 test subjects sniff and rate 476 “structurally and perceptually diverse molecules” using 19 semantic descriptors plus ratings of odor intensity and pleasantness.

In setting up the Dream Challenge, the organizers also “supplied 4884 physicochemical features of each of the molecules smelled by the subjects, including atom types, functional groups, and topological and geometrical properties that were computed using Dragon chemoinformatic software.”

There are several positive aspects to the challenge design. First, instead of recycling the decades-old Dravnieks dataset like so many other attempts at chemometric-based odor prediction, the sponsors supplied a fresh psychophysical dataset. Second, the study included a boatload of odorants, not the handful of smells found in most sensory studies. Third, the odor ratings were gathered from a relatively large number of sensory panelists. Forty-nine is not a super-robust sample size but it’s enough to encompass a lot of the person-to-person variability found in odor perception.

Here’s how the competition worked. Each team was given the molecular and sensory data for 338 molecules. They used these data to build computer models that predicted the sensory ratings from the chemical data. Sixty-nine molecules (absent the sensory data) were used by the organizers to construct a “leaderboard” to rank each team’s performance during the competition. The leaderboard sensory data were revealed to contestants late in the game to let them fine tune their models. Finally, another 69 molecules were reserved by the organizers and used to evaluate performance of the finalized models.

The models were judged on how well their predictions matched the actual sensory data using a bunch of wonky statistical procedures that look reasonable on my cursory inspection. (About the algorithmic structure of the competing models I have nothing useful to say, as “random-forest models” and the like are beyond my ken.) For the sake of argument I will assume that the statistical scorekeeping was appropriate to the task. My concern here is with the sensory methodology, the underlying assumptions, and the claims made for the predictability of odor perception.

Let’s begin with semantic descriptors. The widely used U.C. Davis Wine Aroma Wheel uses 86 terms to describe wine. The World Coffee Research Sensory Lexicon uses 85 terms to describe coffee. The Science paper uses 19 terms to describe a large set of “perceptually diverse” odorants which strikes me as a relatively paltry number. (The descriptors were: garlic, sweet, fruit, spices, bakery, grass, flower, sour, fish, musky, wood, warm, cold, acid, decayed, urinous, sweaty, burnt, and chemical.) Well, you might ask, can’t they just add more descriptors to include qualities like “minty” and “fecal” and “skunky”? It’s not that easy, as I discuss below.

The internal logic of the descriptors presents another issue. Some are quite specific (garlic), other very broad (spices), and still others are ambiguous (chemical). What are we to make of “bakery” as a smell? Is it yeasty like baking bread? Is it the smell of fresh cinnamon buns? (Or would that be “sweet”? Or “spices”?). The problem here is that words that are useful in an olfactory lexicon occur at different levels of cognitive categorization. This is reflected in the wine and coffee examples.

The Wine Aroma Wheel has twelve categories, each with one to six subcategories. For example, the Fruity category includes Citrus which consists of Lemon and Grapefruit. The higher level categories provide overall conceptual structure and are themselves useful as descriptors (e.g. a scent might be citrus-like while not smelling exactly of lemon or grapefruit).

Sensory specialists (including tea tasters, beer brewers, and perfumers) spend a lot of effort setting up lexicons that are concise and hierarchical, and which cover the relevant odor perception space. How were the 19 terms in the Science study arrived at? We do not know. How well do they cover the relevant perception space? We do no know. In fact, the authors state that “the size and dimensionality of olfactory perceptual space is unknown.”

These 19 terms are the basis on which the competing computer models were ranked. Thus a model's success at prediction is locked-in to this specific set of terms (plus intensity and pleasantness). In other words, this is not a general solution to smell prediction: it is specific to these odors and these adjectives. The authors openly admit this:

While the current models can only be used to predict the 21 attributes, the same approach could be applied to a psychophysical dataset that measured any desired sensory attribute (e.g. “rose”, “sandalwood”, or “citrus”).

So if one wants to predict what molecules might smell of sandalwood or citrus, one would have to retest all 476 molecules on another 49 sensory panelists using the new list of descriptors, then re-run the computer models on the new dataset. Easy peasy, right? Alternatively one could assemble a sensory panel and have the members sniff the molecules of interest and rate them on the new attributes of interest. Every fragrance and flavor house has such a panel. That’s how they currently evaluate the aroma of new molecules: they sniff them.

Thus the Dream challenge seems to be tilting at a windmill that the fragrance and flavor industry doesn’t see. The search for new molecules is not done by searching random molecular permutations. It is driven by specific market needs, say for a less expensive sandalwood smell or for a strong-smelling but environmentally safe musk. The parameters are cost, safety, and patentability, along with stability, compatibility in formulations, and (for perfumers) novelty.

Who knows, the smell prediction algorithms of the Dream challenge may turn out to be the first step in automating the exploration of chemosensory space. However I’d be surprised if this approach turns out to be generalizable and amazed if it proves useful in applied settings.

Don’t get me wrong. I like the idea of using Big Data to understand olfaction—have a look at my papers based on the National Geographic Smell Survey. I urged Keller and Vosshall to go big in terms of odorants and the number of sensory panelists for what became our co-authored paper in BMC Neuroscience. At the same time I respect the complexity of odor perception and the effort required to map its natural history. And I think the perceptual side of the equation got short shrift in this study.


The studies discussed here are “Predicting human olfactory perception from chemical features of odor molecules,” by Andreas Keller, et al., published online February 20, 2017 in Science, and “Olfactory perception of chemically diverse molecules,” by Andreas Keller and Leslie B. Vosshall, BMC Neuroscience 17:55, 2016.

Video: AIX Scent Fair Panel at the Hammer Museum

If you couldn’t be there to sniff and see it in person, you still have the chance to watch the May 6, 2016 panel that kicked off the AIX Scent Fair at the Hammer Museum in Los Angeles.

It begins with a welcome by the Hammer’s Darin Klein.

Lizzie “Odette Toilette” Ostrum’s keynote talk begins at 00:07:00.

My presentation begins at 00:29:00

Jacquelyn Ford Morie’s presentation starts at 00:43:00

The presentation by TFFCOTNYT starts at 1:10:00

The panel discussion moderated by the Institute for Art and Olfaction’s Saskia Wilson-Brown begins at 1:25:00.

Enjoy.

The Alleged Limitations of Olfactory Language

That smells can be difficult to name is a commonplace observation. Almost as commonplace as the observation that there exist many specialized vocabularies for smell, such as those used by perfumers, wine tasters, brewers, coffee roasters, tobacconists, potheads, etc., etc. In this battle of off-setting banalities, those who downplay our ability to verbalize odors are often ceded the victory, in keeping with the pessimistic Greco-Freudian view that the human sense of smell is a poor thing, rendered vestigial from lack of use, and substantially inferior to that of other animals.

A more optimistic view is that humans are quite competitive in terms of odor sensitivity (often exceeding that paragon of scent detection, the dog) and that a remarkable amount of information regarding other people (emotional, physiological, and health status, for example) is received and processed via the nose. This positive outlook has been gaining support steadily in recent years but science journalists and assistant beauty editors have been slow to recognize the trend. This is not surprising; they are, after all, science journalists and assistant beauty editors. What is surprising is that a pair of credible scientists have now thrown in with the nasal nay-sayers and offered a theoretical account of why humans must necessarily suck at naming smells.

That might seem a rather rude way to characterize a paper decorously titled “The muted sense: neurocognitive limitations of olfactory language,” but I think it captures the essence of what Jonas Olofsson and Jay Gottfried are attempting to do.

Olofsson and Gottfried begin with three papers from the 1970s that find when you bring people into a psychology lab and have them sniff odors absent any contextual information (visual, auditory, or otherwise), they have a hard time coming up with the correct name. Provide them multiple-choice odor names and they generally choose the correct one. For good measure, Olofsson and Gottfried also reference studies showing that under similar laboratory conditions people suck at picking out individual components from a bouquet of scents.

Atop these rudimentary observations, O&G construct a “biologically informed framework for olfactory lexical processing.” Being neuro-imaging specialists, they sketch connections between piriform cortex, orbitofrontal cortex, and all the other neuroanatomical waystations on the route from nostril to naming. You will enjoy the discussion if this sort of thing appeals to you:

Importantly, odors are already integrated with lexical representations at the third synapse from receptor neuron input. This could put olfaction at a disadvantage compared with the visual system, where multiple subcortical and cortical sites create object representations before lexical–semantic integration by integrating features at different spatial scales.

My concern here is not with O&G’s theoretical edifice, but with their presumption that the difficulty in generating a verbal tag for a context-free odor is somehow fundamental to our understanding of human olfaction, and with their view that this phenomenon benefits from a convoluted account drawing upon “recent behavioral and neuroimaging data.”

Take the laboratory task upon which their entire argument is based: could there be anything more remote from the universal, everyday experience of smell than being confronted with a sniff-bottle and asked to name its contents by smell alone? What real world setting does this resemble? The answer is none.

Smells always occur in a context, and it is only within this context that we try to make sense of them. At the fish market, for example, we sniff to see whether the fish is fresh. Whether we can summon up the name “trimethylamine” is irrelevant. Smells may confirm our visual expectations (“it seems to have rained here recently”) or draw our attention to something that warrants exploration (“what’s burning?”) all without resort to specific lexical representations.

Tagging a smell with a specific lexical term requires high-level abstraction. In contrast, odor identification and description are broader abilities that are exercised more often and with greater functional impact. When olfactory naming happens in real life it also happens in a natural context and the language processing is tuned to an appropriate (and useful) level of generalization. Here’s a non-laboratory example of olfactory language in action:

“Umm. Is someone grilling dinner?”

“Yeah, it smells like hamburgers.”

To me, that is an example of rapid, precise, and biologically useful neurocognitive olfactory processing. But in Olofsson and Gottfried’s model it simply doesn’t exist.

In my view, the O&G model is an elaborate neurocognitive account of a laboratory artifact. It has little bearing on the broader role of olfaction in human behavior and communication. Does this sound extreme? Then ask yourself: does the near-universal inability to name the musical key of a song imply that “people are poor at describing sounds?”

The study discussed here is “The muted sense: neurocognitive limitations of olfactory language,” by Jonas K. Olofsson & Jay A. Gottfried, published in Trends in Cognitive Sciences, 19:314-321, 2015.

F. Scott Fitzgerald: An Olfactory Archaeology

A wonderful literary use of smell to illuminate the thick layers of human detritus that accumulate with time in a New York apartment building. Climbing up the scented stairway is like descending through the layers of an archaeological dig.

Then he would trudge homeward, enter the dark hallway, and climb three rickety flights of stairs covered by an ancient carpet of long obliterated design. The hall had an ancient smell—of the vegetables of 1880, of the furniture polish in vogue when “Adam-and-Eve” Bryan ran against William McKinley, of portières an ounce heavier with dust, from worn-out shoes, and lint from dresses turned long since into patch-work quilts. This smell would pursue him up the stairs, revivified and made poignant at each landing by the aura of contemporary cooking, then, as he began the next flight, diminishing into the odor of the dead routine of dead generations.

Eventually would occur the door of his room, which slipped open with indecent willingness and closed with almost a sniff upon his “Hello, dear! Got a treat for you to-night.”

F. Scott Fitzgerald
“Oh Russet Witch!”
From Tales of the Jazz Age (1922)

The Stench of the Abyss

Dante
Inferno, Canto XI

[Circle Six: The Heretics]

In su l’estremità d’un’alta ripa
che facevan gran pietre rotte in cerchio,
venimmo sopra più crudele stipa;
e quivi, per l’orribile soperchio
del puzzo che ’l profondo abisso gitta,
ci raccostammo, in dietro, ad un coperchio

d’un grand’ avello, ov’io vidi una scritta
che dicea: ‘Anastasio papa guardo,
lo qual trasse Fotin de la via dritta’.

«Lo nostro scender conviene esser tardo,
sì che s’ausi un poco in prima il senso
al tristo fiato; e poi no i fia riguardo».

We came to the edge of an enormous sink
rimmed by a circle of great broken boulders.
Here we found ghastlier gangs. And here the stink

thrown up by the abyss so overpowered us
that we drew back, cowering behind the wall
of one of the great tombs; and standing thus,

I saw an inscription in the stone, and read:
“I guard Anastasius, once Pope,
he whom Photinus led from the straight road.”

“Before we travel on to that blind pit
we must delay until our sense grows used
to its foul breath, and then we will not mind it,”

my Master said.

[Translation by John Ciardi]


A fine example of olfactory literature: the stench of the pit is simultaneously literal and figurative. Virgil’s advice to Dante to pause while their noses adapt draws on a keenly observed bit of odor perception; it also provides a dramatic interlude for him to explain the next levels of the damned. The presence in Hell of Pope Anastasius is nice reminder that popes too can be led astray.