The Past Doesn't Smell Like It Used To

Scientists find ways to study and reconstruct past scents

Apr 2022, phys.org

https://phys.org/news/2022-03-scientists-ways-reconstruct-scents.html

They're trying to develop "an archaeology of scent," which is hard because smells are ephemeral, and last only as long as their source. But because of advances in chromatography, mass spectrometry, sequencing technologies and modern bioinformatics, which include metabolomics, proteomics and genomics; they can identify the organic remains preserved on surfaces like walls, ceramic vessels, incense burners, perfume flasks, cooking pots, dental calculus, mummies, and entire streets:

Advanced biomolecular and ‘omics’ sciences enable more direct insights into past scents, offering new options to explore critical aspects of ancient society and lifeways as well as the historical meanings of smell.

The whole paper is very interesting, and although I encourage anyone interested in ancient history to read it, for those who don't, I bring back only this -- palaeofaeces -- it's a thing in archaeology, and now in olfactory archaeology too: "in an Iron Age roundhouse in Scotland, chemical characterization of floor sediments provided insight into living conditions, hygiene practices and the temporary sheltering of animals in human living areas during this period."

-Mackay, H. et al. J. Archaeol. Sci. 121, 105202 (2020). [pdf]

via the Department of Archaeology at the Max Planck Institute for the Science of Human History in Jena, Germany: Huber, B., Larsen, T., Spengler, R.N. et al. How to use modern science to reconstruct ancient scents. Nat Hum Behav 6, 611–614 (2022). https://doi.org/10.1038/s41562-022-01325-7

https://doi.org/10.1038/s41562-022-01325-7

Image credit: AI Art - Animals in Jars - 2022

via James Gilleard and Justin Gerard

Post Script:

Speaking of human history, the Odeuropa project is taking a completely different angle -- they use computer science to identify, scrape, and coordinate pictures that contain smells in them, either by visually recognizing objects tagged as related to smells, or by reading the text captioned with the image. Sensory mining they call it:

Odeuropa is a European research project which bundles expertise in sensory mining and olfactory heritage. We develop novel methods to collect information about smell from (digital) text and image collections.

https://odeuropa.eu/

Odeuropa's Olfactory Iconographies

€2.8M grant for research project on European olfactory heritage and sensory mining:

Odeuropa bundles expertise in sensory mining and olfactory heritage. We develop novel methods to collect information about smell from digital text and image collections. They will identify and trace olfactory information in text and image datasets using AI, and promote Europe’s tangible and intangible cultural heritage.

Here's one of their ongoing projects, seen in the picture above, an odor wheel based on art historical references to smells: The Odeuropa Art Historical Scent Wheel from the Mediamatic Aroma Lab.

The Odeuropa “Nose first art historical odour wheel” starting from scent families in the centre, connected to odorants in the second ring, and then to artworks and artefacts around that, ending with an outer ring with Iconclass codes. Iconclass is a multilingual online database that museums use to tag historical images and artworks. -link

This wheel is based on imagery like paintings that were then coded with words, allowing a machine-readable dataset of odors, which will become increasingly more popular as we apply this approach to larger and more recent datasets, such as the Wikimedia Commons, or the running corpus of Instagram's zero-liked images.

 

Speaking of datasets, there are many to choose, from dog breeds to aerial photographs to 3 million "Clickbait, spam, crowd-sourced headlines from 2010 to 2015," to "4,000 physical dimensions of abolone." Wikimedia commons has 7.5 million images.

Looking back at the odor wheel, there are some interesting associations here. But they do reflect the dataset. I'm having a hard time finding details on Iconoclass but it was developed by one person in the 1950's, and based I assume on Western Art. It's currently maintained by RKD Netherlands Institute for Art History.

Here's some examples -- "street scenes and horses;" not a common conjunction in today's world. "Unicorns and cinnamon," anyone? "Prostitutes and civet," less unexpected. The two "body odors" are armpit and vagina, fyi. 

The most interesting part of all this? The lead researcher Sofia Ehrich has "become familiar with detecting depictions of smell." She can smell words and pictures. 

And speaking of words and pictures: 

Ocularcentric - like a visual bias, like we as humans tend to have an ocularcentric view of the world, with our trichromatic vision and fancy visual cortices, etc. 

Notes:

Mediamatic (in Amsterdam) is an art centre dedicated to new developments in the arts since 1983. We organize lectures, workshops and art projects, focusing on nature, biotechnology and art+science in a strong international network.

IconoClass dataset -  specialized library classification designed for art and iconography.

Lifting One's Hat

Layers of IconClass system

This is an example of the layers of the IconoClass system, pretty deep stuff. You can see how the image has words attached to it, making it a machine-readable cultural object. This is how we will teach robots of the future how to better understand us, and maybe we can even teach them how to smell.

Iconoclass Browser

Mario Klongmann x BigGAN - 2019

Mostly Unrelated Post Script:

An AI Artist’s Twitter Feed Is an Art Gallery

The images and videos Mario Klingemann posted under the hashtag #BigGAN can only be appreciated by treating his Twitter feed as a digital exhibition. (Images taken from the ImageNet dataset)

Feb 2019, Hyperallergenic

https://hyperallergic.com/481969/an-ai-artists-twitter-feed-is-an-art-gallery/

Dog layers on a bed with a blanket - Sam Hains Zero Likes - 2017

This AI Creates Art From Instagram Posts With Zero Likes

“Zero Likes” is trained to create glitchy visuals from forgotten social media images.

May 2017, Vice

https://www.vice.com/en/article/z4jz3x/this-ai-creates-art-from-instagram-posts-with-zero-likes

Melbourne artist and coder Sam Hains created Zero Likes, an AI trained to respond only to those lost and lonely images that miss out on attention.

Personal Biodata Protection

I've been catching up on the more recent advances in olfactory research, and got lost in a paper from 2003, where they make an olfactory perception database out of co-occurring semantic descriptors from the Sigma Aldrich catalog.

The standard olfactory perception database has grown substantially in the past several years, so we won't go into their results, but I did want to pull an interesting point from their conclusion.

Their data is organized not by chemistry, but by metabolism, and they describe the olfactory system as being able to "recognize metabolism." Our sense of smell is capable of identifying metabolic processes in biological systems. This is one of those things that makes smell so taboo to talk about. Me and you are biological systems.

Throughout the entire book I wrote about the language of smells, not once did the topic of privacy or intimacy come up. But it could have; a major reason why we don't talk about smells is because they remind us of the power held over us by those who get close enough to smell us.

Your scent carries with it information that you wouldn't exactly want to advertise to everyone. The biologic, metabolic activity taking place behind your skin, inside your digestive system, throughout your endocrine network, that information is pretty personal. But I can find out about those things, if I get close enough. Maintaining a general approach to just not talk about smells is probably a good idea all around.

Post Script:

And this focus on metabolism is why I would like to see the metabolome tied-into other olfactory perception databases.

Human Metabolome Database (HMDB) - 40,000 different metabolite entries

Metabolomics

Notes:

"Descriptors used to classify molecules containing nitrogen or sulfur were clearly segregated in the odor perception maps. Because these molecules are key atoms in different metabolic cycles, it was proposed that human olfactory perception reflected the organization of animal and plant metabolism."

Quantifying olfactory perception: mapping olfactory perception space by using multidimensional scaling and self-organizing maps. Mamlouk AM, Chee-Ruiter C, Hofmann UG et al. Neurocomputing 2003;52:591–7.

The biological sense of smell: olfactory search behavior and a metabolic view for olfactory perception. C.W.J. Chee-Ruiter. Ph.D. Thesis, California Institute of Technology, Pasadena, CA, 2000.

Expanding Search Beyond the Semantic Frontier

Alex Pardee’s Escaping Conviction, circa 2010

Here we have a little something about similarity concepts and search.

Why must we limit the internet to text? Sure we have a rudimentary image search function, but I’m referring here to the way we search our own memories. We don't use text only, we use our bodies, our autobiographies, the mental maps in our heads...

Explaining this new method for searching with a sketch vs. keywords:

“In designing the system, the researchers deliberately set a very broad similarity concept and adapted it to different types of sketch; for example, similar colors, shapes or directions of movement.”

Notes:

In a new method for searching image databases, a hand-drawn sketch is all it takes

phys.org, Jun 2016