We still don't know how the mouth works

Every few months, in a laboratory outside Paris, an artificial mouth has its tongue pulled out.

Glowing a resinous orange, as if the mouth had just necked a sports drink, the tongue is cast in silicone from a human volunteer – faithfully reproducing all the minutiae of the tip and papillae tongue buds, though it can’t actually taste anything. The fake mouth will never savour steak frites or tackle something complex such as candy floss. Instead, researchers at France’s National Research Institute for Agriculture, Food and the Environment (INRAE) feed it with a soft solid diet for a study that chews over questions about the mechanics of food and eating.

Over a period of months they have been feeding the mouth, set in a plastic device that looks like a basic coffee filter machine, with chocolate mousse, cream desserts and fondant, to observe how the tongue prepares soft foods into swallowable boli. After the experiments, the mouth is sluiced with water and cleaning solutions until one day the tongue dries up and, past its best, is plucked out and discarded, replaced by a new one, an imitation of one of the most mysterious environments in the human body.

Mysterious may sound like an exaggeration: surely we know what tongues and mouths can do? They’re almost ridiculously useful – heavily involved in our appetites, desires, development and language – while also positioned at a prestigious proximity to our brains. But scientists are still building their understanding of how the mouth works. And, as the world’s population ages and our food supply changes, studies around the world can make a huge difference to how and what we eat.

“There are a lot of questions – most people don’t appreciate how complicated it is – every little thing is very cleverly designed,” says Prof Jianshe Chen of the A*Star Singapore Institute of Food and Biotechnology Innovation, who began researching this field as leader of the University of Leeds’s oral food processing lab. “For example, every individual knows how and when to swallow, but if you ask them, ‘How did you know it was safe to swallow?’ they can’t tell you."

Though most studies have the same starting point of treating swallowing disorders and choking hazards for different demographics, including children and people with Down’s syndrome, the quest is “larger”, says Dr Tan Sze, executive director of A*Star Singapore Institute of Food and Biotechnology Innovation. “What’s mesmerising to me is how sensitive our mouths are in identifying differences that are very challenging to detect in the lab.”

But this ability deteriorates with age. Elderly people often lose the muscle control to properly close their airway while swallowing and if food gets into the lungs, it increases the risk of choking and pneumonia. So scientists are working to understand the creep of sarcopenia – the declining abilities of ageing human bodies – which is writ large in global population forecasts. The UN projects that 2.1 billion people will be over 60 by 2050, double the figure in 2024.

Japan has long been a leader in making food for swallowing disorders – these engay foods (engay means swallow) are pureed, reshaped boil-in-the-bag dishes such as mackerel mousse simmered in miso. They are common in care homes, and specialist companies even hire out colourful dining sets to make the food look better.

As Marco Ramaioli, the lead researcher at INRAE’s biomimetic mouth study, says, research has to be as much about creating meals people want to eat and “to live for”, as much as food they can tolerate physiologically.

To these researchers, the concept of food is much more than what you put in your mouth. It also is your mouth, your saliva and your swallow. Food happens, rather than is simply ingested, when an unthinking orchestra of teeth, jaws, cheeks and tongue readies it for the gullet. After all, if you couldn’t swallow it, would it really be food?

We can’t study the mouth in vivo; ethically, as Chen says, it’s “difficult to put in a probe or camera” in someone’s mouth. And so the bouche artificielle at INRAE’s food and bioproduct engineering research unit does crudely what most humans do instinctively – which is to say manipulate food via a process of tongue swivelling, folding, compressing and moulding.

The fake tongue is computer programmable, and moves according to data collected from ultrasounds gathered by Fujita Health University in Japan. And this research may also be important in refining foods for a planet that has limited resources. It’s also a planet that is waking up to how unhealthy much of its diet is.

Some plant-based protein products, Ramaioli notes politely, still have “sensory issues” and are often “astringent”. To make them taste better, science needs to discover how the mouth translates food into deliciousness. Though much has been made of the food industry’s ability to create hyper-palatable textures from fatty, sugary and salty junk food, it hasn’t yet (arguably) cracked how to make packaged and processed health food.

A recent report by Tate & Lyle called The Future of Mouthfeel highlighted the need for “innovators … to refine their skills in recreating the creamy richness of dairy, the juicy tenderness of meats, and the flaky textures of pastries. That requires skilfully combining plant proteins, fats and stabilisers to achieve the desired mouthfeel without sacrificing flavour.” They picked out Credo Foods’ Oat Milk Spray Cheeze as an example – a synthetic cheddar in a can.

The 20th-century French philosopher Maurice Merleau-Ponty made no distinction between body and mind: the body, he said, was simply “our general medium for having a world”. The centrality of the mouth to our state of being was recognised early on by doctors in the 18th century, who would start their consultations by asking patients to show their tongue, which would be classed as either “clean” or “dirty”, then prescribe funky medicine such as rinses of lettuce juice, according to archivists at the Royal College of Physicians of Edinburgh.

These days, researchers at the University of Edinburgh are using artificial intelligence to show that the human tongue’s surface is exactingly unique – their modelling could predict with up to 48% accuracy the identity of a person just by reading a 3D microscopic scan of a single papilla or tongue bud. Two types of papillae – mushroom and crown-shaped – do different things: taste in the former, and sense and touch in the latter.

This could lead to what Tan describes as a “new frontier” of personalisation in what we eat.

She says: “The days of a ‘one size fits all’ food strategy are fading rapidly. Food must adapt to enhance the eating experience for diverse demographics. A deep understanding of oral food processing – how texture, chewing and oral receptors influence satiety, mood and wellbeing – is crucial for enhancing consumer experience.”

One of the key mysteries, she adds, is “how our genetic differences affect saliva composition. This has key implications on how we taste, digest and absorb nutrients. Could saliva biomarkers eventually guide personalised diets or signal early health risks?”

This conjures visions of a personal food ID that could one day be plugged into your supermarket shop; picking out products that will match to a scan of your tongue and an analysis of your saliva. But it’s a little early to be thinking of this, says Chen – although both he and Ramaioli echo Tan’s view that the days of “one for all” in food could be challenged. “The food industry used to be ‘pick what you need’,” says Chen. “Now we have segments of populations for nutritional needs and capability.”

The other new research frontier for Tan is the oral microbiome – a microscopic swarm of bacteria and fungi that is unique to each person. “This complex, dynamic ecosystem impacts not just oral health, but metabolism and immunity.” It is a mystery, she says, that “we need to unlock. One day, the mouth may no longer be seen as just the starting point of digestion, but as a gateway to precision nutrition and personalised health.”

There are many artificial mouths around the world. The Dijon Centre des Sciences du Goût et de l’Alimentation has one while the Monell Chemical Senses Center in the US has developed an AI tongue model. Another in Clermont-Ferrand, France, focuses on chewing. They each have different setups. “Dijon doesn’t have a tongue,” says Ramaioli. Some work alongside innovations such as electronic noses. “We see how the complexity builds up little by little.”

To eat is a full-body experience, and there is much to still comprehensively understand about how the brain analyses food signals. Chen says: “There are lots of complications there. We taste with our tongue but we also smell a compound smell with the nose, and we do it together: the chemical info goes to the brain, but in the brain we just put it together as a flavour.”

Research into this complex relationship is a multidisciplinary effort. Chen says: “Previously we saw food science as science to form a nicely appealing edible product. But there’s now psychology and neurology that have come into play.”

As Ramaioli says: “The more we study, the more we don’t know – it’s a continuous quest.”

In the tradition of the deepest riddles, one mystery leads to another, but in the meantime, the artificial mouths will keep chomping on this problem in labs around the world.