How to Have Good Taste with Dr. Arielle Johnson

Taking a scientific approach to food writing and recipe development is not the anomaly it was a decade ago. Almost every food publication has a resident nerd who can explain the acid-base reactions that take place during baking, or how one can use starch to thicken a gravy, but very few of those nerds have a PhD in agricultural and environmental chemistry from UC Davis with a focus in flavor chemistry. Even fewer served as the resident scientist at Noma in Copenhagen, or as the head of research at MAD, chef Rene Redzepi’s food symposium and think tank. In fact, I can only think of one food writer who fits that description: Dr. Arielle Johnson.

Her book, “Flavorama: A Guide to Unlocking the Art and Science of Flavor,” is what I wish all science textbooks were: informative, whimsically illustrated and funny, with complex chemical concepts presented in a way that’s completely accessible without even a hint of condescension. After hearing Dr. Johnson get nerdy on an episode of Milk Street Radio, I reached out to her for an extended interview to chat about the surprisingly complex nature of flavor, why bitter needs a chaperone and how everyone can train their palate to be just as sharp as that of your local wine snob.

What are your core laws of flavor?

Flavor is molecules, which is the first one that you hear, but I counted it as the zeroth law just because it was so pervasive. It's fundamental to what flavor is, because flavor is taste and smell—taste and smell are chemical senses. We use them to detect molecules directly. And they create signals that we then translate into our perception of flavor. Everyone understands that taste is a part of flavor. Getting people to understand that smell is also a key part of flavor is very new to most people and a little bit weird to think about. But that's essential.

Next: Flavor follows predictable patterns. This comes from not just thinking about the science of flavor, but also watching how the chefs that I've been working with for years and years worked with flavor, which is a combination of thinking about it as a very material thing, and also training their palates and sort of abstract thinking to recognize flavor notes across dishes and ingredients, and then applying their knowledge in really flexible ways. Fortunately, anyone can learn to think that way.

Then: Flavor can be concentrated, infused and extracted. Flavor is molecules so we can move molecules around and that can be as simple as brewing a cup of coffee in the morning or as complex as making a special smoke-scented oil for making smoky food without a smoker or setting off your fire alarm.

And then lastly, flavor can be created and transformed. We can actually do chemical reactions that will change flavor, and browning is one very big one that most people know. There's others like smoking and caramelization that have to do with changing things with heat. And then another big one is fermentation. It's a very ancient part of human heritage, but has had a renewed focus in the culinary sphere. We encourage microbes to grow in our food, and they eat parts of it and then create new molecules with different flavors.

That covers pretty much anything you would do with flavor. There's obviously very specific iterations of each one, but the idea was once you understand those basic rules of the road, then anything else you do would fall under one or more of those categories.

In the book you say tasting is like hitting a piano key and smelling is like a QR code. Can you expand on that?

So, taste is a piano with five keys. I guess it's pentatonic. It's a little bit more limited that way. Our sense of taste and our sense of smell are both shaped by our evolution and needs. The sense of taste is very much wired to detect a few specific things and get a very clear signal about what's there and how much there is. The five tastes are sweet, sour, salty, bitter and umami. For each of those, there's a relatively small set of molecules that we can sense with each. With something like umami or salty it's really just one primary molecule. Bitterness, we have a bit more flexibility, because bitterness is actually less about trying to find things that we do want and more about trying to avoid things. There's lots of things that could potentially poison us. We have about 20 different bitter receptors, but they're all linked up to the same sort of piano key. They make the same signal.

But setting bitterness aside, something like sweetness—we want to be able to detect easy calories, so we have a very dedicated system for sensing sugars. Saltiness, we need a certain amount of sodium for ourselves and our brains to work at all, so we have a system for detecting the presence of sodium. Umami is a good signal for the presence of protein since it detects glutamate and glutamic acid, one of the 20 amino acids.

“There’s about 40 billion different molecules that we could possibly smell.”

And then sourness is sort of the most multipurpose of the tastes. We can detect vitamin C because vitamin C is ascorbic acid, and that is sour and we need that so we don't get scurvy.

There's some interesting work that came out a few years ago suggesting that the ability to detect sourness has to do with our late-primate, early-human ancestors needing to be able to eat fermented food, or have a better time eating lacto-fermented food, as opposed to spoiled food. So if you could taste acid and like it, you'd be more likely to eat the lacto-fermented food, as opposed to things with less beneficial bacteria, and then also maybe try to deliberately lacto-ferment things and then have a longer life and healthier family, etc.

So yeah, tastes are very evolutionarily-focused. And smell is basically built around flexibility. We actually don't know the number of molecules that we can sense by smell. Some work has estimated it. The number 10,000 as the number of smellable molecules has been thrown around. I don't think any empirical work supported that, it was just kind of like an estimate in the 1920s. The most recent paper I've read has calculated that, based on all of the molecules that could meet the sort of chemical and physical requirements for being able to be smelled—so like, size and polarity, and whatever—there’s about 40 billion different molecules that we could possibly smell.

“We have our emotional reaction and our memories triggered by a smell before we even consciously perceive it.”

But the reason it's so flexible is, unlike taste, where there's a receptor that really only binds to one molecule, and that molecule doesn't really bind to other receptors—salt is not going to activate the sweet receptor—with smell, there are no dedicated receptors. We have about 400 different ones, and any smell molecule that we smell is going to activate several different receptors. And any receptor that we'll use is going to bind to several different molecules. So the reason I say that it's like a QR code is it actually looks quite a bit like a QR code. The sort of first smell impression or smell signal that we generate, after getting impulses from the olfactory receptors that get sent to the brain, they get collected on this thing called the olfactory bulb, and each receptor has a little patch on it, like a patchwork or a grid. And so the actual signal that you generate for any smell is different levels of activation for each of these patches. We actually call it an odor image. So it's a completely different mechanism and purpose behind smelling. Less precise in a lot of ways, but also much more expansive and emotional.

The olfactory cortex, which is sort of like the smell center of the brain, actually incorporates parts of the limbic system, which is the sort of emotional memory, subconscious part of our brains. They're so closely tied together that the signal from smell reaches those emotional centers before it reaches consciousness. We have our emotional reaction and our memories triggered by a smell before we even consciously perceive it.

Is good taste or a good palate something you believe people are born with or something they can learn?

Well, we do have genetic variability in terms of sensitivity. Some people have more taste receptors overall, or more specific taste or smell receptors. But all of that aside, all humans are basically born with the ability to distinguish between flavors, so we're actually very good at telling flavors apart. And when given two or more flavors, we're able to say that they're different.

Where we tend to fall short is describing what those differences are. Fortunately, all of that is just built by experience. I think something that was really important for me to convey is that having a good palate, or knowledge about wine or something like that, and the ability to generate tasting notes is kind of treated as this secret club, or like an esoteric thing—you either you have it or you don't. That's all bullshit. It's all about paying attention and spending time on it. The more that you do it, the more that you sort of force yourself to try to articulate flavors, the better you'll get. I don't just think that because it sounds cool. As an academic sensory scientist, I've had to train dozens of people to be very precise and accurate tasters. So I've seen it happen. I mean, it happened with me. In fact, when I started out, I could not distinguish oak in the sort of bouquet of a wine, and after smelling it, you know, 30 or 40 times and having to try to identify it, finally it clicked. So if it worked for me, and it worked for lots and lots of other people, it's really something that everyone can do.

Would you recommend keeping a journal or describing the flavors to yourself as you're cooking?

I think definitely the first thing is deliberate tasting, tasting your ingredients as you're cooking with them. And just sitting with and trying to articulate what the flavor is is the most important step that you can take to go from not having a palate to having a palate that can recognize things.

If people want to go further than that, adapting how we train sensory panelists in research is not that hard to do. We'll sit them down, we'll taste things, they'll talk about what they think they taste. But we take advantage of the fact that humans are good at distinguishing between things. So, we'll ask them to taste several versions of the same or similar things side by side, so you're not tasting just one olive oil and trying to conjure its qualities out of the ether. If you taste four olive oils next to each other, you can start to understand them and how they're different from each other. So like, “This one is more bitter. This one is more grassy. This one is more fruity.” And those specific nuances will pop out more when they're in this context of similar things.

We also do a lot of tying flavor descriptors to physical references. If we're doing a wine tasting and someone says they smell raspberry, we will then source several different versions of raspberry, like a fresh raspberry, a raspberry jam, maybe a raspberry candy, and then have them taste each one blind. And we ask them “Okay, so when you say raspberry, which raspberry do you mean?” And whichever one is closest to this sort of concept of raspberry, that becomes the definition of raspberry for the entire experiment. We're dealing constantly with this sort of elasticity and inherent imprecision of language, which is good for writing and poetry and tricky for precise tasting.

Why is salt so special?

It's an interesting question because in some ways there is nothing that special about sodium. I mean, it's like a Group 1 metal. Lithium is also one. And one thing that makes sodium special is that we use it for osmotic balance and as a neurotransmitter and stuff like that. But culinarily, the way our salt-perceiving channel gets hooked up to the rest of our flavor processing machinery—it just really perks up other flavors.

One very specific thing it does well is interfere with the bitter taste receptor. So anything you add salt to will taste like the bitterness has been dampened and it'll taste sweeter and more lively that way. And in terms of flavors to get correct, salt is usually the most fundamental one in cooking. Not just because you want it to taste salty, but because it just enlivens everything else.

Do you recommend using salty ingredients over pure sodium chloride?

Yeah, I love complexity and depth of flavor. So, in a sauce or a dressing or a stew, for example, you can achieve correct saltiness by adding salt. But if you know there's lots of salt in capers, anchovies, ham and cheese, things like that, you can add complexity and depth of flavor at the same time as seasoning correctly. I think that cooking with the kind of mindset of “OK, if I'm adding a particular type of ingredient here, I need to add saltiness here. What can I do to maximize the amount of flavor that I'm adding?” And so cured and salted flavorful ingredients is a great way to do that.

What are the two main sources of sour flavor?

That would be fruits and vegetables and fermentation. Plants produce acids for a variety of reasons, including discouraging things from eating stuff that's not ripe and producing vitamins and things like that. The two big groups are the lactic acid bacteria and the acetic acid bacteria. They produce lactic acid and acetic acid as part of their sort of metabolism-slash-growth strategy, because it's toxic to a lot of other microbes.

So acetic fermentation, that's vinegar. Many vinegars started out as not very sour but with alcohol, and the alcohol got converted to acetic acid; and then things like lacto-fermented pickles, cultured dairy, sourdough, etc., are all lacto-fermented sources of sourness.

Lactic acid is one of the fermentation acids. The primary plant acids are tartaric, malic and citric acid. Some people say lactic acid has a bit of a cleaner or slightly creamier sour flavor profile, where citric acid is a little bit more pointed and rough tasting. But the biggest difference is acetic acid is one of the few molecules that's both a taste molecule and a smell molecule, so we taste it as sour but it's small enough so that it can be volatile. The vinegary, pungent smell of vinegar is literally the smell of acetic acid. So that's the biggest taste or flavor difference between between acids.

I had no idea. How does sour affect our perception of other flavors?

Well sourness, in and of itself, is a pointed flavor. It really announces itself. And it's very refreshing. Adding sourness generally will make it feel brighter, more lively, more refreshed. Some of this has to do with [the fact that] acids actually cause us to salivate more, so you'll have this kind of juicy or more dynamic eating experience from your mouth watering when you're eating something sour.

Sourness is also decent at balancing bitterness and good for balancing sweetness as well. And then with any sort of sour ingredient, usually there are aromas involved. A lemon has sourness and lemony aromas. And something like tamarind or sumac has its own sort of associated other flavors. And the way that we generate flavor—taking information from taste and information from smell—we actually get enhancement between taste and smell. Any kind of aroma that you're used to tasting or used to perceiving alongside sour—lemony aroma, for example—that aroma will increase how intense the sourness is. So they amplify each other. If you have lemon sourness plus lemon aroma, the aromas will smell more intense and more lively and the sourness will taste more bright and more complete. Those are all learned associations.

In your book you define sourness not by pH, which is how most people think of acidity, but as “titratable acidity.” Can you explain the difference and why you prefer using that scale?

[Editor’s note: A titration is a common lab method where a solution of a known concentration is used to determine the concentration of another solution. “Titratable acidity” refers to the total amount of acid available for reacting with a base. According to UC Davis, “A titration of sorts is performed in the mouth of the consumer, where basic saliva meets the wine/food and causes increased flow.”]

Frankly, the research shows that pH is not super well correlated to perceived sourness. It's not that something with a pH of 8 or whatever tastes more sour than something with a pH of 4, but when you get into the range of acidic foods and ingredients, pH is a very in the moment snapshot of how many hydrogen ions you have at any given point in time. And so when you're sensing sourness, what you sense is, in fact, the hydrogen or the hydronium ion...you're not actually tasting acetic acid, you're tasting a little tiny hydrogen that's acidic that comes off of the molecule. So pH is like a snapshot of how many hydrogens you have right that second, because most of the acids that we eat don't give up all their hydrogens at once.

So for any amount of lactic acid or citric acid in an apple or a lemon, you're going to have a small pool of acid molecules that have given up hydrogen. So that's the snapshot of how acidic it is at that point in time, but there's a whole bunch that have not come apart. And when you start diluting the food with saliva, or removing hydrogens by putting them into your tastebuds, that pool of potential acids will start to release their hydrogens. So the pH gives you a small snapshot in time and it doesn't predict how much more you will get once you taste it. So a bigger pool of acid at a slightly less acidic pH will actually usually taste more sour than the brief burst at a lower pH.

Sweet is the most boring flavor to me, but maybe that’s just because white sugar is so boring.

Going back to the idea that we were talking about in salty where you're like, “Well, if I need to add salt, I could just add table salt, but I could add a complex-tasting salty ingredient.” In that regard, saltiness and sweetness are for the typical cook kind of the only two where you have access to the pure ingredient molecule. Most people aren't working with powdered acids unless they work for an experimental restaurant or something like that. So we have refined sugar available, but that's really just the tip of the iceberg in terms of sweet flavors. I really like unrefined sugars, basically the sugar of the people wherever they grow sugarcane. Well actually, not just sugar cane, but wherever they're producing sugar, so jaggery or panela or Thai palm sugar. All of these are made by basically taking the sugary juice sap from something and boiling it down.

And where you would make white sugar by bleaching and centrifuging and doing all this crazy stuff, for an unrefined sugar, it just stops at the point of boiled down enough to crystallize. So you get these same flavors from the plant, but also the cooking process creates all these really delicious and surprisingly varied flavors. They don't all just taste like molasses. I often find Indian jaggery is a little bit funky and Thai palm sugar has—I mean, they're all kind of brown and caramelized—but palm sugar has this kind of buttery element to it. I love Japanese black sugar, kuro sato; that just has this juicy, maple-y flavor to it. If you can find those, I love working with them. So saps, like maple syrup, sorghum syrup, birch syrup, honey—those are all much more complicated and I think potentially delicious sources of sweetness. I mean there's places where you want to use white sugar, like if you're making a very nice fruit jam and you want the fruit flavor to be un-interfered with. If you want a white cake, you want white sugar. But in your drinks and your coffee or even in some baking, adding a little bit of complexity with these more unrefined, more complicated sugars is a great way to have fun with it.

Moving on to umami. Why do you think people have such a hard time describing that particular flavor?

Umami comes on much less strong than something like sourness does, or saltiness. Almost all cultures have very umami-rich ingredients, but they're not typically cooking with umami as an isolated component. Unless you're cooking with MSG, or doing a lot of stuff with dashi, then your experience of it is blended in with fermented or cured or whatever other flavors. It's also the most recent taste to be formally identified. It wasn't connected to glutamate until the early 20th century. And then we didn't actually identify the taste receptor until I think the late 90s, early 2000s. Sweet and sour and salty and bitter have been in most lexicons across cultures. Savoriness doesn't always appear. And in the 18th century, a lot of gentleman scientists were interested in this concept called osmazome—I believe they defined it as “the sacred quality of roasted meat.” And so that was really like a combination of the aroma of the Maillard reaction plus umami, but mashed together into one thing.

Tasting pure MSG for the first time, I immediately recognized “this is what savory is,” but the way it kind of blooms on your tongue is much more subtle. It speaks less loudly than something like saltiness or sweetness. You can get quite intense umami, but I feel kind of like the curve of it is much smoother than a kind of a spiky [flavor] like salt. Really high umami feels more like drowning in a bowl of soup than sucking on a lemon. It's a different taste, a different experience, different quality of experience.

I think people tend to equate the Maillard reaction and umami, but they're two different things, they just appear together a lot.

They often co-occur. I guess if they co-occur they can enhance each other at the perceptual level. But if anything, doing the Maillard reaction would reduce the amount of umami as amino acids were exhausted by it. You're literally breaking them for up for parts.

But there is another family of molecules that doesn't taste like umami itself, but it enhances the effects. Can you talk about those a little?

So, the primary umami molecule is glutamate—that is the building block for proteins—and its amino acid. There are umami enhancers that are called ribonucleotides. They are also building blocks, but for RNA, not for proteins. So they don't taste very umami on their own, because they don't interact well with the main active site of the umami taste receptor. Going back to the earlier thing where we don't know how salt works, we actually know this particular mechanism very well. You have a ribonucleotide on its own. Not very umami. But if you have glutamate and the glutamate is interacting with the umami receptor, the ribonucleotides will grab on to the side of the receptor—there's a spot that they accidentally combine with—and that makes the receptor kind of clamp down on the glutamate and fire off up to eight times more signal than you would get with just the glutamate without this interference from the ribonucleotides. So it is a really interesting and literal example of taste synergy. Putting these two things together can let you taste close to an order of magnitude more umami than the components on their own.

Is that why soy sauce tastes like a more complete flavor than pure MSG?

Some of it's also the aroma. We tend to associate umami with these sort of funky, fermented flavors, although that's not the only source of them. But the classic example from Japanese cooking would be dashi. So dashi has kombu, which is the world's most perfect source of glutamate. But then also katsuobushi, which is a dried, smoked, fermented processed fish. The fermentation and aging process releases a lot of these ribonucleotides in the fish. And so putting the two together makes dashi that has way more umami than just kombu dashi.

But generally meats and animal products don’t have a ton of glutamate, but they have a lot of ribonucleotides, especially if they're fresh or slightly aged. So [with] the combination of beef and cheese, you're getting ribonucleotides and glutamate together. Or even cabbage and onion. Dried shiitake mushrooms and almost any other vegetable is also a good synergistic thing. And actually in Japan, there's a group called the Umami Information Center, and some tables in my book have the content of glutamate and ribonucleotides [in foods], but on their website you can find pretty much any ingredient you could think of analyzed for umami.

So bitter, as you said, is actually a flavor that's meant to stop us from eating. Why do humans insist on enjoying bitter things?

Well, I think of acquired tastes, and the way the function of our sense of bitterness is sort of inherently flexible and adaptable. There are lots of different molecules in plants and other potential ingredients that can be toxic to us. We want to make sure we can sense that they're there so that we don't eat too much of them and poison ourselves. But you know, since the dose defines the poison, there's some cases where there's a little bit of a bitter, potentially toxic thing in these Brussels sprouts, not really enough to get us sick. And it would actually be better in the long run for us to eat the vegetable than not eat it. Or there are some bitter compounds that are actually beneficial to have in small amounts—a lot of medicinal things are quite bitter.

So we have, on the one hand, this signal of like, “Well, there's something potentially dangerous here,” but then to adapt that to these sorts of other cases where it's like, “OK, it could be dangerous, but it would benefit us more to be able to eat it.” We can actually, based on bitter being chaperoned and surrounded by other flavors, kind of dampen down the bitter signal and be like, “I know it's bitter, but we've had Campari before and we have not gotten sick from Campari. We know what Campari tastes like. It's got these other aromas.” In that case, we've had enough experience with it that we'll ignore a little bit of the potential danger. Any kind of acquired tastes like coffee, Campari, even dark chocolate, those are cases where by recognizing the aromas from experience with it, we actually develop a taste for it and can kind of quiet that bitter alarm.

What are the best chaperones for bitter?

This is somewhat anecdotal, because it's my own experience, but it really hit home for me how much aroma plays a part in this when I had a cold and ordered a Negroni at a restaurant to drink. And I was like, “It's still got the sweetness and the sourness and so it'll taste balanced.” It was undrinkably bitter. I couldn't smell anything so it was just this mix of sweet and sour together and it was really, really intense and kind of awful, with no aromas to cushion it. Basically, the more information your brain is getting about “What is this context?” the more we are able to let our guard down. So tastes are a big part of that, but also smells because they help kind of flesh out the picture with details that tastes just don't have the capacity to convey.

Is spicy a flavor?

It is a flavor. It is not a taste. It is in fact pain—accidentally chemically activating one of our burning heat pain receptors. It's sort of like a way to release endorphins without causing a lot of lasting damage. Your brain gets endorphins when you feel pain to help you through it. But when you activate the pain receptor with no actual physical injury, you get the painful experience, but then the floaty after effect as well. So the burn itself is pleasant. It's like a little tiny bit of BDSM.

This interview has been lightly edited for clarity,

Join the conversation on Facebook, Instagram, TikTok and Pinterest.

And if you're looking for more Milk Street, check out our livestream cooking classes with our favorite chefs, home cooks and friends for global recipes, cooking methods and more.

Claire Lower

Claire Lower is the Digital Editor for Christopher Kimball’s Milk Street, with over a decade of experience as a food writer and recipe developer. Claire began writing about food (and drinks) during the blogging boom in the late 2000s, eventually leaving her job as a lab technician to pursue writing full-time. After freelancing for publications such as Serious Eats, Yahoo Food, xoJane and Cherry Bombe Magazine, she eventually landed at Lifehacker, where she served as the Senior Food Editor for nearly eight years. Claire lives in Portland, Oregon with a very friendly dog and very mean cat. When not in the kitchen (or at her laptop), you can find her deadlifting at the gym, fly fishing or trying to master figure drawing at her local art studio.