Research ProposalPDF Available

Towards a universal flavor model

  • T.A.S.T.E.


Content may be subject to copyright.
© dr. P.R. Klosse
professor Gastronomy
Hotel Management School Maastricht
The Academy for Scientific Taste Evaluation
Flavor Classification; the why and how Table of contents, page 2
1. Biography ______________________________________________________ 3
2. Introduction: think different, definitions and concepts ___________________ 5
2.1 taste, flavor, tasting ________________________________________________ 5
2.2 are basic tastes really basic? __________________________________________ 6
2.3 flavor and flave ____________________________________________________ 7
2.4 the new paradigm __________________________________________________ 8
3. The flavor model ________________________________________________ 11
3.1 materials and methods _____________________________________________ 11
3.2 universal flavor factors _____________________________________________ 11
3.2.1 mouthfeel: contracting ______________________________________________ 12
3.2.2 mouthfeel: Coating _________________________________________________ 12
3.2.3 mouthfeel: dry ____________________________________________________ 12
3.3 flavor intensity ____________________________________________________ 13
3.4 aromas _________________________________________________________ 14
3.5 countervailing powers finding balance ________________________________ 15
3.5.1 coating contracting: _______________________________________________ 15
3.5.2 coating - dry ______________________________________________________ 16
3.5.3 balance in nature __________________________________________________ 16
3.6 building the theoretical model ________________________________________ 16
3.7 description of the Flavor Styles _______________________________________ 18
3.7.1 flavor styles 1 and 5, ‘neutral’ and ‘robust’ _______________________________ 18
3.7.2 flavor styles 2 and 6, ‘round’ and ‘full’ __________________________________ 18
3.7.3 flavor Styles 3 and 7, ‘fresh’ and ‘pungent’ _______________________________ 19
3.7.4 flavor styles 4 and 8, ‘balanced’, from low to high _________________________ 19
3.8 flavor profiles are dynamic __________________________________________ 20
3.8.1 influence of preparation on the flavor style _______________________________ 21
4. The use of the model ____________________________________________ 24
4.1 deliciousness and liking _____________________________________________ 24
4.2 deliciousness: culinary success factors _________________________________ 25
4.2.1 factor 1: name and presentation fit the expectation ________________________ 25
4.2.2 factor 2: appetizing smell that fits the food ______________________________ 26
4.2.3 factor 3: good balance in flavor components in relation to the food. ___________ 26
4.2.4 factor 4: presence of the umami _______________________________________ 26
4.2.5 factor 5: combination of hard and soft textures. ___________________________ 27
4.2.6 factor 6: flavor richness is high ________________________________________ 28
4.2.7 csf’s are guidelines, not a strait-jacket __________________________________ 28
4.3 wine and food pairing: new guidelines _________________________________ 28
4.4 liking ___________________________________________________________ 29
4.5 conclusion _______________________________________________________ 31
Flavor Classification; the why and how Introduction, page 3
This essay presents a theory about flavor that has developed and matured in about
thirty years, parallel to my professional life. I started my career in 1985 as a restauranteur
with a keen interest in the cooking and in wine. A wine research in 1988 sparked quite a
development which includes launching a training institute for professional chefs and
sommeliers, writing books, obtaining a PhD and ultimately a chair in gastronomy at the
Hotel Management School of Maastricht. It has culminated in the foundation of TASTE: The
Academy for Scientific Taste Evaluation.
The research that started it all took place in the family restaurant: De Echoput near
Apeldoorn in the Netherlands. The restaurant is now housed in a five-star hotel and is still
operated by the family. The kitchen is renowned for the preparation of the products from
the forest: game, berries and mushrooms and has a long Michelin star history (almost forty
years). In a good European restaurant a good advice on the combination of wines and
dishes is expected. As a gastronome by birth I am aware of the importance of this advice.
Wines and dishes can strongly influence each other and not all matches are successful. I
wanted to know why and learn about the fundamentals of flavor.
The world of gastronomy is full of traditional theories and opinions. Many of those are
hardly ever seriously questioned. The idea about zones on the tongue illustrates this. David
Hänig reported in 1901 that there are no zones for the basic tastes on the tongue, but his
work got misinterpreted and today many people believe that we taste sweet only on the tip
of our tongue, salt and sour on the sides and bitter at the back of the tongue. This is not
true, but nevertheless it continues to be mentioned in all kinds of textbooks and on the
internet. On our tongue different structures of papillae can be distinguished, but specific
zones where certain flavors supposedly are tasted, are nonexistent
Culinary education has long been about learning recipes which are basically the result
of culture. Regional differences, beliefs and limited exposure to other ways of working have
given us the beautiful cuisines of the world. This new flavor theory provides a framework
that gives new insights to food professionals, culinarians, and academics. It is powerful in
providing answers and explanations on questions that were never asked because we thought
that it was fruitless to argue about taste.
In our study we searched for fundamental answers on ancient questions; testing them
in practice and ultimately formulating a theory and validate it. This process had all the
pitfalls and deceptions you can expect. But it also led to the unexpected discovery of a
forgotten bottle of old white Burgundy that went well with a roasted leg of lamb. The wine
was ready for the kitchen sink, and you would normally always choose a red wine, but this
combination was great. Now we understand why. We even dare to think that this new flavor
approach works, independent from cultures and traditions. Many chefs and sommeliers use
it in my home country and it is spreading all over the world. Google uses it globally in their
restaurants as a part of their ‘flavor rules’ program to guide chefs in helping them to deliver
delicious dishes. There is a flavor language with parameters that are used for a flavor model
Bartoshuk, L.M. (1978). History of taste research. In Carterette, E.C. & Friedman. M.P. Handbook of
perception, volume VIA, Tasting and Smelling (pp. 3-18). New York, Academic press
Flavor Classification; the why and how Introduction, page 4
that is distinctly different from the way flavor has been addressed traditionally. This new
approach also challenges the current sensory practices in taste research.
This essay is written to explain what flavor classification is about and how it can be
done. Read it with an open mind. If you do, you’re likely to discover aspects of flavor that
you didn’t notice before. As a chef, sommelier or whatever kind of food professional you
may be going to get a better understanding about the composition of the flavor of foods and
beverages. If you want, you’re going to get more enjoyment from pairing beverages and
foods and by finding new combinations. As an academic you’ll learn that there is another
and interesting way of looking at the world of flavor.
Peter Klosse
September 2017
Flavor Classification; the why and how Think different: definitions and concepts, page 5
We don’t blink an eye when we consider color as a phenomenon of nature,
independent of human perception. We know that color is a part of the visible spectrum and
that it can be measured. There are many shades of red and they all have a number. The
Oxford dictionary tells you that color is ‘the property possessed by an object of producing
different sensations on the eye as a result of the way it reflects or emits light’.
The same dictionary describes taste differently. It is defined as: ‘the sensation of
flavor perceived in the mouth and throat on contact with a substance’. That is remarkable;
two senses are treated differently. Color is considered to be a property of an object which
causes a sensation, whereas taste is considered to be just the sensation. As if whatever we
eat or drink doesn’t have any properties. As if the differences between a steak and a
poached salmon, coffee and tea, sauvignon blanc and chardonnay, apple and banana are all
just the result of human perception.
As taste is supposed to be the sensation of flavor …’ we must first look more closely at
the word ‘flavor’. In general, flavor would be considered to be the combination of taste
(mostly referred to as the basic tastes: sweet, sour, bitter, salt, and umami) and smell. The
role of our nose in tasting is very important. First from the outside (orthonasal) and then
while chewing and eating the food, new odorants are liberated (retronasal) that trigger what
is formally called the olfactory system. Smell is an inseparable part of taste: a lot of what is
called taste is in fact registered through the nose. People with a severe cold taste much less
because aromas cannot reach the olfactory receptors high in the nose.
definition includes another aspect as well. It defines flavor as “the complex
combination of the olfactory, gustatory and trigeminal sensations perceived during tasting.
Flavor may be influenced by tactile, thermal, painful and/or kinesthetic effects” (ISO
5492:1992 Glossary of Terms Relating to Sensory Analysis). This definition includes the
sense of touch in the registration of flavor. The word ‘mouthfeel’ is used for the trigeminal
part of flavor. Furthermore the ISO definition uses ‘gustation’ instead of taste to refer to the
registration of the ‘basic tastes’. Indeed, gustation is the better word to use; if only to
separate it from the word taste.
According to this definition, not two but three neural systems are involved in flavor
registration. These neural pathways inform the brain about flavor. It is not the flavor itself
but our perception of flavor that the ISO definition is referring to, as if the visible spectrum
can be found by studying the eye or the brain. In physics, colors and sounds are described
by their objective physical qualities, like wavelength or vibration per second (millimicrons,
Hertz), and intensity (like lux or decibel). The recognition and subsequent definition of these
physical properties showed their structure and made them measurable. In other words, it
revealed their objective side which sets it apart from perception and judgement. So you may
like a certain color or not, and you may listen to the music you like, despite of their
objective qualities.
International Standards Organization, Its slogan is ‘great things happen when
the world agrees’
Flavor Classification; the why and how Think different: definitions and concepts, page 6
To be able to discuss the objective side of flavor we need measurable concepts like
frequency and intensity. For a long time these have been lacking. The focus has always
been on the so-called primary or basic tastes. If you take any basic biology book you are led
to believe that flavor can be reduced to sweet, acidic, salty and bitter. For centuries
research has been conducted on these aspects and yet it has not led to a solid and
fundamental theory. The real reason is that the hypothesis of the basic tastes is rather
Smell and taste are the chemical senses. They have generally been considered to be
inferior to vision and hearing. Perhaps they have always been too closely related to the
pleasure side of life. In ranking the importance of their senses people put sight and hearing
above smell and taste
. Illustratively: the words ‘anosmic’ (total or partial loss of the sense
of smell) and ‘aguesic’ (no perception of taste) do not belong to the daily vocabulary, like
‘blind’ or ‘deaf’. People with poor tasting or smelling abilities might not even be aware of
their problem. Loss of sight or hearing is quickly apparent to the person affected and others.
A loss of smell or tasting capacity in general occurs gradually and is often undetected. A
diminished interest in food can be the result, leading to weight loss and ultimately
undernourishment. There is a great need for methods to treat such chemosensory losses
Scientifically, smell and taste lag behind the other senses. We had to wait until these
last years for a shift in thinking. A neuroscientist, Charles Zuker, challenged the assumption
that taste cells were what is called ‘broadly tuned’, meaning that cells carry sensors for all
‘basic tastes’. Instead he hypothesized that we would have separate receptors for ‘basic
tastes’ and went looking for them
. Sophisticated techniques enable new research. In 2000
a whole range of bitter receptors were identified
and in 2001 the sweet receptors
. It
triggered a whole new way of looking at cells and receptors and led to finding ‘new’ basic
tastes like fat, calcium, carbonic acid, and metallic. Some of their receptors have yet to be
found and there is likely more to come
The chemical senses share yet another characteristic: they are hard to classify. Many
classification attempts in history took the basic four (sweet, salty, sour and bitter) as a base
of taste
. We know now that the receptors on the tongue and in the mouth are much more
Korsmeyer, C. (1999). Making Sense of Taste; food and philosophy. Cornell University Press, Ithaca, NY,
chapter 1
Finkelstein, J.A., & Schiffman, S.S. (1999). Workshop on Taste and Smell in the Elderly: an Overview.
Physiology & Behavior, 66, 173-176
Tsunoda, S, Sierralta, J, Zuker, C.S. (1998). Specificity in signalling pathways: assembly into multimolecular
signaling complexes. Curr Opin Genet Dev. Aug; 8(4):419-422.
Chandrashekar, J. et al. (2000). T2Rs Function as bitter taste receptors. Cell 100, 703711
Nelson, G. et al. (2001). Mammalian Sweet taste receptors. Cell 106, 381390
Kahn, N.A, Besnard, Ph. (2009). Oro-sensory perception of dietary lipids: New insights into the fat taste
transduction. Biochimica et Biophysica Acta 1791, 149155
Chandrashekar, J, et al. (2009). The Taste of Carbonation. Science, vol. 326, 443-445
Bartoshuk, L.M. (1978). History of taste research. In Carterette, E.C. & Friedman. M.P. Handbook of
perception, volume VIA, Tasting and Smelling (pp. 3-18). New York, Academic press
Flavor Classification; the why and how Think different: definitions and concepts, page 7
capable than always thought. With all these ‘new’ basic tastes the focus on the basic four
becomes even more debatable. They cannot be the base for classification. This can hardly
be considered as a surprise. At least three neural systems are involved in tasting. How could
the singular focus on one sensorial system (gustation) and more specifically the papillae of
the tongue ever provide a solid basis for flavor or even tasting?
Indeed, tasting, the perception of flavor, is a multisensorial experience. The question
arises why we shouldn’t also include the other sensorial systems: hearing and seeing. These
senses also have an impact on flavor perception. If yoghurt is colored red with neutral
coloring, people tend to perceive it as sweeter. The crackling sound that you hear when you
eat potato chips influences its perception, just as music or other sounds may enhance the
flavor of a certain food. Even the name of a product may be a strong driver of liking, just as
recollections and prior experiences. Read the book Gastrophysics
of Charles Spence to
learn all about it. Our minds play ‘tricks’ on us in tasting and are capable of altering the
perception of flavor and the liking of foods and drinks.
Although this is true, the same can be said of color. The perception of red is also
influenced by other colors around and by the person who sees it. At sunset the sun may
color the sky orange, but that is not the light that is emitted by the sun. Nevertheless we
know how to measure light, independent from human perception and tricks of the mind. In
Gastrophysics, Spence suggests to start using a new word, ‘flave’ to distinguish what
products have, flavor, from the total experience, flave, which would be the brain image that
we create by tasting. Flave would comprise all elements of a food or drink, including the
ones that influence perception, its taste, aroma, texture, color, sound, package, recollection,
differences between people (culture, age, gender, religion, experience, DNA, etc.) and so
forth. Flave is the total experience, both the brain image and the evaluation: yummy or
The introduction of a new word could be useful. It opens the way to study flavor as a
product characteristic which should be the basis of a model. It is functional to be able to
classify, that is why I elaborate on this. If you pour twelve glasses out of a bottle of wine,
twelve people may have different opinion about the wine, some may love it, others not, and
a few will even be able to identify it, without having seen the label. There are many similar
examples that boil down to the same: flavor is something tangible and it is time to start
developing concepts that help us to identify it. It is the flavor, and not the taste, of a certain
dish or a wine that is being tasted. The word ‘taste’ can be used in regard to a person’s
personal liking. The Oxford dictionary exemplifies: this pudding is too sweet for my taste, or
have your lost your taste for fancy restaurants? Taste is not a sense, and what we often call
taste is the part of flavor that we should call gustation.
Gastronomy is the science of flavor and tasting. It embraces both products and people
in regard to eating and drinking. It constitutes a liaison between natural sciences like (food)
chemistry, physics and bio-chemistry, etc. and human related sciences like psychology,
sociology, marketing, etc. Tasting is the sensorial activity to register flavor. We use all our
senses to do that; it is multisensorial experience: gustation (papillae on the tongue),
Spence, C. Gastrophysics. The new science of eating. Viking/Penguin Random House UK (2017)
Flavor Classification; the why and how Think different: definitions and concepts, page 8
olfaction (receptors in the nose), touch (mouthfeel, by the trigeminal receptors), hearing,
and seeing. Tasting is influenced by many aspects, like color, packaging, price, culture,
experience, etc. A gastronome is a well-educated food professional with a solid
understanding of the processes that are involved in eating and drinking. He or she is able to
guide the creation and development of innovative F&B concepts. His or her expertise will
help others to make the good decisions that lead to a better tasting world.
Products have flavor and people taste them. If we want to develop a classification, we
should look at flavor. As a metaphor, we want to inspire academics to search for the
frequency of flavor; and consequently to develop measures, scales or a model that describes
what it is. The figure below illustrates this.
Figure 1: Flave: combination of product properties (flavor) and people (tasting)
The flavor theory challenges the existing paradigms about flavor. Reductionism rules in
the world of flavor as we have seen, yet it fails to explain what happens in the world of
flavor perception. Can fundamental answers be expected from reductionist, deductive
reasoning, if the whole is more than the individual parts? Surely, a dinner at a table, be it in
our home or in a restaurant, is very different from having a bite or a sip in isolated white
booths with infrared lighting or in the tube of a fMRI scan. The reductionist focus on the
basic flavors has always dominated the science of flavor; in other words: the paradigm used
was far from complete. And even if all senses that are involved in the perception of flavor
were included, the role of the brain and the gut needs to be included
Reductionism also rules in the world of food where focus has shifted from whole foods
to the nutrients they contain. The word nutritionism has been introduced to indicate this
Sclafani, A., Ackroff, K. (2012). Role of gut nutrient sensing in stimulating appetite and conditioning food
preferences. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology. 15 May
2012 Vol. 302 no. 10, R1119-R1133
Flavor Classification; the why and how Think different: definitions and concepts, page 9
shift from real foods to the nutrients they supposedly contain
. The food and health
industry have exploited these ideas to the max. More and more the positive or negative
health effects of our food intake are advertised. The inescapable result is that the control of
the food industry over our food choices has increased. Gastronomy can be considered as the
antonym of nutritionism, as it is focused on the whole experience, including the liking and
enjoying of food and drink. The holistic perspective looks from a different angle to the food
related problems of modern Western societies. It focuses not solely on the food and its
composition, but on the human that eats it as well. And, in doing so gastronomy is
especially interested in why this human likes the things he eats or drinks.
If flavor can be assessed objectively, we can start thinking how different people react
to it. This is valuable step towards getting a better understanding about deliciousness and
liking. We need a model to do that, but defining new concepts is not enough. In the physical
world concepts become real as soon as they can be perceived and measured. Before Isaac
Newton, people believed that white light was colorless and that the prism produced the
colors. This paradigm shifted when Newton’s experiments demonstrated that white light
actually consisted of ‘the colors of the rainbow’ which the prism made perceptible.
Subsequently the visible spectrum was made measurable and now we know that the
wavelengths between 622 and 780 nm correspond to the color red. This is purely functional
and has no relation whatsoever with us being able to see red, our opinion, love, or cultural
or practical uses.
The mission for flavor to ‘become real’ is therefore to develop scales and measures.
Newton started by studying natural, physical phenomena. Likewise it is logical to start
thinking about classifying flavor on the molecular side. We have found that the concepts
mouthfeel and flavor intensity are effective descriptors of flavor. They form the basis of a
three-dimensional model: the flavor styles cube with eight distinct flavor styles. This theory
is scientifically validated in my thesis The concept of Flavor Styles to classify Flavors
fully described in The Essence of Gastronomy; understanding the flavor of foods and
. What we need next is being able to measure flavor. This is not an easy task
and will require many coordinated studies on all kinds of aspects of flavor.
New insights don’t always come easily, but once they are there, everything falls into
place and it is hard to conceive that you didn’t see it before. That is the virtue of a good
theory: it provides a framework that explains the phenomena that we perceive. You need a
theory to explain the facts and the theory stands until facts turn up that call for a critical
evaluation, an adaptation or even a novel theory, a new paradigm. Mostly it takes a while
before new theories are accepted. This is not strange. New insights are often the result of
libertinism and a break with traditionalist views. They question what is generally believed
and accepting something new, implicates changing beliefs, teachings, curricula and so on.
Indeed it is much easier to leave everything as it was and to try to squeeze the new facts
into the old frameworks. However it is not the way to progress. Max Planck pointed out that
Pollan, Michael. In Defense of Food. An eater’s Manifesto. Penguin Books (2009)
Klosse, P.R. The concept of flavor styles to classify flavors. Academic Thesis, Maastricht University | Academie
voor Gastronomie, Hoog Soeren, The Netherlands (2004)
Klosse, P.R. The Essence of Gastronomy; understanding the flavor of foods and beverages. CRC Press/Taylor
Francis, Boca Raton, USA (2014)
Flavor Classification; the why and how Think different: definitions and concepts, page 10
“a new scientific truth does not triumph by convincing its opponents and making them see
the light, but rather because its opponents eventually die, and a new generation grows up
that is familiar with it.”
The flavor styles model is the best available map to embark upon this academic
journey. As with any other expedition, a map is a guide. While travelling you keep your eyes
open, as you never know what comes across. This essay is an invitation: researchers,
institutions and companies are welcomed to get on board, support the expedition and help
to get as far as we can on this mission.
Kuhn, Thomas (1962), The Structure of Scientific Revolutions. Chicago University Press: p. 150 (1970 ed.).
Flavor Classification; the why and how The Flavor model, page 11
We have an interest to have a model that describes and classifies flavor. Such a model
could be used for all kinds of purposes and requires a technical, objective approach. It
should help us to identify flavors and communicate about them. To do so we need to find
and define concepts, powerful parameters that characterize foods and beverages and can be
measured. The flavor styles model has proven in practice to be functional to make the flavor
tangible and intelligible, at least to certain extent. Its concepts, contracting, coating, dry
and flavor intensity have proven to be valuable descriptors. We will discuss them in detail.
The development of the flavor theory didn’t start out as a well-defined mission. It grew
organically. The original study was about the flavor of wines to get a better understanding of
the combination of wines and foods. Together with Dr.Ir. Cramwinckel of Wageningen
University (at the time) in 1989 a control group of twenty wine lovers was formed. During a
year they convened for a series of tastings at restaurant The Echoput in The Netherlands.
Every time a different series of wines was tasted and the group was asked to assess
the differences. What makes one wine taste different from another? Advanced statistical
techniques, principally one that is known as ‘free choice profiling’ were used to interpret the
tasting notes. This research technique is divided in two parts. In the first part subjects use
their own words (free choice ..). With factor analysis words that have things in common are
be clustered. The group is asked to find words that adequately describe each cluster of
words. This is an interesting process. It is not always clear what the common ground is and
which label can best be used to describe it. When consensus is found, it is time for the
second part of the research. The mission is now to use the labels that are found and to
evaluate in tastings if they indeed can be used to indicate differences in wines (profiling ..).
In this process labels are discarded or redefined if necessary. After quite some time it boiled
down to mouthfeel, flavor intensity and flavor type. Last mentioned concept is important for
identifying and matching foods and beverages, but not a part of the classification model. As
we now know, mouthfeel and flavor intensity are useful to describe the flavor of wines,
foods, dishes. In fact all kinds of food products and all types of drinks can be classified and
analyzed in using these terms. All of a sudden interactions between wines and foods could
be explained. The theory gives an insight in the composition of dishes and a structure for
developing or improving them. The terminology turned out to be of great communicative
value. The terms are simple, easily applicable, value free and universal. Let’s see.
Mouthfeel and flavor intensity came out to be the universal flavor factors that help to
describe the flavor not only of wines, but also of chocolate and chips, French fries and
mayonnaise, soft drinks and milk, beer and wine, coffee and tea, meat and fish, fruits and
vegetables. In other words, in everything we could possibly taste, these same flavor factors
can be distinguished, in all kinds of ways, combinations and intensities. Mouthfeel is the
core concept of the classification. Mouthfeel is about the feeling a product gives in the
mouth, whether it is a food or a beverage. Within mouthfeel three classes are distinguished:
Flavor Classification; the why and how The Flavor model, page 12
Think of what you feel in your mouth when you eat a green apple, fresh lettuce, citrus
fruits, leek, turnip, all kinds of radish, raw onions, chives, parsley, vinegar, red pepper,
mustard, wasabi, raw ginger. In general: acidity, salty and all kinds of irritation (pepper,
mustard, horse radish, ginger) cause contraction in the mouth. Just as carbonic acid (CO2)
does in a variety of drinks, such as mineral water, sparkling wines, beer and soda. Light,
fresh white and red wines with a nice acidity are examples of ‘contracting’ drinks. A low
temperature also makes the mouth contract. This implies that serving temperature
influences mouthfeel (and flavor intensity as we will see). Contraction gives the impression
of refreshment, of cleansing the mouth. Contracting elements will often stimulate saliva
Think of how a spoonful of honey will coat the inside of your mouth. Fats, like butter,
cream, animal fats, all kinds of oils and for instance peanut butter do the same. The word
coating is chosen because these elements leave a thin layer behind in the mouth. Saliva
becomes thicker, more viscous. You can imagine that a syrup with a little water is much
more coating than the same syrup with more water. Sweet wines like port and the so called
noble rot or botrytis wines are rich in sugar and therefore mostly coating.
Besides fats and sugar, there is a third group that can be coating: proteins. Just
imagine a soft-boiled egg and you have the best example. At the same time you’ll realize
how preparation influences mouthfeel. The same egg, cooked just for some minutes longer
is no longer coating, but dry, especially the yolk! The same happens when meat or fish has
been cooked just a bit too long, instead of juicy and on the coating side, it lost its juiciness.
Some people like it that way, while others do not. That is why the waiter asks how you like
your meat. Umami will be elucidated later in this booklet. For now we mention that it has
an effect on coating. Think examples like gelatin, meat stocks, dried mushrooms, fermented
sauces, cured hams, and wood fermented chardonnay wines of which the lees have been
stirred regularly. Generally people refer to Umami food as ‘rich’ or savory.
3.2.3 M OU TH FE EL : DRY
Think of biscuits and crackers, puff pastry, crusts, crisps and chips, and - more in
general - about the ‘bite’ of something. Some textures are hard and break easily, and
sometimes the break is somewhat soft, but resistant, it will be more chewy. We call this
elasticity. Bitter will often be a dry or drying component, just as tannins in wines and hop in
beer. Astringency is a form of dryness. It is mostly found in types of bitter substances that
reduce the coating capacity of saliva. The tannin of young red wines is a good examples of
this effect. Other examples include the bitterness of espresso coffee and some types of tea.
Plain chocolate is also a beautiful example: the higher the cocoa content in the chocolate
composition, the harder and drier it will get
Flavor Classification; the why and how The Flavor model, page 13
Components that absorb water (including saliva) are also counted as dry. Think of how
rice, potatoes, or bread absorb all kinds of fluids. Dry is neither contracting nor coating. It
may even rather ‘tasteless’; soft mineral water, plain bread, rice could be characterized as
being ‘neutral’. At wine tastings you often see some lumps of bread for that reason. Toast or
crusts in general absorb as well, but are not as neutral as the crumb of bread. Just think of
how the crust of freshly roasted meat, toast or puff pastry enhances flavor and also how its
crispness influences our perception of quality.
We see the three classes of mouthfeel as forces, vectors if you want. For something to
be tasted, it needs to be above the so-called sensory threshold. That is where the intensity
starts. In essence, flavor intensity is about the force of flavor. Not all intensities are equal
and they mutually react. There are elements that enhance flavor and there is suppression or
masking. It is a real power play. Salt and Umami components are well known as general
flavor enhancers.
Within each class of mouthfeel there are differences in intensity. Take acidity: we can
measure the acidity by looking at the pH and the concentration. The lower the pH and the
higher the concentration, the greater the intensity and the force of contraction will be.
Likewise there are many mustards on the market that may be very different in contracting.
In many cases, mixing ingredients within the same class will also lead to a rise in intensity.
Think of how the flavor intensity increases when you add salt or red peppers to a regular
vinegar. Or how the force of coating increases by mixing sugar and fat. This particular mix is
suggested to be the secret behind many popular industry foods.
Concentration (reducing amount of water) and cooking techniques are clearly ways of
increasing flavor intensity. Dried fruits or mushrooms have a higher flavor intensity than the
regular ones. And cooking techniques like grilling, frying, and especially deep-frying are also
known to raise flavor intensity. A piece of fish or chicken can easily illustrate that. Imagine it
poached or steamed, its fragile flavor is well preserved, and the flavor intensity stays on the
low side. Now fry it and the flavor intensity goes up. Fry it even harder or grill it and further
up it goes. It may even reach a point that the flavor of the preparation gets more important
than that of the primary ingredient: all you taste is the brown crust; it dominates the
original flavor of the meat.
In these processes many bio-chemical reactions take place. In the Maillard reaction
between proteins and carbohydrates hundreds of new flavors come to life. In fermentation
and aging flavor intensity is also likely to be an important issue. In these cases enzymes
often play an important role. They break down chains of e.g.. proteins freeing amino acids
like glutamate that enhance flavor. Think of how the flavor intensity (and often mouthfeel
as well) change in these processes. Just compare young cheeses, ham, wines, brandy, etc.
to the older ones.
Some products plainly have more flavor than others. If you compare apples and
oranges the flavor intensity of most oranges will be higher flavor than most apples. It may
seem strange to compare the fruits in this respect, but it can be done. It can even be
measured quite simply. Have a bite of an apple, then of an orange, then go back to the
apple. The second time around, you’ll hardly taste the apple at all; its flavor is dominated by
the orange that you tasted before. It implies that the flavor intensity of the apple is indeed
Flavor Classification; the why and how The Flavor model, page 14
lower. This is called a triangle test. It is a very useful technique to evaluate the flavor
intensity of two products.
It is important to note that a difference in flavor intensity doesn’t say much about
quality. Just as the volume knob of the stereo does not say anything about the quality of the
music. You may even take this a step further, there are many examples of products with a
lower flavor intensity that are much appreciated in the kitchen. Take poultry for example, its
flavor intensity is lower than that of beef. Yet it is highly appreciated across the world.
Products with a high flavor intensity can be rather dominating, pushing away interesting,
but fragile flavors. On the other hand, lower flavor intensity demands attention from the
taster. Metaphorically speaking, flavors that are low in intensity don’t come to you, you have
to go to them. This means that a high flavor intensity is ideal in all those situations where
you don’t want to give yourself the trouble of tasting. This is exactly why the flavor intensity
of successful fast-foods is high.
The figure below illustrates the classes of mouthfeel and the role of intensity.
Figure 2: the vectors of mouthfeel
3.4 A RO MAS
Earlier we mentioned that everything we eat and drink has gustatory (taste), olfactory
(smell) and tactile (touch) components. The tactile and gustatory elements are obviously a
part of the classification, how about the olfactory components? There is no doubt that they
are equally important as well. Mouthfeel and intensity give structure to flavor, make up its
framework; the aromas give flavor its character. Think of music. Hertz and decibel are the
scales that we use to measure sound. And then there is music with its melody, harmony and
instrumentation. Aromas are the music of flavor.
Aromas are versatile. To reduce the complexity of naming aromas we define the two
extremes: fresh and ripe. Apple and citrus fruits are nice examples of fresh. Just as parsley
or mint, or cucumber and fennel. Ripe flavor tones are those of ripe melon or pear, or of
rosemary and garlic. Also nuts, honey, caramel, mushrooms (especially dried ones) and
spices like vanilla, nutmeg, cinnamon and the sultry aromas of musk and roses are
examples of ripe flavor tones. The word 'ripe' as flavor tone should not be confounded with
Flavor Classification; the why and how The Flavor model, page 15
the maturity of the fruit. A ripe apple or lemon is always fresh. In general we use ‘fresh’ and
‘ripe’ to characterize the flavor type. Although it is not a part of the basic classification, as
we will see, it is useful make this distinction, especially in flavor composition and food and
beverage pairing.
It is interesting to note how forces interact; some compensate, others intensify. Here
are some examples to illustrate:
sweet sour: many beverages
This is probably the most well-known of all countervailing powers: sour can be masked
by sweet. It is used extensively in daily products. There is even an algorithm for it: the
German wine law states that white wines can be called trocken (dry) on the label when
the wine has a maximum of 9 grams of sugar per liter with a maximum of 7 grams of
acidity. It is generally called the ‘acidity-plus-two-rule’. Likewise, Champagne can be
called brut (dry) with a maximum of 13 grams per liter. In both of these cases the
sweetness is not noted; the wines would be characterized as being on the contracting
side. It is important to note that the acidity masks the sweetness; it doesn’t take
either away.
Also note the role of CO2 (contracting). In wines and other drinks balance is found in
playing with the contracting elements acidity, carbonation, temperature and sweetness
as coating element.
fat sour: vinaigrette:
In this case the oil (coating) and vinegar (contracting) find a balance somewhere,
aided by mustard, salt and pepper (all contracting). Types of oil differ in viscosity and
therefore in coating, just as types of vinegar and mustard can be (very) different
in contracting. You need your tongue and personal preference to find the ‘right’
It has become quite common to add sugar (coating) to the vinaigrette. It indeed helps
to mask the acidity and adds on the coating side, as we have seen above. This means
that if you find your vinaigrette not balanced enough because it is too acidic
(contracting), there several solutions: add more oil, or use a different type that is
more viscous, or add sugar or some other binding agent that would support the
coating side. It is also good to mention that balance is not necessarily found when the
parts are equal. Most vinaigrettes are made with at least 70% oil.
sugar/fat cold: ice cream
The basic mixture of any ice-cream is certainly sweet and may also contain fat (cream)
of proteins (eggs in a parfait). So: very coating in the unfrozen form. By freezing the
mixture, the contracting power of cold changes the mouthfeel. The size of the crystals
has an impact as well: you don’t want a hard lump of ice, that why you freeze it while
constantly stirring the mixture.
sugar spicy: chili sauce
In the Asian cuisines spiciness, the contracting force of all kinds of peppers is used to
Flavor Classification; the why and how The Flavor model, page 16
balance either the coating of e.g. fat of sweet. The Scoville scale can be used to
measure the intensity of the spiciness.
3.5.2 C OA TING - DRY
cream crust: many desserts
Even though many doughs are made with butter and sugar, the end result is often dry
because of the starch. Think of the balance of a well-made millefeuille: the crisp and
dry layers of puff pastry mixes beautifully with all kinds of creams that are rich and
coating. The dryness of the pastry balances the coating richness. In fact, many
desserts rely on a good balance between dry and coating elements. Now also realize
the importance of keeping your crust crispy. As soon as the crispness is gone, the
balance is gone and the products loses much of its appeal. That is also a reason not to
wait to eat your French Fries: if they lose their crust, they become soggy and
sugar/fat bitter: chocolate
Chocolate is a beautiful illustration of finding balance on all kinds of levels. Basically it
is a mixture of two coating element, sugar and fat and one dry element, cocoa. The
mouthfeel of chocolate has everything to do with the amount of cocoa in the mixture
and varies from coating (white chocolate, no cocoa) to dry and hard (plain chocolate)
and all the mouthfeels in between. Note that there are lovers of chocolate in all styles.
The function of adding cream or sugar to coffee or tea can also be mentioned in this
respect: they mask the bitterness and therefore make it less dry
In the examples above balance is found by mixing substances. Nature has its own way
of giving us balance. In a green apple there is acidity (contracting), sugar (coating) and it
has a bite (dry). Different variety of apples have other balances some are less acidic and/or
more sweet, others are less hard. A melon, peach or avocado is hard to start with but gets
juicy and more sweet (coating) in ripening. Also consider how the crema-cover on an
espresso is coating, whilst the coffee itself is dry especially when a little water is used, as in
a ristretto. The quality of the brew is judged by the thickness and color of the crema.
Another illustrative example is found in olive oil, which is essentially coating. The best olive
oils, however, have a spicy, peppery note that is contracting, and makes that those quality
oils are not as viscous as the simpler and cheaper types.
Adding intensity to the vectors of mouthfeel, makes a three-dimensional model: the
Flavor Styles Cube (figure 3). The name is derived by making eight subcategories, flavor
styles, that can be distinguished by looking at their basic characteristics as is shown in the
following table:
Flavor Classification; the why and how The Flavor model, page 17
neutral, light, balanced, subtle
round, smooth, supple, creamy
fresh, sour, contracting,
Robust, solid, powerful
full flavor, ripe flavor, filling
pungent, spicy, hot, explosive
Complex, differentiated, satiating
Table 1: The basic characteristics of the flavor styles
We now have eight subcategories, the flavor styles. Every style has specific
characteristics. They can be used to identify the location of a food or drink. This model is a
depiction of the world of flavor and serves as an instrument for classification. It doesn’t
imply that there are only eight flavors left. Compare this to light for example. The flavor
styles correspond to the colors of the rainbow. Every color has many shades and they fade
beautifully into each other. There is an obvious difference as well. The exact position of any
color is precisely known. You can give a number to get the desired color. In flavor we are as
yet nowhere near such exactitude.
Figure 3: The Flavor Styles Cube
Products that are positioned close to a specific corner can be considered to be ‘icons’
of that specific flavor style. Flavor styles within the cube that lie above each other are
closely associated. This implies that these flavors are similar in mouthfeel but differ in
intensity; their basic characteristics are discussed below.
Flavor Classification; the why and how The Flavor model, page 18
In this section the Flavor Styles are discussed in detail. Table 2 gives an overview of
practical examples of each style.
Flavor style 1 is called ‘Neutral’. The icons are plain water or white bread. The bottom
left corner represents ‘tasteless’ or very little taste both in mouthfeel and intensity. In the
history of flavor classification Fernel proposed at one time the term ‘insipid’
. A term that
still exists these days but it has a negative association. We prefer to use value-free terms as
indications for the specific flavor styles. ‘Neutral’ would be an appropriate term to
characterize this flavor style. It is important as well because this is the group absorbing
starches and the juiciness or moistness, that is a highly appreciated element of many foods.
Flavor style 5 is the extension of style 1 and is called ‘Robust’. The icons are dark
chocolate and ristretto. Mouthfeel is basically dry. Several influences may account for
dryness. The first is absorption of saliva as happens with dry solid foods such as toast,
biscuits, crackers, and chips. Next, tannins are known to affect saliva making the mouth
. Astringency is the appropriate term for this kind of bitterness. Some young red
wines can be ‘drying’ or ‘puckering’
. Likewise some other bitter tasting substances like
strong coffee or tea and dark chocolate with a high percentage of cocoa can be astringent,
bitter or acrid
Flavor style 2 is called ‘Round’ and the higher intensity, flavor style 6, ‘Full’. The icons
are respectively simple oil and caramel. Both styles are characterized by the coating layer
that is left behind in the mouth like fats and/or dissolved sugars do
. However different,
fats and the mono- and di-saccharides share that they are coating in mouthfeel.
Physiologically these aspects of flavor are sensed by the mechanoreceptors of the trigeminal
system. Their acuity is quite remarkable: even small differences in viscosity can be
perceived and particles as small as 5µm (that is 0,05 mm, just slightly bigger than one
bacterium) can be detected
. The difference between flavor style 2 and 6 is explained by
flavor intensity.
Bartoshuk, L.M. (1978). History of taste research. In Carterette, E.C. & Friedman. M.P. Handbook of
perception, volume VIA, Tasting and Smelling (pp. 3-18). New York, Academic press
Kallithraka, S., Bakker, J., Clifford, M.N. Vallis, L. (2001). Correlations between saliva protein composition and
some T-I parameters of astringency. Food Quality and Preference, 12, 145-152
Gawel, A., Iland, P.G., Francis, I.L. (2001). Characterizing the astringency of red wine: a case study. Food
Quality and Preference, 12, 83-94
Drewnowski, A. and Gomez-Carneros, C. (2000). Bitter taste, phytonutrients, and the consumer: a review.
Am. J. Clin. Nutrition, 72, 1424-1435
Rolls, E.T., Critchley, H.D., Browning, A.S., Hernadi, I., Lenard, L. (1999). Responses to the sensory properties
of fat of neurons in the primate orbitofrontal cortex. Journal of Neuroscience, 19 (4), 1532-1540
Guinard, J-X., Mazzucchelli, R. (1996). The sensory perception of texture and mouthfeel. Trends in Food
Science & Technology, 213-219
Flavor Classification; the why and how The Flavor model, page 19
Flavor style 3 is called ‘Fresh’. The icons are respectively vinegar and red peppers. The
flavors in this section are basically characterized by contracting elements in foods and
drinks. Acids and salts are well-known substances that have a contracting influence, just like
CO2. Acidity is also reported to increase saliva flow
. Physiologically these substances are
reported to enter the ion-channel of taste cells directly
Flavor style 7 is called ‘Pungent’. It is the style where we principally find the oral
irritants, like capsaicin, gingerone, which all activate trigeminal pain pathways. Those
reactions are mediated by the nociceptors of the trigeminal system that also enter directly
into ion-channels as well
Flavor Styles 4 and 8 are both ‘Balanced’ but the flavor richness is different. The icons
are respectively yoghurt and well-made dish or Roquefort. Flavor becomes balanced if
coating and contracting forces more or less compensate each other. Furthermore, the so-
called fifth basic taste, Umami, is principally active in these flavor styles. Basically Umami
stands for the taste that is enticed by glutamic acid. It is widely used in the food industry to
enhance flavor, palatability and overall preference
. Statistical analysis of terms that are
used to describe umami flavor, showed a correlation with ratings for complexity, mildness,
brothy, and meaty
. These are descriptions of flavor that would perfectly suit these styles.
Guinard, J.-X., Zoumas-Morse, Chr. & Walchak, C. (1998). Relation between parotid saliva flow and
composition and the perception of Gustatory and Trigeminal stimuli in foods. Physiology & Behavior, 63(1),
Gilbertson, T.A., Damak, S., & Margolskee, R.F. (2000). The molecular physiology of taste transduction.
Current Opinion in Neurobiology, 10 ( 4), 519-527
Carstens, E., et al. (2002). It hurts so good: oral irritation by spices and carbonated drinks and the underlying
neural mechanisms. Food Quality and Preference, 13 (7-8), 431-443
McCleskey, E.W., Gold, M.S. (1999). Ion channels of nociception. Annual Review on Physiology 61: 835-856
Schiffman, S.S. (2000). Intensification of sensory properties of foods for the elderly. Supplement to the
Journal of Nutrition, 130, 927S-930S
Fuke, S. & Ueda, Y. (1996). Interactions between umami and other flavor characteristics. Trends in Food
Science & Technology, 7, 407-411
Flavor Classification; the why and how The Flavor model, page 20
Table 2
Flavor Style
Typical Examples
1. Neutral
Rice, white bread, zucchini
poached/steamed fish
Plain Water
2. Round
Spaghetti, milk chocolate,
banana, melon, avocado,
young Gouda cheese,
butter, plain oil
Milk, wines with a touch of
3. Fresh
apple, orange, lemon,
green salad, pickles,
oysters, tomato, fresh
goat cheese, vinegar
orange juice, acidic white
wines, like Sancerre,
Muscadet, Riesling
4. Balance low
Vinaigrette, creamy
tomato sauce
yoghurt, balanced daily
5. Robust
Dark chocolate, grilled red
Espresso, strong tannic red
wines with wood character
6. Full
vanilla, caramel, whipped
cream, vegetable oil,
peanut butter
Sweet, concentrated
wines, fortified wines
7. Pungent
Peppermint, mustard,
horseradish, red pepper,
ginger, spicy dishes
Some sparkling wines
8. Balance high
braised meat, mushrooms,
Rich, full-bodied yet
balanced wines (not
Table 2: The Flavor Styles and typical examples
There many variables that ultimately give a certain flavor profile. Therefore it is hard
to say something beforehand about what the mouthfeel and the intensity will be; it is an
interplay of many factors.
Flavor can for instance be strongly influenced by temperature. Temperature is
registered by the trigeminal system. Besides the real warm and cold that you measure,
there are also substances like spearmint and menthol that give a cool sensation. Likewise,
there are substances like ethanol and capsaicin cause a burning sensation.
We already mentioned how the perception of sweetness is enhanced by higher
temperature and is suppressed when something is very cold. Therefore, the serving
temperature of wines can influence their flavor, especially if the wines have some residual
sugar. If they are served on the cool side (<10 °C) the balance moves toward contracting
because the sweetness is suppressed. The acidity shows more strongly. Serve it a few
Flavor Classification; the why and how The Flavor model, page 21
degrees warmer and the mouthfeel moves toward coating. It is an interesting experience.
Even more so when this same wine at the low temperature gives a better match with a
certain dish, than when it is slightly warmer or the other way around. A sommelier in a
restaurant can actually improve a match just by playing with the temperature of a wine.
Balance in mouthfeel can also be altered by a change in dryness. The crispness of the
crust of many products is an important aspect of the mouthfeel. If the crispness gets lost,
they lose their appeal as well. That is why you want your bread to be freshly toasted and
there are many similar examples. In these cases the crispness is even essential for liking
the product.
Flavor intensity and flavor type are also influenced by temperature. Many aroma
compounds need a certain temperature to become volatile. Anything that is served too cold,
will appear to be less rich in flavor than when served at the proper temperature. Most
cheeses are best when served at room temperature. When chilled, the flavor of a creamy
cheese like Brie or Camembert is not as rich as when the same cheese is savored at room
temperature (around 22 °C.). Room temperature is often advised for red wines as well, but
in fact many are better at 18 °C.
The flavor of many cheeses is also strongly influenced by ageing, just as many other
products. In natural maturation processes enzymes work on proteins. Such a process may
take a year or more. Think of how the flavor of cheeses like Gouda or Cheddar develops. In
some cases weeks of difference can already cause a big difference in flavor. Beverages like
wines, whiskey or brandy can also gain flavor richness in ageing. This may be due to
complex interactions between aromas and alcohol or evaporation. In general flavor intensity
tends to increase and mouthfeel changes as well.
Consequently, if you want to evaluate the flavor profile of a certain product you need
to do it on the moment of tasting. And if you want to make general suppositions about the
flavor of products you need to take into account the physical conditions like temperature,
age, amount that is tasted, changes in texture, etc.
Preparing food is all about transforming molecules. This an unromantic approach to
cooking and some even consider it as ‘not done’ to consider cooking from a physical and
chemical point of view. Cooking techniques are likely to have an impact on mouthfeel,
intensity and the aromas. A great chef will always be critical in the selection of his
ingredients. Bad ingredients will never lead to something great. On the other hand, a good
chef is able to transform even average ingredients into a delicious dish. It is essential to
know about and master techniques, and especially what their influence is on flavor. The
following examples illustrate this.
Whitefish (for instance turbot):
Respects the pure flavor of the fish; mouthfeel: rather neutral (if the fish was cooked
correctly); Flavor intensity: low to medium, neutral in flavor type
Mouthfeel will be a little more contracting; more flavor richness, with ripe flavor tones.
The flavor of frying should not dominate the flavor of the fish
Flavor Classification; the why and how The Flavor model, page 22
Mouthfeel will be more contracting, and flavor richness will be higher again, as will the
ripe flavor tones. The flavor of grilling risks to dominate the flavor of the fish.
Red meat:
(Thin slices of raw meat, marinated in oil/vinaigrette). Flavor intensity is on the low
side and quite dependent on the vinaigrette and the garnish. The use of Parmesan
cheese or roasted pine nuts is more influential than often thought.
The meat is cooked in the stock and not been roasted, so there is no Maillard effect,
nor a crust. Flavor intensity will be medium, depending on seasoning, herbs and spices
This preparation is closed to what people expect with red meat. If done perfectly,
there is a flavorful crust and a juicy inside. Flavor intensity is above average and with
ripe flavor tones.
Mouthfeel is somewhat dry, absorbing. Flavor intensity is low to medium. Varieties of
potatoes will show differences. Some keep their texture after boiling, others fall apart.
It is good to be aware of these differences and to select the right one for the purpose.
Also, the texture of the potato variety appears to have an influence on the way they
are cooked. When you compare several varieties that are boiled with the same amount
of salt, you are likely to experience differences in saltiness.
Mouthfeel will be coating, especially when butter, olive oils or eggs are added. Such
additions also influence Flavor intensity, which gets higher.
French fried potatoes
Mouthfeel is contracting because of the crust. Flavor intensity is high, with ripe flavor
Bell peppers:
Mouthfeel contracting because of the specific acidity. The flavor tones are fresh and
flavor intensity is high and rather dominating, one-dimensional.
Mouthfeel is notably less contracting and may even have turned to coating. This partly
dependent on the color of the pepper: the red ones are riper harvested later and
therefore basically sweeter. Flavor intensity is high, but now with ripe flavor tones.
Mouthfeel is on the contracting side, despite the apparent sugars. Flavor intensity is
rather high and the flavor tones are fresh.
Canning requires sterilization and therefore a prolonged period of high temperatures.
Flavors change, partly because of the syrup that is added and partly because of
transformations in the acids, which makes them softer. In short: canned fruits,
Flavor Classification; the why and how The Flavor model, page 23
including pineapples, are very different from the fresh ones. The Mouthfeel is notably
more coating; texture changes as well. Flavor intensity is still high, but less fresh.
(This applies to other canned fruits as well, think of peaches or pears).
Thin slices can be dried in an oven or a microwave. Texture is again very different, it
has turned crisp. Mouthfeel is therefore more contracting; Flavor intensity remains
high, but the flavor type is more ripe.
Just to finish with a practical example, we’re theoretically going to prepare a chicken
breast with a creamy sauce with tarragon. Let’s see how we can manipulate the flavor
profile of the dish? Its mouthfeel is mainly influenced by the sauce. If it is reduced more, it
gets more coating. If we want it more on the contracting side we should, for instance, start
with a reduction of some wine and shallots. When the cream is added, the level of reduction
determines how coating the sauce will be. Flavor intensity is controlled by the level of frying.
If we fry it a little too long, the mouthfeel of the chicken will turn dry. The tarragon in our
sauce is added for the fresh flavor tones. If we want to pair it with a wine that is aged in
wood we might want to use rosemary or thyme instead, or add some mushrooms and/or
garlic. This demonstrates that the flavor profile of any dish is not a given fact. It is in the
hands of chefs, or winemakers and brewers in the case of wines and beers.
Flavor Classification; the why and how The use of the model, page 24
Food quality can be considered both the most well-defined and ill-defined concept
in the food industry today. Food scientists or professional chefs are likely to define food
quality from a product point of view. Their definition does not necessarily correspond with
consumer opinion. For commercial food products it is essential that a product has a high
quality from a consumer point of view. For commercial food companies, restaurants
included, it is essential to know the ‘drivers’ of product acceptance
In the introduction flavor and tasting were launched as two different concepts.
Although they are closely related, it is useful to make the distinction and use the words
correctly. The same can be said from deliciousness and liking. Deliciousness relates to
flavor as liking relates to tasting. The first is product related, the second human related.
The logical consequence would be that deliciousness could also be approached
objectively, just as flavor. Liking is a personal judgement and by definition subjective.
Deliciousness becomes a mission and is in the hands of the producer, chef, winemaker,
even of nature itself, while liking is the desired consumer response.
An interesting perspective opens up. Is ‘yummi’ a coincidence? Is there is a right
way of doing things? To say the least, there is a lot of objective knowledge involved.
Harold McGee’s ‘On Food and Cooking’
unlocked many secrets. It is likely that more
people like foods that have been prepared perfectly than the ones that have been ruined
by a bad cook. We can also approach the question from the other side: if yummi would
be a coincidence, it would be reasonable to suppose that the group of people that like a
certain product or combination would always be find a similar number of people that
dislike it. We know that this is not the case. Although there are in my knowledge no
foods that are universally liked it is clear that there are foods that are liked more than
others and even more importantly: it is safe to assume that the ones that are well
prepared are better liked than the ones that are less well made.
Liking on the other hand, is certainly subjective. It is a personal judgement. People
are not the same which implies that the same food may be liked by the one and disliked
by the other. Tasting capacity differs from person to person and there are many other
influences. Tasting is learning. Some preferences are ‘acquired tastes’. The liking for
beer, Brussels sprouts, coffee, and dark chocolate takes time to develop. Wine tasting is
can also serve as an example. People can learn to recognize flavors and build up
experience. In the process it is likely that preferences and liking are going to shift.
Elements that experts consider as positive, may be negative drivers of liking for ‘normal
consumers’, and the other way around: amateurs may like flavors that experts consider
to be defects
Cardello, A.V. (1995). Food quality: relativity, context and consumer expectations. Food Quality and
Preference, 6, 163-170
McGee, Harold. On Food and Cooking. The Science and Lore of the Kitchen. Scribner, New York (2004)
Hughson, A.L. & Boakes, R.A. (2002). The knowing nose: the role of knowledge in wine expertise. Food
Quality and Preference, Vol. 13: 463472
Flavor Classification; the why and how The use of the model, page 25
External factors like a brand, a package, a color or a shape can influence flavor
registration. A strong well-known brand is likely to have a positive impact on the flavor
perception. Just as the name and fame of a chef influences the flavor perception of a
dish. In a negative way, foods or drinks that supposedly made you sick one day are not
liked and even avoided.
So there is every reason to look at deliciousness and liking separately.
Up to now the fundamentals of flavor composition have not been formulated.
Without a solid backbone, cooking and food-product development can easily get the
character of ‘cook and look’. The failure rate of many industrial food products is high.
Possibly the absence of knowing the fundamentals of deliciousness may well be part of
the explanation. In art, music and architecture laws of composition have been developed.
Deliciousness in taste is comparable to ‘beauty’ in art. The ancient Greek had the
same word, techne, for both ‘art’ and ‘technique’. Art was defined as ‘the right way of
making things’. In all art, technique is essential. There may be technique without art, but
there is no art without technique. Formulating CSFs can be seen as a first step in getting
a better understanding of flavor and the components that drive liking.
The CSF’s were developed by analyzing successful dishes of eighteen of the best
chefs in The Netherlands were analyzed. Six product characteristics, Culinary Success
Factors (CSFs), were found and these were tested several times at different locations.
This was done by having chefs develop three series of similar dishes. Every series
consisted of one dish based on the CSFs and two variants, in which systematically one of
the CSFs was left out, under the condition that the dish was still restaurant worthy. In a
tasting these nine dishes were served to focus groups. In the tasting, the ‘perfect’ dishes
were preferred over the variants.
The results show that the deliciousness of dishes is not a coincidence. It is the
predictable outcome when the CSFs of food are present. The formulated CSFs will help
chefs in the development of new dishes and improvement of existing ones. An interesting
prospect for future research is to verify if these factors apply to industrially produced
foods and in other cultures. The Culinary Success Factors are discussed below.
Package, color, presentation, and image influence expectation. The description on
the menu should be precise and in line with what can be expected; no false expectations
should be raised. Changing the name and presentation during the years can hardly be
done except for minor adjustments and evident improvements. In many cases the main
ingredient is clearly visible and in general the presentation is pleasing and appetizing.
Delgado, C., Guinard, J.-X. (2011). How do consumer hedonic ratings for extra virgin olive oil relate to
quality ratings by experts and descriptive analysis ratings? Food Quality and Preference, Vol. 22: 213-225
Klosse, Peter, et al. (2004). The formulation and evaluation of culinary success factors (CSFs) that
determine the palatability of food. Food Service Technology, 4, pp. 107-115
Flavor Classification; the why and how The use of the model, page 26
Evidently, the flavor of the dish should be stable. If the recipe changes drastically the
name should change as well.
Smell is an important driver of deliciousness. It is even plausible that smell is
potentially the most important contributor to pleasure. Foods are extra attractive with a
pleasant smell. Deliciousness increases if smell and flavor fit together. On the other
hand, bad smells can destroy the pleasure of eating. The importance of smell in flavor is
also seen in the decline of olfactory capabilities of older people, which tends to begin
around 60 years of age and becomes more severe in persons above 70 years of age. The
enjoyment of eating is reported to go down, potentially leading to inadequate dietary
intake. A study involving elderly showed that when the smell of food was enhanced,
natural intake increased
In palatable foods the flavor components are well balanced, never too sweet, salty,
acidic, or bitter. The right dose of a certain component is food-dependent. A salad is
expected to be fresh and acidic, and a cake is expected to be sweet. Contrary to what is
generally assumed, acidity and a certain bitterness were often mentioned as being
important gustatory components. Many chefs consider bitter to be an essential element
making food easier to consume and more interesting. Food products that are universally
popular: beer, chocolate, coffee, tea and wine are all bitter to a certain extent.
Finding the right balance is important. The word ‘too’ is often used in regard to
balance. Too bitter or acidic is not appreciated, neither is too salt or too sweet. Such
preferences are personal which leads to the suggestion that chefs may better stay on the
safe side, certainly with salt and sugar. It can easily be added up to the personal level of
pleasure. Too ripe’ or ‘brown’ was also mentioned as a potential disturber. To illustrate:
meat that has been fried or grilled too long, bread that has been toasted too long, French
fries that have been fried too long or a wine that has been aged in a wooden cask too
long. In each of these cases ripe flavor notes prevail, while taking away the original
flavors; the complexity of the flavor diminishes.
The results of this study confirm the importance of umami in relation to palatability.
The name and the scientists that discovered umami are Japanese, but as a flavor
component umami has no specific relation with Oriental cuisine. In fact it is a new word
for an old flavor component that has always been valued as important. Umami has not a
distinct flavor of itself. It is described as ‘brothy’, ‘savory’ or ‘meaty’. Depth and fullness
are also mentioned. Moreover, umami gives a strong aftertaste. An agreeable aftertaste
is an important determinant of the overall pleasantness of a meal. Umami also influences
the perception of other ‘basicflavors. Saltiness is perceived more intensely, which
implies that the actual amount of salt used can be lower in combination with a umami
Schiffman, S.S. (2000). Intensification of sensory properties of foods for the elderly. J. of Nutrition, 130,
Flavor Classification; the why and how The use of the model, page 27
compound like MSG (E621). MSG reduces the perception of sourness and bitterness, and
has no apparent influence on the perception of sweetness. Preference for umami is likely
to be innate - just as sugar and fatty acids - as breast milk is rich in glutamic acid.
It was striking to see how often natural ingredients were used that are rich in
glutamate acid or fermented products that acquire glutamate in the process. The chefs
used these products intuitively. At the time of the research many of them didn’t know
what Umami was. In the table below the ingredients of success dishes related to umami
are listed.
Examples of ingredients contributing to Umami
Natural ingredients
Mushroom (all kinds)
Green pea
Green asparagus
Alaska King Crab
Chicken meat
Soy sauce
Oyster sauce
Aged cheese (Parmesan!)
Emmental cheese
Cured Ham
‘Glace de viande’
Table 3: Examples of ‘Umami ingredients’
Chefs look for a contrast in mouthfeel, with a combination of hard and soft
textures. The combination of ‘crispy’ or ‘crunchy’ on one side and ‘juicy’, ‘creamy’, or
‘moist’ on the other seems crucial in palatability. Crispy is associated with freshness. For
example: within a relatively short time the crispness is gone in the crust of freshly baked
bread, roasted steak, toast, French fries or even a salad. As soon as it is gone, these
products have lost much of their appeal
As the soft textures are concerned, fat is an important factor. Foods owe much of
their flavor to fat. Fats serve a variety of functions. In some cases fat contributes a
desirable mouthfeel or texture, in others fat enhances the flavor. Both influences
contribute to deliciousness and liking. For a food product or a dish to be successful
Duizer, L. (2001). A review of acoustic research for studying the sensory perception of crisp, crunchy and
crackly textures. Trends in Food Science & Technology, 12, 17-24
Flavor Classification; the why and how The use of the model, page 28
though, the softness in whatever textural form apparently needs to be compensated by
textures that have ‘bite’, according to the professional chefs interviewed.
The last factor that was identified was a high flavor richness of the dishes. Within
flavor richness ‘flavor intensity’ and ‘ripe’ were the factors that characterized the flavor of
palatable dishes. It is conceivable that this result is biased by our study design. Delicious,
but non-complex dishes and flavors are not to be expected on the menu of restaurants.
The guests of these restaurants are often in a certain frame of mind: relaxed and
enjoying themselves: ideal circumstances to savor complexity. Situational aspects have
an influence on preference: flavor profiles that are fresh and slightly acidic fit warm
weather. The perception of complexity is also likely to be situational, as it requires to be
relaxed. The atmosphere in good restaurants is therefore an important ingredient in the
overall enjoyment of the food.
It is important to note the CSFs do not dictate how a food product should be made,
nor do they impair creativity. Chefs that have been trained and instructed to compose
dishes based on the CSFs, have no problem to do so. CSFs make flavor tangible, and are
as such useful in modifying or rather improving existing products of formulating new
ones. They could play a role in changing food behavior and promote a diet that is healthy
or sustainable.
In order to find good matches we need to have a notion of the flavor profile. Foods
and beverages don’t just accidentally have a certain mouthfeel and flavor richness. Their
flavor profile is the result of basic qualities, and of what has been done and/or added.
What will the mouthfeel be: neutral, dry, contracting or coating? And what about the
flavor intensity? Modern guidelines on combining foods and beverages follow quite
logically from the flavor profiles of them. Overall, successful combinations can be
expected when the profiles food and beverage match or at least have strong resemblance
on the most important elements. They are formulated as follows:
Mouthfeel needs to match, i.e. contracting fits contracting, coating fits coating
Flavor intensity needs to match
Aromas need to match: fresh fits fresh, ripe fits ripe
As far as the order is concerned, the following may be added:
Contracting before coating
Low intensity before high intensity
Fresh before ripe
Flavor Classification; the why and how The use of the model, page 29
This is the essence of matching foods and beverages. The basic premise is logical
and easy to understand. Application will take some practice. Fortunately, eating and
drinking are daily activities, and therefore there is plenty of opportunity to test the
suggestions. Experience shows that it is easier than you first may think. Learning by
doing is not too bad in this respect and as always, practice makes perfect.
The Culinary Success Factors Food and the guidelines for food and beverage pairing
are based on flavor and show that there is a right way of doing things. But this is not a
guarantee for liking. At most you can say that the chances of liking increase if the things
are done right. Even the best dish does not guarantee liking. In the more advanced
sensory fields, vision and sound, the term ‘fluency’ has been introduced to assess why
our brain likes things. Our brain constantly gets an enormous amount of sensory
information to process. Therefore it is organized to discard what it cannot quickly
understand or organize. If fluency is high, signals can be quickly processed by the brain.
It has been shown that this is an essential part of liking something. Coherent,
harmonious signals stimulate the brain and can give us pleasure. From a neurological
point of view, this happens when It leads to a certain degree of arousal and desire. In
that case, a certain part of our brain, the orbitofrontal cortex, will release the pleasure
hormone dopamine. This also explains the problem with harmony: if we know exactly
what is coming next, we don’t get excited. Perfect harmony will ultimately be boring,
because it is predictable
We taste with our eyes, ears, tongue, nose and touch. All these sensory pathways
deliver parts of the information and they interact. The word to use is cross-modal.
Human perception is influenced and liking is hard to control. Yet, everything start with
awareness. We mention some striking examples, ambient lighting to start with. In a
tasting of Riesling wines in a winery in Germany the color of the room was changed from
white to blue, red, or green, while 200 wine buyers rated a certain wine. The wine itself
was served in a black glass to prevent the influences of the color of the room upon the
perceived color of the wine. Under blue and red lighting the quality of the wine was rated
significantly higher than under green or white lighting. People were even willing to pay
50% more for the wine tasted under red lighting compared to the same wine under green
or white lighting. Further research revealed that blue and green room lighting made
wines taste spicier and fruitier, while the red lighting made the wine taste nearly 50%
. Good-looking food will induce eating it, just as seeing people enjoy a certain
food will do so. Just plain seeing and having foods on hand leads to an increased
Ishizu, T., Zeki, S. (2011). Toward A Brain-Based Theory of Beauty. PLoS ONE 6(7): e21852.
Oberfeld, D., Hecht, H., Allendorf, U. & Wickelmaier, F. (2009). Ambient lighting modifies the flavor of
wine. Journal of Sensory Studies, 24, 797832
Painter, J.E., Wansink, B. & Hieggelke, J.B. (2002). How visibility and convenience influence candy
consumption. Appetite, 38, 237-238
Flavor Classification; the why and how The use of the model, page 30
Other experiments show the influence of environmental sounds. Music in
restaurants, for instance, may well prove to influence the flavor experience. People
enjoyed eating oysters much more when hearing breaking waves than farmyard sounds.
Conversely, a dish of bacon and eggs was tasted more ‘bacony’ when listening to the
sound of sizzling bacon than to a farmyard of clucking chickens. At Heston Blumenthal’s
restaurant The fat Duck in the UK you can enjoy ‘the sound of the sea’, a dish with
sound. It is a good example of how this knowledge can be set to use
How about influences of season, climate, weather, can they have an impact as well?
The results of an experiment in The Netherlands held in October 2000 suggest that
indeed they can (Klosse, unpublished data). Five independent groups of subjects (non-
professional volunteers, total N=238), across the country, were first asked to rate the
flavor of a mousse of smoked salmon and a tartar of fresh salmon (summary of the
results: table 4).
Sensory judgment
(N= 238)
Mousse smoked
Tartar fresh
Flavor Profile:
Mouthfeel Contracting
Mouthfeel Coating
Flavor richness
Fresh flavor tones
Ripe flavor tones
Deliciousness (scale 1-10)
My choice for a hot day on
a terrace (N)
26 (11%)
212 (89%)
Table 4: Sensory judgment of two salmon dishes
: not at all present;
hardly present; 0 indifferent; + present; ++ very present)
The mousse was clearly more coating in mouthfeel, rich in flavor with ripe flavor
tones. The tartar was less coating, and the flavor richness was somewhat lower and more
fresh. In terms of flavor styles, the mousse could be classified as ‘full’ (segment 6 of the
Flavor Styles Cube), while the tartar can be characterized as ‘fresh’ (segment 4 of the
Flavor Styles Cube). Both dishes scored relatively high on palatability (7,8, resp. 7,3 on a
10-point scale). When asked the question ‘which of the two would you prefer on a hot
day on a terrace’: 89% of the subjects chose the tartar. Consider that this was asked,
after the two versions of salmon were tasted and in October. Apparently, people have a
strong feeling about when and in what situations certain dishes are appreciated. Consider
as well that the mousse as such was rated higher in palatability.
If this experiment would have been about warm apple pie and ice cream this result
would hardly be surprising. The ‘fresh’ ice cream would be a logical choice for a hot day
Spence, C. & Shankar, M.U. (2010). The influence of auditory cues on the perception of, and responses to,
food and drink. Journal of Sensory Studies, 25, 406430
Flavor Classification; the why and how The use of the model, page 31
on a terrace. A tartar of salmon can hardly be considered to be such a ‘logical’ and well-
known choice. Nevertheless, people are very clear in their preference, which leads to the
conclusion that objective product properties are at the origin of this preference.
Apparently, the tartar of salmon and ice cream have things in common, making them
instrumentally preferable for a hot day on a terrace. The common denominators of ice
cream and the tartar are their contracting mouthfeel and freshness. Our conclusion of
this experiment is that people know instinctively that the ‘fresh’ flavor style, would suit
them best on a hot day. Apparently, liking is not just a function of product
characteristics. Flavors need to fit the situation as well. Flavor styles have a function in
this respect as they communicate about the flavor to be expected.
Flavor can be studied from a human and a product angle. Deliciousness is defined
on the product level and therefore they are closely related to flavor. The Flavor Styles are
functional in communicating the flavor. Liking is all about a favorable judgment on what
is eaten or drunk. It has to do with both nourishment, replenishing deficiencies (we like
what we need) and with plain pleasure which is the result of the release of dopamine.
Differences between humans and cross modal sensorial interferences influence
registration and tasting. If the physiology underlying the transduction of single sensory
modalities is already complex, the physiology underlying a multi-sensorial experience
such as tasting, is much more complex. The ‘four basic taste hypothesis’ has dominated
the field and research methods. This implies that attention has been focused on the
gustatory part of flavor registration. Although the human angle is intriguing, we need the
product angle for classification.
We need objective terms to describe phenomena. It is a prerequisite for
classification. Flavor, defined as a product characteristic, opens the door to do this.
Coating, contracting and dry are the three vectors of mouthfeel that provide the basis for
a model. Based on these parameters a model can be built, the Flavor Styles Cube.
Everything that we eat or drink finds its place somewhere in this model.
The challenge that faces us now is to find an entity for mouthfeel that ultimately
defines where a certain food or drink is located in Flavor Styles Cube. The better we are
able to describe the coordinates of flavor, the more it becomes something real,
measurable and predictable and a useful tool to match product characteristics to human
preferences and liking. Therefore we need to be able to measure to the forces of
mouthfeel and their intensity and especially how these forces interact. In other words
we want to find and define the algorithms of flavor. If your first reaction is that this is a
crazy idea, realize that every color has a number. Lets see how far we can get. Flavor
modelling would help all kinds of educators and professionals in the food and beverage
world to understand the specifics, from growers and farmers to brewers and wine makers
and from professionals in the food industry to chefs in the culinary domain; marketeers,
food designers, and packaging experts would all benefit. Just as health councils and
dieticians that need to find a way to motivate people to make different food choices.
Nobody will ever object eat something delicious that is liked. Even if it is a dish that
contains little meat, salt, or sugar and is predominately composed of vegetables.
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.