Spence et al.

Flavour (2017) 6:4

DOI 10.1186/s13411-017-0053-0

REVIEW Open Access

Pairing flavours and the temporal order of

tasting
Charles Spence1* , Qian Janice Wang1 and Jozef Youssef2

Abstract
There can be little doubt that plating food beautifully is becoming ever more important in the world of high-end
cuisine. However, there is a very real danger that all the attention to how a dish looks (or photographs) can end up
obscuring the importance of temporal order to flavour perception. That is, there is an increasing tendency to prioritize
plating elements in a dish so that they please the eyes of the beholder, rather than necessarily because they deliver
the optimum balance of sensations to the palate. Relevant here is the fact that certain combinations of ingredients
taste better when sampled simultaneously, or in a particular sequence, than when sampled in a different order. In this
review, we examine the importance of sequencing and pairing taste/flavour sensations, both in the design of the meal
itself, and when attempting to combine (or match) food and drink. We address the chemical, psychological, and
computational strategies that have been suggested by those wishing to combine flavours for maximal impact. We
evaluate three general principles of flavour matching: similarity—matching components based on common flavour
compounds (or similar flavour profiles); contrast—combinations that are purposely chosen because they differ from
each other (a strategy that is more common in the cuisine of some countries than others); and synergy (or emergence)
—those combinations that together deliver new flavour experiences or else harmonize with one another. We argue
that the psychological account (informed by an awareness of cultural differences), and to a lesser extent the chemical
account, provides meaningful suggestions as far as effectively combining flavours is concerned.
Keywords: Taste, Flavour, Pairing, Similarity, Contrast, Harmonization, Emergence, Computational gastronomy

Review various ingredients simply been mixed together, the

Introduction diner’s experience would likely have been quite different
The two Michelin-starred chef, Denis Martin, used to (see [2–6], on the temporal ordering of taste and odour
serve an intriguing dish in his namesake restaurant in mixtures).
Vevey, Switzerland (see Fig. 1). It came to the table on a Much the same idea (that there is pleasure in the dy-
spoon, and in a single mouthful, the attentive diner namically changing sequence of tasting sensations) is
would experience a sequence of distinctive flavours/ hinted at in the following description of Heston Blu-
mouthfeels, each revealing themselves one after the menthal’s surprising Caviar and white chocolate disc
other. First came the sharp “hit” of wasabi, then the ten- combination: “To serve, spoon some caviar (about half a
der texture of the tuna itself, and, finally, the slow coffee spoon) on top of the disc, and eat. You can ex-
creamy melting mouthfeel of the white chocolate. (The periment with the quantity of caviar, depending on your
dish itself contains raw tuna, white chocolate, raw Thai taste. The sensation of these sweets is heightened if you
chilli, toasted peanut oil, wasabi, and a pinch of fleur de place the chocolate and caviar disc on the tongue, close
sel; see [1], pp. 82–83). It can be argued that part of the your mouth and leave to melt. As the chocolate melts,
pleasure of this dish was the very temporal evolution of the caviar flavour comes through gradually. You will be
the distinctive mouthfeel/flavour experiences. Had the amazed by the pleasure of the changing flavours and
sensations.” (quote from [7]). Once again, it is the chan-
ging sequence of flavours and mouth sensations that is
* Correspondence: charles.spence@psy.ox.ac.uk
1
Department of Experimental Psychology, University of Oxford, Oxford OX1 deemed to be so pleasurable.
3UD, UK
Full list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Spence et al. Flavour (2017) 6:4 Page 2 of 15

of molecular gastronomy would seem to imply that

exact control over the tasting experience ought to be
important (see [10, 11]). That said, while some chefs
work hard to control the temporal order and dynamic
changes in the tasting experiences that they deliver,
others would seem to go to the other extreme and
deliberately give up as far as controlling the sequence
of tasting is concerned.
Chef Andoni of Mugaritz in San Sebastian is interest-
ing on this theme. Just take the following description of
one of the dishes served at the restaurant: “The dish
‘Roasted and raw vegetables, wild and cultivated shoots
and leaves’, … consists of hundreds of vegetables, leaves
Fig. 1 The Thon au chocolat blanc et piment Thaï dish served at and herbs – something nobody would think of trying at
Denis Martin’s namesake restaurant in Vevey, Switzerland (http://
www.denismartin.ch/; [1]). A dish to be consumed in a single
home. One hundred ingredients make it impossible for
mouthful, thereby delivering a distinctive series of flavour the person preparing the dish to plate any two in the
experiences, extended over time (but, importantly, under the control same way. It also makes it almost impossible for two
of the chef). [Picture from http://www.cavesa.ch/blog/wp-content/ diners to eat it in exactly the same way. This is one of
uploads/archives/b/bl/blog.cavesa.ch/aoctobre/thonlivre10_small.JPG. the evocative powers of this recipe.” ([12], p. 42). In this
Author: Denis Martin]
case, it would appear that the chef has designed the dish
in light of the awareness that they cannot control how
the diner will eat it.
In terms of sequencing tasting experiences during In the three fine-dining examples mentioned so far,
more mundane food encounters, there may well be a we are undoubtedly talking about the very high end of
relevant link here to Hyde and Witherly’s [8] notion of modernist cuisine. It is important to note that this situ-
“dynamic contrast,” defined as “moment-to-moment ation might be in some sense special, given the diner’s
sensory contrast from the everchanging properties of likely mindset in such a venue to want to search out
foods manipulated in the mouth.” Focusing on everyday different flavour experiences with each consecutive
foods such as ice cream, these researchers argue that dy- mouthful (see [9]). More generally, one might ask how
namic contrast during consumption, typically somato- important it is to get the timing and sequence of fla-
sensory contrast, such as the temperature change when vour sensations within a dish “right” during more regu-
consuming a spoonful of ice cream or the textural lar everyday consumption episodes. Is it enough simply
changes when eating a piece of chocolate or masticating to plate “for the eye” as the Roman gourmand Apicius
a crunchy food such as a potato chip is especially rewar- [13] would have us believe (see also [14])? Thinking
ding—that is, we find it particularly palatable. They about this, and properly designing the sequence of fla-
hypothesize that it is just such dynamic contrast that is vour experiences, is a topic that is becoming increas-
such a distinctive feature of ice cream (together with its ingly important. Many chefs are becoming ever more
cooling/refreshing nature) that explains its place as the obsessed with the eye appeal of the dishes they prepare
final course on the menu, at a time when many diners (see [15, 16], for reviews). At the same time, diners
might well say that they are already full. This they call often want to choose the sequence in which they taste
the “ice-cream effect.” There is probably also a link to the various elements in a dish (see Fig. 2 for one such
the recently introduced notion of “hedonic escalation” colourful example). Of course, one might argue that it
[9]. This term describes the increased liking of each add- is often actually the chefs who are happy to leave the
itional bite of a palatable food that contains multiple temporal ordering of tasting, and taste sensations, in
identifiable flavours that are revealed successively (we re- the hands of their diners.
turn to a fuller discussion of the notion of hedonic escal- In any case, perhaps it can be argued that for much of
ation later). the time, the order in which flavours, and the various
It should come as little surprise to find that many components of a dish, are sampled, and hence experi-
top chefs are increasingly starting to consider how enced, does not significantly impact the final flavour ex-
exactly the diners interact with their dishes. After all, perience. The corollary of such a view, if correct, being
even if the food on every plate is the same, the order that the chefs can continue plating for the eye without
in which the diner samples the various elements may worrying too much about the sequence in which the fla-
fundamentally change the ensuing experience. Cer- vours they create on the plate happen to be experienced.
tainly, all the talk of precision cooking in the world After all, with a mouthful of food and drink, how does
Spence et al. Flavour (2017) 6:4 Page 3 of 15

the mind of the diner (see [17, 18], for contrasting per-
spectives on this theme). As we will see later, multisen-
sory flavour perception turns out to be much more
complex, and culturally determined, than some com-
mentators would seem to realize.

Some classic combinations to avoid

Red wine and seafood
Traditionally, certain food-wine combinations were
thought to be best avoided, such as red wine with
seafood (e.g. [19]). For, while both may taste wonder-
ful on their own, when combined, diners would often
Fig. 2 The Sole, Olive Oil, and Mediterranean Flavors dish served be left with a rather unpleasant sensation. Reports of
at the El Cellar de Can Roca restaurant in Girona, Spain (http:// a ferrous taste, an unpleasant fishy and/or metallic
cellercanroca.com/index.htm). Simultaneously plating for the eye
odour, and possibly also bitterness in the mouth are
and, at the same time, allowing/encouraging the diner to explore
the various flavour combinations in the dish. (Notice also the not uncommon [20]. Intriguingly, the latest research
disambiguation of the meaning of the colours through the use of has identified that it is the ferrous ion in some red
flowers, etc. in each of the sauces; see [94]) [Image from http:// wines that gives rise to the unpleasant fishy aftertaste
gastronomyblog.com/2010/08/14/el-celler-de-can-roca-girona/; Photo [21].
credit: HYPERLINK "http://gastronomyblog.com" GastronomyBlog.com]
In an elegant series of studies, Tamura et al. demon-
strated that it was the concentration of total iron and of
ferrous ion in red wine (not the phenolic tannins that
one really know what to bind together and what to keep some had suspected to be responsible) that correlated
separate? Once we break food down and mix it up in the with reports of a fishy aftertaste when participants were
mouth, when all texture cues have gone, what is left in given samples of dried scallop to evaluate. Chelating the
terms of distinctive elements in the tasting experience? ferrous ion in the wine reduced reports of the unpleas-
In this review, we argue that getting the order of ant fishy aftertaste when paired with the scallops. By
mouth sensations right is important, at least for certain contrast, adding ferrous sulphate to a model wine in-
combinations of ingredients/flavours. We will start by il- creased such reports. The researchers suggested that it
lustrating what not to do in terms of sequencing tasting was the potent volatile compounds such as hexanal and
experiences with a few classic everyday examples. Hav- heptanal that are formed (by the combination of dried
ing analysed these “worst-case” scenarios, we will then scallop and red wine) and that were responsible for the
be in a much better position to address what to do when unpleasant sensations in this case. (Interestingly, though,
it comes to trying to optimize the diner’s multisensory this potential clash is less likely to occur nowadays given
tasting experience. We will review chemical, psycho- the change from iron to steel in wine-making. It was the
logical, and the (newly emerging) computational gas- former that would result in much of the ferrous matter
tronomy approaches to the problem of effective flavour making its way into the wine; see [22].)
combination. We will then look at the various strategies On the flip side, in terms of good food and wine
that have been put forward over the years in order to ac- matches, one oft-made recommendation is to pair red
centuate, contrast, or harmonize flavours ([17], p. 26). wine with meat while pairing white wine with fish. This
All of these strategies have been suggested at some point suggestion can be explained, at least in part, by interac-
in print, regardless of whether one is talking about the tions taking place between components that are found
various elements that might be combined together in a in the wine and in the food. So, for instance, the food
dish, or when trying to optimize food and drink (e.g. proteins in meat are thought to reduce the bitterness
wine or beer) pairings. and astringency of tannic red wine. They may also help
One of the tensions that will be highlighted in this lit- to alleviate the sourness and astringency of a dry white
erature review relates to the widespread assumption that wine (see [23]). Other combinations that have been rec-
those foods that share flavour compounds will pair well ommended, so far on the basis of personal opinion ra-
together. For contrary to what many who are interested ther than any obvious evidential footing, include sherry
in computational gastronomy would have us believe, the and dry white wine with kippers and mackerel [20, 24].
fact that a selection of ingredients share some number Note here that highly acidic wines have iron chelating
of flavour compounds is certainly no guarantee that they properties, while both fino and manzanilla sherries have
will necessarily combine well, perceptually speaking, in a relatively low ferrous ion concentration [21].
Spence et al. Flavour (2017) 6:4 Page 4 of 15

Orange juice after toothpaste The slow-developing, yet relatively long-lasting oral
Many of us have had the unpleasant experience of drink- tingle that one gets on contact with the Szechuan
ing orange juice that tastes exceedingly bitter, not sweet, peppercorn or flower (in this case triggered by the pres-
after having just brushed our teeth (e.g. see https:// ence of the hydroxy-alpha-sanshool molecule) presum-
www.youtube.com/watch?v=9X5_gtel-c0; [25, 26]). In ably raises similar issues around the ordering of taste
this case, it is the sodium lauryl sulphate (SLS), a deter- sensations within a meal (e.g. see [35]). Szechuan pep-
gent added to many toothpastes, that blocks the sweet percorns are a commonly used ingredient in regional
taste receptors in the mouth. SLS apparently also de- Chinese cuisine. Finally here, another relatively common
stroys the phospholipids that inhibit bitter taste percep- example of a sequential effect on tasting occurs when
tion. Of course, eating an orange before washing your water tastes sweet after eating globe artichoke (see [36–
teeth does not have anything like the same effect—that 38]). In this case, it is a substance known as cynarin that
is, this is a case where the temporal ordering of sensa- latches onto the sweet receptors without activating
tions really does matter (about which, more later). them. Nothing happens until a person drinks water, at
Notice how in the first of the above-mentioned exam- which point the cynarin molecules are washed away,
ples, it is the chemical reaction between components of thus releasing the receptors and triggering a sweet per-
the two substances that is best avoided. In the orange cept in roughly 60% of those tested by Blakeslees [37]. It
juice case, by contrast, it is the change in sensitivity of a has also been reported that artichoke can modify the
subset of gustatory receptors in the oral cavity that then taste of wine, making it appear sweeter to some people
changes how a subsequently tasted food is perceived. (e.g. [39–41]).
These can be thought of as chemical and physiological Finally, in this section, one other more mundane se-
accounts, respectively, for why certain combinations of quence to avoid can be illustrated by thinking about
oral sensations just do not work together. And while the something like eating a slice of chocolate cake, say,
toothpaste example might not seem especially relevant followed by the taste of a drink of hot chocolate. The
to the restaurant setting, bear in mind here only that latter is likely to lack flavour, probably tasting more like
there are a number of other ingredients that also sup- warm milk instead. Tasting the cake and sauce the other
press various aspects of taste perception. In fact, any- way around is, we would argue, less of an issue. One
thing that irritates/stimulates the trigeminal nerve, might be able to explain this asymmetry in terms of one
such as chilli and menthol, will probably have just such ordering of these similarly flavoured stimuli giving rise
an effect, though the literature suggests that this kind to more dynamic contrast than the other way round (see
of cross-sensory interference may be restricted to the [8]).
perception of sweetness (e.g. see [27, 28]). (No one,
after all, would think of serving a mint before dinner, Olfactory adaptation
would they?) Sometimes, a combination of flavours may work espe-
So, for example, miracle fruit (miraculin, derived from cially well together (or not), due to the effects of adapta-
berries of the West African plant synsepalum dulcificum; tion. Olfactory adaptation can be defined here as a
[29]) is known to suppress the sour receptors and hence temporary, normal inability to distinguish a particular
make whatever is tasted for an hour or so thereafter odour after prolonged exposure to that airborne
seem much sweeter than normal (e.g. a slice of lemon compound (e.g. Cometto-Muñiz & Cain, [42]; see also
will taste sickly sweet; see [25, 30–32]). Miraculin is [43, 44]). The notion that by adapting to a flavour com-
thought to attach itself to the taste receptors on the pound in one item (be it the food or drink), another fla-
tongue [33]. When a person subsequently bites into a vour may be experienced differently was elegantly
slice of lemon, lime, or perhaps a packet of salt and vin- captured in a dish trialled by the chef Heston Blu-
egar crisps, the pH in their mouth drops, and somehow menthal a few years ago ([45], pp. 232–237; see Fig. 3).
(though researchers are still uncertain quite how), the A two-flavoured cinnamon/vanilla ice cream was pre-
sweet receptors are activated. Relevant here, miracle sented together with two squeeze bottles, one containing
berry pills have featured on the menu at some high-end sticks of cinnamon and the other a vanilla pod. The idea
modernist restaurants (see [34]), though always as pretty here was that simply by sniffing one of the bottles for a
much the last thing that the diner is given to taste. few seconds, the nose/brain would start to adapt to the
Meanwhile, gymnemic acid, which can be found in the aroma contained within. Hence, when the ice cream is
leaves of the tropical plant (Gymnema sylvestre), de- actually sampled, it should taste more strongly of the
presses the perceived sweetness of foods. So, for ex- other flavour. If one repeats the procedure, this time
ample, Eisner and Halpern [31] have reported that an sniffing the other bottle first, then the apparent flavour
orange eaten after chewing on Gymnema leaves will of the ice cream may well switch in the other direction.
tastes like lemon or lime. While the dish itself has more of a novelty-value feel to
Spence et al. Flavour (2017) 6:4 Page 5 of 15

and having considered the role of short-term adaptation

in multisensory flavour perception, let us now take a
look at some of the strategies for effective flavour pairing
that have been suggested over the years.

Flavour combination strategies

Eschevins [53] interviewed 20 sommeliers and beer ex-
perts in order to assess their recommendations concern-
ing the pairing of two wines and two beers. An analysis
of their responses revealed three main principles of fla-
vour pairing—perceptual (harmonizing the dish as a
whole or enhancing specific sensory properties), concep-
tual (pairing by tradition or by region), and individual
(based on the diners’ own personal preferences and ex-
periences). Interestingly, wine and food matches tended
Fig. 3 Two-flavoured (cinnamon and vanilla) ice cream by Heston
Blumenthal. Sniffing and hence adapting to the smell of either spice
to obey the conceptual principle whereas beer matches
apparently changes the perceived flavour of the cinnamon-vanilla ice tended to focus more on personal liking.
cream for about 85% of people [Image from http://sageappliancesclub. There is, in fact, a long-established art to pairing wine
co.uk/article/cinnamon-and-vanilla-ice-cream; Photo credit: Angela with food, with scientist Francoise Chartier, one of the
Moore, Bloomsbury's Heston at Home] best-known researchers working in this space. Chartier
([17], p. 211) describes what he does as molecular som-
it than anything else—as Heston Blumenthal himself mellerie, defining it as: “the practice of developing food
puts it: “The cinnamon/vanilla milkshake still has the pairings and food and wine pairings based on dominant
look of an eating experiment rather than an eating ex- aromatic compounds” (see also [18, 54–56]).2 Over the
perience. Until I can change that, it stays off the menu.” years, people have adopted a number of different pairing
([45], p. 237)—it nevertheless beautifully illustrates the strategies: One suggestion here being to pick those com-
principle that might help explain the success of at least binations of ingredients/flavours that are either similar
some food-wine pairings.1 That said, a systematic ana- to, or contrast with, one another. Another strategy in-
lysis of the approach and of the role of adaptation defin- volves picking those combinations of flavours that
itely awaits further study. One of the challenges with harmonize, possibly giving rise to new flavour sensations
conducting research in this area being the role of con- in the process (see [57]).
textual and sequential effects on people’s tasting judg- When talking about similarity, though, the question
ments (e.g. see [46, 47]). soon becomes one of whether it is chemical similarity
It turns out that combining flavours, as when pairing (i.e. shared flavour compounds) that matters or else per-
food and wine, may affect one component of the tasting ceived similarity (psychologically speaking; see [58, 59])
experience more than the other. A recent study illustrat- that we are talking about. Some commentators would
ing this point involved the analysis of multi-bite wine appear to believe that one is, in fact, reducible to the
and cheese combinations using the increasingly popular other. However, the critical point to stress here is that
temporal dominance of sensations (TDS) technique. The food and/or drink combinations that share flavour com-
participants tasted Pacherenc, Sancerre, Bourgogne, and pounds need not necessarily be rated as perceptually
Madiran wines after nibbling on epoisses, comte, Roque- similar. Furthermore, it is worth bearing in mind that
fort, and crottin de chavignol cheeses. Amongst other co-exposure—that is, the regular pairing of aroma with a
things, the results of this research revealed that cheese taste in the food we eat—can lead to an increase in per-
was more apt to alter the flavour of wine (increasing ceived similarity without having any impact on chemical
fruitiness ratings for the Bourgogne and Madiran wines), similarity (e.g. [58, 60–62]). Aromas may take on the
than vice versa [48, 49]. In this case, wine and cheese taste qualities of the foods that they have been associ-
pairings might be seen as a way to enhance or decrease ated with previously. For instance, most Westerners will
specific sensory properties in a wine, rather than the rate the aroma of vanilla as sweet, not because the spice
wine necessarily being used to affect the cheese-tasting is itself sweet (it is not, anyone who has bitten into a
experience. Importantly, eating the cheese helped reduce vanilla pod will know just how bitter it is) but rather be-
the perceived astringency of the subsequently tasted cause in many Western cuisines, it is paired with sweet-
wine (see also [46, 50–52]). tasting foods (such as cola and ice cream; [58]).
So, having taken a look at a number of the classic Blank and Mattes [58] had 70 North American adults
combinations to avoid when pairing tasting experiences, rate the perceived similarity of 10 spices (salt and sugar
Spence et al. Flavour (2017) 6:4 Page 6 of 15

granules, vanilla and anise extract, dried bay leaf, dried Italy, potatoes in Germany, or chilli peppers in Thailand.
crushed spearmint leaves, ground nutmeg, ground cloves, Nowadays, there are simply so many “new” ingredients
ground cinnamon, and ground ginger), which they available that were not part of a country’s traditional do-
first smelled and then tasted. The intensity of the mestic culinary repertoire. Consequently, chefs and
odour, its compatibility with sweetness, and the bitter food/drink companies are increasingly looking for in-
taste were the three major characteristics that partici- novative new ways in which to combine the range of in-
pants used to arrive at their similarity judgments. Un- gredients that they now have access to, especially given
surprisingly, vanilla was rated as most compatible the interest of many consumers in fusion/exotic cuisines.
with sweetness. Intriguingly, these researchers also re- Indeed, as Visser ([65], p. 124) presciently noted, when it
ported some individual differences, with the non- comes to cuisine, the contemporary taste for novelty of-
white participants tested attributing a greater degree fers “a wonderful marketing milieu” (see also [34], Sec-
of sweetness to nutmeg and less sweetness to anise tion 1.3).
than their white counterparts (see also [63]).
Here, in passing, it is worth noting that the topic of in- The computational gastronomy approach to pairing
novative food/flavour pairing has become increasingly flavours
important recently due to the growing accessibility to The recently emerging field of computational gastron-
foreign ingredients. For most of our time here on earth, omy (e.g. [66–76]) is premised, at least in part, on the
recipes were developed based on ingredients which were notion that one can make recommendations for novel,
native to the land/culture. Consequently, most national but purportedly successful, flavour pairings based on the
cuisines have always had distinctive flavour profiles: identification of common flavour compounds (see
Think Mexico—lime, coriander, and chilli; France—gar- Fig. 4).3 So, for example, it turns out that a chemical
lic, rosemary, and butter; etc. [64]. That is not to say, compound like acetal can be found in everything from
however, that new ingredients do not become incorpo- apple and orange juice, through whiskey and beetroot
rated quickly into a region’s cuisine—think tomatoes in [77]. Meanwhile, there are flavour compounds in

Fig. 4 The backbones of the flavour network according to Ahnert (2013; [66]). Culinary ingredients are represented by circles and their chemical
relationship by lines. The colour of an ingredient indicates the food category that it belongs to. The size of the circle indicates how frequently the
ingredient is used. Ingredients are connected if they share a number of flavour compounds [Figure reprinted from [67].]
Spence et al. Flavour (2017) 6:4 Page 7 of 15

strawberries that one also finds in white wine.4 This ap- https://www.foodpairing.com/en/science-behind) to the
proach, involving data mining and network analysis, now test in an experiment with untrained participants. Im-
uses the large body of publically available recipes to- portantly, however, no support for the theory was found.
gether with an understanding of flavour chemistry (see That is: “food pairings with more aroma overlap did not
[67, 78]).5 According to Zolfaghifard [79], there are 381 taste better than food pairings with less overlap. For ex-
ingredients in the world that are used for cooking, each ample, chocolate and tomato (43% overlap) did not taste
containing an average of a little over 50 flavour com- better than cauliflower and pear (no overlap).” (quoted
pounds ([80]; with that value ranging from tens to hun- in [18], p. 58). de Klepper [18] continues: “…food pairing
dreds of different flavour compounds per food based on aromatic overlap is not a guaranteed recipe for
ingredient; see Ahn et al., 2011). success. Balancing flavors is what does the trick.” (see
Computational gastronomy builds on the food pairing also [83], for a similarly pessimistic take). So it would
hypothesis that: “If two ingredients share important fla- appear that flavour pairing is more useful to the chef as
vour compounds, then they will go well together.” (quote a tool for generating some novel ideas for ingredient
from [78]). Or take de Klepper’s ([18], p. 55) slightly combination rather than necessarily as a system for pick-
more nuanced definition: “The more aromatic com- ing guaranteed winners [18]. As Blumenthal ([45], pp.
pounds two foods have in common, the better they taste 171–172) puts it: “I soon realised that the molecular
together. This effect is particularly strong when two profile of a single ingredient is so complex that even if it
foods share aromas that make up their characteristic fla- has several compounds in common with another, there
vour.” World-famous chef Heston Blumenthal was ini- are still as many reasons why they won’t work together
tially a vocal proponent of molecular flavour pairing. As as reasons why they will… Molecular profiling is a great
Ahnert ([67], p. 2) notes: “The chef Heston Blumenthal, tool for creativity, but it supports intuition, imagination
together with flavour scientists has suggested that two and emotion rather than replacing them.”
foods that share chemical flavour compounds are more There is a strong claim behind much of the computa-
likely to taste good in combination [45].” tional gastronomy research. Just take the following quote
In the early days of interest in this new approach, com- from Jain et al. ([84], p. 3): “Molecular composition of
binations such as white chocolate and caviar, chocolate food dictates the sensation of flavour [77].” It would ob-
and blue cheese, and pork liver and jasmine were offered viously be nice for the computational gastronomist were
as surprisingly good, yet unusual, combinations that also such a claim to be proved correct. However, that is simply
happened to share flavour molecules ([7, 45], pp. 171– not the case: Just take coriander as an illustrative exam-
172; [18]). White chocolate and caviar, for instance, both ple—the population is genetically divided in terms of
contain amines, while pork liver and jasmine share the whether they perceive it as herby or soapy and
volatile compound indole. Other popular combinations hence pleasant or unpleasant (e.g. see [85]). One could
that share volatile compounds include salmon and make much the same argument about androstenone [86]
liquorice, bananas and parsley, oysters and passion fruit, and a number of other volatile compounds. In other
and garlic, coffee, and chocolate, the latter trio sharing the words, there are a number of cases where molecular struc-
volatile compound 2-methylfuran-3-thiol [18]. ture clearly does not dictate the exact flavour sensation.
However, despite Heston Blumenthal’s initial enthusi- When thinking about the practical implementation of
asm, it is worth pointing out that the chef soon recognized the computational gastronomy approach, one also needs
the shortcomings of the flavour pairing approach. As he to factor in that the volatile compounds that are present
stated in an article that appeared in The Times newspaper in a food may change during cooking. Chartier ([17], p.
in 2010: “Looking back at my younger self I’m almost 203), for instance, mentions the generation of new sapid
embarrassed at my bumptious enthusiasm, not least be- molecules such as beta-ionone (with a fragrance of vio-
cause I now know that a molecule database is neither a lets) when carrots are cooked (see also [87]). And as if
shortcut to successful flavor combining nor a failsafe way that was not enough, one also needs to consider the in-
of doing it Any foodstuff is made up of thousands of dif- fluence of masking (just think about how salt masks bit-
ferent molecules, that two ingredients have a compound terness while at the same time acting as a flavour
in common is a slender justification for compatibility. If enhancer; e.g. [45], pp. 232–237; [27, 88, 89]). And, as if
I’d known then what I know now, I would probably never that was not enough, there are some combinations of
have tried this method of flavor pairing: there are simply odorants that seem to give rise to the perception of a
too many reasons for it not to work As it was, in my naiv- different odour quality (i.e. where the component odours
ety I just got stuck in.” ([81], p. 45). are no longer identifiable; e.g. see [90]).
In a paper entitled “Food Pairing from the perspective Also relevant here is the fact that what a key com-
of the ‘Volatile Compounds in Food’ Database,” Miriam pound that gives one ingredient its distinctive flavour
Kort et al. [82] put the Flavour Pairing Theory (see may well be barely noticeable in another, or else it may
Spence et al. Flavour (2017) 6:4 Page 8 of 15

contribute to a different aroma profile, thus making it perhaps healthy.” (Varshney et al. [65, 94], p. 14). On the
unidentifiable. Another potential problem relates to the other hand, however, those who have tried Watson’s bar-
perception of suprathreshold olfactory qualities. It has tending skills have not always been that impressed [95].
been known for many years that when the intensity of Watson makes suggestions based on an algorithm in
an olfactant changes, it can sometimes change the per- which the system uses traditional pairings, regional pair-
ceptual quality of the associated experience and not just ings, flavour profiles, aroma (flavour) pairing, and learnt
the rated intensity (e.g. [91]). The computational gas- pairings from recipes developed by chefs who have used
tronomy/flavour pairing approach also struggles with the system. A list of ingredients is then produced that is
those flavour compounds where there are profound gen- computed to work well together based on all of this infor-
etically determined perceptual differences. Finally, it mation. Chef Watson has been described as a brilliant tool
should always be borne in mind that ingredients are not to exercise a chef’s imagination and challenge them to use
always incorporated into a recipe solely for the flavour seemingly random, or at the very least unconventional,
they impart. Sometimes, they might be added to enhance combinations of ingredients in dishes. However, it is im-
the colour or change the texture of the final product. portant to note that it is still only pairing ingredients to-
In summary, the problem with the computational ap- gether and does not actually take final taste, flavour, or
proach to flavour pairing, searching as it does for ingredi- mouthfeel into account nor presumably the chemical
ents that share common chemical constituents, is that it transformations that might be induced by cooking. Ultim-
simply does not work as a general principle.6 The ap- ately, then, it is still the chef’s responsibility to figure out
proach neither guarantees nor does it necessarily predict how exactly to make something that actually tastes good
many good unusual ingredient combinations. The main and to develop recipes that really work.
problem with this approach to flavour pairing, at least as
it was originally formulated, is that it is based solely on Cross-cultural aspects of flavour matching
shared compounds, without taking perceptual thresholds The latest computational gastronomy research has
into consideration. Note here only that detection thresh- highlighted a cultural element to the way in which differ-
olds for certain compounds differ dramatically from one ent flavours are combined. Specifically, the ingredients
individual to the next [59]. And even if different ingredi- that co-occur in many Western cuisines tend to
ents share aroma compounds that does not guarantee that harmonize (that is, they share similar flavour compounds).
they will necessarily be present at a level that is detectable. By contrast, an analysis of more than 2500 online Indian
One attempt to address this issue comes from focusing on cuisine recipes from [96] by researchers from the Indian
the volatile compounds in food (as reported in [92]; see Institute for Technology in Delhi revealed that ingredients
also [93]) that are shared, rather than merely focusing on tend to be combined in recipes (the average Indian dish
the presence/absence of specific compounds (as captured contains seven ingredients) because they are dissimilar.
by the Fenaroli database). This intriguing research revealed that ingredients with
While it is certainly true that later attempts to develop dissimilar flavour compounds were combined more
the computational gastronomy approach based on the frequently than would have been expected by chance (e.g.
principle of flavour pairing do now tend to take perceptual [79, 80, 84]). A similar trend toward combining ingredi-
thresholds into account, it is unclear how much success ents with different flavour profiles has now been shown to
the approach actually has. For while proponents argue that, be shared by the cuisine of a number of other Asian coun-
on average, the approach is likely to succeed at an above- tries too [66].7 For instance, Ahn and Ahnert [78] reported
chance level (and that the fact that some combinations that that an analysis of South Korean recipes also shared fewer
do share molecules do not pair well together does not compounds that expected if ingredients were combined
prove that the approach, in general, does not work), those randomly. This appears a much more productive direction
independent attempts to assess the approach do not appear for computational gastronomy approach than the flavour
to have met with much success. The latter negative assess- pairing route discussed earlier.
ment also seems to gel with the anecdotal reports of those
chefs who have experimented with the approach. Availability and habit as key determinants of the
appropriateness of flavour pairing
Chef Watson As mentioned already, there are, of course, examples of
What should we say about IBM’s Chef Watson (https:// culturally appropriate (or recognized) flavour combina-
www.ibmchefwatson.com/community)? On the one hand, tions that seem to have been based primarily on those
one hears those working on the system suggesting that: particular combinations of ingredients that an individual
“…we have developed a computational creativity system has been exposed to previously ([64]; starting presum-
that can automatically or semi-automatically design and ably in the womb, see [97, 98]). As to why those particu-
discover culinary recipes that are flavourful, novel, and lar elements should be combined, that is presumably
Spence et al. Flavour (2017) 6:4 Page 9 of 15

partly explained by what is available in the region, just biscuit side is what they consider the “top” (see [103]).
think of the traditional British combination of salmon However, if you watch how people actually eat their
with watercress (the latter, a common feature of the biscuits, it would appear that the majority eat them
streams in which the salmon would have been found). with the chocolate side facing up. Perhaps, people
This notion of combination based solely on co- want to see the chocolate as they eat the biscuit (ap-
availability is also captured by the phrase “What grows pealing to the eye; see [16]), even though, given the
together goes together,”8 a common idiom in the world distribution of receptors in the oral cavity, it could be
of wine-food pairing. Intriguingly, the latest computa- argued that a better tasting experience would be had
tional gastronomy research to contrast the relative im- were the biscuit to be eaten in the “proper” orienta-
pact of shared climate vs. geographical proximity tion (i.e. with the chocolate side facing downward).
suggests that proximity is the more important factor in Here, it should be stressed that we do not yet have
terms of determining the similarity of various regional any evidence to support the argument regarding the
cuisines in China [99]. initial orientation of the biscuit in the mouth. Never-
While much of the literature on computational gastron- theless, related work from the food industry has
omy and flavour pairing has undoubtedly focused on the stressed the importance of the initial tasting experi-
simultaneous delivery of flavours, it is worth noting how ence, given the assumption of homogeneity (of a
under the majority of everyday conditions (e.g. when sev- given food product), across a particular tasting experi-
eral different elements are presented on the plate or when ence (e.g. see [104–107]).
food is combined with a drink such as wine), one could Sushi provides another interesting example here. For
well imagine how tasting experiences are more likely to according to tradition, it is the fish that should be
occur sequentially. Hence, before closing, it is worth dipped into the soy sauce (in the cases where the chef
returning to the sequencing of flavour sensations. deems it ok to dip in soy sauce). Next, the sushi should
be inserted into the mouth fish-side-down, so that the
Sequencing flavour experiences diner experiences the flavours of the fish/soy sauce more
While much of the discourse around pairing flavours directly on their tongue (see [108]).10 Once again,
seems to be premised on the simultaneous tasting of, though, presenting the sushi fish side down would not
e.g. food and wine, in practice, it is presumably more be very attractive, as all you would see is white rice (i.e.
often the case that tasting experiences occur sequen- the variety between different sushi would be obscured).
tially. The question then becomes one of whether one There is something here about signalling choice visually
order of tasting is better than another and what factors and perhaps putting the most expensive or freshest in-
influence the temporal ordering of our tasting experi- gredients on top.
ences. We started this article thinking about sequential Taken together, then, these examples of layered tasting
flavour experiences in one of Denis Martin’s dishes. experiences help to highlight, once again, the tension be-
However, there are a number of factors that can influ- tween serving a food so that it appeals maximally to the
ence the experienced order of tastes that are worth men- eye and serving it so that it delivers the optimal multi-
tioning briefly before closing. sensory tasting experience.

Layering tasting sensations The role of attention in individuating and binding together
One might think here about the layers of lettuce, cheese, the components of flavour
bacon, etc. in a burger. Indeed, some researchers have Sometimes, changing the sequence in which mouth sen-
already put their mind to optimizing the layering of the el- sations are delivered/experienced does nothing more
ements in this most popular of foods (e.g. [100, 101]; than that. Think about it, if a light is seen before a sound
though see [102]). One might also consider how most ice- is heard, it will be perceived in much the same way as if
cream sundaes, individually served trifles, etc. encourage the sound is perceived before the light. That is, the per-
the dinner into experiencing the elements in the dish in a ception of the component stimuli remains unchanged;
certain sequence.9 In fact, the layering and orientation of all that seems to change is the perception of the tem-
various foods can also be seen as highlighting the contrast poral relationship between them.11 When it comes to
between what pleases the eye and what might be expected the chemical senses, though, a reading of some of the
to please the palate. Layering has also become interesting older literature might be taken to suggest that things
to the food industry as a way of asymmetrically distribut- may, at least sometimes, be importantly different. For, as
ing certain ingredients (e.g. salt) in order to enhance the reported by Georg von Békésy [109], when an olfactory
perception of this quality in the food. stimulus was delivered shortly before a gustatory stimu-
For instance, most chocolate McVitie’s biscuits have lus, the ensuing mixture was typically experienced as an
the chocolate on the “bottom” of the biscuit. The plain odour. However, when the temporal order of stimulation
Spence et al. Flavour (2017) 6:4 Page 10 of 15

was reversed by the order of a few milliseconds, the ex- out in their recent paper, hedonic escalation is more
perience was of a taste localized to the mouth instead likely to occur when a palatable food consists of a com-
(due, presumably, to the phenomenon of oral referral; plex combination of flavours. It is also more likely to
see [110], for a review). occur when someone is motivated to taste additional fla-
This degree of precise temporal control of stimulus vours on each successive bite. To us, the latter descrip-
delivery most likely lies beyond the influence of the chef. tion sounds very much like the mindset of the typical
Nevertheless, what is important to the diner’s tasting ex- diner eating in a modernist restaurant these days. As
perience, at least at the fine temporal scale (i.e. in the Crolic and Janiszewski put it: “…hedonic escalation is
millisecond range), is attention. What the diner volun- more prevalent when people can identify more flavors,
tarily chooses to attend to can bias, and even reverse, attend to additional flavors on each taste trial, have an
the order in which stimuli are perceived (e.g. see [111], opportunity to identify an additional flavor on each taste
for a review of the literature on prior entry; cf. [112]).12 trial, and isolate distinct flavors on each taste trial.”
Here, it is worth noting that even the very description of Thus, given the limited ability of diners to attend se-
the elements in a dish may help direct the diner’s atten- lectively to the elements in a simultaneously experienced
tion in a certain way during a tasting experience. Hence, flavour (due to gestalt grouping and exogenous atten-
it can be argued that attention needs to be considered as tional capture by the mouth), and given the existence of
an important factor when it comes to any consideration the phenomenon of hedonic escalation [9], especially
of sequential tasting experiences taking place in the amongst many contemporary diners, the question in
timescales of seconds [113]. Furthermore, the attentional which order will tasting experiences deliver maximum
disposition of the taster, specifically whether they choose palatability/pleasure becomes more important than ever.
to take an analytic vs. synthetic approach to tasting, has That said, the mystery still remains as to the conditions
also been shown to play an important role in people’s re- under which all components of a flavour experience will
sponse to food stimuli (e.g. see [114]). bind together into a multisensory gestalt, and when they
Here, though, it is important to bear in mind that the remain distinctive, as in the dishes served by Denis
published research also suggests that people can find it Martin and Heston Blumenthal that we started with.
very difficult to pick apart (or individuate) and attend One might wonder whether these, then, are quintessen-
to the individual components of flavour stimuli because tial examples of hedonic escalation but delivered in a
of the brain’s tendency to group the component inputs single mouthful.
into a perceptual gestalt [115, 116]. Binding the compo-
nent stimuli in this way (i.e. as a multisensory flavour Conclusions
gestalt) may well limit the ability of a taster to direct The order and sequence in which people taste foods
their attention selectively to just one element in a com- (and drinks) matters more than many of us realize.14 In
plex tasting experience (see [111]). The laboratory re- this review, we started out by highlighting the potential
search clearly suggests that people exposed to odour tension that exists between artistic plating—what looks
mixtures really struggle to identify any more than three good to the eye [13, 14]—and the likely sequence in
odours in mixtures (of up to six odorants, say) even if which diner will sample/experience the elements on the
they are familiar with all of the component odours plate. We examined the literature concerning the princi-
[115, 117, 118].13 Furthermore, as pointed out by Ste- ples of pairing, combining, and sequencing tasting expe-
venson [113], the exogenous attention capture by gusta- riences. We highlighted examples (of both good and bad
tory and oral-somatosensory stimuli in the oral cavity pairing) reflecting chemical interactions taking place be-
may be a key part of the reason why people find it so tween the component stimuli, giving rise to a resulting
difficult to attention selectively to the olfactory compo- tasting experience that itself may be much better/worse
nent of a flavour stimulus, much though they might than predicted from the mere sum of the parts. We also
wish to do so (see also [119]). summarized those cases where eating something
In summary, therefore, the oft-mentioned limitations changes the functioning of the receptors in the mouth
in the ability of people to direct their attention to separ- (think globe artichoke, miracle fruit, Szechuan pepper-
ate components of a simultaneously experienced flavour corn, etc.).15
may well be part of the reason why the sequential deliv- Having looked at these more extreme examples of the
ery of tasting experiences might be preferred as a way in which physico-chemical interactions can influ-
method of delivery for the chef. Furthermore, the se- ence sequential tasting, we then outlined some of the
quential delivery of tasting experiences also offers the psychological factors relevant to sequencing flavour ex-
opportunity for hedonic escalation (see [9]). This is the periences, such as dynamic contrast and hedonic escal-
name given to the increased liking of each additional ation. We highlighted some of the attentional limitations
bite of a palatable food. As Crolic and Janiszewski point associated with trying to attend separately to the
Spence et al. Flavour (2017) 6:4 Page 11 of 15

components of complex flavour experiences. Further, we reviewed here show it often does, then we would argue
contrasted that with the hedonic escalation that is some- that a more careful analysis of how diners typically inter-
times seen when flavours are experienced (and attended act with the food on their plate (i.e. do they start at the
to) sequentially. Of course, while different flavours may, side, or in middle, with the food on the left, or the right,
on occasion, be experienced simultaneously (this being etc.) becomes increasingly relevant in terms of designing
the situation that much of the flavour pairing literature the optimal plating arrangements. Optimality, here, be-
talks about), the more common situation under everyday ing defined in terms both of what is good for the eyes of
consumption conditions is the sequential experiencing the diner and also for their palate.16
of the elements in a dish or when combining food and And looking to the future, while in this review, the
drink. As this review has tried to make clear, a number focus has been squarely on what is going on within a
of factors may influence the likely success (or failure) of mouthful or a single dish. One can, of course, broaden
sequential tasting experiences. things out to consider the temporal sequencing of the
That said, the majority of the literature on combining courses within a meal, say, or the meals within a day
flavours tends to focus on simultaneous flavour pairing. (e.g. [8, 120–123]). However, addressing that issue is def-
As such, we reviewed some of the general principles of initely the subject for another day.
flavour matching that have been suggested in the litera-
ture. These included the following: similarity—that is, Endnotes
1
matching components based on common flavour com- A somewhat similar approach can be used to resolv-
pounds (or similar flavour profiles); contrast—combina- ing the tannins in a young red wine by first chewing on
tions that are purposely chosen because they differ from a coffee bean. The latter also contains tannin [124], and
each other (a strategy that, as we have seen, is more so the tannic wine will not be perceived as being quite
common in the cuisines of some countries than others); so astringent as it otherwise might (this a trick intro-
and synergy (or emergence)—those combinations that duced by Prof. Barry Smith).
2
together deliver new flavour experiences (see [90]) or Chartier’s [17] Taste buds and molecules was perhaps
else harmonize with one another. the first book to really discuss food-wine pairing at the
However, beyond these perceptual approaches to pairing, molecular level. According to Chartier, wine/food pairings
we also came across examples of conceptual pairing (i.e. are akin to note-by-note cooking in that he believes pair-
based on tradition or by region), as well as examples of in- ings can be made between the two at the molecular level.
dividual pairing (based on an individual’s own personal However, while Chartier’s book is packed with apparently
preferences, experiences, and food history/culture; [53]). successful food-wine pairings, and the approach has been
Indeed, the latest research shows culture can also exert an enthusiastically endorsed by Ferran Adrià and Juli Soller
influence, both on the particular combinations of ingredi- of elBulli fame, the text itself is short on detail about how
ents an individual becomes familiar with (through prior exactly such molecular matches are established.
3
exposure) and through the underlying principle of either This is not quite all though; while Chef Watson does
picking harmonizing (Western cuisine) or distinctive ele- access a database of recipes containing tens of thousands
ments (as in Indian cuisine; [84]; see also [67, 78]). of existing dishes, this information is then connected with
There can be little doubting that the concept of fla- a second database that provides information concerning
vour pairing (based on shared chemical composition) the flavour compounds in thousands of ingredients. Chef
and the rapidly emerging field of computational gastron- Watson also has information about how people respond
omy have become increasingly popular in recent years as to different combinations of flavours (see [125]).
4
possible routes to the effective combination of flavours In fact, it turns out that strawberries have more in
(that are not normally found together). However, as the common with white wine than they do with apples, or-
limited research makes clear, there are a number of anges, or honey (see [80]). Scientific American produced a
problems and limitations with this approach that mean great interactive graphic a couple of years ago that allows
that its main use may be in terms of fostering creativity one to see how many flavour compounds are shared by
in the kitchen rather than necessarily predicting unusual different ingredients (see http://www.scientificamerican.-
flavour combinations that will work well together. As com/article/flavor-connection-taste-map-interactive/).
5
such, we would argue that the effective combination of This all sounds simple enough if you believe the claim
novel elements in a dish or in a food/drink combination that “pleasantness is an approximately linear property of
still remains as much an art as a science. compounds [126]. If two compounds are mixed together
In closing, then, let us return to the question with and smelled, the hypothesis is that the odor pleasantness
which we started this piece: Does the order in which of the mixture is approximately a linear combination of
chemical sensations are experienced when tasting a dish the pleasantness values of the individual compounds.
matter? Given that as a number of the examples With such linearity, one can predict the pleasantness of
Spence et al. Flavour (2017) 6:4 Page 12 of 15

food ingredients that contain several flavour compounds most delicate and subtle of flavors and textures, such as
and of dishes that in turn contain several ingredients.” a few slices of raw sashimi. This is followed by soup or
(Quoted in Varshney et al., [75], p. 10). Hold on for a simmered vegetables in broth. The flavors and textures
moment, though. What about dried scallops and red then get progressively more substantial; perhaps some
wine? Two flavours that many people like individually crispy tempura, followed by grilled fish or meat. The
but together do not combine well. Strawberry aroma meal then winds down with rice, soup and pickles. Des-
and chicken stock aroma? Perhaps, this prediction about sert is sometimes served as well, and is always light; a
pairing works better for unknown aromas and/or odours perfect slice of melon, or perhaps a refreshing cold tofu
that are unrelated to food? Or maybe this claim holds custard.” (quoted at http://www.savoryjapan.com/learn/
more as a generalization despite there being a number culture/power.of.five.html).
15
of exceptions. Olfactory and gustatory adaptation effects can be
6
This despite the huge success of companies such as thought of operating along much the same lines.
16
Flavour Pairing (see https://www.foodpairing.com/; One intriguing culinary challenge here, then, is to try
https://www.foodpairing.com/en/home), whose poster is and find a combination of elements that individually
to be found on the walls of many a top restaurant kitchen. taste just fine and in combination taste much better in
7
Just take the following quote from Ahnert ([67], p. 2): one order than another.
“By comparing the network of ingredients to a body of
Acknowledgements
56,498 online recipes, downloaded from epicurious.com, None.
allrecipes.com, and menupan.com, we were able to show
this hypothesis is confirmed in most Western cuisines, Funding
CS would like to acknowledge the AHRC Rethinking the Senses grant
but not in Eastern ones. This result indicates that shared (AH/L007053/1).
compounds may offer one of several possible mecha-
nisms that can make two ingredients compatible.” Availability of data and materials
8 There is no data or material to make available.
Here, one might wonder whether what grows to-
gether also tends to share more chemical compounds Authors’ contributions
than might be expected by chance. CS, Q(J)W, and JZ wrote all parts of this review. All authors read and
9 approved the final manuscript.
And while we have focused on the temporal order of
sensations during a mouthful or dish, it is worth noting Authors’ information
that there are also interesting questions to be addressed Charles Spence is an experimental psychologist and gastrophysicist working
about the temporal ordering of sensations across the dif- out of Oxford University who is fascinated by the design of multisensory
dining experiences. In 2014, he published the prize-winning The perfect meal:
ferent courses in a meal, say [121, 122]. Unfortunately, The multisensory science of food and dining (Oxford, UK: Wiley-Blackwell)
however, addressing this intriguing topic lies well beyond together with Dr. Betina Piqueras-Fiszman. In March, 2017, he will publish his
the scope of the present article. latest book, Gastrophysics: The new science of eating, with Penguin.
10 Qian (Janice) Wang is a PhD student currently completing her PhD at the
It is considered wrong to dip the rice part of the Crossmodal Research Laboratory, Oxford University. She is also the captain of
sushi in soy sauce (as most Westerners intuitively seem the Oxford University Blind Wine tasting team.
to do), as the rice would soak up too such sauce and Jozef Youssef is a professional chef and director of Kitchen Theory. In 2013,
he published Molecular cooking at home: Taking culinary physics out of the lab
may disintegrate. and into your kitchen with Quintet Publishing.
11
That said, there are a few exceptions: Thunder be-
fore lightning would likely be characterized as two separ- Competing interests
The authors declare that they have no competing interests.
ate events, whereas in the reverse order, it may well be
treated as a single event. Consent for publication
12
There is also an intriguing link to the awareness of Approval to publish has been given where required.
mouthfeel characteristics here too, as part of the total tast-
Ethics approval and consent to participate
ing experience (see [127, 128]). Remember that in the se- No participants were tested in this review paper/opinion piece.
quentially experienced dish from Denis Martin that we
started this piece with, there was an initial wasabi hit, Author details
1
Department of Experimental Psychology, University of Oxford, Oxford OX1
likely localized to the bridge of the nose, followed by the 3UD, UK. 2Kitchen Theory, London, UK.
taste/texture/flavour of the tuna, and ending up with the
lingering creamy mouthfeel of the white chocolate tablet. Received: 20 October 2016 Accepted: 7 January 2017
13
This finding, by itself, might nudge the chef toward
trying to deliver different flavours sequentially rather References
than all at the same time. 1. Martin D. Évolution. Lausanne: Editions Favre; 2007.
14 2. Halpern BP. More than meets the tongue: Temporal characteristics of taste
The Japanese were onto this long ago. Just take the intensity and quality. In: Lawless HT, editor. Sensory science theory and
following quote: “Kaiseki meals usually start with the applications in foods. New York, NY: Marcel Dekker; 1991. p. 37–105.
Spence et al. Flavour (2017) 6:4 Page 13 of 15

3. Jinks A, Laing DG. Temporal processing reveals a mechanism for limiting 34. Spence C, Piqueras-Fiszman B. The perfect meal: The multisensory science
the capacity of humans to analyze mixtures. Cogn Brain Res. 1999;8:311–25. of food and dining. Oxford, UK: Wiley-Blackwell; 2014.
4. Kobayakawa T, Gotow N. Interaction between olfaction and gustation by 35. Hagura N, Barber H, Haggard P. Food vibrations: Asian spice sets lips
using synchrony perception task. i-Perception. 2011;2(8). http://i-perception. trembling. Proc R Soc B. 2013;280:20131680.
perceptionweb.com/journal/I/article/ic964. 36. Bartoshuk LM, Lee CH, Scarpellino R. Sweet taste of water induced by
5. Kuznicki JT, Turner LS. Temporal dissociation of taste mixture components. artichoke (Cynara scolymus). Science. 1972;178:988–90.
Chem Senses. 1988;13:45–62. 37. Blakeslees AF. A dinner demonstration of threshold differences in taste and
6. Marshall K, Laing DG, Jinks AL, Effendy J, Hutchinson I. Perception of smell. Science. 1935;81:504–7.
temporal order and the identification of components in taste mixtures. 38. Woods WM. Artichokes. Science. 1971;193:1195.
Physiol Behav. 2005;83:673–81. 39. Beard J. Gourmet. 1971;31:30.
7. Blumenthal H. Weird but wonderful. In: The Guardian, May 4th. 2002. p. 84. 40. Centelles F. Can wine and artichokes ever be friends? May 4th. 2014. www.
https://www.theguardian.com/lifeandstyle/2002/may/04/foodanddrink. jancisrobinson.com.
shopping. 41. Graves E. Life. 1967;62:114.
8. Hyde RJ, Witherly SA. Dynamic contrast: A sensory contribution to 42. Cometto-Muñiz JE, Cain WS. Olfactory adaptation. In: Doty RL, editor.
palatability. Appetite. 1993;21:1–16. Handbook of olfaction and gustation. New York: Dekker; 1995. p. 257–81.
9. Crolic C, Janiszewski C. (in press). Hedonic escalation: When food just tastes 43. McBurney DH, Pfaffmann C. Gustatory adaptation to saliva and sodium
better and better. J Consumer Res. chloride. J Exp Psychol. 1963;65:523–9.
10. Myhrvold N, Young C. Modernist cuisine. The art and science of cooking. La 44. O’Mahoney M, Heintz C. Direct magnitude estimation of salt taste intensity
Vergne, TN: Ingram Publisher Services; 2011. with continuous correction for salivary adaptation. Chem Senses. 1981;6:
11. Youssef J. Molecular cooking at home: Taking culinary physics out of the lab 101–12.
and into your kitchen. London, UK: Quintet Publishing; 2013. 45. Blumenthal H. The big Fat Duck cookbook. London, UK: Bloomsbury; 2008.
12. Aduriz AL. Mugaritz: A natural science of cooking. New York, NY: Phaidon; 46. Nygren IT, Gustafsson IB, Johansson L. Perceived flavour change in blue
2014. mould cheese after tasting white wine. Food Service Technol. 2003b;3: 143–
13. Apicius. Cooking and dining in Imperial Rome (c. 1st Century; translated by 150.
J. D. Vehling). Chicago, IL: University of Chicago Press; 1936. 47. Schifferstein HNJ, Frijters JER. Contextual and sequential effects on
14. Michel C, Velasco C, Gatti E, Spence C. A taste of Kandinsky: Assessing the judgments of sweetness intensity. Percept Psychophys. 1992;52:243–55.
influence of the artistic visual presentation of food on the dining 48. Galmarini M. Use of temporal dominance of sensations (TDS) to better
experience. Flavour. 2014;3:7. understand wine-cheese associations. Paper presented at Eurosense 2016,
15. Deroy O, Michel C, Piqueras-Fiszman B, Spence C. The plating manifesto (I): 11-14th September. Dijon, France; 2016.
From decoration to creation. Flavour. 2014;3:6. 49. Galmarini MV, Loiseau A-L, Visalli M, Schlich P. Use of multi-intake Temporal
16. Spence C, Okajima K, Cheok AD, Petit O, Michel C. Eating with our eyes: Dominance of Sensations (TDS) to evaluate the influence of cheese on wine
From visual hunger to digital satiation. Brain & Cognition. 2016;110:53–63. perception. J Food Sci. 2016;81:S2566–77.
17. Chartier F. Taste buds and molecules: The art and science of food, wine, and 50. Madrigal-Galan B, Heymann H. Sensory effects of consuming cheese prior
flavor (translated by Levi Reiss). Hoboken, NJ: John Wiley and Sons; 2012. to evaluating red wine flavour. Am J Enology Viticulture. 2006;57:12–22.
18. De Klepper M. Food pairing theory: A European fad. Gastronomica: J Food 51. Nygren IT, Gustafsson IB, Johansson L. Perceived flavour changes in white
Culture. 2011;11:55–8. wine after tasting blue mould cheese. Food Service Technol. 2002;2:163–71.
19. Cooper D. Wine with food. 2. In: Wine with food. London, UK: Octopus 52. Nygren IT, Gustafsson IB, Johansson L. Effects of tasting technique –
Publishing; 1980. p. 10–9. sequential tasting vs. mixed tasting – on perception of dry white wine and
20. Simon J. Rules, & how to break them. In: Wine with food. London, UK: blue mould cheese. Food Service Technol. 2003a;3: 61–69.
Octopus Publishing; 1996. p. 10–9. 53. Eschevins A. From beverage to companion food: Experts’ principles for
21. Tamura T, Taniguchi K, Suzuki Y, Okubo T, Takata R, Konno T. Iron is an pairing wines and beers with food. Paper presented at Eurosense 2016, 11-
essential cause of fishy aftertaste formation in wine and seafood pairing. 14th September. Dijon, France; 2016.
J Agric Food Chem. 2009;57:8550–6. 54. Dornenburg A, Page K. What to drink with what you eat: The definitive
22. Ribéreau-Gayon P, Glories Y, Maujean A, Dubordieu D. Chapter 4: Dry extract guide to pairing food with wine, beer, spirits, coffee, tea - even water -
and minerals. In: Handbook of enology, The chemistry of wine stabilization Based on expert advice from America’s best sommeliers. USA: Little, Brown,
and treatments, vol. 2. 2nd ed. West Sussex, UK: Wiley; 2006. p. 91–108. & Company; 2006.
23. Jackson S. Wine, health, and food. In: Wine science: Principles, practice, 55. Page K, Dornenburg A. The flavor bible. 1st ed. USA: Little, Brown, &
perception. 2nd ed. San Diego, CA: Academic Press; 2000. p. 600–5. Company; 2008.
24. Walton S, Glover B. Drinking wine with food. In: The ultimate encyclopedia 56. Segnit N. The flavour thesaurus: Pairings, recipes and ideas for the creative
of wine, beer, spirits & liqueurs. London, UK: Lorenz Book; 1998. p. 21. cook. London, UK: Bloomsbury; 2010.
25. Greenwood V. How to hack your taste buds. In: BBC, December 15th. 2014. 57. Shepherd GM. Neurogastronomy: How the brain creates flavor and why it
http://www.bbc.com/future/story/20141214-how-to-hack-your-taste-buds. matters. New York, NY: Columbia University Press; 2012.
26. Woollaston V. So THAT’S why toothpaste ruins the taste of food! Chemicals 58. Blank DM, Mattes RD. Sugar and spice: Similarities and sensory attributes.
in paste destroy receptors on the tongue and cause a bitter sensation. In: Nurs Res. 1990;39:290–3.
Daily Mail Online, October 7th. 2014. http://www.dailymail.co.uk/ 59. Smith BC. Same compounds: Different flavours? In: Chassagne D, editor.
sciencetech/article-2783404/So-THAT-s-toothpaste-ruins-taste-food- Proceedings of Wine Active Compounds 2008. University of Bourgogne:
Chemicals-paste-destroy-receptors-tongue-cause-bitter-sensation.html. Oenopluria Media; 2008. p. 98–102.
27. Lawless HT. Sensory interaction in mixtures. J Sensory Stud. 1986;1:259–74. 60. Schifferstein HNJ, Verlegh PWJ. The role of congruency and pleasantness in
28. Lawless HT, Stevens DA. Effects of oral chemical irritation on taste. Physiol odor-induced taste enhancement. Acta Psychol (Amst). 1996;94:87–105.
Behav. 1984;32:995–8. 61. Stevenson RJ, Boakes RA. Sweet and sour smells: Learned synaesthesia
29. Daniell WF. On the Synsepalum dulcificum, de Cand.; or miraculous berry of between the senses of taste and smell. In: Calvert GA, Spence C, Stein BE,
western Africa. Pharmacol J. 1852;11, 445–448. editors. The handbook of multisensory processing. Cambridge, MA: MIT
30. Anon. Gymnema sylvestre, an Indian plant which, when chewed, destroys Press; 2004. p. 69–83.
the power of tasting sugar. Pharmacol J. 1847;7:351–2. 62. Stevenson RJ, Prescott J, Boakes RA. The acquisition of taste properties by
31. Eisner T, Halpern BP. Taste distortion and plant palatability. Science. 1971; odors. Learning Motiv. 1995;26:433–55.
172:1362. 63. Williams JA, Bartoshuk LM, Fillingim RB, Dotson CD. Exploring ethnic
32. Kurihara K, Beidler LM. Taste-modifying protein from miracle fruit. Science. differences in taste perception. Chem Senses. 2016;41:449–56.
1968;161:1241–3. 64. Rozin E. Ethnic cuisine: The flavor-principle cookbook. Brattleboro, VT: The
33. Koizumi A, Tsuchiya A, Nakajima KI, Ito K, Terada T, Shimizu-Ibuka A, Briand L, Stephen Greene Press; 1983.
Asakura T, Misaka T, Abe K. Human sweet taste receptor mediates acid- 65. Visser M. The rituals of dinner: The origins, evolution, eccentricities, and
induced sweetness of miraculin. Proc Natl Acad Sci U S A. 2011;108:16819–24. meaning of table manners. London, UK: Penguin Books; 1991.
Spence et al. Flavour (2017) 6:4 Page 14 of 15

66. Ahn YY, Ahnert SE, Bagrow JP, Barabási AL. Flavor network and the 92. Nijssen LM, van Ingen-Visscher CA, Donders JJH. Food database – Version
principles of food pairing. Sci Rep. 2011;1(196):1–6. 13.2 – Zeist. The Netherlands: TNO Triskelion; 1963–2012.
67. Ahnert SE. Network analysis and data mining in food science: The 93. Van Gemert LJ. Flavour threshold values in water and other media. 2nd ed.
emergence of computational gastronomy. Flavour. 2013;2:4. Canton, GA: Leffingwell; 2011.
68. Davis AP. Digital gastronomy: When an IBM algorithm cooks, things get 94. Velasco C, Michel C, Youssef J, Gamez X, Cheok AD, Spence C. Colour-taste
complicated—and tasty. Wired. 2013;20(10). September 17th. https://www. correspondences: Designing food experiences to meet expectations or to
wired.com/2013/09/digitalgastronomy/. surprise. Int J Food Design. 2016;1:83–102.
69. IBM in collaboration with Institute of Culinary Education. Cognitive cooking 95. Trout C. I trusted my gut to IBM’s Watson and it gave me a fowl old-
with Chef Watson: Recipes for innovation from IBM & the Institute of fashioned. In: Engadget, May 15th. 2015. http://www.engadget.com/2015/
Culinary Education. Institute of Culinary Education; 2015. 05/15/drinking-with-watson/.
70. Kinouchi O, Diez-Garcia RW, Holanda AJ, Zambianchi P, Roque AC. The non- 96. Dalal T. Tarla Dalal. 2014. http://www.tarladalal.com.
equilibrium nature of culinary evolution. New J Phys. 2008;10(7):073020. 97. Schaal B, Durand K. The role of olfaction in human multisensory
71. Morris RG, Burton SH, Bodily PM, Ventura D. Soup over beans of pure joy: development. In: Bremner AJ, Lewkowicz D, Spence C, editors. Multisensory
Culinary ruminations of an artificial chef. In: Proceedings of the International development. Oxford, UK: Oxford University Press; 2012. p. 29–62.
Conference on Computer Creativity (ICCC 2012), May 2012. 2012. p. 119–25. 98. Schaal B, Marlier L, Soussignan R. Human foetuses learn odours from their
72. Singel R. Google recipe search cooks up next gen of search. In: Wired, pregnant mother’s diet. Chem Senses. 2000;25:729–37.
February, 24th. 2011. http://www.wired.com/epicenter/2011/02/google- 99. Zhu YX, Huang J, Zhang ZK, Zhang QM, Zhou T, Ahn YY. Geography and
recipe-semantic/. similarity of regional cuisines in China. PLoS One. 2013;8(11):e79161.
73. Taft DK. Cognitive cooking: Inside IBM’s Watson food truck. In: eWeek, 100. Lee J. How to make the perfect burger: Oxford food scientist claims to have
March 14th. 2014. http://www.eweek.com/database/slideshows/cognitive- answer: Oxford University chef says perfect burger is 7cm tall, should be
cooking-inside-ibms-watson-food-truck.html. eaten to music, given a name and should feel as good as it tastes. In: The
74. Teng CY, Lin YR, Adamic LA. Recipe recommendation using ingredient Telegraph, August 16th. 2015. http://www.telegraph.co.uk/news/science/
networks. In: Proceedings of the 3rd Annual ACM Web Sci. Conf. science-news/11823677/How-to-make-the-perfect-burger-Oxford-food-
(WebSci’12), June 2012. 2012. p. 298–307. https://arxiv.org/pdf/1111.3919. scientist-claims-to-have-answer.html.
pdf. 101. Morgan M. A 7cm wide patty, crunchy lettuce and a warm bun… eaten
75. Varshney KR, Varshney LR, Wang J, Myers D. Flavor pairing in medieval European with your hands: Chef reveals formula for the perfect burger (and how it
cuisine: A study in cooking with dirty data. In: Proceedings of the International SOUNDS is as important as the taste). In: Daily Mail Online, August 25th.
Joint Conference on Artificial Intelligence Workshops. 2013. p. 3–12. 2015. http://www.dailymail.co.uk/femail/food/article-3210117/Chef-reveals-
76. Varshney LR, Pinel F, Varshney KR, Bhattachrjya D, Schörgendorfer A, Chee formula-perfect-burger-SOUNDS-just-important-taste.html.
YM. A big data approach to computational creativity. In: IEEE International 102. Sinnerton J. Burger battle: Boffins can kiss my buns. In: Brisbane Courier-
Conference on Cognitive Informatics and Cognitive Computing, New York, Mail, August 5th. 2015. http://www.couriermail.com.au/news/queensland/is-
July. 2013. this-queenslands-best-burger-ohio-cafe-owner-yogesh-koshe-reckons-his-
77. Burdock GA. Fenaroli’s handbook of flavor ingredients. 6th ed. Boca Raton, recipe-cant-be-beat/story-fnihsrf2-1227514375022.
FL: CRC Press; 2010. 103. Enoch N. Crumbs, we’ve been eating McVitie’s Digestives and Hobnobs all
78. Ahn YY, Ahnert SE. The flavor network. Leonardo. 2013;46:272–3. wrong! Firm says chocolate part is the BOTTOM. In: Daily Mail Online, March
79. Zolfaghifard E. Why Indian food is so delicious: Scientists say our love of 28th. 2014. http://www.dailymail.co.uk/news/article-2591803/McVities-
curry is down to a lack of overlapping flavours. In: Daily Mail Online, March Digestive-Hobnobs-shock-Weve-eating-wrong-way.html#ixzz4KIw1L5e5.
4th. 2015. http://www.dailymail.co.uk/sciencetech/article-2979794/Why- 104. Dijksterhuis GB, Le Berre E, Woods AT. Food products with improved taste.
Indian-food-delicious-Scientists-say-love-curry-lack-overlapping-flavours.html. In: Patent Identifier No. EP2451289 A1. 2010.
80. Ferdman RA. Scientists have figured out what makes Indian food so 105. Dijksterhuis G, Boucon C, Le Berre E. Increasing saltiness perception through
delicious. In: The Washington Post, March 3rd. 2015. https://www. perceptual constancy created by expectation. Food Qual Preference. 2014;
washingtonpost.com/news/wonk/wp/2015/03/03/a-scientific-explanation-of- 34:24–8.
what-makes-indian-food-so-delicious/. 106. Le Berre E, Boucon C, Knoop M, Dijksterhuis G. Reducing bitter taste
81. Blumenthal H. Naivety in the kitchen can lead to great inventions, but too through perceptual constancy created by an expectation. Food Qual
much can take you to some strange places. In: The Times, August 19th. Preference. 2013;28:370–4.
2010. p. 45. 107. Woods AT, Poliakoff E, Lloyd DM, Dijksterhuis GB, Thomas A. Flavor
82. Kort M, Nijssen B, van Ingen-Visscher K, & Donders J. (n.d.). Food pairing expectation: The effects of assuming homogeneity on drink perception.
from the perspective of the ‘volatile compunds in food’ database (pp. 589- Chemosensory Perception. 2010;3:174–81.
592). https://home.zhaw.ch/yere/pdf/Teil146%20- 108. Lowry D. The connoisseur’s guide to sushi: Everything you need to know
%20Expression%20of%20Multidisciplinary.pdfb. about sushi varieties and accompaniments, etiquette and dining tips and
83. Bredie WLP, Petersen MA, Hartvig D, Frøst MB, Risbo J, Møller P. Flavour more. Boston, MA: Harvard Common Press; 2005.
pairing of foods: A physical-chemical and multisensory challenge for health 109. von Békésy G. Interaction of paired sensory stimuli and conduction in
promotion. Eur Sens Netw. 2015. http://www.esn-network.com/index. peripheral nerves. J Appl Physiol. 1963;18:1276–84.
php?id = 1034. 110. Spence C. Oral referral: Mislocalizing odours to the mouth. Food Qual
84. Jain A, Rakhi NK, Baglerb G. Spices form the basis of food pairing in Indian Preference. 2016;50:117–28.
cuisine. 2015. http://arxiv.org/ftp/arxiv/papers/1502/1502.03815.pdf. 111. Spence C, Parise C. Prior entry. Consciousness Cognition. 2010;19:364–79.
85. Mauer L, El-Sohemy A. Prevalence of cilantro (Coriandrum sativum) disliking 112. Ashkenazi A, Marks LE. Effect of endogenous attention on detection of
among different ethnocultural groups. Flavour. 2012;1:8. weak gustatory and olfactory flavors. Percept Psychophys. 2004;66:596–608.
86. Wysocki CJ, Beauchamp GK. Ability to smell androstenone is genetically 113. Stevenson RJ. The role of attention in flavour perception. Flavour. 2012;1:2.
determined. Proc Natl Acad Sci U S A. 1984;81:4899–902. 114. Prescott J, Lee SM, Kim KO. Analytic approaches to evaluation modify
87. Mottram D. How flavours develop and change during cooking. In: hedonic responses. Food Qual Preference. 2011;22:391–3.
Blumenthal H, editor. The big Fat Duck cookbook. London, UK: Bloomsbury; 115. Laing DG, Link C, Jinks AL, Hutchinson I. The limited capacity of humans to
2008. p. 501–2. identify the components of taste mixtures and taste-odour mixtures.
88. Breslin PAS, Beauchamp GK. Salt enhances flavor by suppressing bitterness. Perception. 2002;31:617–35.
Nature. 1997;387:563. 116. Spence C. Cross-modal perceptual organization. In: Wagemans J, editor. The
89. Keast RSJ, Breslin PAS. An overview of binary taste-taste interactions. Food Oxford handbook of perceptual organization. Oxford, UK: Oxford University
Qual Preference. 2002;14:111–24. Press; 2015. p. 649–64.
90. Mitchell MJ, McBride RL. Effects of propanol masking odor on the olfactory 117. Laing D, Francis G. The capacity of humans to identify odors in mixtures.
intensity scaling of eugenol. J Exp Psychol. 1971;87:309–13. Physiol Behav. 1989;46:809–14.
91. Kruger L, Feldzamen AN, Miles WR. A scale for measuring supra-threshold 118. Laing DG, Glenmarec A. Selective attention and the perceptual analysis of
olfactory intensity. Am J Psychol. 1955;68:117–23. odor mixtures. Physiol Behav. 1992;52:1047–53.
Spence et al. Flavour (2017) 6:4 Page 15 of 15

119. Stevenson RJ, Mahmut MK, Oaten MJ. The role of attention in the
localization of odors to the mouth. Atten Percept Psychophys.
2011;73:247–58.
120. Lahne J, Zellner DA. The great is the enemy of the good: Hedonic contrast
in a coursed meal. Food Qual Preference. 2015;45:70–4.
121. Rolls BJ, Laster LJ, Summerfelt A. Meal order reversal: Effects of eating a
sweet course first or last. Appetite. 1991;16:141–8.
122. Shukla AP, Iliescu RG, Thomas CE, Aronne LJ. Food order has a significant
impact on postprandial glucose and insulin levels. Diabetes Care. 2015;38(7):
e98–9.
123. Spence, C., Wang, Q. (J.), & Youssef, J. (2017). On the temporal structuring of
the meal. Manuscript in preparation.
124. Savolainen H. Tannin content of tea and coffee. J Appl Toxicol. 1992;12:191–
2.
125. Wakefield J. What would a computer cook for dinner? BBC News Online,
March 7th. 2014. http://www.bbc.co.uk/news/technology-26352743.
126. Lapid H, Harel D, Sobel N. Prediction models for the pleasantness of binary
mixtures in olfaction. Chem Senses. 2008;33:599–609.
127. Gallace A, Spence C. In touch with the future: The sense of touch from
cognitive neuroscience to virtual reality. Oxford, UK: Oxford University Press;
2014.
128. Haggard P, de Boer L. Oral somatosensory awareness. Neurosci
Biobehavioural Rev. 2014;47:469–84.

Submit your next manuscript to BioMed Central

and we will help you at every step:
• We accept pre-submission inquiries
• Our selector tool helps you to find the most relevant journal
• We provide round the clock customer support
• Convenient online submission
• Thorough peer review
• Inclusion in PubMed and all major indexing services
• Maximum visibility for your research

Submit your manuscript at

www.biomedcentral.com/submit