As mentioned in my previous post about the flavor pairing presentation given by Wender Bredie as part of the Copenhagen seminar on molecular gastronomy I’m really happy that the topic has been brought into the scientific community. At the same time is has also become very clear to me that the term flavor pairing needs some clarification. First of all I have come to realize that the the term flavor pairing is slightly misleadning, and I wonder if aroma similarity perhaps is a more precise term. As I see it, today the term flavor pairing is used in a range of different ways:

1. Empirically based pairings. These are the good combinations of ingredients (or even food and wine) that more or less all cooking relies on. An excellent source for such flavor pairings is The Flavor Bible. If one can talk about any chemical principles at all it would be related to balancing the basic tastes and using contrasting elements. But the pleasantness of the empirical based pairings are probably also often subject to the exposure effect – you get used to something and eventually start liking it.
2. Predicted aroma similarity. This is the broad category that I’ve previously referred to as flavor pairing. It can be further subdivided into the following categories:
   1. Predicted aroma similarity based on overlap of one or a few volatiles
   2. Predicted aroma similarity based on the number of overlapping volatiles
   3. Predicted aroma similarity based on overlap between one or more of the high concentration volatiles
   4. Predicted aroma similarity based on overlap between high impact odorants (volatiles to compare are first selected based on OAV, AEDA, CHARM, FD …)
   5. Predicted aroma similarity based on similar neurological responses (as judged by fMRI or any other technique)

The reason I propose aroma similarity for category 2 is to avoid confusion with category 1. Also, such a term does not suggest that only combining the two will yield heavenly combinations. To a chef, it may even be the opposite, as foods or ingredients selected based on aroma similarity may be too similar if used by themselves. The predictive power in category 2 increases from 2a, 2b and 2c (where it is zero or very close to zero) to 2d and 2e (where predictions make sense). Personally I have always thought that what I previously referred to as flavor pairings should at least be of category 2d based on this, but I realize that I probably havent expressed this clear enough before. Another obvious reason to clearify the terms is of course that some of the discussion that arises around flavor pairing stems from different understandings and definitions of what flavor pairing is.

From what I’ve heard the flavor pairing examples Heston Blumenthal has come up with (or at least some of them) belong to category 2d/2e thanks to proprietary research that was made available to him. But since the background data has not been published (and probably never will be) it is difficult to evaluate this further. I also seem to remember that Heston has talked about flavor pairings of category 2a, 2b and/or 2c based on data from the VCF database (I think it was in one of the RSC videos). I do not know in which category the example from François Benzi (indole in jasmine and pork liver) belongs, but since indole was mentioned specifically it maybe category 2a unless indole is in fact present in high concentrations or even is a high impact odorant.

In the food blogging event They Go Really Well Together (TGRWT) I have tried to argue that flavor pairing at least should be based on category 2d types of pairings. But due to the very limited amount of data available it has not been possible to asses whether the pairings published by Heston actually do have an overlap of the impact odorants. From what I’ve heard the foodpairing website (old .be site, new .com site) put together by Bernard Lahousse and Lieven De Couvreur does incorporate threshold values and thereby is based on the impact/activity of the odors and hence falls into category 2d, but regrettably very little has been published about the underlying methodology used to create the website.

Let me also add that although I firmly believe that it makes sense to talk about activity or impact of odorants, I have previously addressed some of the limitations with odor activity values, and I quote myself:

It’s only fair enough to add that the concept of odor activity values has it’s limitations. Some are related to matrix effects, because thresholds are not necessarily recorded in a matrix mimicking the food product. Possible synergies between flavour compounds are disregarded (examples are known where sub-threshold concentrations are detected in the presence of other volatile compounds). Also, the underlying assumption that the odor intensity increases linearily is not quite correct. The typical intensity vs. concentration curve is more ‘S’ shaped with an expansive, linear and compressive region as shown below. At low concentrations (expansive region) synergism (also known as hyperadditivity or mutual enhancement) is observed. At high concentrations (compressive region) antagonism (or subadditivity or mutual suppresion) is observed. This means that a high OAV overestimates and a low OAV underestimates the impact of the individual compounds.

The fact that Wender Bredie in his studies found absolutely no correlation between the hedonic score of the food pairings and the overlap of volatile flavors could at first be a little disappointing. But I believe the only conclusion that can be drawn using data from VCF (which has information about volatiles and some concentration data) is that type 2b/2c predictions are very weak. I strongly believe that one should take into account some kind of metric to filter out the odors that do not contribute to the overall aroma of the food.

For fun I checked Wender’s list of “the good, the bad and the ugly” using The Flavor Bible (a good source of category 1 flavor pairings) and the foodpairing website (presumably category 2c and/or 2d predictions). This is what I found:

It is interesting that The Flavor Bible (being purely empirical) actually mentions three of the “good” pairings. That it also mentions one of the “ugly” pairings is a good reminder that binary mixtures are not actually food and this real life is more complex. And it is also peculiar that all the garlic combinations were perceived as “ugly”. In a moussaka for instance one would combine cinnamon and garlic (with many other spices and herbs of course), and it is indeed quite delicious. Moving on to the predictions from foodpairing.be it is not so easy to draw any conclusions, but it is interesting that the four predicted aroma similarity pairs fall into the whole range of hedonic categories (i.e. ugly, bad and good). These could be interesting cases in a follow up study at lower concentrations as Wender mentioned in his presentation.

Based on the distinction between the different types of flavor pairing suggested above the flavor pairing hypothesis may be reformulated as follows: For foods with a predicted aroma similarity based on the analysis of it’s volatiles there is a good chance that they can be used together in a dish. Needless to say, the prediction should be of type 2d or 2e. Experience from the TGRWT rounds further suggests that some experimentation may be needed to find the right balance between the two and that contrasting elements are very important, otherwise the combination may turn out quite bland.

So to conclude:  What I’m after from a gastronomical perspective is a tool to pair foods and suggest extra ingredients based on aroma similarity. This first and foremost becomes interesting if the combinations are novel or have a surprise element. Because of the aroma similarity the complexity of the combination will presumably be less than anticipated. Thinking about the Wundt curve (which I learnt about in Michael Bom Frøst’s presentation in Copenhagen – more on that in a separate post) which suggests that pleasantness first increases and then decreases with increasing complexity, a less than anticipated complexity combined with novelty/surprise could perhaps be what we are looking for when we try to create new and delicious food.

Finally it’s time for a new round of TGRWT. It’s the 18th round and the host this time is Aidan Brooks, a trainee chef who works in Spain. In his blog he’s touched upon flavor pairing several times and also wrote a blog post on the same topic for “Word of mouth”, the food blog of The Guardian. The foods to pair this time are plum and blue cheese, and as usual you can read more about how to participate in the announcement post. The deadline for submissions is September 1st.

TGRWT is not a competition, but Aidan wanted to add a little competitive element to round of the meal. So his challenge to you (in addition to the normal TGRWT contributions) is to make a dessert using plum and blue cheese which additionally incorporates any two of the following flavors: salty, sour, bitter, umami, pungent, astringent. He will pick a winner and the prize is quite cool: he’ll make a restaurant level version of it, with the name of the winner incorporated in the title of the dish and try to get this on the menu at Restaurante Ferrero where he works as a Pastry Chef on the team of Paco Morales, Spain’s “Chef of the Year 2009″.

Photo by mykrar from flickr.com (CC BY-NC 2.0)

Regarding the chemistry behind this pairing there is (for once) actually some informartion available. From a paper on gorgonzola I’ve learnt that 2-nonanone, 1-octen-3-ol, 2-heptanol, ethyl hexanoate, methylanisole and 2-heptanone are the most important impact odorants of natural and creamy Gorgonzola cheese [1]. For plums I haven’t found anything on impact odorants in fresh plums, but there is a paper on impact odorants in candied plums [2], and comparing the two papers I find that phenyl acetaldehyde, ethyl benzoate and ethyl octanoate are present in both Gorgonzola cheese and candied plums and have odor activity values greater than 1. There’s another paper as well comparing two plum varieties were the abstract mentions odor unit values, but alas – I don’t have access to the full text (maybe someone with access could check it and give me a hint about the impact odorants?) [3]. There are probably some differences between fresh and candied plums, but at least this shows that you’re safe if you want to use candied plums when cooking up something with blue cheese.

[1] Moio, L.; Piombiono, P.; Addeo F. J. Dairy Research 2000, 67, 273 “Odour-impact compounds of Gorgonzola cheese”. DOI: 10.1017/S0022029900004106
[2] Nunes, C.; Coimbra, M. A.; Saraiva, J.; Rocha, S. M. Food Chemistry 2008, 111, 897 “Study of the volatile components of a candied plum and estimation of their contribution to the aroma”. DOI: 10.1016/j.foodchem.2008.05.003
[3] Gómez, E.; Ledbetter, C. A. J. of the Science of Food and Agriculture 2006, 65 (1), 111. “Comparative study of the aromatic profiles of two different plum species: Prunus salicina Lindl and Prunus simonii L”. DOI: 10.1002/jsfa.2740650116

The Canadian sommerlier François Chartier (he has an extensive website featuring several blogs, including a section named Sommellerie moléculaire) is out with a new book on food and wine pairing. It’s not just another (superfluous) book on the subject. As the title Papilles et molécules (= Tastebuds and Molecules, unfortunately not available in English) suggests there is some science involved. It turns out in fact that he has applied the principles of flavor pairing to food and wine. With help from Richard Béliveau from Agriculture and Agri-Food Canada and Martin Loignon from PerkinElmer he has analyzed wines and food and comes up with the following suggestions for lamb, as described in the article “Chemistry-set wine pairing”:

Having roast lamb? Don’t waste it on an ill-advised red Bordeaux, the old standby trotted out by generations of sommeliers. Lamb’s characteristic flavour comes from thymol, an aromatic compound found in the oil of, yes, thyme. It’s also a flavour note associated with red wines from the southern Languedoc region of France, such as Minervois, Corbières or St. Chinian.

Other combinations mentioned in the article include:

rosemary – white wines from northern Alsace
pork – oaked red wines
curries – viognier
cinnamon – pinot noir, grenache, ice cider, oloroso sherry

François Chartier also introduces “bridge ingredients”. Mint, which goes well with sauvignon blanc, shares aroma compounds with parsley, fennel and tarragon. Based on this he theorizes that sauvignon blanc should also pair well with dishes based on these ingredients.

As far as I can see (with my very limited high school French) there are no links or references to all the other activities in the field: Heston Blumenthal’s pioneering of the concept together with François Benzi from Firmenich, the Food pairing website – not even to the TGRWT food blogging event Nevertheless it’s nice to see that the concept has now been applied to food and wine as well. As I don’t own the book yet I can’t tell whether François Chartier includes odor activity values in his discussion or not (but I certainly hope he does!).

Update: The book is now available in English as Taste Buds and Molecules: The Art and Science of Food With Wine.

I had a wonderful trip to Brugge/Bruges to attend the foodpairing seminar The Flemish Primitives. I got to meet many interesting people including Heston Blumenthal, Peter Barham, Andrew Dornenburg, Karen Page, Ben Roche and Tony Conigliaro to mention a few. I also finally had the opportunity to talk to my fellow Swedish food bloggers Lisa Förare Winbladh (Matälskaren, Swedish only but Google can translate) and Malin Sandström (Matmolekyler, Swedish only but Google can translate) who’ve recently been awarded money to write a Swedish book about molecular gastronomy for home cooks. I even talked to several people who read Khymos! It’s always nice when I can attach some faces to the crowd out there in the big, unpersonal blogosphere.

As you see from this long post the day was packed and believe it or not – there will be a couple more posts in the next few days. One on the surprise “chocolate box” (for me this was the highlight), a summary of the interview with Heston Blumenthal and some info on the chemistry behind the glowing lollipops! I’ll also try do dig up the recipe for the chocolate dip that came with our lunch fries.


In case you wondered: Peter always wears penguins!

Peter Barham (physicist, author of The science of cooking) started the day by giving an introduction to the scientific approach to cooking. Cooking started a lot of science, and chemistry (and alchemy) has origins that can be traced back to the observation of how food changed when cooked. And as a side note: even the word chemistry is linked to food through the greek word Khymos Until about 100 years ago there was a clear link between cooking and chemistry, but then came a period where scientists mostly did not bother much about food. Until Nicholas Kurti entered the scene: that’s the guy who said that “I think it is a sad reflection on our civilization that while we can and do measure the temperature in the atmosphere of Venus we do not know what goes on inside our soufflés”. Four examples were used to illustrate how science can help chefs.

1) Conflicting reasons are given for why salt should be added to blanching water to retain the green color of vegetables. Contrary to popular belief salt does not raise the boling temperature of the water with more than a fraction of a degree. Science tells us that there really isn’t any good reason for adding salt to the water when blanching vegetables.

2) Science also provides insight to what happens when we heat up meat. When heated the meat first gets tougher. At around 40-45 °C the meat proteins start to denature, and at 55 °C the meat goes from tender to tough. Prolonged heating above 55 °C however will cause the collagen to dissolve into soft gelatin.

3) Thanks to science we now have a vide range of gelling agents (or hydrocolloids if you like) available for use in the kitchen. They can be used to create gels that are hard, soft, tough, elastic, brittle, hot and so on. We can also explain the crunchy and crispy texture of caramel which technically is classified as a “glass”. And we can also understand why ice cream has such a soft texture. This has to do with the very small size of the solid particles in ice cream (they’re typically < 1/10 mm). And a very nice way of achieving this is by making the ice cream with liquid nitrogen so it freezes very fast.

4) Science helps us understand what flavor is and how we perceive flavor using all our senses. For instance our memory plays an important role when our brain interprets a flavor. Your history of eating will influence how your next meal tastes! Sight is also extremely important. This can easily be demonstrated with some white wine and blue/red food coloring. If an unsuspecting subject is asked to describe “white” white wine and white wine colored with blue and red food coloring the descriptions will probably be typical of a white and red wine. Even if you know your are drinking a red colored white wine it is hard to believe that it is the same wine. The color of plates also influences our perception of flavor, so most people find eating from a blue plate unpleasant. Sound influences how we chew and a crunchy sound actually stops our jaws from closing to fast. If the chewing sound is played back with a 0.5 second delay it is almost impossible to eat! It has also been shown that the beat of music influences how fast guests chew, and apparently there is a restaurant in Australia that utilizes this. It has also long been known that touching sand paper or smooth skin can affect the texture of what we eat.

Andrew Dornenburg and Karen Page

Next up were Karen Page and Andrew Dornenburg who presented several of their bestselling books on cooking and flavor, including Culinary Artistry, What to Drink With What you Eat and The Flavor Bible. They made an important point that cooking is the sum of ingredients and technique/preparation. Food science and molecular gastronomy has provided a lot of information on technique, and the example par excellence is Harold McGee’s On food and cooking. However, for ingredients and flavor pairing in particular there hasn’t been any authoritative source available, forcing chefs to cross read a number of cookbooks when looking for new flavors and pairings. It was this realization that motivated them to write Culinary Artistry which is really a thesaurus of classic pairings. I can testify to it’s usefulness – and one example I’ve presented here on the blog are the cherry jams I made with pepper and other spices. Their most recent book, The Flavor Bible, is an updated and expanded edition of Culinary Artistry which includes and reflects the changes in consumer preference in the last decade.

I have emailed a litte with Karen and Andrew previously and got the opportunity to chat with them in the break. – We’re excited to learn about new approaches to flavor pairing, Karen let me know. Andrew added that they hadn’t had the opportunity yet to sample molecularily based food pairings before so they we’re looking forward to try this.

Lorenzo Cerretoni gave a presentation about flavor compounds in olive oil and how they can be paired with Mantis shrimp. After this informative, yet slightly less inspiring talk the Belgian chef Bart de Pooter presented his 4 taste model and how he uses this as a creative approach to create dishes and menus. One concept he promoted was to use ingredients from the same environment. A pheasant for instance walks on grass and eats maize and carrot, so naturally it could also be served together with flavors such as grass, maize and carrot. He also talked about physical and psychic needs and gave a slightly wrong explanation of the flavor pairing hypothesis.

In fact during the whole day there was no proper scientific explanation and discussion of the hypothesis, but Bernard Lahousse – project leader of The Flemish Primitives – did give a popularized version of the hypothesis and an account of how it all started. He explained how the Firmenich scientist François Benzi first stubled across a new way of pairing foods. At one of the Erice meetings he smelled jasmin in a garden and being a flavor scientist knew that this smell was mainly due to indole. Knowing that pork liver also smells of indole he got the idea that jasmine and pork liver might taste nice together – and they did! Bernard did show a picture of a GC-MS and explained how the University of Leuven has been doing food analyses and that only odors with concentrations above the odor thresholds were taken into account. Since he didn’t mention odor activity values (OAV) specifically I asked Bernard about that later and he confirmed that all the data in the food pairing website are now based on odor activity values. Since the first version of the food pairing web site was based on concentration and number of odorants I would expect that this would change some of the pairing charts and Bernard confirmed that this had been the case.

For the event 10 chefs where each given a food product from one of the sponsors. These products were then analyzed and a list of suggested pairings was provided to all the chefs. It should be mentioned that these were top class chefs, but I have to admit that watching chefs cook in real time actually became quite boring after a while, especially since some of the dishes needed seemingly endless steps of preparation. For most of the preparations presented it was also very unclear which of the ingredients used were picked based on the flavor pairing and which were picked by the chef to complete the dish. But the dishes looked absolutely wonderful! Too bad we couldn’t sample them I found two of the dishes particularily interesting (and I think the recipes will be made available on the web). Sang Hoon Degeimbre, chef at L’Air du temps had chosen Leffe beer, but instead of doing a flavor pairing he did a flavor substitution. To achieve this he used several ingredients (shown in the figure below) which when combined would resemble Leffe. In fact he made a dish completely without Leffe that was intented to taste like Leffe – it’s hard to say whether he succeeded or not, but the concept is very interesting! And what warmed the heart of an organic chemist was that Sang Hoon had equipped his kitchen with an erlenmeyer flask and used a glass syringe (looked like a luer lock Hamilton type syringe to me).

Figure from food pairing website. (C) Creax 2008.

The food pairing I found most interesting was the one with Oud Brugge (a cheese), coffee and vanilla. To bind these flavors together chef Gert de Mangeleer from Hertog Jan used potatoes. The surprising element of the dish was the coffee – he sprinkled his dish with freshly ground coffee. The vanilla was applied as a grape seed oil extract of natural vanilla – a nice example of how oil can be used for flavor extraction.

Ben sitting on the stage watching his own video.

- If Einstein had been a chef, he’d probably be like Ben Roche, said presenter Gene Bervoets. Ben is the pastry chef at Homaro Cantu’s Moto restaurant in Chicago – it’s the restaurant where you can chose between a 10 and 20 course menu and then start your meal by eating the menu. Ben’s topic was the Chicago style hot dog and he started off by presenting a mind map of this, resembling the food pairing charts at the food pairing website. Most of his presentation was in fact prerecorded video clips which were really entertaining! His deconstruction/construction of the Chicago style hot dog resulted in a chips & salsa. Evolution of flavor turned the hot dog into pop corn (“less calories than when eating the hot dog”) sprayed with hot dog infused oil and sprinkled with freeze dried garnish. Translation of culture resulted in a Mexican version of “encased meat” using chorizo and a soft taco shell. This dish was also made into a dessert version. Lastly the hot dog was transmogrified into a cartoon dessert version with strawberry sorbet, mint and pistacchios. Hopefully the videos will be made available, but in the mean time you can check out the video of Ben at Taste3.

Although we couldn’t sample the chef’s preparations there a number of tasting samples of different food pairings available in the breaks:

Of these my favorites were the almond thins with mozzarella, fig and ham, and the cooked beef with peanut, chocolate vinagrette and peas. The potato cream was nice, but was a litle overpowered by to much cheese.

Baked beef, cream of peas, peanuts and vinagrette with 70% chocolate

The program session was closed by Albert Adria, younger brother of Ferran Adria. He showed stunning pictures from his coming book natura, accompanied by videos showing how the different elements of the dishes are created. He also showed a short teaser documentary for A day at el Bulli.

Next year’s meeting will be devoted to tools, including one which was presented at the very end of the day with music, ballet dancers and fire works: a reinvention of Ford’s assembly line for mass production of haute cuisine.

Tri-2-cook has announced the foods to pair in the twelfth round of “They go really well together” (or TGRWT for short): apricot and chanterelle. More information on how to participate can be found in the announcement post. If the ingredients are out of season where you live, remember that you can use ingredients that are dried, canned or preserved. The heating and/or air exposure can of course alter the flavor composition, but it’s still worth giving it a try.

Regarding the chemistry behind this flavor pairing I’ve found the following. Based on quantitative measurements Greger and Schieberle identified 18 compounds with odor activity values (OAVs) greater than 1 in apricot (Prunus armeniaca). I have not been able to locate any studies of chanterelle (Cantharellus cibarius) with OAV values, but there are a couple of articles which list volatile compounds. Comparing these lists with the OAV data for apricots there is not much overlap. The only compounds which had an OAV > 1 in apricots and were also found in chanterelle are 1-octen-3-one (OAV in apricot = 55) and hexanal (OAV in apricot = 15) shown in the figure below.

It’s interesting to note that OAV studies often come with certain surprises regarding flavor compounds. As Greger and Schieberle point out in their abstract:

certain lactones, often associated with an apricot aroma note, such as gamma-undecalactone, gamma-nonalactone, and delta-decalactone, showed very low OAVs (<5) (...) Omission experiments indicated that previously unknown constituents of apricots, such as (E,Z)-2,6-nonadienal or (Z)-1,5-octadien-3-one, are key contributors to the apricot aroma.

Some compounds that are present at higher concentrations are less important because they have a high odor threshold, whereas other compounds which are present in minute quantities play important roles because we can detect them at very low concentrations. Once again this shows how important it is to use OAV values when looking for flavor pairings!

It’s time for a new round of food pairing! Robert of lamiacucina is hosting TGRWT #9 and the foods to pair this time are parmesan and cocoa. I’ve previously blogged about this combination and odor activity values (OAV) are available for both parmesan and cocoa. These are the molecules that significantly impact the odor of both parmesan and cocoa:

Don’t forget to check out Chad’s roundup of TGRWT #8 where several professional chefs participated!

8. Experiment!

Dare to experiment and try new ingredients and procedures. Do control experiments so you can compare results. When evaluating the outcome, be aware that your own opinions will be biased. Have a friend help you perform a blind comparison, or even better a triangle test to evaluate the outcome of your experiments.

In a scientific context, an experiment is a set of actions and observations performed in the context of solving a particular problem, in order to support or falsify a research hypothesis. In a kitchen context, the problem to solve would typically be related to taste, aroma, texture or color. And the required actions and observations would be cooking and eating.

An essential part of the scientific method is that new knowledge is gained when, based previous knowledge, an assumption is made and tested. In the kitchen, this is exactly what you do when you taste your concoctions repeatedly as you cook. And it is also what makes you an experienced cook, because you remember and learn from your previous successes and mistakes. It might sound very complicated, but here’s how it goes:

1) Observation: soup lacks flavor
2) Hypothesis: adding something with flavor might help
3) Experiment: add more spices
4) New observation: soup tastes more (or less)
5) Hypothesis is either supported (or rejected)

Of these steps, I think observation is the easiest. Coming up with a hypothesis however can sometimes be difficult. If you have lumps in your custard or a sauce that’s separating, it isn’t always easy to think of what to do. This is where books on popular food science and molecular gastronomy might help you.

Think outside the cook book! I mentioned in previous post that you should always question authorities and cook books. And even when you have a recipe that works, remember that it’s nothing more than a suggestion. For instance, it can be useful to know when to be sloppy and when to be accurate with measurements. The smaller amount you measure, the greater the precision should be. Let’s consider a hypothetical recipe that calls for 1000 g flour and 1 g of saffron. Whether you use 999 or 1001 g of flour makes no difference, but using 1 or 2 g of saffron will be quite noticable. A good rule of thumb is that you should measure to within +/- 10% of the given amount. But again, don’t follow this blindly. Experience will show when you can be even more sloppy.

Thinking of good experiments to do requires both creativity and experience, and there are many sources of inspiration. The molecular gastronomy movement has come up with a number of books and blogs which point towards new ingredients and procedures. There are several approaches to flavor pairing (i.e. a general one based on experience and a chemical one based on impact odorants). Further more there’s a lot of inspiration to get from regional cooking – also for molecular gastronomists! Lastly, I think considering not only the food but the whole atmosphere and the setting of the meal is important, because our senses are connected!

The best way to judge the outcome of a new procedure or ingredient is to compare it with the original. I’ve previously termed this “parallel cooking”. In scientific contexts it’s very common to do control experiments and I can’t see why this shouldn’t be done in the kitchen routinely. Im convinced that this could have saved us from many kitchen myths!

Once you’ve done your parallel cooking, you have to taste it. If you did the cooking, you’ll probably have an opnion or expectation that the new dish is better or worse than the original. The big problem here is that due to confirmation bias, if you know what you are eating, this will influence your perception of it. Therefore it’s crucial to do a blind tasting (or a double-blind tasting). Have friend help you label each dish with random three digit numbers (to avoid thinking about ranking) and give them to you. If the dishes can easily be recognized due to color, it’s important that the lights are turned down or that you are blindfolded. State which dish you prefer and have your friend reveal the identity of the dishes tasted.

A slightly more sophisticated test is the triangle test which is commonly used in the food industry. The tester is presented with three samples of which two are identical and the task is to pick the odd one out. Using statistics, it’s possible to evaluate the outcome of repeated tests. The number of correct assignments in a number of triangle tests required for you to be 95% sure there is a difference are given in the table below. Read more about simple difference tests here.

Bionomial distribution for a triangle test (p=1/3) at 0.05 probability level. A more extensive table can be found here.

It seems that this would be the ultimate way to determine whether or not there is a difference between pepsi and coke. It’s more than 50 years since the first experiments were conducted. The theory is simple, but in the real world things aren’t always that simple. Read the entertaining story about Fizzy logic.

There are several examples of experimental cooking out on the net, and I thought I’d share some of them with you as this might illustrate my ideas on the subject.

Many cooks have strong opinions about how garlic should be treated. Should it be minced, crushed or microplaned? And does this really influence the taste and aroma? Or does it only affect the degree of extraction and hence the intensity of the flavor? Dominic of Skillet Doux had a excellent post on this subject in 2006, Deconstructing garlic. The task was formulated as follows:

The subject of this experiment is the effect that various methods of breaking down garlic have on its flavor when used to prepare a dish. The hypothesis is that not only does mincing garlic create a different flavor than crushing it, but also that mincing is the preferred method for pasta sauces. Furthermore, the experiment is intended to determine if microplaning garlic achieves a character different from mincing or crushing.

In his conclusion, Dominic writes ” I was surprised to discover that the difference between the minced and crushed garlic sauces was even more significant than I had previously thought”. Check out his post to find out which kind of garlic treatment he prefers for his pasta sauces. As a side comment it can be mentioned that a group of researchers in 2007 studied the effect of cooking on garlics ability to inhibit aggregation of blood platelets. They found that crushing could reduce the loss of activity upon heating. But unfortunately they didn’t report anything about the flavor.

Other food bloggers have also adopted experimental cooking with emphasis on systematic and thorough testing. Examples include Chad’s experiments with gellan, konjac and iota/kappa carrageenans, Michael Chu’s parallell cooking of bacon and his eggplant test and Papin’s comparison of orange juices – to mention but a few! And I shouldn’t forget Dylan Stiles either whom I mentioned in part 5 of this series:

A challenge with aroma molecules is that they should remain intact during storage and not be released until cooking (or even better, until consumption). A example would be to install a Liebieg condenser over your pot. Dylan Stiles has explored this in his column Bench Monkey by placing a bag of ice on top of the lid. He claims that his roommates preferred the curry which has been cooked under “reflux conditions”. The study was performed in a double blind manner (which I will come back to in part 8 of this series).

An extreme example of the application of the scientific method to cooking appeared in the news last spring when the recipe for the ultimate bacon buttie was revealed by scientists from Leeds University. Commissioned by Danish Bacon, the study evaluated more than 700 variations of a bacon buttie. They even came up with a “formula” for the perfect bacon buttie and quantified the required crispiness and crunchiness. The news story was picked up by many news agencies, so although it wasn’t necessarily ground breaking science, at least it was clever marketing.

*

Check out my previous blogpost for an overview of the 10 tips for practical molecular gastronomy series. The collection of books (favorite, molecular gastronomy, aroma/taste, reference/technique, food chemistry, presentation/photography) and links (webresources, people/chefs/blogs, institutions, articles, audio/video) at khymos.org might also be of interest.

As a late (but just in time for the deadline) response to TGRWT #8 which was announced by Chadzilla in December last year – here is finally my write up on a recipe and a little on the background of this flavor combination which has become a classic in molecular gastronomy.

Heston Blumenthal introduced it around 2002 at The Fat Duck. It’s well worth reading what Heston wrote about this combination back then. He describes how salt can help bring out the flavor of many desserts. At one point he tried caviar and white chocolate – the effect was stunning. He then wanted to find out why this combination was so successful:

I gave some caviar and chocolate to François Benzi, who works for Firmenich, the flavourings and perfumes company based in Geneva. He was so surprised at the way that the caviar and chocolate melded together that he excused himself for half an hour while he tried to discover the reason behind the success of this union.

When he returned, the response was that both the chocolate and caviar contain high levels of amines. These are a group of proteins that have broken down from their amino acid state but not so far as to become ammonia. Amines contribute to the desirable flavours that we find in cooked meats and cheeses, among other things.

Some might object to using caviar but remember that there is no need to turn to sturgeon caviar as this species is endangered. I used caviar from Capelin which costs less than $4/€3 for a box of 50 g. As I have never tasted the “real” stuff I’m not the right person to judge about similarity or difference in aroma. And in case you also wondered about the terminology – roe is the fully ripe egg masses of fish whereas caviar refers to processed, salted roe. I decided to make a soufflé and based the recipe loosely on one of the soufflé recipes in my Larousse Gastronomique.

White chocolate soufflé with caviar
40 g white chocolate
30 g flour
1 dL milk
35 g caviar
3 eggs, separated
nutmeg

Melt chocolate on very low heat. Add 1/3 of the flour and stir, heating gently. Add a 1/3 of the milk and mix thoroughly. Add another 1/3 of the flour, then more milk and so on. Add finely ground nutmeg. Add 3 egg yolks and heat until right before the mixture sets (yeah – I admit – this is not very precise…). Then add the caviar. Beat egg whites stiff and fold them in. Pour into greased soufflé dish and bake at 220 °C for about 15 min.

Verdict: Aromas blend well together, but when eaten alone it’s perhaps a little bland. But I’m quite sure that it could be succesfully incorporated into a menu together with something acidic. The texture was nice, but the soufflé quickly falls together once it’s removed from the oven (I’ll have to post more on the chemistry of soufflés some other time – Hervé This has written a lot about this).

If you try to make this – note that white chocolate doesn’t behave excately like butter when you add the flour. It all got very thick, very fast – that’s why I started adding milk early. I also guess you have to be really careful when heating the whtie chocolate, but I didn’t do any stress tests here.

This is what the mix looks like before I folded in the egg whites.

For my first attempt at this recipe I used 20 g flour and 15 g caviar. The result was that the caviar sedimented before the soufflé had set, besides the fact that one could hardly taste the caviar at all. On my second attempt however, there was enough flour to keep the caviar suspended until the soufflé set. And one could actually also taste the caviar.

And now on to the chemistry behind:
I promised that I would come back with more information about the chemistry behind this pairing, but there isn’t very much information out there. There is one paper on aroma development in block-milk which used in the production of white chocolate. This paper lists a couple of volatiles, but only with their relative peak areas. Turning to caviar (or roe), there is a recent paper on flavor characterization of ripened cod roe, and this paper includes qualitative information about odor intensity.

Comparing the list of volatiles, the following volatiles which contribute substantially to the odor of ripened cod roe are also found in block milk (followed by odor thresholds in water, given in ppb, taken from this page):

2-butanone (50000 ppb)
2-methylbutanal (1 ppb)
3-methylbutanal (0.2-2 ppb)
pentanal (na)

Of these, the first has a high odor threshold, so it’s not likely to be an impact odorant in block-milk (and white chocolate). The methylbutanals however probably contribute to the overlapping aroma of roe and white chocolate. I didn’t find any threshold value for pentanal.

One group of compounds which was not mentioned in the paper on cod roe odor from 2004, but which was mentioned in a Russian paper from 1967 are amines (Golovnya: “Gas-chromatographic analysis of amines in volatile substances of salmon caviar”). Considering the fact that trimethylamine has a threshold in the range of 0.37-1.06 ppb, and that trimethylamine is found in block-milk suggests that it might contribute significantly to the odor of both white chocolate and roe. I guess the reason trimethylamine (and the whole range of other, closely related amines) is not found in the odor analysis in the 2004 paper has to do with the analytical method used.

The fact that amines are crucial is further supported by the Guardian article I quoted from in the beginning where Heston Blumenthal describes how he turned to François Benzi, a flavor chemist at Firmenich, to find out why white chocolate and caviar is such a good match. Benzi concludes that it is due to the presence of similar amines in white chocolate and caviar.

The long awaited website on foodpairings has now been launched, and they’ve also registred the corresponding blogspot name (which isn’t online yet as of today). The beautiful photos, great design and easy maneuvering makes it an excellent place to start if you are looking for some new and perhaps surprising combinations of foods. The foods are grouped into categories such as cocoa (?), dairy, fruits, meat, sea food and vegetables. One of the vegetables listed is cauliflower, and clicking it reveals that the topic of TGRWT #7 (caramelized cauliflower and cocoa) is one of several possible combinations. This is how it is displayed (an important detail is that the shorter the distance between the names, the more flavours they have in common):

(click to open the full picture from the foodpairing.be site)

As an added bonus interchangeable herbs and spices are also listed. This is how it works:

A food product has a specific flavour because of a combination of different flavours. Like basil taste like basil because of the combination of linalool, estragol, …. So if I want to reconstruct the basil flavour without using any basil, you have to search for a combination of other food products where one contains linalool (like coriander), one contains estragol (like tarragon),… So I can reconstruct basil by combining coriander, tarragon, cloves, laurel. The way to use it is to take from each branch of the plot one product and make a combination of those food products.

It should be noted that the proximity of the foods in the diagrams is based on the number of volatile compounds they have in common, not the actual key odorants. As I have elaborated on previously, pairings like these should preferably be based on odor activity values (OAV). Or to put it differently, if the volatiles shared by two foods are not the ones that actually contribute to the overall flavor, there is no reason to expect that they go well together from a chemical perspective (which is not to say that they won’t match, only that if they do, it is for some other reason). This is a limitation both of the foodpairing site, but of course also of the food blogging event They Go Really Well Together (or TGRWT) which I have initiated. Having said this, I still believe that the foodpairing site is an excellent place to start, especially if you like to improvise in the kitchen. I sincerely believe that the site will spark the creativity both of professional and amateur cooks (just like TGRWT already has)! I should add that the website is set up by the people behind Food for Design, so no wonder it looks so good!

Amrita of Le Petite Boulanger has announced the foods to pair for the fifth round: chocolate and meat! And in case you didn’t notice, Dennis has written an excellent summary of the mustard-mint recipes of round four.

I was not able to figure out which odorants actually formed the basis for the mustard-mint pairing (and perhaps there is none… as I’ve touched upon before, some of the data is hard to come at so it’s difficult to check all the entries of my compiled list). At M’s blog however you can find more info on the cold receptors which are triggered by both mint and mustard.

Fortunately data for chocolate and meat is available: odor activity values for cocoa and flavour dilution values for boiled beef and roasted beef. All flavour compounds were ranked and compounds given the same color coding as before. As you can see, there is considerable overlap between chocolate and meat.

(click to enlarge)

And here’s what the molecules that are found both in cocoa and beef look like. Notice that this pairing is dominated by furanones and pyrazines. The molecules are ranked according the the odor activity values in cocoa.