Flavors and Ingredients : Analytical Method (part 1)

Note:

A disclaimer: picking pleasing flavors—or at least ones that invoke an emotional response or trigger a memory—is somewhere between an art and a science, so no scientific equation can capture the entire picture. Still, understanding how such a “flavor compatibility algorithm” would work can provide you with a way of organizing your thoughts on food, and for geeks, it’s fun to see how far one can take these sorts of things. If you really want to geek out and need a food project to work on, an open source version of this concept would be fun.

To start, we need a model of how to describe individual flavors, before considering how to combine them. Odors can be categorized in a few ways, most commonly either chemically or descriptively.

Chemical taxonomies classify compounds by their odors. Such a taxonomy is essentially a database of chemicals that each map to distinct flavor sensations. For example, Flavornet (http://www.flavornet.org), created by two researchers at Cornell (Acree and Arn), describes some 700+ chemical odorants detectable by the human nose. Listing compounds such as citronellyl valerate (smells like honey or rose; used in drinks, candies, and ice cream), the database is useful for generating certain flavors artificially, but not so useful outside of laboratory kitchens.

Descriptive taxonomies apply labels to odors as a way of classifying and grouping foods. For example, both lemon and orange are generally classified as “fruity/citrus.” Lacking the precision of a chemical taxonomy (the compound is either present or it isn’t), descriptive taxonomies suffer from the subjectivity of human judgment. Most of us would agree that a lemon smells “fruity/citrus,” but how much does a food like chocolate smell of the odors in celery? Not much, but certainly more than chocolate smells of fish.

Note:

The simplest descriptive taxonomy, from the 1950s by J. E. Amoore, proposes just seven primary odors: camphoric (like mothballs), ethereal (like cleaning fluid), floral (like roses), musky (like aftershave), pepperminty, pungent (like acetic acid in vinegar), and putrid (like rotten eggs).

One modern descriptive taxonomy can be found in the American Society for Testing and Materials’ Atlas of Odor Character Profiles – DS61, by Andrew Dravnieks. While you might not necessarily think of all of the terms included as pleasant, it’s certainly a diverse set, which is useful in thinking about smells. With 146 terms, Dravnieks’s list also provides enough granularity to begin to form a meaningful model for food flavors.

Dravnieks’s 146 odor terms, broken down into main categories, provide a good basis for thinking about odors. If you’re heading out on a date and want to impress, this list is a pretty good starting point for describing wines!

Note:

Another adjective classification system, Allured’s Perfumer’s Compendium, is used by the perfume industry, the fine folks responsible for the smells of products from laundry detergent to toothpaste. Think that new car smell is accidental? Trained employees smell the materials that go into the interior of a new car to make sure that it smells just right. (To quote The Matrix: “You think that’s air you’re breathing now?”) Allured’s taxonomy uses more descriptive and narrow scents—familiar items such as banana, peach, and pear—but also specific items like hyacinth, patchouli, and muguet (lily of the valley), making it less useful to the layperson.

Let’s start by defining a flavor profile as the weighted scores of a collection of terms in a classification system, such as Dravnieks’s 146 odor terms. For every term, imagine taking an item of food—say, a pear—and scoring it on a scale from 1 to 5, where a score of 1 indicates “doesn’t smell like it at all” and 5 is “the very definition of the word!” Given a pear, how much does it smell like a “heavy” odor? 1. Fruity? Maybe a 3? Fragrant? Say it’s a ripe pear, so 4.

The scoring is not asking if it is a compatible smell, just if the odor label describes the smell. Are the odors you sense in a pear (are the chemoreceptors that fire off in your nose) the same as when you smell other things that are considered fragrant? Given the weights of all of these odor descriptions for a pear, you can plot a graph (almost like a histogram) that you can then compare to similar graphs for other foods.

Some of Dravnieks’s odor terms associated with banana and pear, as scored by a few thousand Internet voters (taller bars indicate a larger degree of agreement between food and odor). These voters were not trained or verified to be familiar with the definitions of those odors; for these reasons, this graph should be treated as a conceptual demonstration only.

Given such a graph for each individual ingredient in a recipe, you can imagine a combined graph that describes the overall profile of a dish, showing all the “frequencies” present in the smells of each ingredient. Think of it like the various instruments that contribute to a piece of music: each has its own set of frequencies, and the combination of all the instruments makes up the overall song’s frequency distribution. When in tune, the frequencies line up and balance one another; when out of tune, the combination of sounds can be jarringly dissonant, even if each sounds fine individually. Of course, this music analogy isn’t a perfect fit for thinking about flavors: chemical changes brought about by cooking or by reactions between foods change the histogram, and the music analogy doesn’t cover other variables in foods, such as texture, weight, or mouth-feel.

Many chefs—often pros, but also non-pros who’ve been cooking for years—can imagine flavor combinations in their heads, doing something similar to this process mentally. Just as a composer imagines each voice and track in a piece of music, an experienced cook imagines the profile of the entire dish. Good cooks think about which notes are missing or are too soft and figure out what ingredients can be added to bring up those values.

What about achieving entirely new pairings, combinations that have no precedence in tradition? This same concept of matching up foods by their flavors can be done via the chemical taxonomy method, given enough time. The high end of the luxury restaurant industry spends an inordinate amount of time working on new flavor combinations, often with upward of two years spent working on a concept before it’s presented. Chef Heston Blumenthal of The Fat Duck (UK) maintains three distinct kitchens, one of which is devoted to laboratory work and is staffed by individuals holding both masters-level degrees in hard sciences like physics or chemistry and degrees from first-tier culinary institutions such as Le Cordon Bleu. Here is a partial list of pairings Chef Blumenthal has used: strawberry and coriander, snails and beetroot, chocolate and pink peppercorn, carrot and violet, pineapple and certain types of blue cheese, banana and parsley, harissa and dried apricot. Give them a try!

In addition to conducting their own private research, high-end chefs interested in creating new flavor combinations sometimes work with researchers at universities. Both Flavornet (http://www.flavornet.org) and FoodPairing (http://www.foodpairing.be) include such research in their sites. If you’re interested in exploring some of the chemical commonalities between ingredients, look at FoodPairing, which uses a chemical flavor database in order to suggest what ingredients to try together. (FoodPairing claims to be used by Chefs Heston Blumenthal and Ferran Adrià.)

Food Pairing diagrams for chocolate and chicken. Their database is based on chemical analysis, and it gives suggestions based on both chemical similarity and chemicals known to be complementary. GRAPHS USED BY PERMISSION OF BERNARD LAHOUSSE

The analytical approach tends to be very abstract. There’s little here that helps one select what ingredients to toss into a bowl together to make dinner. For this reason, these sorts of tools have yet to become particularly successful. This technique doesn’t generate recipes. While a set of odors might go together from an aromatic perspective, there are other variables in cooking that prevent mixing and matching various ingredients indiscriminately. For example, one ingredient might require cooking, while another might break down in high heat.

You can work around these constraints by separating the two ingredients into different components that are prepared separately and combined on the plate—say, a meat with a sauce. Or try using cooking methods that are, in essence, about conveying the perfume in food. Soups, ice creams, even soufflés: all are methods of transporting the flavors and aromas of ingredients without carrying the texture or volume of the original ingredient. A number of more recent, novel flavor pairings have used this solution. At the very least, you might find these types of tools a fun source of inspiration to try new things. Go experiment!