Jan 152013
Spread the love

The fascination of northern Americans and northern Europeans by “foreign and exotic food” is unbroken. Judging from my personal experience in Germany this fascination may have started with the Italian cuisine that was imported by Italian immigrants after the Second World War. Italian cuisine was regarded as more creative, as much better tasting and as superior in the quality of ingredients used.

If we try to prove or disprove the felt superiority of Italian cuisine that brings us down to the fundamental question of food science: How are different ingredients mixed in culturally different cuisines and how do they influence the taste of the resulting dish?

This question is much more complex than it seems, so complex that it intrigued a group of scientists of the “Center of complex network research” in Boston, recently. In their study “flavor network and the principles of food pairing” the authors found a surprising discrepancy between the number of potential recipes of ?1015 and the number of the actual applied ones, which they estimate to be 106, nine orders of magnitude less. Taken into account together with some frequently reoccurring combinations these findings imply the existence of principles that rule our choice of ingredient combinations.

But do these principles exist and if so, what are they and do they differ for regionally and culturally distant cuisines? Alas, food sensation is a complex experience, confronting the scientist with a variety of aspects that are hard to quantify, including colors, texture and temperature of the dish. That leaves the profile of chemical flavor compounds as the only reasonable starting point. However, the authors state that even this starting point is not unbiased due to lack of data on the concentration of flavor compounds in different ingredients.

Consequently, the authors created a rather qualitative but statistically demanding and tremendously complex flavor network, revealing connections between different ingredients. We can get an idea of the complexity of this network by noting that every ingredient contains 51 flavor compounds on average and that the total number of ingredients in the network is 381, while containing 1,021 flavor compounds. By this flavor network the authors tested the “food pairing hypothesis” which claims that ingredients sharing flavor compounds are more likely to taste well together.

Using more than 56,000 recipes from American and Korean web-based databases the number of ingredients per recipe was determined to be ?7 for Northern American and Northern European cuisine, while being ?11 for Southern European, Latin American and East Asian cuisine. Surprisingly “exotic” cuisines do not only use more ingredients, they also routinely mix ingredients that are independent in flavor compounds, thereby generating a more tasty food experience. For example the six most used North American ingredients were found to be butter, cane molasses, egg, milk, vanilla and wheat, which share many flavor compounds. East Asian cuisine however uses ginger, rice, scallion, sesame oil, soy bean and soy sauce, ingredients which are almost totally unrelated in flavor.

Does that mean that our western food taste is dull? For me I can only say that I really enjoy German food, be it the traditional stews or farmer-inspired brewery food. However, kind of mood-dependent, I equally enjoy Italian or for example Chinese cuisine. Although some statistics may say so, I would not label our food as dull. This may be especially true for the celebrated French cuisine, which may have escaped close scrutiny by the authors. Still some questions remain: Why do we prefer to eat similar ingredients together and why do other people avoid it? And why do we still crave for exotic food and do not change our traditional recipes?

Felix Spenkuch


Read more:

Y. Ahn, S. E. Ahnert, J. P. Bogro and A. Barab?si Scientific Reports 2011, 1, 196.

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>