Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

FOOD INNOVATION ASIA CONFERENCE 2010

“INDIGENOUS FOOD RESEARCH AND

DEVELOPMENT TO GLOBAL MARKET”

KEYNOTE SPEAKERS

1 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

In Developing Indigenous Anti-aging Formulae in Taiwan James Swi-Bea Wu National Pingtung University, Taiwan

ABSTACT

The development of anti-aging formulae is an important task for health food industry. Certified anti-dementia or skin-care products are scarce on the market. There is tremendous demand from the consumer for these products once they are successfully developed. A projected entitled “Developing anti-aging formulae from therapeutical materials and food processing products” is therefore being carried out. The raw materials used in the experiments are those domestic food materials from plant origins, by-products in food processing and the officially approved therapeutic food materials which are supported by literatures or pre-tests to have the possibility being functional. The purpose of the project is to develop functional formulae for the counteraction against vascular dementia, Alzheimer’s disease or skin-aging. The product will preferably be in the similar configuration as a common food. However, products in capsules or tablets are also acceptable if necessary. The first phase is “Screening for Anti-aging Materials and their Components in Cell Models.” It is primarily the extraction, refining, and purification for the functional components, followed by cell study. The second phase is “Reconfirmation of the Anti-aging Components in Animal Models or 3D Human Skin Model.” It is to reconfirm the function of those active components picked up in the first phase using animal models or an equivalent. The third phase is “Process Development and Pilot Production.” The results in the preceding phases will be reviewed first, and then further investigation will be done in establishing an adequate process of production. The formulation, processing parameters and their effects on the physicochemical and sensory qualities of the product will be investigated. The product specifications will be set. The stability of functional components in processing and storage will be evaluated. Pilot production in the plant will be performed. The last phase will be to assess for the safety and the functionality of products in referring to Health Food Control Act in Taiwan.

2 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

New Zealand Indigenous Food Research David Pooch President, The New Zealand Institute of Food Science & Technology Inc.

ABSTRACT

New Zealand’s indigenous food research main focus is on marine products. The paper discusses three indigenous seafoods and research with commercial application; sea urchin Evechinus chloroticus, GreenshellTM mussels Perna canaliculus and Abalone Haliotis iris. It also discusses two native honeys; honeydew and manuka honey, and a polysaccharide extract from a tree fern Cyathea medullaris. The paper draws on research work done at New Zealand’s leading universities and research institutes. It gives the name and contact details of the scientists involved.

3 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Improve Food Security and Nutrition via Food Science and Technology Duo Li Department of Food Science & Nutrition, Zhejiang University, Hangzhou, China 310029 [email protected]

ABSTRACT

Food security is a global issue, especially natural seafood is harmed by deterioration ecosystem. In this paper, I would like to review what’s happened about seafood, and what the solution which has adapted in China is. Fish (finfish or shellfish) has been classified as healthy by health professionals despite containing contaminants, since fish is high in long-chain n-3 polyunsaturated fatty acids which have multiple beneficial health effects such as decreased risk of stroke via anti-thrombotic and vasodilative effects, increased heart rate variability, reducing serum triacylglycerol and blood pressure, anti-inflammatory activities, improving visual function, improving attention-deficit conditions/ hyperactivity disorder, schizophrenic and dementia; and may be effective in managing depression in adults. All these beneficial effects are thought to be mediated through altering cell membrane composition, fluidity, receptors and membrane-bound enzymes, gene expression and eicosanoid production. However, natural marine and freshwater fish populations are declining as a result of over-fishing, temperature and climate changes etc. To re-establish and maintain the fish population in China, fishing has been banned for 2-3 months during specified periods of the year, which differs depending on the area, since 1995. The fish population has recovered since implementation of these banned fishing periods, and thereby maintaining the sustainability and affordability of fish. Aquaculture products have had a significant contribution to China’s food system, with significant increase in output over the past few decades, from one million tons in 1978 to 32 million tons in 2007. Aquaculture fish represents a higher portion of total aquatic products compared with natural marine and freshwater fish, which has only been achieved in China, and this has contributed greatly to food and health security. China’s success in this area is a good model for other developing countries.

4 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Potential Health Attributes of Indigenous Australian Foods

Izabela Konczak

CSIRO Food and Nutritional Sciences; 11 Julius Avenue, Riverside Life Science Centre,

North Ryde, NSW 2113, AUSTRALIA

e-mail: [email protected]

ABSTRACT

Native Australian flora are represented by a large number of fruiting plants which are used for both food and medicine by the indigenous population. Cooper (2004) described 2440 of fruiting rainforest species growing in the tropical regions of Australia alone. In the 1990’s, commercialization of selected fruits have been undertaken (Ahmed & Johnson, 2000), and the subsequently obtained products are available from local growers and producers. However, information on the phytochemical composition of these fruits is limited. The knowledge of composition and potential physiological activities of the indigenous fruits may lead towards their utilization as health-promoting foods.

Within this study twelve native Australian fruits were evaluated for antioxidant capacity (Oxygen Radical Absorbance Capacity, ORAC-H assay; Ferric Reducing Antioxidant Power, FRAP assay) and levels of phenolic compounds (Folin-Ciocalteu assay and HPLC-DAD). The fruit of an evergreen rainforest conifer from the east coast of Australia, the Illawarra plum (Podocarpus elatus R. Br. ex Endl., Podocarpaceae), exhibited an enhanced antioxidant capacity in the reagent-based assays (1708.9 µMol Trolox Eq/gDW; ORAC-H assay) and 1502.5 µMol Fe+2/gDW; FRAP assay). These results were consistent with the Folin-Ciocalteu values (447.1 µMol GA Eq. g/DW), which indicated a high content of total phenolics. Compositional analysis by liquid chromatography-mass spectrometry (LC-MS) confirmed that the main phytochemicals of the fruit extract are phenolic compounds with the domination of cyanidin 3-glucoside, quercetin and luteolin/kaempferol glucosides. Traces of 3-glucosides of delphinidin and peonidin were also detected.

Potential pro-apoptotic anticancer and anti-inflammatory activities of polyphenolic-rich Illawarra plum extract (IPE), purified on XAD-16 column, were evaluated in cell culture-based experimental models.

IPE suppressed the proliferation of colorectal (HT-29) and gastric (AGS) adenocarcinomas, bladder carcinoma (BL-13) and human promyelocytic leukemia (HL-60) cells in a dose-dependent manner. The growth inhibition occurred with differing degrees of potency

5 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND between cell lines; the most pronounced effect was observed for the BL-13cell line (IC50 value: -1 - 76.8 µg.ml ), followed by AGS, HT-29 and HL-60 (IC50 values: 203.8, 242.2 and 337.0 µg.ml 1, respectively). The mechanism of growth suppression, studied in HL-60 cells, has been identified as apoptosis based on morphological assessment, DNA analysis, flow cytometry and induction of caspase-3 enzymes. IPE applied at a concentration of 750µg.ml-1 over 3 hours induced apoptosis in approximately 45% of cells. Enhanced activity of caspase-3 was detected after 3 hours of treatment and remained constant for 24 hours of the treatment. IPE suppressed the activity of cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) enzymes in RAW 264.7 cells in a dose-dependent manner. Accordingly, the levels of their products (nitric oxide (NO) and prostaglandin E2 (PGE2) respectively) were significantly reduced.

In conclusion, the major phytochemicals of the Illawarra plum are phenolic compounds; anthocyanins and flavanols. Fruit extract evaluated in vitro exhibited modulatory activity towards selected enzymes involved in the apoptosis of cancer cells and inflammation. Further studies have been undertaken with an objective to confirm these results in in vivo studies, which could subsequently lead towards the utilization of Illawarra plum in the development of health- promoting foods.

6 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

References

1. Cooper, W. 2004. Fruits of the Australian tropical forest. Nokomis Editions Pty Ltd., Melbourne, Victoria, Australia, p. 616. 2. Ahmed, A.K., & Johnson, K.A. 2000. Horticultural development of Australian native edible plants. Australian Journal of Botany, 48, 417-426. 3. Konczak, I., Zabaras, D., Xiao, D., Shapira, D., Lee, G. 2008. Screening native Australian fruits for health-promoting properties. Anti-proliferative and pro- apoptotic activity of Illawarra plum. Journal of Clinical Biochemistry and Nutrition. Vol. 43, Suppl.1: 543-547. 4. Tan, A.C, Konczak, I., Ramzan, I, Sze, D.M.Y. 2010. Antioxidant and cytoprotective activities of polyphenols from native Australian fruits. Manuscript in preparation. 5. Tan, A.C., Hou, D-X, Konczak, I., Tanigawa, S., Ramzan, I., Sze, DM.Y. 2010. Anti- inflammatory activity of native Australian fruit polyphenols in LPS-activated murine macrophages. Manuscript in preparation.

7 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Sensory and Consumer Methods for Measuring Acceptance and Quality of Indigenous Foods with Local and Non Local Consumers David H Lyon Director of Sensory Analysis, Firmenich SA, Switzerland

ABSTRACT

Many new products launched onto the market fail. The key reasons for failure are linked to poor analysis of the market for the product with target consumers, and inherent quality problems with the product. Sensory and consumer research methods offer tools to help us to understand products better, and what is right for target consumers, but this requires the application of the right tools in the right way to meet these objectives. This presentation will outline some of the key sensory methods available and their use at different stages in the product development cycle, especially the use of descriptive analysis to understand product characteristics and how this information can be linked to consumer preference data to uncover liking and disliking drivers. An application of descriptive method for online process quality control will also be described. For long term product success, especially for indigenous foods which encompass traditional and cultural values in their acceptance by consumers, increasingly we need to look beyond liking as offering main drivers for acceptability. This presentation will explore some of the new techniques being developed to measure the ‘emotional profile’ of products which may have relevance for determining acceptance of innovations in indigenous foods by consumers.

8 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Achieving Food Security through Reduction in Postharvest Losses of Indigenous Foods and Enhancing Manpower Training Opportunities: Canada-Ethiopia Experience Hosahalli Ramaswamy Professor, Department of Food Science, McGill University, Ste Anne de Bellevue, Quebec, Canada, Tel: 514-398-7919; e-mail: [email protected]

And

Tessema (Tess) Astatkie Professor of Statistics, Department of Engineering, Nova Scotia Agricultural College, Truro, NS, Canada, B2N 5E3, Phone: (902) 893-6694 [email protected] Presented at Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010 at BITEC, Bangkok, Thailand, as a part of ProPak Asia 2010.

ABSTRACT

It is generally estimated that over 25% of what we produce is wasted in the post-harvest chain and never reaches the consumer’s pot. Wastage of food represents an economic loss as well as food insecurity. The economic loss increases as the food moves down the food pipeline, because to the cost of food that is lost must be added the cost of handling, transportation, storage, etc., up to that stage. The first challenge is therefore to minimize the losses during the various post-harvest operations and keep the food as fresh as possible. In order to do so, one must first understand the various causes of post-harvest spoilage and the various factors which influence these causes, and then, secondly, use those post harvest conditions/operations which will result in extending the shelf-life of the produce. It should also be recognized that food (whether it is agricultural or animal product) is highly perishable. In spite of taking all precautions for post- harvest handling, packaging and transportation it is impossible to keep all available food in its farm-fresh state. The only way to make them available for later use in distant markets is to convert them into more stable form. Thus, the major emphasis of food processing is preservation or shelf-life extension by preventing undesirable changes in the wholesomeness, nutritive value and sensory qualities. This is done by controlling chemical, biochemical, physiological and microbiological activities. The food is most likely converted to other stable forms because it would be difficult to control the above. Various techniques can be used for this purpose:

Removal of heat: Since most of the biological, biochemical, physiological and microbial activities increase or decrease with temperature, temperature control (refrigeration) remains perhaps the most widely used method today to keep the food fresh. Since the spoilage activities are not completely stopped, it only provides temporary shelf-life extension. On the other hand, freezing terminates most of these activities (except chemical and some enzymatic changes), the freezing process can be expected to provide a long storage-life especially when the product is frozen and stored at temperatures below -18oC.

Addition of heat: Heat can preserve foods by inactivating the enzymes, destroying the microorganisms of both spoilage and public health concert. If they are appropriately packaged to prevent recontamination, the food can be stored for extended periods of time. Pasteurization processes only deal with mild heat aiming to provide short-term extension of shelf-life in combination with refrigeration while the commercial sterilization process (canning) produces shelf stable products. 9 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Removal of moisture: All life sustaining activities require the use of water available as free moisture in foods. By removing or reducing the moisture content, the food can be rendered stable since most of the spoilage activities are stopped or retarded. This is the principle used in such processing applications as drying, concentration or evaporation.

Controlling water activity: Water activity is a measure of the available moisture. A water activity level of 0.75 is considered a minimum required for most activities. Water can be bound to salts, sugars or other larger molecules which makes it unavailable. Such conditions may exist in dried products, intermediate moisture foods, concentrates, etc.

Addition of preservatives, sugar, salt, acid: These have specific roles in different products. Preservatives can selectively control the activities of microorganisms and enzymes. Sugar and salt can control the water activity. Some acids (example acetic acid – vinegar) have antimicrobial properties. Products such as jams, jellies, preserves, pickles, bottled beverages, etc. make use of such concepts.

Other techniques: Other techniques such as irradiation, exposure to ultraviolet light, high intensity pulsed light, pulsed electric field, high pressure, etc. have different mechanisms for controlling the spoilage activity in foods and have been used for shelf-life extension.

Secondary objectives: There are secondary objectives for food processing as well. They include: diversification of products to provide variety, taste, nutrition, etc., to provide end use convenience, to facilitate marketing, to prepare food ingredients through isolation or synthesis, and to produce non-conventional foods.

These are science based techniques available for extending the shelf-life of foods. Education programs make the difference in terms of training people for implementation of these concepts. Such programs also need to be established on a sound basis for successful implementation of the science based technologies; and with a sustainable manpower to handle them and create knowledge networks for problem solving. Canadian International Development Agency’s UPCD program provides a platform for creation of such programs linking Canadian institutions from where knowledge can be transferred with institutions in developing countries. This presentation will highlight one such successful program effectively creating a cooperative link between two Canadian Institutions (Nova Scotia Agricultural College and McGill University) and Jimma University in Ethiopia. Resources have been established to launch gender and environment sensitive BSc and MSc programs in post-harvest management, staff upgrades through short term and long term trainings, enhancement of laboratory equipment, and library and information technology upgrades as well as creating satellite stations for linking with the user community

10 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Village Level Processing: Empowerment through Enterprise Skills Development in Cambodia, Lao PDR, Myanmar and Viet Nam Alastair Hicks Food and Agriculture Organization of the United Nations, Regional Office for Asia and the Pacific, Thailand

ABSTRACT

As in many parts of the developing world, the agriculture sector is the primary source of income for most of the working population and there is an urgent need to diversify and create additional on and off-farm enterprises; as the income to rural households is regularly disrupted often falling below expectations, and the rural poor in the countries must deal with the vagaries of agriculture without the luxury of an economic safety net. Across the four CLMV countries of Cambodia, Laos, Myanmar and Viet Nam there is seen a clear need for enterprise development, marketing skills and means of coordinating households/communities into cooperative action. The capability in these countries to develop and operate rural businesses needs strengthening. In Cambodia and Laos PDR, there is a strong need for rural industrialization and crop diversification programmes, now starting up. In Myanmar a sound entrepreneurial spirit exists and there is a demand for necessary skill sets to take full advantage of rural income diversification. A key feature of Viet Nam’s rural industry base is the presence of Occupational Community Villages – (OCVs) in which households work closely together to produce one product. This collective approach allows groups to develop some degree of economy of scale and greatly improves their sustainability and efficiency. This paper highlights the role of practical training, providing case studies in rural income generation and enterprise development.

11 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

An Overview on Safety Concern of Some Indigenous Foods Songsak Srianujata, Ph.D. Institue of Nutrition, Mahidol University and Knowledge Network Institute of Thailand

ABSTRACT

Indigenous foods are the food that people originally use for consumption in their community for long time. It may be slightly or some modified but still keeps the essence of the origin. There are many kinds and items of indigenous foods around the world. Some are having similarity and some are totally different. Those which are quite similar are parts of plants, insects and small local animals. The food may be eaten as raw and some are process in different ways. For example, raw fish has been eaten in northeast Thailand for centuries, fresh water fishes are fermented (called Pla Ra) and kept for consumption all year round, even some people who are not used to it will say that it is very strong smell. Fish sauce (Nam Pla) is one of the condiment that has been used in many countries as Thailand (as a table sauce and for cooking in almost all Thai dishes in all Thai kitchen and restaurants) Loas, Cambodia China and some countries in Europe. Some of the indigenous foods are consumed wider and spread into different community and country. These mostly due to the migration and traveling of people in and out of the local community where the foods originated. Raw fish, sushi, is eaten wider in the world other than in Japan. Plara can be found in western countries as in the U.S.A., imported from Thailand. However, with the development and changes around the world, the utilization and the consumption of the indigenous foods lead to some safety problems. Canned bamboo shoot is one of the very good examples that caused serious hazard to people in some area of Thailand due to the botulism. The problem occurred because the traditional canning process was not appropriate then allowed the clostridium botulinum to grow and produces botulin or botulinun toxin whish is the very potent neurotoxin. Plara has been studied and it was shown that some items were added too much of Nitrate which can be dangerous when it reacted with amines in fish to form dangerous compounds as N-nitroso compounds that can cause cancer. The consumption of raw fishes in the northeast Thailand is the main cause of liver fluke infestation that may lead to liver bile duct cancer later on. These are only some examples that the consumption of indigenous foods posted food safety problems. In conclusion, the food safety of indigenous foods comes from about three main reasons as. 1) The change in the environment and the life style of the world population that causes the contamination of the indigenous foods. 2) The economic development and the knowledge of the food processing is not known to the indigenous people in the preparation and preservation of the foods, which may lead to hazardous condition in the food consumed. 3) The food habit of the consumers that do not change with the state of the development and the environment surrounding the area of the production and preparation of the indigenous foods.

12 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Application of Five Senses Communication Model to Design the Palatability of Green Tea Beverage Yasuyuki Sagara Food Kansei Communications, Corp. (FKC) e-mail address: sagara@foodkansei .or.jp

ABSTACT

The Food “Kansei” Model has been developed to formulate the causal relationships between the analyzed characteristics and perceived quality of food products. It was applied to correlate the physicochemical properties with the food perception, preference and pleasantness quantitatively. The model was applied to the practical design of flavor- and taste-active components in green tea beverages. Instrumental analyses using gas chromatography and high performance liquid chromatography were carried out for test samples or green tea beverage. Sensory evaluation was also performed by three consumer groups belonging to different social categories. The perception of tastes and flavors was summarized in new sets of uncorrelated sensory factors by applying principal component analysis to sensory data. Preferences were indicated by regression coefficients estimated to correlate hedonic ratings with the sensory factors. The relationships among the green tea components and the sensory factors were analyzed by the use of an artificial neural network. The preferable intensities of sensory factors and the concentration ratio of the components were estimated for each consumer group. Relatively high levels or roasted and sweet flavor components were found to cause higher smoothness that was preferred by female students. In contrast, female and male office workers preferred the samples containing a greater amount of flowery flavor and total amino acid, which offered greater thickness and fragrance. It was demonstrated that the design and product information of plastic bottle increased dramatically the palatability obtained from sensory evaluation.

Keywords: Kansei model, green tea, sensory evaluation, product information

13 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Advances in R&D for Halal Indigenous Food Yaakob B. Che Man, Ph.D Halal Products Research Institute (HPRI) Universiti Putra Malaysia 43400 UPM Serdang, Selangor, Malaysia

ABSTRACT

Global halal food market is estimated at USD 580 billion while the total halal market is over USD 2.1 trillion annually. The increasing demand for halal food is expected to continue in the coming years. There are numerous types of indigenous food in different countries, and the halal status is normally based on the culture and religious belief of the local community. In Malaysia, indigenous food such as dodol, baulu, rendang, serunding, satay, keropok etc. are usually halal. However, with the advent of science and technology, even the indigenous food had undergone many processes during their production. This trend has raised concerns among Muslim consumers regarding the halal status of processed food, including indigenous food they purchased. This is especially critical when the source of the components in food are of animal origin. These food items are only halal if the entire value chain is processed and handled in accordance to halal standards or guidelines. Therefore, reliable products analysis are crucial for verification and certification of halal food. This paper also highlights the advances in the R & D of analytical techniques for detection for the presence of non-halal components in halal indigenous food.

14 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

15 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

16 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

17 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

18 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

19 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

20 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

21 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

22 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

23 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

24 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

25 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

26 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

27 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

28 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

29 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

30 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

31 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

32 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

New Zealand Indigenous Food Research David Pooch President, The New Zealand Institute of Food Science & Technology Inc. *Corresponding e-mail address: [email protected]

ABSTACT

Introduction

New Zealand was one of the last places in the world to be discovered by mankind. It is a long way from anywhere else and is surrounded by cold ocean. When the indigenous people, the Maori, arrived less than a thousand years ago, they found a country just over half the size of Thailand and no people. These people came south from the tropics in canoes; they could not bring much with them. They brought sweet potato but it did not grow in the colder parts of the country. They found plenty of marine and birdlife, but no animals, not even snakes and there were no grains they could cultivate As a result Maori were mostly coastal people: New Zealand has 15,000 km of coastline to get food from, five times greater than Thailand’s 3,000km coastline. Even today, my country’s indigenous food research focuses on marine products. Today I talk about three seafood products sea urchin, mussels and abalone. In addition I talk about two native honeys and a polysaccharide extract from a tree fern.

Sea urchin

In New Zealand, sea urchins (kina in the Maori language) Evechinus chloroticus, are commonly found growing wild around our rocky coasts. I can gather them myself about 5 minutes walk from my house. People gather them for home consumption and there is limited small scale processing of the roe. In Japan though, sea urchin rose is a delicacy and commands high prices. So, sea urchins have low value in New Zealand and high value in Tokyo. This situation has good longer term potential for business. Some early trial exports were disastrous because of inconsistent products. Research was done at the university of Otago to bridge the gap. I rather liked the common sense design of their research on sea urchin roe, which was headed up by their Pat Silcock and Phil Bremer.

Define the questions that you want your research to answer •What are the sensory properties and how do they vary? •How do the sensory properties relate to composition? •How does handling, processing and storage impact upon the final quality? •How does this relate to what the market wants?

33 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Define what the market wants. In this case the market discriminates on the basis of origin, colour, flavour and size.

Understand the raw material. Gather information on yield, composition, colour and flavour of the roe. Analyse roe for a wide range of parameters. Carry out descriptive sensory analysis with trained panels. Look for correlations.

Conduct trials. Lab and field based feeding trials; the effect of processing, the effect of gender. Measure physical properties, composition and sensory properties.

Analyse the results and come to conclusions. They found the sea urchin’s seasonal cycle has a major impact on roe flavour. They identified the components responsible for bitter taste. They concluded that the keys for this industry to be successful are selective harvesting, aquaculture and smart processing. And one finding … the female sea urchins are sweeter than the males.

Since 2002, NIWA has worked to get consistent yields of high quality roe from wild- harvested sea urchin. Their scientist, Mr Philip Heath, has established that although roe content of wild sea urchin varies seasonally, the amount of roe can be enhanced from 2-3% to 12-16% of total live weight in a 10-week period regardless of season. Specialist diets developed at NIWA are used to bulk up the roe, followed by feeding seaweeds to help with flavour and colour. A number of specialist tasting panels and sophisticated colour analysis software has allowed them to define flavour and colour and assess the impact of diet and culture conditions on these attributes. Industry-linked research has shown that transferring sea urchin from low food areas to high food areas can also improve roe quality.

The question on sea urchin What low value indigenous materials do you have in Thailand that you can research and make to a high value processed product.

Green shell mussels

Introduction

GreenshellTM mussels (kuku, kutai in the Maori language ) Perna canaliculus, are native to New Zealand. They have, as the name suggests, emerald green markings on the shell. The flesh of the female is a soft orange colour, and the male’s flesh is a pale creamy colour. The New Zealand green lipped or green shell™ mussel has become a major aquaculture success story. Wild mussels are no longer commercially harvested and in 2008 the country had over 1,000 farms covering 11,500 of shallow coastal water. This yielded 233,000 tonnes of mussels valued at 5,000m Baht. Mussels are New Zealand’s biggest aquaculture industry and our biggest single species of marine product exports.

Mussel farming

A mussel farm consists of a grid of floating buoys that are roped together and also anchored to the bottom of the sea. The mussels themselves grow on lines that hang down from

34 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND the buoys. The farms are always in sheltered water and not too deep. When the mussels reach a desired size, the rope is lifted and the mussels stripped from it. Trials have now been successfully completed on deep-water mussel farming. Mussel production and area of farms is expected to increase in the future and aquaculture is predicted to become a 24 billion baht industry by 2025. Processed added value retail products include natural and flavoured half shell products, mussel meat, marinaded product and shelf stable mussel chowder packed in a pouch. A green mussel extract of lipids is also made which is claimed to have anti-inflammatory benefits.

What research is being done? Mussel research is being conducted at several sites; Plant and Food Research Ltd in both Auckland and Nelson, the Cawthron Institute in Nelson and the University of Auckland.

Selective breeding programme Currently the industry relies on the collection from the wild of spat (juvenile mussels) that are used to seed the ropes that the mussels will grow on. The risks with this practice are obvious; for example irregular supply, mortality and variability in attributes of the adult mussels. Nick King of the Cawthron Institute in Nelson is working on selective breeding of mussels and mussel hatcheries. Economic prospects are bright. To give a comparison, similar research on the Pacific oyster (Crassostrea gigas) by the Cawthron institute has seen the recent implementation of a large scale hatchery that is hoped to triple New Zealand’s current oyster export turnover. At the same institute Aditya Kesacodi-Watson is searching for probiotics: microbes with a positive effect on shellfish larvae. The aim is to improve the larval rearing process in hatcheries.

High pressure processing Sravani Gupta of University of Auckland is conducting research at Plant and Food Research on shucking (opening) mussels using high pressure processing. The conventional way of opening the shell of the mussel is by cooking, as you will know from eating fresh them at restaurants. This ongoing research in Auckland aims to see if high pressure processing is a good alternative. They have found that it works. High pressure processing improves the yield of meat, allows easy removal of the shell and inactivates both pathogenic bacteria like Listeria and non- pathogenic bacteria. At the same time there are effects on autolytic enzymes. It decreases peptidase activity, which improves shelf life but may increase lipase activity, which is bad for oxidative rancidity. Ongoing work also includes predictive kinetic modelling to choose the best optimal conditions for using high pressure processing on GreenshellTM mussels Listeria Graham Fletcher of Plant and Food Research Auckland, is one of New Zealand’s senior sea food microbiologists. He researched control of Listeria in mussel processing plants. This is important research because high value food products must have high consumer confidence and must be able to prove a high standard of food safety. The bacteria Listeria can cause serious food poisoning and some markets enforce a zero tolerance policy for Listeria monocytogenes in frozen shucked mussels. He found firstly that Listeria is generally absent in mussels arriving at the factory so Listeria brought in is the problem. They tested the effectiveness of 24 different commercial sanitisers on Listeria monocytogenes grown in suspension. All sanitisers bar one were effective against the 20 strains that were selected. They tested the same organism grown in biofilm. They found this was generally more resistant to sanitizers and the most effective ones were peroxyacetic acid, chlorine dioxide and acidified sodium chlorite. All sanitisers tested were

35 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND less effective in the presence of protein. Rapid testing kits for Listeria were also evaluated. All the commercially available products gave comparable results.

Sensory Andrew Boey of Auckland University is making an in depth study on sensory aspects of Greenshell™ mussels. In essence, what makes a mussel good to eat? This work is being done in order to determine specific consumer-desirable characteristics and then be able to breed for them. This research is expected to result in new and significant findings. This is the first time instrumental texture analysis has been done on mussels. It is a complicated material to work on because of the mussel anatomy. For example, its abductor mussel is tough. Its lips are chewy. Its flesh is very soft. Rupture measures how much force is needed to pass through the component Shearing is like a biting action done and the whole mussel was tested Compression is more like chewing and the lips and tongue were tested The numeric results from the texture profile analysis tests are being compared with information gained from taste panels. The taste panel focus groups tested mussels from different locations, different cooking methods, condition (maturity), gender and time in storage. When the focus group work is complete it will be possible to correlate results with the texture analysis machine Consumers preferred the colour of the orange female mussel much more than the cream male mussel. And, in terms of value addition, freshness, taste and texture were much more important than price.

The question on mussels Does Thailand have the interest or capacity to invest in large scale mussel farming?

Abalone There are three species of indigenous New Zealand abalone (paua in the Maori language); the common black foot Haliotis iris, the yellow foot and the white foot. Uniquely, New Zealand abalone has a most beautiful iridescent shell, which is prized in the jewelry trade. Tourists often buy a polished shell or item of jewelry made from the shell. New Zealand fishermen earn 50 to 250 Baht per kg for the shell and 1200 baht to as much as 3,000 Baht per kg for the meat. Pearl farming is also underway with Eyris Blue Pearls the key player in both production and marketing. About 1,000 tonnes of wild abalone are caught each year while farmed abalone amounts to about 10% of this amount. There are a number of small abalone farms scattered around New Zealand, some focusing on producing juveniles for enhancement of the wild fishery. The market leader in abalone farming is OceaNZ Blue Ltd. At Bream Bay in northern New Zealand, their business includes the spawning, settling, on-growing, processing and marketing of premium quality abalone products. New Zealand farmed abalones are sold predominantly at 75 to 95 mm shell length, at around 4 years of age. The farmed abalone grows an attractive blue outer shell. This is often used in the presentation of an abalone dish. Most are cryogenically frozen whole in the shell or sold live. The meat is high in protein, low in fat but a good source of omega-3’s and is exceptionally rich in taurine. Both OceaNZ Blue and the National Institute of Water and Atmospheric Research (NIWA) are studying abalone. Jane Symonds of NIWA is working on brood stock genetics. This incorporates both family line breeding and mass selection with DNA fingerprinting. NIWA is also studying the role of water quality on growth, and particularly the effects of dissolved oxygen, water motion and temperature, as these can severely impact growth rates, especially in larger animals. They are also looking at sperm cryo-preservation, genetic markers and wild stock enhancement. The selective breeding research aims to greatly increase profitability by improving key production traits. For example, they hope to cut 6 months off the current growing time of 3-4 years. Shellfish aquaculture is probably the only major primary production industry in NZ that produces young from wild parents rather than from high performing domesticated broodstock. The potential for big gains is already evident, but it is a long-term project when generation time 36 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND is so long. Abalone farming has enormous potential for growth in New Zealand and this applied research will greatly assist the industry.

Honey New Zealand is very largely consists of grassland and forest. As a result the honey industry is well advanced and we get good prices for exported product. Originally we made honey from mixed flora. It became apparent that higher prices could be obtained from named floral sources. For example, premium prices are obtained from high altitude honeys, and honey made from honeydew, which is the sugary secretion produced by a scale insect living on the bark of indigenous beech forest.

Honeydew Honeydew is a classification of honey that refers to honey produced by honeybees collecting nectar that is exuded from another insect such as an aphid or scale insect. It is quite common in a number of countries and the best known is honeydew from the Black Forest in Germany. In the South Island we have large forests of black beech (Nothofagus solandri). Two different honeydew insects live on them. Ultracoelostoma assimile favours the upper branches and twigs while Ultracoelostoma brittini inhabits the trunks and larger branches. The black colour of trees and plants with a honeydew source is due to the growth of Capnodium, a black sooty mould (fungus)which feeds on the surplus nectar exuding over the plant and sometimes even the ground. Particles of this fungus are typically found in honeydew and are used as a part of the identification as honeydew. The insects suck sap from the host plant and exude a sweet sticky nectar: essentially a slightly modified sap. This is then collected by honeybees as a nectar source and is "ripened" into honey. This pathway is quite different from that of normal flower honeys and the composition is different too. Honeydews are normally higher in fructose and maltose and lower in glucose than other honeys. There is additional mineral content and the oligosaccharide content is higher too. Oligosaccharides are helpful in maintaining and promoting beneficial bacteria in the gut, particularly after treatment with antibiotics. Here, there is lots of potential for both researchers and marketing people to promote special benefits.

The question on honeydew Do any trees in Thailand produce honeydew? There may be an opprtunity there

Honey from Tea tree Tea tree (manuka in the Maori language) Leptospermum Scoparium, is a common native tree in New Zealand and is not cultivated. It was originally called by that name because its small leaves, when seen lying on the forest floor, look like black tea. It does NOT make a nice tea. Manuka honey has been long prized for its rich smell and strong flavour and has always commanded a higher price. Research done by Professor Molan and others at the University of Waikato (date?) showed that some manuka honey showed special properties in terms of antiseptic use. Today, what is now called ‘active’ manuka honey commands a much higher price than any other honey in the country. Honey is one of the oldest known medicines and continues to be used up to present times in folk medicine. Interestingly, the ancient physicians were aware of differences in the therapeutic value of different honeys: Aristotle (384-322 BC), discussing differences in honeys, referred to pale honey being "good as a salve for sore eyes and wounds. Its use has been "rediscovered" in later times by the medical profession, especially for dressing wounds. Any honey can be expected to suppress infection in wounds because of its high sugar content and its osmotic effect. In addition, all honeys have an antibacterial activity, due primarily to hydrogen peroxide formed in a "slow-release" manner by the enzyme glucose oxidase, which 37 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND is present in honey. The difference in potency of antibacterial activity found among the different honeys is very big. Some honeys can be diluted more than 100-fold and still halt the growth of bacteria. Determining the level of antibacterial activity in honey is a simple procedure in a microbiology laboratory. Some manuka honey, called ‘Active Manuaka Honey’, additionally has non-peroxide, antibacterial activity. The anti-bacterial components come from the nectar and research is currently identifying them.This makes manuka honey about twice as effective as other honey against Eschericihia coli and Enterococci, common causes of infection in wounds. It is much more effective than other honey against Helicobacter pylori, a common cause of peptic ulcers. Professor Molan also found that manuka honey can be sterilised by gamma irradiation without loss of activity. Sterilised Active Manuka Honey is now commercially available for use in manufactured wound dressings. Clinical and histological studies showed the honey has an anti-inflammatory action; this might be due to antioxidants. The honey also stimulates growth of new blood capillaries, granulation tissue and epithelial cells. The components responsible for these effects have not been identified but growth may be due to nutrients in the honey, as nutrification of wounds is known to hasten the healing process.

The question on honey Commodity honey gets commodity prices. Special-variety honey gets higher prices. Functional/neutraceutical honey gets higher pries again and prices are even higher when a food finds it way into a medical use. We are prepared to apy more for medical benefits than gastronomic benefits. What commodity priced foods do you have in Thailand that you can identify and research a unique flavour or variety or benefit and market at a higher price?

Tree fern polysaccharide The indigenous New Zealand black tree fern (mamaku in the Maori language) Cyathea medullaris, is our largest tree fern and grows to 20 metres high. The Maori people used an extract from the pith of the fern for treating wounds and diarrhea but they did not use it for food. Professor Kelvin Goh of Massey University worked with an interesting polysaccharide he found in its fronds. It showed complex rheological behaviour including rod climbing and self- siphoning. It is early days yet but there is potential for use as a food thickener and in the cosmetics industry. Current research involves purifying the extract, determining the size and conformation of the large molecular weight fraction and the effects of solvents on rheological properties This material had never been characterised before so it is breaking new ground.

The question on tree ferns A year or two ago I went walking in the rain forest north of Chiang Mai and I saw you have tree ferns too. Have they have been researched for interesting properties?

Finally The science I have talked about this morning is in the public domain. The scientists who did this work were happy to talk to me about it and they will be happy to talk to you about it also. I have given their names and email addresses. You are welcome to contact them; true science is international; this conference is international too. I wish you all success and thank you for inviting me.

38 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Application of Five Senses Communication Model to Design The Palatability of Green Tea Beverage Yasuyuki Sagara Food Kansei Communications, Corp. (FKC) *Corresponding e-mail address: sagara@foodkansei .or.jp

ABSTACT

Introduction

Kansei engineering was founded 30 years ago as an ergonomics and consumer-oriented technology for new product development (Nagamachi, 2002). Recently, the specified paradigm and methodology of Kansei engineering related to foods has been proposed as "Food Kansei Engineering" by Sagara (2001). The measurement methods of intrinsic food attributes concerning food palatability have been proposed for each of the five senses, however, the approach of psychophysical relationships among physicochemical food characteristics and perceived quality is scarce for complex food products. Thus the development of models has been desired to correlate these physicochemical attributes and food perception quantitatively based on the causal association. Sijtsema gave an overview of conceptual models about factors influencing food choice, acceptance or preferences (Sijtsema et al., 2002). The need for food in relation to its function in groups of people is studied by anthropologists, psychologists, sociologists, marketers and economists who place food in a societal perspective. In psychology, most literature about food that deals with eating concerns quantity: regulation of food intake and disorders related to food intake, such as obesity, anorexia nervosa, and bulimia. The cultural aspects of food and the function of food in a group were first studied by anthropologists. Recently, sociologists also started to study the differences of food use and choice in different countries, though, in those studies, research that analyzes the linkage of physicochemical characteristics to quality attributes is scarce. 39 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

The psychophysical literature contributes to the fundamental insight into the relationship between physical product characteristics and consumer perceptions (Steenkamp and Van Trijp, 1996). A serious limitation of traditional psychophysical research, however, is that the overwhelming majority of these studies involve at most a couple of narrowly defined physicochemical characteristics. Therefore, the psychophysical functions obtained from these studies have limited practical relevance for complex foods. Since the tastes and flavors are dependent upon a number of natural components, the quality prediction in food processing is difficult, especially from the viewpoints of consumer-oriented design. Quality recognition in humans is thought to be generally unsupervised pattern recognition. The sensory attributes can originate through the interactions of complex nonlinear physical and chemical properties that can individually be quantified only by instruments. Artificial neural network (ANN) was developed almost four decades ago as a tool that has the ability to handle information-processing problems (Ni and Gunasekaran, 1998). ANN has recently gained more attention because of advanced technology in computer hardware and software (Bomio, 1998). ANN was capable to generalize and deduce essential characteristics from inputs containing chaotic data, and handle complex nonlinear, and even unknown, relationships. The ANN has been used as an effective method for constructing a highly accurate model of relationships between instrumental measurements and sensory evaluations (Wilkinson and Yuksel, 1997; Ni and Gunasekaran, 1998; Bomio, 1998; Wailzer et al., 2001; Boccorh and Paterson, 2002; Tominaga, 2002). The ANN models could be used to predict the intensities of tastes and flavors immediately, however, these represent little about the interactions among perceived quality and preference for the flavors. On perceptual levels, the models generated in these studies cannot be classified as causative in terms of the determinants influencing food perception and preferences. It seems that it is important for consumer-oriented product development to integrate the preferences for tastes and flavors into the model, which are considered to be different among individuals. Since food perception is an abstract construct and is of a multi-dimensional nature, the development of mensuration scales is a question of vital importance (Jover et a1., 2004). Many authors recognize the interest of scales directed to a product, suggesting that quality is specific to a single good or service. The fact that the intrinsic attributes are specific to each product means that choosing one specific product is recommendable. Recently, many types of tea beverages have been produced and consumed in Japan, including green tea, oolong tea, black tea and scented tea (Kobayashi, 1995). In the marketing of tea beverages, the development of new methodologies to design taste and flavor adapted to consumer preference is desired for creating new demand and an advance in sales performance. We have chosen green tea beverage as a specific case, for the reasons mentioned and due to the fact that it is one of the products which is enjoyed in household consumption allowing different preferences among individuals. The objectives of this study are (a) to develop the Food Kansei Model to correlate the physicochemical properties of food with the food perception, preference and pleasantness quantitatively, and (b) to predict preferable taste and flavor design of green tea beverages for consumer groups from varying social backgrounds.

Theoretical

1. Factors affecting food preference and pleasantness In the study of factors influencing food preferences by Randall and Sanjur, three groups of characteristics: food, individual, and environmental were clearly divided (Sijtsema et al., 2002). The characteristics of food deal with the taste, appearance, texture, cost, food type, methods of preparation, form, seasoning, and food combinations. On the other hand, the characteristics of the individual are represented in terms of age, sex, education, income, nutritional knowledge, cooking skills and creativity, and attitudes toward health as well as the 40 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND role of food. The characteristics of environment refer to season, employment, mobility, degree of urbanization, size of household and stage of families. The variables were selected based on the criteria of the frequency with which they were isolated in past studies and/or the strength of the proposed association with food preference (Sijtsema et al., 2002). The variables in the three characteristics, for which operation or measurement can be available in the process of product development, should be incorporated into the present model. Except for the characteristics of the environment because of their impossibility to control, the other two causal factors of pleasantness could be classified broadly into three categories, viz., perceptual, appetitive and cognitive factors (Fig. 1), based on the classification of factors relating to the mechanism of affective response (Fiske, 1982; Ito, 1994). Perceptual factor represents external stimuli caused by physicochemical characteristics of food such as components, structure, temperature and appearance. They contribute materially to sensory pleasantness by the perception of taste, flavor and texture through sensory organs. Appetitive factor represents internal motivation and attitudes, the characteristics of the in-dividual, which include hunger, preference, knowledge, wish and demands for a food product. Cognitive factor represents the external stimuli from the information about the food product indicating price, manufacturer, calorie and health-promoting benefits, which are also included in the characteristics of food. These influence the pleasantness through cognition and judgment of benefits at the time of the food consumption.

Figure 1 Three factors affecting the pleasantness of food ingestion

41 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

2. Food Kansei Model The "Food Kansei Model"(Fig.2) is primarily intended to formulate the causal relationships between the characteristics and perceived quality of food products.

Figure 2 Food Kansei Model: Five senses communication model for palatability

The model integrates the key factors and causal relationships discussed above. In the model, the food product was assumed to have "intrinsic" and "extrinsic" attributes. Intrinsic attributes work as perceptual factors, which represent the characteristics of food such as chemical components and physical structure analyzed instrumentally by gas chromatography-mass spectrometry (GC- MS), high performance liquid chromatography (HPLC) and texture measurement. On the other hand, extrinsic attributes deals with the contents described on packages and advertisements such as product name, manufacturer and health promoting benefits work as cognitive factors of pleasantness. The upper routes of the model represent the causal relations between intrinsic attributes, perception and pleasantness. Perception represents perceived quality of taste, flavor and texture felt by tongue, nose and teeth, respectively. The intrinsic attributes of a food are perceived firstly as taste, flavor and texture through sensory organs. The link between physicochemical characteristics and the perceptions is generally referred to as a psychophysical relationship (Steenkamp and Van Trijp, 1996). It is assumed that these perceived qualities are converted to pleasantness based on the preference. Preference shows the hedonic scale for perceived quality of tastes and flavors, acquired through biological inheritance, dietary habit and aging process. It works as an appetitive factor in the determination of pleasantness for perceived taste and flavor during a brief period of tasting food. The pleasantness through sensory perceptions, which is determined at the moment of ingestion, is regarded as experienced pleasantness. In a similar fashion, the lower routes represent the way extrinsic product characteristics are related to pleasantness through quality cognition. Cognition indicates mental image in the mind caused by the recognition of extrinsic attributes through eyesight and hearing. Parallel to the upper procedure, extrinsic attributes of the product are recognized as imaginary which subsequently evokes the pleasantness (Fiske, 1982). Attitude is also an appetitive factor providing criteria for the images, evaluating the benefits of the food product. The pleasantness resulting from the images based on extrinsic attributes is referred to as expected pleasantness, which is expected at the point of purchase and consumption. Both procedures are presumed not to be independent but correlated to each other. Sensory quality of taste and flavor might be 42 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND assimilated by the expectation of high quality from extrinsic attributes of food products (Cardello, 2003). It should also be mentioned that the overall pleasantness for food consumption is demonstrated to be the result both of the experienced and the expected pleasantness (Accebron and Dopico, 2000).

Materials and Methods

1. Test samples Eight green tea samples were prepared: one standard sample (std.), three taste- and four flavor-controlled samples, namely T1-T3 and F1-F4, respectively, by brewing the green tea and adding natural extracts and materials differently. The flavor component of T1-T3 was designed to be the same as F4. These were filled and sealed mechanically in 500 ml plastic bottles, preserved at room temperature for a few weeks. For sensory evaluation, the chilled samples in the refrigerator were poured and served in translucent disposable plastic cups.

2. Instrumental analyses Volatile compounds of f1avor-controlled samples were analyzed with a gas chromatograph (GC, Agilent 6890) to determine the absolute concentrations of the odor-active compounds in green tea. Gas chromatography/olfactometry (GC/O) was also conducted to identify the characteristics of odor. A sniffer, an expert sensory technician, sat at the GC outlet and recorded the characteristics of each flavor released in the stream of purified and humidified air at a linear velocity. Non-volatile compounds including amino acids and catechins of the eight samples were analyzed by high performance liquid chromatography (HPLC, Shimadzu LC-10Avp; Waters AccQTag™ Amino Acid Analysis System 2690 xe). These characteristics are generally used in the field of chemical studies to measure the quality or tea beverages, taking multiple measures of standardized procedures.

3. Human sensory evaluation Sensory evaluation was performed by a panel of 240 consumers; 80 female senior high school or university students, 80 female office workers in the age group of 20-39 yrs and 80 male office workers in the age group of 20-59 yrs. The panel size is sufficient for the evaluation of the consumer response to the product samples (Poste el al., 2001). The panel, who were neither hungry nor sleepy, profiled the samples using a sensory 7-point descriptive analysis after being informed about the test. Both sniffing and tasting were performed in order to al1ow evaluation of odors, flavors and tastes. After the preliminary studies we found that we could measure the perceived quality of green tea beverage using 17 items. The samples were described and discriminated by the following 17 sensory attributes: fragrance intensity of green, floral and roasted, taste intensity of bitter, sweet, after-tastes, authentic and ten other attributes (shown in Table 1). Hedonic ratings were obtained from an overall impression of each sample. Each of the 240 consumers performed the evaluation of 4 samples in a fully randomized order.

43 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

3. Application of Food Kansei Model Since no differences in the appearance were presented among samples, the influence of extrinsic attributes for pleasantness through cognition, the lower routes of the model, was neglected to design taste and flavor. As shown in Fig. 3, the upper routes of the Food Kansei Model were applied to specify the causal relationships between intrinsic attributes, perception, preference and pleasantness. It contains both food perception and acceptance phases, which express the causal effects of intrinsic attributes on perception and the effects of perception on pleasantness through preference, respectively. The concentrations of volatile and non-volatile components were characterized as the variables of intrinsic attributes, while the described tastes and flavor intensities in sensory analysis were assumed to be the variables of perception.

Figure 3 Simplified Food Kansei Model dealing with intrinsic attributes.

44 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

In Fig. 4, an instance of causal relationship in the model was illustrated from the aspect of quantity. In our empirical application, the relationship of food perception phase was estimated on the use of ANN in consideration of the presumed multiple and synergistic effects in the sensory perception, which have been observed in the relationship between perceived and actual intensity of many stimuli (Wilkinson and Yuksel, 1997). The latter relationships of food acceptance phase were estimated by multiple regression analysis (MRA) for the empirical linearity among the ratings of sensory intensity and hedonic impression. In the food acceptance phase, the gradient of hedonic ratings was defined as preference for perceived quality, which facilitates the understanding of the different preferences among consumer groups. It seems advantageous to combine ANN and MRA from the viewpoint of product development.

Figure 4 Causal relationship among intrinsic attributes, perception and pleasantness in quantity

4. Statistical analyses Given the theoretical evidence that consumers' information processing capacity is limited and that they generally use only three or four more general dimensions in judging products (Steenkamp and Van Trijp, 1996), a principal component analysis (PCA) was performed on the sensory data to condense the evaluated response into simple and few variables. The principal components (PC) show the directions of greatest variance in the data. Additionally, in order to form orthogonal combinations of these principal components, rotated components were obtained by the method of varimax rotation, which is the same method traditionally used in factor analysis. These analyses were carried out using commercially available software called JMP 5 (SAS Institute Inc.). Using the rotated PC scores and the hedonic ratings data, the relationships described in the preceding sections as food acceptance have been analyzed by means of MRA for each consumer group. The hedonic ratings were used as the dependent variables and rotated PC scores used as the independent variables. One of the assumptions of multiple regression is that there is no linear relationship between any of the independent variables in the model (Accebron and Dopico, 2000). If such a relationship does exist, variables are co-linear, which sometimes leads to instability of estimated coefficients. Since the rotated PC scores are orthogonal with each other, then the correlations among independent variables do not exist in this case. The regression coefficients were used to investigate preference for the perceived quality.

45 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

An ANN modeling was carried out to elucidate food perception on the dataset obtained by instrumental and sensory analysis, using JMP software package. This network has the architecture of a characteristic multilayer perception (MLP) network. This consists of one hidden layer between the input and output layers (Fig. 5), modules of which use an error back propagation algorithm for weight adjustment, determining the contribution of each weight to prediction error. The weights are then adjusted by a fixed proportion of that contribution using the gradient descent method. To avoid finding local minima, several runs were carried out with different sets of initial weights. In ANN modeling, the whole dataset was used in calibration for the shortage of evaluated samples to be used for cross validation. Number of nodes in input layer of each network was set to the number of extracted components by GC/O and HPLC. In the output layer, the number of nodes was same as the number of optimal PC factors. For other network parameters, an over fit penalty was set to avoid the erroneous solution by holding down the weight estimates. The root mean square error (RMSE) and coefficients of determination were used to evaluate prediction performance of the models.

Figure 5 An example of 3-layer artificial neural network (ANN)

5. Estimations of perceived quality The estimations of perceived quality were performed to determine the most preferable concentration ratio of components using the ANN model. Geometrical average of PC scores weighted with their coefficients to hedonic ratings was calculated repeatedly in order to explore which combination of components gives the perceived quality preferred by each consumer group leading to the greatest pleasantness for them. It determines the most appropriate design of intrinsic attributes and perception within the limits of evaluated samples for consumer groups having different social backgrounds.

Results and Discussion

1. Sensory analysis Using the normalized sensory data for each attribute, the PCA revealed four principal components which accounted for 68.1 % of the variance in the dataset, where PC #1, PC #2, PC #3 and PC #4 accounted for 34.4, 22.1, 5.9 and 5.5% of the variance, respectively, while PC #5 accounted for 4.0%. Table 1 shows the configuration matrix of high factor loadings for sensory attributes, obtained by applying the method of varimax rotation. The first rotated component showed factor loadings on the items of "appropriate to the food", "good to gulp down", "well- balanced" and "refreshing". Thus the first perceptual dimension was identified as smoothness which accounts for 19.7% of the variation in the data. Similarly, the following basic perceptual

46 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND factors were uncovered: thickness (authentic, leaves a clear impression, aftertaste, intensely flavored and distinctive), fragrance (roasted, green and floral fragrance), and sweetness (sweet and soft). The following factors accounted for 23.8, 14.1 and 10.4% of the variance, respectively. Figure 6 and 7 show average rotated PC scores of the four factors for each sample in the sensory space, the output data for the ANN. In these figures, the taste- and flavor-controlled samples were indicated with characters of T and F, respectively. The Standard sample, Std. was positioned almost in the center of the perceptual space as expected. The perceived tastes and flavors of the samples ranged widely in the sensory space with the central focus on the standard sample.

2. Food acceptance The results of MRA obtained for each 320 observations of female students, female and male office workers are shown in Table 2. The values of β are the standardized regression coefficients, showing the consumer preference. As has been explained, the pleasantness is based on the perception. It became clear that all four factors had significant influence on pleasantness in every consumer group. The results confirmed that smoothness (β= 0.696, p < 0.00 1) has a positive effect on pleasantness by female students, indicating that it is used as a quality cue by female students to infer pleasantness. It was also demonstrated that sweetness (β= 0.382, p < 0.001) had a positive effect on pleasantness for female students, however, it weakly affected pleasantness for male office workers (β= 0.103, p < 0.05). For female students, thickness (β= 0.218, p < 0.00 I) and fragrance (β= 0.182, p < 0.(01) were indicated as be less preferred, compared with the relative importance of thickness (β= 0.299, p < 0.(01) and fragrance (β= 0.308, p < 0.001) for male office workers. The overall accuracy of the adjustment was quite good, and a significant part of the evaluation of pleasantness was explained by these factors. Results showed a need to maintain smoothness and sweetness levels for female students due to their positive sensory implications in the processed product. For male office workers, thickness and fragrance were recognized to be more important than sweetness. The preference of female office workers was found to exist between female students and male office workers.

Figure 6 PCA plots of first and second factors in sensory space of sensory evaluation data from eight tea samples. T1 to T3, taste-controlled: F1 toF3, flavor- controlled: Std., standard sample.

47 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Fig 7 PCA plots of third and fourth factors in sensory space of sensory evaluation data from eight tea samples. T1 to T3, taste-controlled: F1 toF3, flavor-controlled: Std., standard sample.

3. Instrumental analysis Table 3 and 4 document the results of the HPLC measurements per sample. Table 3 shows the concentration of caffeine and eatechins. Taste-controlled samples T2 and T3 exhibited significantly higher levels of catechins. Since the concentrations of catechins were highly correlated with each other, the total catechins were set to the input of the ANN as representative of catechins. The concentration of caffeine showing small different among samples was neglected to reduce the number of variables to correlate with the perceptual factors. Table 4 shows the concentration of amino acids, representing the taste description of umami, sweet or bitter. The taste-controlled sample T1 showed notably high levels of amino acid, especially in the concentration of glutamic acid. For ease of understanding the result, the total amount of amino acids was used for the input of ANN. Table 5 shows the concentrations of volatile components in flavor-controlled samples measured by GC, representing the odor descriptions by GC/O. Among the 20 GC peaks, 12 were recognized by sniffing in the GC-O analysis. Total concentrations of similar odor descriptions, "flowery", "roasted", "green" and "sweet", were determined (Table 5). Flavored sample F2 indicated the highest flowery and green odors, while sample FI showed the highest concentration of roasted and sweet odors. 48 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

49 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

4. Food perception As shown in Fig. 5, the ANN modeling was carried out to correlate the four factors of perception with the concentrations of total catechins, total amino acids, and the flowery, roasted, green and sweet odors. Table 6 shows the summary of estimation by ANN in terms of RMSE and the coefficients of determination or R squared for each sensory factor, which indicated sufficient accuracy in calibration considering the dependent variables evaluated by human sensory analysis; three of four sensory factors indicated R squared over 0.90. Figures 8 and 9 show the preferable combination of components and intensities of sensory factors, respectively, estimated 10 maximize pleasantness by ANN modeling for three consumer groups having different social background. In Fig. 8 the preferable combinations of components were shown in the concentration ratio or components to standard sample. For female students, relatively high levels of roasted and sweet flavor components were estimated as a preferable combination, which was assumed to lead to higher smoothness and lower thickness and fragrance as shown in Fig. 9, according to the causal relationship formulated in the Food Kansei Model. Similarly, Fig. 8 shows that female and male office workers preferred the samples containing greater amount of flowery flavor and total amino acid, which offer higher thickness and fragrance through food perception. Female office workers preferred sweetness rather than thickness compared with male office workers (Fig. 9).

50 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

Conclusions

As a consumer-based approach to the improvement of quality of food products, the Food Kansei Model has been developed to formulate the causal relationships between the analyzed characteristics and perceived quality of food products. The model is applied to understand how consumers form pleasantness about green tea perception and how this issue can help us to address optimum levels of components. The combination of PCA, ANN and MRA is proposed as a technique that is particularly suited for estimating the relationships specified in the conceptual model in practical applications. The empirical application for green tea beverages illustrates the usefulness of the model and methodology. The results showed the preferable intensities of tastes and flavors and concentration ratio of components estimated for three consumer groups belonging to different social categories. In the present case, flowery, roasted and sweet flavor and total amino acids were identified as key components that could be modified to suit the preferences of different consumer groups by the perception of smoothness, thickness, fragrance and sweetness. Future research might test the model on other food products. Further, the interrelations among perceptual and cognitive deserve attention. The quantitative evaluation of extrinsic attributes and cognizable images of food products would be essential to incorporate the influence of cognition into the model. The application of the Bayesian network and the other graphical modeling methods is expected to help formulation of the integrated structure of food perception and acceptance. It is likely that the application of this model is not restricted to tea or beverage in general. Finally, the Food Kansei Model may be extended to include the effects of extrinsic attributes such as advertising and brand name as well as other elements of the marketing mix. 51 Food Innovation Asia Conference 2010: Indigenous Food Research and Development to Global Market, June 17-18, 2010, BITEC, Bangkok, THAILAND

References

Accebron, L. A. and Dopico, D. C. (2000). The importance of intrinsic and extrinsic cues to expected and experienced quality: an empirical application for beef. Food Qual. Prefer., 11, 229-238. Boccorh, R. K. and Patcrson, A. (2002). An artificial neural network model for predicting flavor intensity in blackcurrant concentrates. Food Qual. Prefer., 13,117-128. Cardello, A'v'. (2003). Consumer concerns and expectations about novel food processing technologies: effects on product liking. Appetite, 40, 217-233. Fiske, S. T. (1982). Schema-triggered affect: Applications to social perception. In "Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition", ed. by M. S. Clark & S. T. Fiske, Hillsdale, NJ: Erlbaum, pp. 55-78. Ito, M., Umemoto, M., Yamadori, A., Ono, T., Tokosumi, A. and Ikeda. K. (1994). "Cognitive Science 6." , Iwanami Shoten, Tokyo (in Japanese). Kobayashi, A. (1995). Various Teas: Their aroma characters and analysis. Food and Food ingredients Journal of Japan, 163, 70-77 (in Japanese). Nagamachi, M. (2002). Kansei engineering as a powerful consumer-oriented technology for product development. Applied Ergonomics, 33, 289-94. Paste, L. M., Mackie, D. A., Butler, G. and Lannond, E. (2001). Laboratory methods for sensory analysis of food: Part I. Nippon Shokuhin Kagaku Kogaku Kaishi, 48, 311-320 (in Japanese). Sagara, Y. (2001). "Kansei" engineering for investigating food preference. Jpn. J. Taste Smell Res., 8, 153-159 (in Japanese). Sijtsema, S., Linnemann, A., Van Gaasbeek, T, Dagevos. H. and Jongen, W. (2002). Variables influencing food perception reviewed for consumer-oriented product development. Crit. Rev. Food Sci., 42, 565-581. Steenkamp, J. B. E. M. and Van Trijp, H. C. M. (1996). Quality guidance: A consumer- based approach to food quality improvement using partial least squares. Eur. Rev. Agric. Econ., 23, 195-215. Tominaga, O., Ito, E., Hanai, T., Honda, H. and Kobayashi, T. (2002). Modeling of consumers preferences for regular coffee samples and its application to product design. Food Sci. Technol. Res, 8, 281-285. Wailzer, S., Klocker, J ., Buchbauer, G., Ecker, G. and Wolschann, P. (2001). Prediction of the aroma quality and the threshold values of some pyrazines using artificial neural networks. J. Med. Chem., 44, 2805-2813. Wilkinson, C. and Yuksel, D. (1997). An overview; Using artificial neural networks to develop prediction models for sensory-instrumental relationships. Food Qual. Prefer., 8, 439-445.