FLAVOR PERCEPTION: SENSORY METHODOLOGY.

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1004484
• Project Start Date: Nov 6, 2014
• Project End Date: Feb 6, 2017
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIVERSITY OF CALIFORNIA, DAVIS
410 MRAK HALL
DAVIS,CA 95616-8671

Performing Department
Food Science and Technology

Non Technical Summary
It is estimated that approximately 90% of products fail in the marketplace in spite of rigorous sensory testing, resulting in the loss of millions of dollars. When the methods used for the sensory testing are examined, it becomes obvious why failures occur. Many of the testing procedures are misapplied and based on outdated 1950s psychology. We know much more about how the senses work and how the brain processes information. Using more modern approaches to experimental psychology we can find out how the brain processes the information that is used during sensory and consumer testing and how the senses react during such testing. From this we develop more efficient and valid tests for researchers both in academia and the industry to use.

Animal Health Component

45%

Research Effort Categories

Basic

45%

Applied

45%

Developmental

10%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
503	7220	3070	80%
503	7310	3070	20%

Knowledge Area
503 - Quality Maintenance in Storing and Marketing Food Products;

Subject Of Investigation
7310 - Experimental design and statistical methods; 7220 - Sensory evaluation/analysis methods (human senses);

Field Of Science
3070 - Psychology;

Keywords

sensory measurement methods

consumer acceptance testing

experimental psychology

cross-cultural studies

sensory function

brain information processing

methodology innovation

Goals / Objectives
The first thrust of this program involves basic research. It is to study sensory mechanisms and function and how the brain processes information. From this can be developed models and theories of how to use humans for sensory evaluation. If you are going to use an instrument; learn how it works. It also provides understanding of consumer behavior. The second thrust of the program is applied. The basic knowledge obtained is used to develop more sensitive and efficient protocols for the measurement of sensory perception, for the sensory evaluation of food and for measures of consumer behavior and acceptance. The main areas of focus are (1) Developing more efficient and powerful discrimination tests. (2) Developing less biased and more powerful methods of scaling (numerical estimation). (3) Developing more rapid ways of training 'expert' judges and developing methods that require less training. (4) Developing less biased methods of predicting consumer acceptance; current methods can give seriously misleading results: high levels of false preferences

Project Methods
The methodological approach will be broad based depending on the particular problems being solved. Measures will be made of the effects of sensory adaptation and memory loss during sensory testing. Theoretical models like Signal Detection Theory, Thurstonian modeling, Sequential Sensitivity Analysis will be developed and predictions from these models will be tested experimentally. This information will be used to develop protocols for more powerful discrimination tests and intensity measurement procedures. Models of fast thinking, subroutines of information processing, will be applied to the problems of false preferences in paired preference tests. Protocols used for hedonic scaling will compared in terms of inaccuracies caused by memory failures. The mechanisms of concept formation will be investigated to provide further information on the concept alignment mechanisms required for accurate description of sensory attributes. Other applications for these measurement techniques like the measurement of consumer concepts for marketing are also investigated.The next stage is to gather this information together to develop better protocols for sensory and consumer testing. This is usually performed initially with model system under careful experimental control. In this way, the variance that is involved is caused by the judges and their interaction with the new method, rather than the stimuli being tested. If these steps are successful, then the new method is tested under the less controlled conditions of industrial testing with food and beverage products, as a measure of the robustness of the protocol.

Progress 11/06/14 to 02/06/17

Outputs
Target Audience:The target audience consists of academic researchers into the mechanisms of decision-making during sensory and consumer acceptance tests, sensory professionals in the food, personal and household product industries and students studying food sensory science and any of the general public who show an interest. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?All this research provides valuable training for students of food sensory science. The approaches taken here are what are called "thinking outside the box". It is important for our students to question current methods of sensory measurement, because they were not developed using any scientific approach. They were generally the result of guesswork without any validation studies. My lab studies the sensory system and how the brain processes information during sensory testing. This information is then applied to develop more efficient sensory tests. This is excellent training for students, who in their professional careers will need to be independent thinkers and always ready to question current methods. How have the results been disseminated to communities of interest?Results are disseminated to communities of interest using journal papers, seminars and classes. What do you plan to do during the next reporting period to accomplish the goals?I shall retire but continue my research and teaching privately.

Impacts
What was accomplished under these goals? The statistics associated with the dual-reference duo-trio method provided tables of d'. From this it could be seen that the test was not equivalent to the single-reference duo-trio. This corrected the error of Hye-Seong Lee. However, it did confirm her hypothesis but not as strongly as she had previously thought. Disruptive protocols for the paired preference test were applied in the high variance conditions usually associated with those encountered in the food industry. They were successfully applied to investigations both in the USA and China, indicating their fitness for purpose. A much needed review was published, giving the current status of disruptive protocols for preference tests. Notwithstanding, a completely different approach to preference tests, using a signal detection analysis, was also developed, tested and published. This provides an alternative to newly developed disruptive protocols.These results continue my program of the development of more valid discrimination and preference test methods and provide replacement for those currently used but developed during the 1950s.

Publications

• Type: Journal Articles Status: Published Year Published: 2015 Citation: Bi, J. O'Mahony, M. Lee, H-S. The performance of the dual-reference duo-trio (DRDT) method using a balanced reference mode. Journal of Sensory Studies 2016, 48, 303-313.
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Xia, Y. Zhong, F., O'Mahony, M. Paired preference testing: False preferences and disruptive protocols. Food Science and Biotechnology 2016, 25, 1-10.
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Xia, Y. Zhong, F., O'Mahony, M Applying disruptive preference test protocols to increase the number of "No Preference" responses in the placebo pair, using Chinese consumers. Journal of Food Science 2016, 81, S2233-S2239.
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Zhang, X., Halim, J., Wichchukit, S., O'Mahony, M., Hautus, M. Paired preference tests: A signal detection based analysis with separate d' values for segmentation. Journal of Sensory Studies 2016, 31, 481-491.

Progress 11/06/14 to 09/30/15

Outputs
Target Audience:The target audiences for this research are: academics concerned with sensory and consumer research, sensory professionals in the food and personal products industries, students that are the future of food sensory science and any of the general public who show and interest. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The research performed this year, provided valuable training in the design of sensory experiments, their accomplishment, analysis and reporting, as well as a deeper understanding of the cognitive strategies associated with decision making during these testing protocols. It contributed to the organization of research theses and subsequent papers to be submitted for publication in peer reviewed journals. The methods tested required understanding, using lateral thinking and creativity, that is so essential in a young science like food sensory science. They provided an excellent training for success in an academic or an industrial career. How have the results been disseminated to communities of interest?As mentioned above, by publication in peer reviewed journals, and lectures. What do you plan to do during the next reporting period to accomplish the goals?The immediate goals are to perfect and add to the new methods already developed. These will be tested under various conditions, to test their robustness and ease of use and understanding by academics and industrial professionals, for whom they have been developed. Experiments will examine the efficiency of variations the methods already developed and completely new methods for solving the same methodological problems. There is also a need for further understanding of the cognitive strategies involved in the decision making that takes place during these testing protocols. The development of the precise design of these studies will depend on the results of current experimentation.

Impacts
What was accomplished under these goals? Two new and more efficient preference tests were developed last year, based on the application of a systems analysis of information processing in the brain. They were applied under the higher variance conditions usually associated with those encountered in the food industry. They were successfully applied to investigations both in the USA and China, indicating their fitness for purpose. Last year, a more statistically powerful discrimination test protocol based on self specification was developed. This also underwent successful testing under industrial conditions, to determine robustness and fitness for purpose. For industry, this means a more rapid and efficient test, which translates into considerable cost savings. Last year a newly developed method for hedonic (liking, wishing to buy) testing, based on simple ranking and signal detection theory was developed. It was applied to global comparisons between American and Chinese consumers and was found to be free of the range constriction artifacts that are inherent in currently used scaling methods. This now provides a reliable tool for cross-cultural comparisons of consumer acceptance. Finally, in collaboration with statisticians, tables were published for the new methods of preference testing.

Publications

• Type: Journal Articles Status: Published Year Published: 2014 Citation: Choi, J-H., Gwak, M-J., Chung, S-J., Kim, K-O., OMahony, M., Ishii, R., Bae, Y-W.Identifying drivers of liking by investigating the reasons for (dis)liking using CATA in cross-cultural context: a case study on barbecue sauce. Journal of the Science of Food and Agriculture, 2014, 95, 1613-1925
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Calder�n, E., Angulo, O., OMahony, M., Wichchukit, S. Liking and Take Away preferences for Mexican consumers: cross-cultural comparison with Thais for Psychological style. Journal of Sensory Studies, 2015, 30, 77-84.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Xia, Y., Zhang, J., Zhang, X., Ishii, R. Zhong, F. OMahony, M. Triads, tetrads and pairs: experiments in self specification. Food Quality and Preference, 2015, 40, 97-105.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Wichchukit, S., OMahony, M. The 9-point hedonic scale and hedonic ranking in Food Science: some reappraisals and alternatives. Journal of the Science of Food and Agriculture 2015, 95, 2167-2178.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Feng, Y-H., Guti�rrez-Salomon, A. L., Angulo, O., OMahony, M., Wichchukit, S. Data from words only and numbers only 9-point hedonic scales are not interchangeable for serial monadic as well as rank-rating protocols: aspects of memory and culture. Food Quality and Preference 2015, 41, 12-19.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Bi, J., Lee, H-S., OMahony, M. A Thurstonian model and statistical inference for the 2-Alternative Choice test with both test pairs and placebo pairs. Journal of Sensory Studies 2015, 30, 10-20.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Xia, Y., Zhong, F., OMahony, M. Pairing detection of off-flavors in orange juice with preference tests. Journal of Sensory Studies 2015, 30, 259-268.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Xia, Y., Zhong, F., OMahony, M. Is the discrepancy between numbers derived from the verbal and numerical protocols for the 9-point hedonic scale an artifact of product choice. Journal of Sensory Studies 2015, 30, 269-279.