=Paper=
{{Paper
|id=Vol-2285/ICBO_2018_paper_45
|storemode=property
|title=Ontological Framework for Representation of Tractable Flavor: Food Phenotype, Sensation, Perception
|pdfUrl=https://ceur-ws.org/Vol-2285/ICBO_2018_paper_45.pdf
|volume=Vol-2285
|authors=Tarini Naravane,Matthew Lange
|dblpUrl=https://dblp.org/rec/conf/icbo/NaravaneL18
}}
==Ontological Framework for Representation of Tractable Flavor: Food Phenotype, Sensation, Perception ==
https://ceur-ws.org/Vol-2285/ICBO_2018_paper_45.pdf
Proceedings of the 9th International Conference on Biological Ontology (ICBO 2018), Corvallis, Oregon, USA 1
Ontological Framework for representation of Tractable Flavor:
Food Phenotype, Sensation, Perception.
Tarini Naravane Matthew Lange
Biological and Agricultural Systems Engineering Food Science and Technology
UC Davis UC Davis
Davis, California,US Davis, California, US
tnaravane@ucdavis.ed mclange@ucdavis.edu
repulsion/desire for future consumption. Learned
consequences of ingested foods continue to influence food
Abstract. Among all sensory sciences, flavor remains a wicked choices in humans, ubiquitously known as the multi-modal
problem. Sight, sound, and touch have all been digitized, and vast sensation of flavor. [4–6] Challenges for designing
resources exist around their computation.. While the biological computational flavor systems are effectively highlighted by
basis for food consumption is primarily to nourish bodily
functions, it fulfills a greater second function of sensory pleasure. comparison to more developed computational neuroscience
Flavor, and the pleasure it engenders, is the primary driver of food systems of vision and sound, where scientific research and
choice. Moving toward a semantic web of food that enables technology successfully mapped physical properties of
personalization of food and flavor experiences requires an stimuli to their perceptual characteristics. We argue that
interoperable ontological model of flavor. This paper proposes a these systems were comparatively easy to digitize due to the
framework of several ontologies to model a comprehensive view continuous nature of their data. In vision, wavelength
of flavor, by partitioning it into three interoperable matrices of translates into a RGB color model; in audition, frequency
interacting variables: objective characteristics of food, subjective
sensory experience, and interpretive communication of that and wavelength translates into amplitude/pitch model. [3]
experience. The objective matrix details the properties and This information digitisation provides unambiguous
behaviour of food molecules. The subjective matrix represents the identification of colour and sound, without influence of
multilayered and highly individualised consumption and sensory perception or hedonic response. We utilize an analogous
perception variables. The interpretative layer deals with the approach to solving the wicked flavor problem, albeit the
communication and language used to describe the food experience. dimensionality of flavor is orders of magnitude greater than
Together these three matrices represent an initial ontological model for sound or colour, and requires multiple layers (matrices)
for the flavor and sensory experience portion of the emerging
semantic web of food. of variable separation. The reference to “matrix” in this
paper is not the algebraic matrix, but a complex state of
I. INTRODUCTION interacting variables. The ontology-based model has 3
In 1973, two social scientists, Horst Rittel and Melvin principle matrices: Objective characteristics of food (Food
Webber defined a class of problems they called “wicked Phenotype), Subjective Sensory Experience, and
problems”.[1] Wicked problems are messy, ill-defined, Interpretive Communication of the perceived experience.
more complex than we fully grasp, and open to multiple These broadly correspond to the knowledge domains of
interpretations based on one’s point of view. [2] Flavor Food Science, Sensory/Neurophysiology, and
among all sensory neurosciences remains a wicked problem. Anthropology/Psychology/Linguistics respectively.
While many researchers have proposed methods for digital II. TRIPARTITE FLAVOR MODEL
replication of specific tastes and aromas [3], to date there
exists no semantic or ontological models for operating over The model in Figure 1 shows the three matrices. The first
food flavor and the sensory experience. matrix enclosed by a curve dashed line represents the Food
Selection of food for nourishment in animals is an Phenotype Matrix, unbiased by individual response. The
evolutionary process, influenced by habitat and ecological second matrix, enclosed in the human body boundary,
conditions, whereby recognition of tastants and especially represents the sensory capture and modulating factors in
odorants are associated with (dis)pleasurable eating and decoding the ingested food. The third layer still partly
post-prandial experiences, and highly influence
ICBO 2018 August 7-10, 2018 1
Proceedings of the 9th International Conference on Biological Ontology (ICBO 2018), Corvallis, Oregon, USA 2
enclosed in the human boundary is the interpretation of the an aqueous medium which includes other soluble and
experience which is finally communicated. insoluble compounds. Chemical components are all
atoms/compounds in foods classified by molecular structure.
Biological properties are the bioactivity roles, Chemical
properties characterize the reactability and aroma. Physical
properties include Rheological, Morphological, Surface,
Acoustic, Volumetric, Reflective/Refractive properties to
name a few. Within the objective matrix, the biological,
chemical and physical properties are expressed by three
vectors [B,C,P]. This notation connotes the state of a food at
a given point in the timeline of its transformation.
“Organoleptic properties are the characteristics of the
Phenotypic classes detectable by electrical, mechanical,
chemical, and temperature bio- mechanisms and felt as the
sensation of touch, sight, smell, taste, sound, inflammation,
and lacrimation. Hence the Organoleptic Ontology has
relevance to the consumption of food and is at the boundary
of the objective and sensory matrix.” [7] It is expressed by
the variable [Organoleptic] and is associated with a given
[B,C,P].
The Objective matrix illustrates the transformation of a
given [B,C,P] into another [B’,C’,P’] as a function of all or
any of the variables ;an added ingredient represented by
[B,C,P], the passage of time for example in the ripening or
rotting of a fruit, a food altering process, and variables for
the environment the food is in for example environmental
conditions at high altitude or at sea level on the ground.
Fig. 1. Tripartite flavor Model.
Boundary lines separate three matrices; Objective
characteristics of food (Food Phenotype), Subjective
Sensory Experience, and Interpretive Communication of the
perceived experience, explained in section II A,B,C resp.
Fig. 2. Food Phenotype Ontology model
A Objective Matrix This [B,C,P] representation is in early stages and known
Food is classified into components and properties shown caveats should be mentioned explicitly to avoid any
in Figure 2. Biological components include living cells like confusion or misrepresentation of present capabilities.While
bacteria and morphological features of the food, like germ, it has been stated earlier that the Matrix is not the algebraic
bran and endosperm in a grain or the milk fat globule matrix, it should be mentioned that the formulaic
membrane in milk which is a structure composed primarily representation in its current form is highly simplified and
of lipids and proteins that surrounds milk fat suspended in
ICBO 2018 August 7-10, 2018 2
Proceedings of the 9th International Conference on Biological Ontology (ICBO 2018), Corvallis, Oregon, USA 3
will evolve to several algorithms connecting properties, physiological, psychological and neurological factors. These
their transformational variables and phenotypic outcomes. Sensory Phenotypes are in turn modulated by the Sensory
Interpretation layer which includes emotive responses and
An ongoing project to characterise dough is a use case for associative learning.
this model and a means to vet and develop it. The project
proposes to define the [B,C,P] model for flour and added
ingredients like salt, water, yeast, quantify the C Interpretative Matrix
transformational energy of force and time and
environmental conditions of temperature and humidity , and Across human existence, social constructionism has
define the [B,C,P] model of the resulting dough. given rise to varied informal vocabularies across
socio-cultural demographics. These folksonomies represent
collections of words utilized by humans to model their
B Sensory Matrix varied experience arising from their social and cognitive
processes [15]. Fenko et al describes expressions divisible
The sensory apparatus and neural processing is a into three groups: sensory descriptors (hard, red, noisy);
highly-nuanced combination of psychological and symbolic descriptors (interesting, expensive, modern); and
physiological factors shown in the second layer of the affective descriptors (pleasant, beautiful) [16]. More
matrix framework. The olfactory apparatus is approximately recently, social constructionism popularised “freshness”.
400 odorant receptors, but each individual has a unique set Judgments of freshness vary based on colour and smell cues
of genetic variations [3]. Factors like ancestry, age, and and generally have little to do with the temporal aspect of
gender accounted for over 70% of the explainable variance “freshness” [17]. Ontological modelling of Food
for some odors (guaiacol, diacetyl, and nonyl aldehyde) and Phenotypes, and especially their Organoleptic Descriptors,
less than half of the explainable variance for others[8]. The remains challenging due to the fact that these folksonomies
taste papillae in the tongue vary in density across have percolated through layers of sensation and perception
individuals and throughout the life span.[9, 10]. A whose context is culturally dependent. This effort to
comparative study of groups, with varying higher taste bud distinguish interpretation from content can be appreciated in
densities reported these perceptions; sucrose (196%), NaCl the context of the constantly growing world wide web where
(135%) ,PROP (142%), Citric acid (118%) and quinine HCl user-tag based folksonomies are used to catalog web content
(110%) than the lower density group [11]. Anosmia and and drive personalised search strategies. [18]
hyposmia, the inability or decreased ability to smell, is
estimated to afflict 3–20% of the population and is linked to
old age,chronic sinonasal diseases, severe head trauma, III ONTOLOGICAL REPRESENTATION
upper respiratory infections, or neurodegenerative diseases The logical matrix flavor model connects the inherent
[12]. properties of food to its sensory perception. The
On the psychological front, stress causes changes in representation maps to focussed disciplines that have
neuroendocrine balance (high cortisol and insulin) thus remained isolated: objective characterisation of food
impeding the more reflective cognitive control over eating phenotype, sensory analysis and consumer perception. The
that is distinct to humans leading to non-homeostatic eating development of high throughput technologies and emerging
patterns. Associative learning acquired from repeated AI applications presages the need for an integrated
exposure to a specific organoleptic stimulus drives changes ontological framework. This trend bears similarity to the
to the peripheral sensory organs themselves [13]. Emotive events and developments in molecular biology that lead to
responses add a further variable in the interpretative the OBO Foundry, being instrumental to the success of the
process.Moods and emotions ranging from neuroticism, to Gene Ontology. [19] In alignment with objectives of OBO
conscientiousness influence eating styles and food choices to foster and organise ontological development, and the
[14]. foundational Continuant-Occurrent architecture, the matrix
The Sensory matrix has three distinct interacting representation is formalised into a modular ontological
components. The peripheral sensory organs relevant to system. It is important to point out that the future work is to
organoleptic stimulus are modulated by the Sensory develop the Phenotype and Interpretative ontologies.
Phenotype variables which include aforementioned
ICBO 2018 August 7-10, 2018 3
Proceedings of the 9th International Conference on Biological Ontology (ICBO 2018), Corvallis, Oregon, USA 4
The proliferation of applications for computational flavour
may cause unruly ontology creation and development, and
this suggested architecture could guide in creating
ontologies of varying granularities; top level ontologies and
specialised ontologies that link and harmonise consistently
and efficiently. ChEBI is not intended for culinary
application, since the ‘has role’ relationship which links
chemical entities to their roles and ‘has part’ which links
composite entities [20] has some incomplete or incorrect
coverage of culinary data. For example Molasses “has part”
glucose, “has part” fructose, and “has part” sucrose and “has
role” flavouring agent is incomplete since the constituents
are not quantified and the role of “flavouring agent” is too
broad and hence non informative.. Another limitation is the
lack of a causal relationship between the structural
properties and role. The specific chemical structural
property linked to the the role of emulsifier is essential
from a culinary perspective for the next step of defining
reactions. FOODON must be recognised as an upper level
ontology that organises food products from the
LanguaL-indexed SIREN database into subclasses like food
safety, food processing and agricultural and animal
husbandry practices. However the subclasses do not explain
the specific dynamics and reactions of the food process,
which is better left to specialised ontologies.
In conclusion this architecture disambiguates objective
properties of food from its subjective experience while also
suggesting an architecture to organise this vast information.
4 CONCLUSION
The digital model for flavor is an important part of the
semantic web of food. The suggested design enables References
capabilities like the prediction of flavor outcomes resulting
from specific processes and ingredient combinations,
1. Rittel, H.W.J., Webber, M.M.: Dilemmas in a general
personalization of experiences, and integration of flavor
theory of planning. Policy Sci. 4, 155–169 (1973).
variables with those related to health outcomes and
2. Gawande, A.: Something Wicked This Way Comes,
sustainability metrics to promote behaviour change without
https://www.newyorker.com/news/daily-comment/som
sacrificing desirability of foods. Framing the flavor model in
ething-wicked-this-way-comes.
modular sections considers the (future) role of
3. Mainland, J.D., Lundström, J.N., Reisert, J., Lowe, G.:
measurements to support reasoning and decision making in
From molecule to mind: an integrative perspective on
any food processing sequence toward a desired phenotypic
odor intensity. Trends Neurosci. 37, 443–454 (2014).
outcome. The Food Phenotype model can also be applied
4. Prescott, J., Taylor, A., Roberts, D.: Psychological
towards quality/grading standards of commodities; for
processes in flavour perception. Flavor perception.
example, characterizing and differentiating products like
256–277 (2004).
tea, bread, and cocoa based on ingredients and processing
5. Guichard, E., Salles, C., Morzel, M., Le Bon, A.-M.:
methods--thus establishing bases for price premium via
Flavour: From Food to Perception. John Wiley & Sons
quality standards, thereby giving recognition to
(2016).
artisanal/speciality segment products.
6. Spence, C.: Multisensory Flavor Perception. Cell. 161,
ICBO 2018 August 7-10, 2018 4
Proceedings of the 9th International Conference on Biological Ontology (ICBO 2018), Corvallis, Oregon, USA 5
24–35 (2015). N., Muthukrishnan, V., Turner, S., Swainston, N.,
7. Naravane, T.: OrganolepticAndSensoryOntology, Mendes, P., Steinbeck, C.: ChEBI in 2016: Improved
http://ceur-ws.org/Vol-2050/ODLS_paper_8.pdf. services and an expanding collection of metabolites.
8. Keller, A., Zhuang, H., Chi, Q., Vosshall, L.B., Nucleic Acids Res. 44, D1214–9 (2016).
Matsunami, H.: Genetic variation in a human odorant
receptor alters odour perception. Nature. 449, 468–472
(2007).
9. Arey, L.B., Tremaine, M.J., Monzingo, F.L.: The
numerical and topographical relations of taste buds to
human circumv allate papillae throughout the life span.
Anat. Rec. 64, 9–25 (1935).
10. Shimizu, Y.: A histomorphometric study of the
age-related changes of the human taste buds in
circumvallate papillae. Oral Medicine & Pathology. 2,
17–24 (1997).
11. Miller, I.J., Reedy, F.E.: Variations in human taste bud
density and taste intensity perception. Physiol. Behav.
47, 1213–1219 (1990).
12. Boesveldt, S., Postma, E.M., Boak, D.,
Welge-Luessen, A., Schöpf, V., Mainland, J.D.,
Martens, J., Ngai, J., Duffy, V.B.: Anosmia-A Clinical
Review. Chem. Senses. 42, 513–523 (2017).
13. McGann, J.P.: Associative learning and sensory
neuroplasticity: how does it happen and what is it good
for? Learn. Mem. 22, 567–576 (2015).
14. Keller, C., Siegrist, M.: Does personality influence
eating styles and food choices? Direct and indirect
effects. Appetite. 84, 128–138 (2015).
15. Gergen, K.J., Gergen, M.: Social Construction: A
Reader. SAGE (2003).
16. Fenko, A., Otten, J.J., Schifferstein, H.N.J.: Describing
product experience in different languages: The role of
sensory modalities. J. Pragmat. 42, 3314–3327 (2010).
17. Fenko, A., Schifferstein, H.N.J., Huang, T.-C.,
Hekkert, P.: What makes products fresh: The smell or
the colour? Food Qual. Prefer. 20, 372–379 (2009).
18. Vallet, D., Cantador, I., Jose, J.M.: Personalizing Web
Search with Folksonomy-Based User and Document
Profiles. In: Lecture Notes in Computer Science. pp.
420–431 (2010).
19. Smith, B., The OBI Consortium, Ashburner, M.,
Rosse, C., Bard, J., Bug, W., Ceusters, W., Goldberg,
L.J., Eilbeck, K., Ireland, A., Mungall, C.J., Leontis,
N., Rocca-Serra, P., Ruttenberg, A., Sansone, S.-A.,
Scheuermann, R.H., Shah, N., Whetzel, P.L., Lewis,
S.: The OBO Foundry: coordinated evolution of
ontologies to support biomedical data integration. Nat.
Biotechnol. 25, 1251–1255 (2007).
20. Hastings, J., Owen, G., Dekker, A., Ennis, M., Kale,
ICBO 2018 August 7-10, 2018 5