BY: JO-EZER LAU

Image from: http://rdteam.com/sensory-evaluation/

A panel of expert specialists routinely trudge towards the laboratory. They have years, even decades, of experience in sensory evaluation[1], a scientific method that evokes, measures, analyses, and interprets responses to products as perceived through the five senses. Also masters of descriptive analysis, they aim, in a single day, to taste a few hundred-odd samples with varying flavour profiles, thresholds, and ratings.  

These processes are vital aspects of production because the collected data is continuously interpreted to study how and why food products evoke certain responses from the senses, which collectively serve as the gatekeeper of a body’s nutrition. This field of work requires maximum progress in the morning prior to  significant influence on their taste buds, whether by external factors or adaptive response. Post meal evaluation is also possible, but the data would generally have lower confidence levels due to lack of motivation or endocrine-induced fatigue.

The parameters set out by Product Research and Development are strict, enforced by the experiment design team. To decrease variability and bias, as well as increase data reliability, the samples have to be randomly numbered, and presented in duplicates in no particular order; at most, they may evaluate x samples per session, cleanse their palate, and wait for y minutes before starting the next session. All of these vary according to a plethora of factors. These include, for example, size and the facilities of the sensory evaluation suite, or the polarity and solubility constants of the product i.e. water-based samples require less waiting time in between consecutive sample tests compared to oil-based samples.

After crunching the gathered data, food products are developed accordingly. Therefore, the human role is undoubtedly a cornerstone in sensory evaluation of food from the perspective of innovation, developing product variety, sales and marketing, as well as increasing consumer loyalty. This realisation may linger on in your mind as you casually swing by the Pet Food aisle and pick out your pet’s favourite homemade beef and cheddar dog biscuits – ‘improved aroma!’ the bag screams at you in extra large font, ‘With high-quality dog food, you can now offer your pet a wide variety of tastes and textures.’

How else could ‘wide variety of tastes and textures’ or ‘improved aroma’ have transpired? Max the corgi telling his owner that the treats contained too much sodium nitrate and were not cheesy enough? Like a bitter pill, the truth is hard to swallow, but so is pet food, literally.

Humans, too, are responsible for taste testing pet food, especially those for dogs and cats. Therefore, in light of sensory differences and existing animal-only palatability tests, this piece emphasises the significance of human beings in pet food tasting and provides an overview of how an actual tasting session is conducted.

How do dogs and cats’ sense of taste differ from that of human beings?

According to a study by Xia Li et al., recordings from domestic cats’ (Felis silvestris catus)  taste nerve fibers and from units of the geniculate ganglion innervating taste cells demonstrated responses to salty, sour, and bitter stimuli as well as to amino acids and nucleotides[2]. Compared to other mammals, the sense of taste in cats showed no significant difference, except an inability to taste sweet stimuli. However, they avoided stimuli that tasted either bitter or very sour to humans, while having preference and need for 11 essential amino acids, especially taurine, explaining why cat food contains a relatively high concentration of animal protein[3].

image from: http://www.vegasplayermagazine.com/2010/11/british-chef-gordon-ramsay-at-serendipity-3/

Humans can have up to 10, 000 taste buds, 1700 for dogs, whereas cats, only 500. Hence, while dogs’ tasting abilities exceed that of cats, including being able to taste sweetness, their sense of taste (and perspective of their world)[4] is constructed primarily from smell. Dogs possess up to 300 million olfactory receptor cells in their noses, compared our 6 million; while the part of its brain devoted to analyzing smells is, proportionally speaking, 40 times greater than ours. To this extent, food product developers have even designed methods such as. false-bottomed bowl tests to focus solely on dogs’ odor preference, as published[5] in the Journal of Sensory Studies.

Palatability tests are conducted using animal test subjects

Image from: http://palatantsplus.com/blog/animalvalidation

It makes complete sense for pets to taste test food meant for them in the first place, since animals are the panelists instead of humans, the difference in test design and parameters is primarily based on the nature of pets’ nutrition. It’s partly passive i.e. pets are fed by their owners, pets eat, but do not ‘feed themselves’. Hence, there is a limited choice and accessibility to food, if they have grown to dislike whatever their owner has bought, it is unlikely they practice selectivity like humans do. Coupled with minimal understanding of the concept of time, i.e. pets are most likely to eat when fed since the next period of access to food is not within their control.

Associate professor of evolutionary anthropology at Duke University and founder of Dognition, Dr. Brian Hare, is revolutionising dog food palatability testing with methods used in research on canine cognition.[6]

Discounting tasks can be used to test for pet food preferences, where there is either an immediate reward or a delayed reward, somewhat a parallel to economic decision making. Considering a dog’s motivation to eat, if it would rather refrain from one sample and suffer through waiting in anticipation of the second, sufficient discounting data may give a strong indication of the product’s potential marketing performance relative to other leading brands.

Meanwhile, in the Risk-taking paradigm, a dog has the option of either eating a given amount of a particular sample (A), or taking a risk by refusing to eat sample A. As a result, it either receives a much smaller portion of the same sample (A), or, a different sample (B). Again, considering dogs to be highly motivated-eaters, if the dog is willing to chance against the fixed option in hopes of a variable, preference data for a new product can be established.

Despite these sensory caveats and palatability tests, why do humans taste pet food nevertheless?

Science

#1 Behavioural studies or palatability tests can be conducted continuously to determine the effects of changes in processing, raw materials, flavourings and presentation. However, these tests are nevertheless different forms of preference and acceptance tests; they are time consuming, not cost-sustainable, and most questionably – yield equivocal data of a binary nature. For example, accepted or not, preferred or not. Simply put, limited meaning can be constructed no matter how much data is collected and construed. Furthermore, previous experience or diet, and individual animal variation and lateral bias complicate the protocols. To compound the problem, in flavour research involving animals only, large amounts of resources have to be employed just to isolate specific quantities of aroma and fractions required to recreate concentration-dependent variables.

#2 Although the physiological and perceptual systems involved in taste and olfaction differ between humans and pets, sufficient experimental evidence suggests that human sensory data can be useful in assisting with pet food formulation. Unlike data gathered from animal testing where outcomes are binary, human sensory responses can be designed to provide hedonic data points as feedback, for example, from a scale of 1-9.

Language

Image taken from: https://www.facebook.com/mayapolarbear/videos/2305701889457493

#3 Dogs are opportunistic and willing eaters of food – often more than required – whereas cats are very sensitive to flavour differences in their food, and given ample choice, very discriminative in food selection. Yet, pets are clearly unable to express and communicate the ‘what’ and ‘why’ behind their likes and dislikes, challenging pet food manufacturers, animal nutritionists and behaviourists to optimise products through data based on human feedback.

Nutrition & Economics

#4 In ‘The Pursuit of Minimal Trade-off’, nutrients, flavour, and cost are factors that product developers have to juggle, regardless of human nutrition or pet nutrition. Food that light up sweet, salty umami fireworks on your tongue are not necessarily the best for your health, like that heap of steak fries with extra salt, mayonnaise and ketchup. Conversely, steamed sweet potato, broccoli, and avocado may not be tasty no matter how healthy these foods are. Simply put, nutritious food does not always appeal to our senses, and this issue co-exists in the sphere of pet food too. So if pet food tastes like how raw kale would to humans, no matter how nutritious it is claimed to be, it will not sell. As superior and more intelligent creatures, humans with the likes of animal nutritionists and sensory experts are committed to innovating products that are flavourful yet nutritious.

Moreover, pets simply perceive the world according to their needs. If Rover the German Shepherd  sniffs out a meat treat in a pile of decaying, pathogenic junk, he might just consider eating it nevertheless simply because of nuances in discernment on what is beneficial and harmful. In that same manner, dogs and cats sometimes cannot even tell if a food product has gone bad, oblivious to what humans would instantly identify as a health hazard.

#5 To make matters worse, production costs force product developers to ensure pet food contains all the necessary nutrients with minimal decrease in the profit margins. With standard references, human sensory panels can help to ensure that the taste of pet food doesn’t change significantly if the product is subject to food fortification e.g. minerals, vitamins or oils.

The Consumer

#6. Knowing that pet food is tested by humans appeases pet owners. Rover doesn’t drive to the grocery store, push a trolley into the store, select premium pet food, and open the can. The pet owner purchases the product, Rover simply digs in. As such, the pet owner is, by definition and by practice, a consumer of pet food products as well, since the ability, choice, and rights of purchasing lie with the former. As you, the pet owner, rip open that bag of beef and cheddar dog biscuits you bought earlier, and smell the meaty and cheesy fragrance that wafts out, you feel satisfied getting what you paid for.

#7 Furthermore, knowing that their pet food is made from human-grade ingredients and fit for human consumption increases the confidence of pet owners in a particular product. After all, the Federal Food, Drug, and Cosmetic Act (FFDCA) requires that all animal foods, like human foods, be safe to eat, produced under sanitary conditions, contain no harmful substances, and be truthfully labelled.[7] If this sounds far-fetched, punch in “Simon Allison dog food taster” into your Google search box, or “Phillip Wells every batch every recipe every day”. 6-figure salary per annum to eat dog food, why not?

Image from: https://www.petfoodindustry.com/articles/5637-the-honest-kitchen-expanding-in-dehydrated-pet-food

The Honest Kitchen, though, is on another level. Founded by Lucy Postins, a premium pet food company, in 2004, obtained a Statement of No Objection from the US Food and Drug Administration (FDA) to use the term ‘Human Grade’ pet food on its product labels – a first in the industry. They have renewed this Statement regularly since then, and in 2007, won a court case versus the Department of Agriculture Ohio who refused to register the products with ‘human grade’ statement on the labels[7]. DEDICATION.   

Enough said. How is pet food tasting performed?

Currently a Professor of Biological Sciences and Psychology/Wine Science at Brock University in St. Catharines, Ontario, Dr. Gary Pickering’s research offers much insight into the nitty gritty (literally) of cat food tasting.  In 2007, Pickering set out to develop a methodology for using human taste panels to optimise the sensory characteristics and acceptance of canned cat food products (CFP). [8] The panel of taste testers was drawn from the staff and student populations of Charles Sturt University-Riverina campus in Australia, where Pickering taught at the time.

The prospective panelists were screened with four simple exercises in which their sensory acuity was assessed, distinguishing the basic tastes by blind presentation, discriminating different levels of hardness, and ranking bitterness. In the last exercise, subjects had to taste CFP, rating it on a 9-point balanced hedonic scale, enabling Pickering’s research team to filter out subjects with a “strong negative attitude towards the products” as they may have had reduced motivation, concentration, which ultimately would affect data reliability.

Image from: https://www.suntory.com/sic/research/t_oolong/

About one-third of the subjects chose to pull out, simply by reason of dislike for the taste of cat food. The final panel consisted of 11 individuals within an age range of 19-60 years old, 4 of which had experience in tasting panels. The panel tasted samples of cat food in different forms, including meat chunks, gravies/gels, and meat-gravy mixes over a 2 week-course of six 1.5-hour training sessions.

Prior to entering the tasting booths, panelists familiarised themselves with the reference samples. Each sample was then evaluated as follows: (i) mouth rinsed with water; (ii) 0.5–1 teaspoon of sample taken onto a teaspoon and placed in mouth; (iii) sample moved around mouth and chewed for 10–15 seconds until at a consistency ready for swallowing; (iv) a portion of the sample swallowed and the remainder expectorated into a spittoon; (v) intensity ratings for each attribute made on 15-cm line-scale; (vi) mouth  rinsed with water. A 1–2 min break was enforced between samples.

Using a descriptor generation form provided by Pickering, a list of 119 flavour and 25 texture descriptors was produced, further levelled down to 18 flavor descriptors: sweet, sour, tuna, herbal, spicy, soy, salty, cereal, caramel, chicken, methionine, vegetable, offal-like, meaty, burnt, prawn, rancid and bitter. The resulting four texture dimensions included: hardness, chewiness, grittiness and viscosity.  Moving on to each of the 13 canned cat food products, with the same sampling methods, (yum!) intensity and hedonic impression of each flavour on the list was scored relative to the standard reference flavour attribute.

Companies that own these products can then use this data as a precursor in the experimental design of In-House Usage Testing with animals, or with palatability tests according to the pet animal of interest.

That said, scientific evidence on the extent of how these methods translate to optimising dog food products with human sensory evaluation is still being studied extensively. Nevertheless, the technique implemented by Pickering and his team is more than adequate to be improvised upon in terms of considering other variables like increasing vocabulary for describing odour profiles, and range of textures depending on data gathered from animal preference tests.

Is it the way forward?

Definitely. Although the usefulness and limits of sensory data gathered from human panels could be more well-defined, it has already been cross-applied to other products (exactly what Pickering did with dried cat food products shortly after)[9], including modifying existing research methods for human food.

Ultimately, experimental data generated by the human senses in this regard facilitates more efficient methods for optimising pet food flavour and texture while predicting the effects of composition and processing changes on feeding patterns. To this extent, science, production economics, consumers and most importantly, pet dogs and cats, raise their glasses in favour. Who’s a good boy? The amazing humans who are committed to swallowing pet food samples each day.

Image from; https://gph.is/1UTd1UU

References:

[1] Choi, Sung Eng. Food Science: An Ecological Approach. Edited by S. Edelstein. Vol. Chapter 3: Jones & Bartlett Learning, 2018. Science. Page 84

[2] Xia Li et al., ‘Cats Lack a Sweet Taste Receptor’, The Journal of Nutrition, 136(7), pp. 1932S-1934S. Available at: https://academic.oup.com/jn/article/136/7/1932S/4664718. (Accessed 23 August 2018)

[3] Coates, J., Dr (2015) ‘What You Need to Know About the Ingredients in Your Cat’s Food | petMD’. Available at: https://www.petmd.com/blogs/nutritionnuggets/cat/dr-coates/2015/april/everything-you-need-know-about-cat-food-and-protein-32590 (Accessed 25 August 2018).

[4] Horowitz, A., How do dogs “see” with their noses? – Alexandra Horowitz (2018). TedEd.

[5] C. Pétel, C. B., M. Thomsen.  L. Callejon.  F. Péron (31 January 2018) ‘A new method to assess the influence of odor on food selection in dogs – Pétel – 2018 – Journal of Sensory Studies – Wiley Online Library’, Journal of Sensory Studies, 33(1).

[6] Wall, T. (30 April 2018) ‘2 scientific techniques for dog food palatability testers’. Available at: https://www.petfoodindustry.com/articles/7157-scientific-techniques-for-dog-food-palatability-testers?v=preview (Accessed 24 August 2018].

[7] Petfood Industry,  “The Honest Kitchen Renews Fda ‘Human Grade’ Claims Approval.” (19 June 2014 2018). Available at: https://www.petfoodindustry.com/articles/4375-the-honest-kitchen-renews-fda-human-grade-claims-approval. (Accessed 24 August 2018)

[8] Pickering, G. J. (2008) ‘Optimizing the sensory characteristics and acceptance of canned cat food: use of a human taste panel’, Journal of Animal Physiology and Animal Nutrition, 93(1), pp. 52-60.

[9] Pickering, G. J. (2008) ‘Optimisation of dried cat food using a human taste panel: Methodology and characterization of flavor’, Journal of Animal Physiology and Animal Nutrition, 61(1/2), pp. 30-36. Available at: https://www.researchgate.net/publication/275963203_Optimisation_of_dried_cat_food_using_a_human_taste_panel_Methodology_and_characterization_of_flavor/citations