Flavor Science: The Human Element


By Olivia Auell

Humans are complicated.

When we drink a cup of coffee, cues from all five senses turn into signals in the brain, travel through complex circuitry and produce what we call flavor.

As you’ve probably realized if you’ve worked in coffee for a while, not everyone perceives the same thing when tasting coffee. Individuals vary based on their level of experience, their genetic makeup, how they are feeling that day, and many other factors. So, is it possible to form an agreement on exactly how a particular coffee tastes?

In other words, can humans, whose perception of flavor is inherently subjective, produce data with almost machine-like precision? Are we fooling ourselves when we aim for agreement between people who have different memories, emotions, and experiences? Let’s dig a little deeper into a few of these sources of variation.


First, it is well-known that genetic variation exists in taste sensitivity. If you’ve tasted one of those paper strips in high school biology class, you know what I’m talking about. There’s a gene for the TAS2R38 receptor that determines how strongly a person perceives the bitterness of a compound called phenylthiocarbamide (PTC). Based on how intolerably bitter the strip is to a person, they are categorized as a “taster,” a “nontaster,” or a “supertaster” (Bartoshuk 1993).

However, the degree to which a person can taste PTC does not predict their sensitivity to other bitter compounds, let alone other tastes (Delwiche and others 2001). There is some evidence that PTC taster status can influence coffee preference (Masi and others 2015).

Taste bud distribution on the tongue also varies genetically. Some people taste more intensely because they have more taste receptor sites (Stuckey 2012, p. 21). Some people are “smell-blind,” or anosmic, to specific odorants (Stuckey p. 62). Even our affinity for cilantro is partially genetic: people with a certain genotype more frequently report an unpleasant, soapy taste (Eriksson and others 2012).

Memory and Experience

A person’s previous experience can affect which flavor attributes they notice when tasting a coffee. There are multiple elements to this, from subconscious associations to cultural culinary preferences.

Our past food experiences can influence our reaction to new flavors, including both how we describe them and their hedonic valence, or pleasantness. As any cupper knows, the more familiar we are with a particular food, the more nuances we notice.

Language is a key part of experience. Especially across cultures, we often use different vocabulary to describe flavors, or cook with different flavor combinations. The foods we have consumed in the past form flavor images in our brains, which then determine whether we perceive certain pairings as congruent (Small and Prescott 2005; Shepherd 2013, p. 160). This, combined with the fact that humans struggle to describe flavor attributes (Shepherd, p. 211), means that two people can use the same word to describe two very different ideas, or a different word to describe the same idea. Lexicons are created by sensory scientists for this very reason: to help unify the language between evaluators. The World Coffee Research Sensory Lexicon is a huge step toward helping us speak about coffee with a common tongue.


If you were at Re:co Symposium this year, you might have participated in the experiment where different types of music were played while you drank coffee. Slow, haunting music brought out bitterness, while bright sounds accentuated sweetness.

Everything from the color of the cup to background noise can influence how we perceive a coffee’s flavor. One recent experiment found that the arrangement of food on a plate influences its perceived monetary value (Michel and others 2015). Coffee professionals know this intuitively; it is the reason we design our retail environments so carefully. But environment is also tremendously important to control in cupping settings. The details of cupping lab design are important. (Emma Sage wrote a great article about atmospherics; find it here.)

Perhaps the most influential aspect of context is expectation (Okamoto and Dan 2013). What we anticipate informs what we taste. It’s the same concept as the phrase, “you see what you look for.” Cupping samples without knowledge of their origin is important for this reason.

What It Means for Coffee Professionals

So, how much does this matter for everyday operations, and what can we do about it?

Minimizing variation from other sources is also crucial in balancing individual variation. The more we can dial in the variables between cuppings, the more precise our sensory data and the more meaningful our conclusions.

The important part is not necessarily standardization across the entire industry, but clear communication within companies and within supply chains. Many coffee companies have developed extensive cupping protocols and standards. The terms and references in the Lexicon can serve as a useful complement to these. Emma Sage, Coffee Science Manager at SCAA, remarked, “What is most important is that you can communicate within your own supply chain about what your product, the coffee, tastes like.”

Academic sensory science, while a different exercise than cupping, can provide helpful principles. Here are a few practical tips, courtesy of Molly Spencer, one of the lead developers of the new flavor wheel:

  1. Establish a training standard and calibrate yourself.

Consider implementing a procedure to make sure you’re all on the same page. When someone is learning cupping, test their accuracy. There is a lot of background flavor in a cup of coffee. Joe Marrocco, Director of Education at Café Imports, suggests spiking in flavor defects to a cup of coffee. This helps a novice cupper learn how the defects show up against the other flavor complexities of the cup.

  1. Use warmup samples and references.

Everyone who has evaluated flavor knows there are just some days when you’re more “tuned in” than others. Get in the zone before cupping by warming up with a few samples before you begin scoring. If possible, have some WCR Sensory Lexicon references available to familiarize yourself with the flavor attributes. Even for familiar tastes, it’s helpful for everyone on your cupping team to experience the same reference. Molly says, “When they are describing a certain word, like blueberry flavor, they’re all on the same page about what the definition of that really is.” Training to a common standard helps mitigate individual variation.

  1. Take frequent breaks.

In sensory science, it is standard to evaluate no more than 6-8 samples at once. Molly says, “Coffee is so complex, there is physiological fatigue because your tongue and nose can only take so much.” If you’re evaluating a lot of samples, try to space them out in time to preserve acuity.

Emma Sage adds that consistency in protocol is key. Minimizing the variation in the cupping process details can help decrease the noise in your data. This can be especially important for companies with staff and roasteries in multiple locations.

Variation between tasters is a significant factor in coffee cupping, but it’s one that can be partially overcome by honing our process. Even simple practices like coding cups and taking a few more breaks can vastly improve the precision of our data. This precision helps us learn even more about the coffees we roast and serve, and ultimately communicate more specifically about their uniqueness and value.

auell_oOlivia Auell is a chemist with a background as a barista and a focus on flavor. She holds degrees in Biochemistry and Spanish and currently studies flavor perception as an MS student at the University of Minnesota. Specialty coffee is an industry where her scientific background, her passion for great coffee, and her concern for equity and sustainability can coexist. After graduating, she hopes to be involved in the effort to develop coffee that is both resilient to climate change and high in flavor quality. 


Bartoshuk, L. M. (1993). The biological basis of food perception and acceptance. Food Quality and Preference, 4(1), 21-32.

Delwiche, J. F., Buletic, Z., & Breslin, P. A. (2001). Covariation in individuals’ sensitivities to bitter compounds: evidence supporting multiple receptor/transduction mechanisms. Perception & psychophysics, 63(5), 761-776.

Masi, C., Dinnella, C., Monteleone, E., & Prescott, J. (2015). The impact of individual variations in taste sensitivity on coffee perceptions and preferences. Physiology & behavior, 138, 219-226.

Eriksson, Nicholas et al. “A Genetic Variant near Olfactory Receptor Genes Influences Cilantro Preference.” Flavour 1.1 (2012): 1–7.

Stuckey, B. (2012). Taste: surprising stories and science about why food tastes good. Simon and Schuster.

Small, D. M., & Prescott, J. (2005). Odor/taste integration and the perception of flavor. Experimental Brain Research, 166(3-4), 345-357.

Shepherd, G. M. (2013). Neurogastronomy: how the brain creates flavor and why it matters. Columbia University Press.

Michel, C., Velasco, C., Fraemohs, P., & Spence, C. (2015). Studying the impact of plating on ratings of the food served in a naturalistic dining context. Appetite, 90, 45-50.

Okamoto, M., & Dan, I. (2013, March). Extrinsic information influences taste and flavor perception: A review from psychological and neuroimaging perspectives. In Seminars in cell & developmental biology, 24(3), 247-255.