top of page


Meet Gloria, my home-research mascot. I found her sleeping in my microscope when I pulled off the cover to count cells in a starter culture for our sparkling wine. Tree frogs frequently venture into the winery, where they hunt the earwigs who hitch a ride into our cellar on the grapes during crush.
Meet Gloria, my home-research mascot. I found her sleeping in the winery lab microscope.

Picture yourself enjoying a beautifully presented dinner in a hip new restaurant with your favorite song playing softly in the background. Now picture yourself waiting an hour in line to get the same food slapped haphazardly on a paper plate which you struggle to balance on one hand while you grasp for your sticky plastic cup of beer at a festival where you can't hear yourself think over the music. You would have vastly different expectations, emotions, and engagement with the food in the two scenarios. This can translate to differences in your direct experience of the food that are more significant than you might expect.

Research shows that our senses of taste and smell are influenced by our environment and disposition. We all know airplane food tastes bad, but it's not just the poor quality of the food- various things about the airplane cabin, including pressure, humidity, and noise, negatively impact how food tastes to passengers (links to an article pdf that was later published in the International Journal of Gastronomy and Food Science). Depression seems to occur together with a reduced sense of smell. And conversely, the euphoria of victory makes ice cream taste sweeter.

It's important to consider these facts when developing an experimental method. I wish to assess the influence of sound on wine without any other factors screwing up my results. This means eliminating biases and controlling my environment as much as possible.

Pitfalls of tasting without a method

If I worked without any particular method, it might look something like this:

I open a bottle of Merlot, sit down with a bunch of sound files, and think: what might be good with Merlot? Wait, I'm supposed to be unbiased! I move my mouse around the screen, trying to land on a file at random. But the "cicadas" sound clip keeps catching my eye, so I click on that one first.

I like cicadas, so I'm expecting something good to happen. I record my observations honestly, but my mood and expectations have already been primed for a positive experience. As I click through different sound files, I start to feel tired. It's hard work testing a dozen sounds in a row. The feeling of burn-out shows in my tasting notes. I compensate by swallowing more wine. Now the level of inebriation shows in my notes.

After several sessions, I develop an unconscious habit of always starting with the same few sounds, and saving others until last, reinforcing the biases above. I also develop expectations for specific sounds over time, based on what I've seen before.

Hell, the wine isn't static either. Wines open up with oxygen exposure, and if I'm always trialing the same sound last, I may start to believe the sound softens the wine and brings out aromas that actually emerged as a result of exposure to the atmosphere.

That scenario is an epistemic disaster. I don't pretend I can eliminate all possible biases from my subjective experience, but the whole point is to understand how sound influences the wine, not how wine tastes when I'm burnt out from tasting. By applying a little scientific rigor, I hope to shield myself from some of the factors that would bias me.

Blinders and controls

Here's the detailed run-down of how I plan to make my tasting sessions focused and consistent, with minimized bias. If it seems over-the-top, know that this method only approximates what is done in actual sensory analysis labs.

Location: Always at my dining room table.

Handling of wine: So that I can taste the same wine in different sessions, I'll split it up into several smaller vessels, all the same size with identical labeling. I will keep bottles in the fridge but allow them to reach room temp before tasting.

Identity of wine: I will blind myself somewhat to the identity of the wine. I'll pick either six red or six white candidate wines of different varieties. I'll let my girlfriend choose one at random, transfer it to smaller vessels, label those vessels with a random 3-number code, and hide the bottle until I'm done with my trials and ready to report results.

Tasting quantity: I will measure out the same amount of wine for each tasting.

Palate cleansing: I will eat a neutral cracker between sound/wine pairings.

Reps: I will taste the same wine with the same set of sounds on several occasions (however many small bottles I have- probably three or four). I will not review any of my notes from previous tastings until the trial is over.

Identity of Sounds: I will amass an archive of a few dozen sound clips. I will then let my awesome girlfriend pick six at random (I don't think I can handle more than six in a session), copy them into a new folder, and re-name them with random 3-number codes. She will note the real file names in her secret notebook which she keeps wherever she put the empty bottle of wine.

Order of sounds: I will program the names of the sound files into the flashcard program Anki and let it present them to me in randomized order. I will also include a flashcard for "no sound" and one for "earplugs".

Duration of tasting: I want to take my time with each pairing, but I should avoid giving excessive attention to some over others. I'll limit myself to five minutes each.

Notes and ratings: I'll rate descriptors on a scale, such as astringency, heat, fruitiness, earthiness, length, flavor intensity. I will reserve time for additional notes in case something entirely new pops up in the presence of the sound.

Evaluating my results

Did I write "hey, this noise makes this wine taste like strawberries" just once, or every time?

Did I consistently rate the wine as more earthy when playing a given sound, or did I only do it when I was annoyed by something so-and-so said?

Does the same sound always have the same general effect across many red wines? Or is there some particular, a specific aromatic note, that the sound manipulates?

I'll try to answer some of these questions with my data by looking at variance vs. consistency in the numbers I record.


What do you think of all this? Do you seen any glaring flaws in my method? Got some helpful advice to share? This is a work in progress, so I'm glad for any feedback you may have.

1 Comment

Sophia Mantheakis
Sophia Mantheakis
May 19, 2020

It would be interesting to catalogue the different sounds/tastes to study potential relationships, like if machinery sounds tend to impart a metallic taste or string instruments are usually associated with fruit notes.

bottom of page