DOES ADDING MILK TO COFFEE REDUCE BITTERNESS? – OR IS IT JUST AN ILLUSION?
A typical cup of coffee with milk often feels milder than black coffee. Bitterness is less pronounced, acidity may become more pleasant, and milk, caramel, or vanilla flavors become more apparent. But what actually happens when milk is added to coffee?
Does milk reduce volatile compounds or simply change the way we perceive them?
A 2025 study published in the Journal of Agricultural and Food Chemistry approached this question by combining temporal dominance of sensations (TDS) with headspace analysis of volatile compounds in the breath of coffee drinkers. The results showed a significant difference between what people perceive and what the analytical equipment records.
1. How was the study conducted?
The study, titled Effects of Roasting Level and Milk Addition on In Vivo Aroma Release and Perception of Coffee, was conducted by Nina Cleve and colleagues, published online on May 20, 2025, and featured in the June 2025 issue of the journal.
The research team investigated the impact of two factors: roasting level and type of milk added to coffee.
- Coffee: Arabica, medium roast, and dark roast.
- Milk: cow's milk and oat milk.
- Milk ratio: 30% by weight (w/w).
- Perception assessment: Temporal Dominance of Sensations (TDS), recording which sensation dominated at each time point during consumption.
- Flavor analysis: PTR-ToF-MS, measuring volatile compounds in the nose-space region during consumption.
The key aspect of this design is that participants were evaluated while drinking coffee, rather than simply smelling samples or evaluating after consumption. This allowed the researchers to observe changes in perception over time and correlate them with the chemical signals measured.
2. Roasting level determines the initial flavor profile
Before considering the impact of milk, it's essential to understand that medium-roast and dark-roast coffees have different sensory profiles.
In the medium-roast sample, roasty and bitter sensations were prominent. When switching to dark roast, bitter and burnt sensations became more pronounced.
This is consistent with the chemical changes that occur during roasting. As temperature and roasting development increase, many aroma compounds are formed or transformed. Some compounds related to smoky, strong roasty, and bitter flavors become more pronounced, while fruit and light flavor characteristics may decrease.
Therefore, even with the same type of milk, adding it to medium-roast and dark-roast coffee does not necessarily yield the same results.
3. How does adding milk affect the perception of bitterness?
The TDS results showed that, without milk, bitterness was the dominant sensation in both roasting levels.
In the medium-roast sample, 64% of participants reported bitterness as the dominant sensation in the no-milk sample. When adding cow's milk, this percentage dropped to 30.8%; with oat milk, it was 36%.
In the dark-roast sample, the corresponding percentages for the no-milk sample were 69%. When adding cow's milk, the figure was 46.2%; with oat milk, it was 31%.
These numbers reflect the results of the perception test under controlled conditions, not the reduction in the concentration of bitter compounds in coffee. They also do not imply that all coffee milk recipes outside of this study will reduce bitterness at the same rate.
What's worth noting is that milk doesn't just make coffee taste milder. It also introduces new sensory characteristics, making the overall flavor perceived in a different direction.
4. Milk from cows and oats create different flavor notes
Research found that cow's milk adds a characteristic feeling to the product, while oat milk can highlight caramel and vanilla associations.
In the medium-roasted sample, flavor notes related to cow's milk were recorded in 28% of the tasters, while the dark-roasted sample was 36%. With oat milk, caramel or vanilla notes were recorded in 38% and 33%, respectively.
This is the percentage of respondents who felt a sensation in the experiment, not the actual amount of caramel or vanilla in the milk.
This also helps explain why some people feel that coffee with milk tastes sweeter, even if they don't necessarily add sugar. The aroma can affect how the brain interprets the taste. A scent that evokes caramel or vanilla can make the overall taste feel sweeter and more pleasant.
5. What does the aroma detector show?
This is the most interesting part of the study.
The research group used PTR-ToF-MS to analyze volatile compounds in the breath through the nose when participants drank coffee. This method allows for the tracking of the release of flavor compounds in real consumption conditions.
In the study, 47 volatile compound signals were surveyed. The results showed that adding milk significantly changed the perception of flavor, but did not create a uniform decrease in the amount of coffee-derived volatile compounds released in the breath.
Some individual signals still changed depending on the type of milk and the experimental conditions. Notably, the research group recorded differences in some sulfur-containing compounds, including a signal related to the burnt smell.
Therefore, it's not correct to interpret the results as adding milk having no effect on the release of flavor. A more suitable conclusion is that the effect of milk is not uniform on all measured compounds.
6. Why the amount of flavor detected and the taster's perception don't match?
Coffee flavor doesn't just depend on the amount of volatile molecules released. It also depends on how these molecules interact with each other, the presence of fat and protein, temperature, the structure of the drink, and the processing of sensory information by the olfactory-taste system.
When adding milk, fat and protein can interact with some flavor compounds through mechanisms such as hydrophobic interaction, hydrogen bonding, and molecular force. These interactions can change the distribution, availability, and release rate of each compound.
At the same time, the unique scent of milk also participates in the overall perception. When the milk scent, caramel, or vanilla appears, it can reduce the prominence of some inherent characteristics in the coffee.
This is why one person may feel that the coffee smell is milder or masked, while the device does not record a corresponding decrease in the overall group of volatile compounds.
In other words, the released aroma is not entirely synonymous with the aroma perceived by humans.
7. A fascinating result about oat milk and protein
In the comparative presentation of the study, the oat milk sample had a lower protein content than cow's milk, while the fat content of the two samples was relatively close.
This raises a question: it's not possible to just look at the total amount of protein or fat to accurately predict how milk will change the flavor of coffee.
The protein composition, emulsion system structure, type of fat, stabilizers, and product formula can all affect the interaction between milk and flavor compounds.
Therefore, it's not possible to conclude that milk with more protein will certainly retain or lose more coffee flavor. To determine this, separate experiments are needed on each milk formula and each type of coffee.
8. What does this study mean for roasters and brewers?
The study provides a useful perspective when developing coffee for milk-based drinks.
For the roaster: do not evaluate a roast profile solely by tasting black coffee. A good-smelling roast may not necessarily create the same balance when combined with milk.
For the barista: the ratio of milk, type of milk, temperature, and coffee base can all change the final experience. Tasting the same coffee with cow's milk and oat milk is a practical way to notice the differences.
For the manufacturer: when building a coffee milk product, it is necessary to evaluate both taste and aroma perception. Do not rely solely on a single chemical index to predict sensory quality.
9. Limitations to be noted
This study provides valuable data but does not represent all types of coffee and milk formulas.
- The experiment focused on medium-roasted and dark-roasted Arabica, not directly proving the same results for Robusta or Arabica-Robusta blends.
- The two types of milk surveyed are cow's milk and oat milk, not representing the entire range of plant-based milks on the market.
- The milk ratio used is 30% by weight, which cannot be applied directly to all latte, cappuccino, or Vietnamese coffee milk formulas.
- PTR-ToF-MS analysis recorded signals of volatile compounds in breath, not a direct measurement of the entire remaining aroma in the coffee cup.
- The results of a single sensory study are not sufficient to confirm the preferences of all consumers.
Therefore, the results should be considered as evidence of how milk affects perception and the release of aroma in a specific experimental condition, rather than a rule applicable to all coffee drinks.
Coffee aroma is not just in the cup
A good coffee milk drink is not simply a cup of black coffee diluted with milk. It is a new flavor system, in which compounds from coffee and milk participate to create the experience.
A 2025 study showed that bitterness perception may decrease, new notes may appear, but the change in perception does not necessarily accompany a corresponding decrease in the total volatile compound content.
This is also a reminder that when researching coffee quality, it is necessary to distinguish between three factors: chemical composition, release of volatile compounds, and real-world consumer perception.
References
- Cleve, N., Gonzalez-Estanol, K., Khomenko, I., Pedrotti, M., Cappellin, L., Biasioli, F., & Beauchamp, J. (2025). Effects of Roasting Level and Milk Addition on In Vivo Aroma Release and Perception of Coffee. Journal of Agricultural and Food Chemistry, 73(22), 13792–13808.
- DOI: 10.1021/acs.jafc.4c12852.
- Full-text article: PubMed Central – Effects of Roasting Level and Milk Addition on In Vivo Aroma Release and Perception of Coffee
