As coffee cools, bitterness or other odors may stand out more than its roasted notes. Warm tea may feel comforting, while cold tea seems more invigorating. One study explored such differences by serving the same beverages at different temperatures.
In two tasting experiments conducted in the United States, coffee and green tea served at 65°C, 25°C, and 5°C differed in the aroma and flavor terms participants selected, the emotions they reported, and how much they liked them. The aim was to examine what changes with serving temperature when people evaluate a beverage, including relationships with gender and liking.
Hot-brewed beverages served at three temperatures
The coffee experiment involved 79 participants, and the green tea experiment involved 78. All regularly drank and liked the relevant beverage without added sugar, milk, or other condiments. The coffee group comprised 51 women and 28 men; the green tea group, 55 women and 23 men.
The researchers hot-brewed coffee from the same blend of beans and green tea from the same tea-bag product, cooling the samples as needed in water baths to reach 65°C, 25°C, and 5°C. Each 55-mL sample was served in a foam cup to reduce thermal stimulation of the hands. Every participant experienced all three temperatures, in randomized order. This was neither a study following a single hot cup as it cooled naturally over time nor a comparison of cold brewing with hot brewing.
Participants first evaluated emotions, then took a break before evaluating sensory attributes. After drinking each sample, they selected all applicable terms from 39 emotion words, such as “satisfied” and “calm.” Sensory evaluation used the same check-all-that-apply approach, with 49 terms for coffee and 57 for green tea covering aroma, flavor, taste, mouthfeel, and other attributes. They also rated liking on a nine-point scale.
This method reveals the proportion of people who selected a particular attribute. An increase in the number selecting a roasted note should not be confused with a measure of how many times stronger that note became.
More participants selected roasted flavor for hot coffee
For coffee, selection rates for six sensory terms varied with temperature. For example, roasted flavor during drinking was selected by 63% at 65°C, 37% at 25°C, and 33% at 5°C, making it more common at 65°C. Roasted aroma perceived through the nose was also selected more often at 65°C than at 5°C.
By contrast, bitter taste was selected by 84% at both 5°C and 25°C, compared with 67% at 65°C. More participants selected metallic and skunky flavors in the cold condition. Pungent aroma was selected more often at 5°C than at 25°C, but the difference from 65°C was not clear.
Selection rates for 16 emotion terms changed with temperature. At 65°C, terms such as “happy,” “satisfied,” and “pleased” were selected more often than at 25°C or 5°C, while “bored” was more common at 25°C than at 65°C. However, the cold condition was also associated with words such as “active” and “energetic,” so the relationship between temperature and emotion cannot be reduced to a single pleasant–unpleasant dimension.
Mean liking scores were 6.0 at 65°C, 4.2 at 25°C, and 4.0 at 5°C. For these participants and this coffee, the 65°C sample was liked more than either of the other two.
Green tea showed changes in floral aroma and mouthfeel too
For green tea, selection rates for 18 sensory terms varied with temperature. Floral aroma was selected by 27% at 65°C, 13% at 25°C, and 12% at 5°C, making it more common in the hot condition. Roasted aroma followed the same pattern, at 21%, 6%, and 1%, respectively.
Compared with 5°C, more participants selected herbal/herb-like aroma, mild/mellow flavor, and sweet taste at 65°C. Conversely, astringent mouthfeel and bitter aftertaste were selected more often at 5°C than at 65°C. The changes thus involved the broader experience in the mouth, not aroma alone.
Nineteen emotion terms differed with temperature. Hot green tea prompted more selections of terms such as “pleasant,” “satisfied,” and “secure” than the other temperatures, while cold green tea prompted more selections of “active” and “energetic.” “Joyful” was also selected more often at 5°C than at 25°C. Treating cold green tea simply as a condition that produces negative feelings would miss this part of the findings.
Mean liking scores were 6.3 at 65°C, 5.3 at 25°C, and 4.6 at 5°C, with significant differences in that order. However, only one product of each beverage type was tested, and the participants were US consumers. Whether the same ranking would hold with different tea leaves, roasts, brewing methods, or drinking cultures lies beyond this study’s scope.
Attributes associated with liking also depended on temperature
The researchers also compared liking scores between people who selected a given term and those who did not. For coffee, selecting roasted flavor was associated with higher liking at all three temperatures. For green tea, smooth mouthfeel was associated with higher liking at every temperature.
Other associations varied with temperature. For example, “active” and “energetic” were associated with higher liking for coffee at 5°C, whereas terms such as “pleasant” and “satisfied” were associated with higher liking at 65°C. These are relationships between responses, not direct tests showing that inducing an emotion causes liking to increase. They nevertheless suggest that what people like about the same product may change with temperature.
Gender-specific analyses need careful interpretation
Gender differences were also an explicit research objective. In separate analyses of women and men, more emotion and sensory terms varied with temperature in the female group, and both emotional and sensory responses were associated with liking. In the male group, significant associations with liking were found for emotion terms.
However, roughly twice as many women as men participated, and finding significance in only one group does not itself establish a significant difference between groups. For liking scores, neither the main effect of gender nor the interaction between temperature and gender was significant for either beverage. Rather than generalizing that men do not base liking on taste or aroma, these findings are better read as group-specific patterns requiring further investigation.
In this experiment, aroma, taste, and the temperature experienced in the mouth changed together. Volatile compounds and brain activity were not measured. Rather than isolating which sensory pathway produced an emotional change, the study captured how the experience of actually drinking a beverage varied with temperature.
Summary
For the coffee and green tea tested here, samples at 65°C were liked more than those at 25°C or 5°C. Roasted flavor was reported more often for hot coffee, and floral and roasted aromas more often for hot green tea. Terms expressing satisfaction and pleasantness were also more likely to be selected. Cold green tea, meanwhile, was associated with feelings such as being active and energetic.
The sensory attributes and emotions associated with liking varied with temperature and in the group-specific analyses. Evaluating a cup only while it is hot may not be enough to understand its characteristics. What aspects of the same beverage emerge at different temperatures, and how are they liked? The study offers a way to consider aroma, taste, and feelings together.
The study was supported in part by Hatch Act funding from the United States Department of Agriculture’s National Institute of Food and Agriculture. The authors declared no commercial or financial relationships that could constitute a potential conflict of interest.
Reference
Pramudya, R. C., & Seo, H.-S. (2018). Influences of Product Temperature on Emotional Responses to, and Sensory Attributes of, Coffee and Green Tea Beverages. Frontiers in Psychology, 8, 2264. Original research article
