Research note 017 — 2026-08-31 (expanded 2026-09-08)
Does any seasoning remove bitterness? — Sugar and salt
It can be tempting to add sugar or salt to make bitterness go away. Research exists showing both weaken bitterness, but the answer changes depending on what you add it to. In aqueous solutions with sugar added, the bitterness rating dropped for all five bitter substances tested. In aqueous solutions with sodium salt, many bitter substances weakened, but some did not change. Rather than deciding by the seasoning's name alone, you need to read what the research tasted separately from your own cooking.
On this page (11 sections)
- What this page establishes
- What were sugar and salt tested on?
- Sweetness weakened five kinds of bitterness
- Bitterness weakens sweetness back, too
- Some bitterness is not erased by salt
- Raising concentration does not always strengthen the same direction
- "Umami-bitter" splits into three layers
- The tool of suppression is not always taste
- Multiplication becomes an experiment table, not an answer
- Frequently asked
- What is not known
What this page establishes
- Adding sucrose has been confirmed, across multiple studies, to lower the bitterness ranking of several bitter substances, including urea, caffeine, and quinine.
- Conversely, one study has reported the opposite direction of suppression: adding caffeine makes sweetness feel weaker.
- Bitterness suppression by salt works differently depending on the substance, and one study has recorded an example with tannic acid in which raising the concentration reverses the direction from suppression to enhancement.
- A result showing that umami compounds inhibit a bitter receptor has been demonstrated at the cellular level, but whether bitterness weakens by the same degree on an actual tongue has not yet been confirmed.
What were sugar and salt tested on?
| What is added | What the study tasted | The answer obtained | The boundary when reading it for cooking |
|---|---|---|---|
| Sugar | An aqueous solution of five bitter substances, including caffeine | The bitterness ranking dropped in children and adults, across all five substances tested | Not a test that decided the right amount per dish, or the overall likability of a dish |
| Sodium salts | Aqueous solutions of five bitter substances, and their two-component mixtures | Bitterness weakened in most combinations, but tetralone, and mixtures containing it, did not change | It cannot be said that adding table salt removes the bitterness of every dish |
To check this in the kitchen, one option is to set aside a small portion of a dish and compare it with sugar or salt added against the plain portion. This is not a suggestion of the correct amount to add. It is an editorial way of testing the balance of the whole dish — sweetness and saltiness included, not bitterness alone.
Soaking, boiling, and pressing ingredients are kept separate from the story of adding a seasoning. For a comparison of food-based prep steps, see Kitchen Prep, Read Through Measured Bitterness.
Sweetness weakened five kinds of bitterness
In a study of children and adults, sucrose weakened the bitterness of urea, caffeine, denatonium, PROP, and quinine — all of them. Even so, the effect was not uniform. In adults, the size of the reduction varied with individual differences in sweet-taste perception and with the type of bitter substance. Rather than "add sugar and bitterness disappears," which bitterness, tasted by whom, under which conditions is itself part of the data.
Scope: Forced-choice comparison in aqueous solution with children and adults. Adults also completed a gLMS
Note: Not a claim that adding sweetness erases every kind of bitterness, or that it raises preference for a dish.
Show data
compound-sucrose —can_reduce_perceived_bitterness_of→ "Urea, caffeine, denatonium benzoate, PROP, and quinine"Scope: Mennella et al.'s child and adult study, and Keast et al.'s bitter-mixture study
Note: The genotype effect differs by compound, and the difference for PROP did not reach statistical significance.
Show data
concept-taste-mixture-suppression —has_perceived_bitterness_affected_by→ "The bitter substance tested, age, and individual differences related to TAS1R3 and TAS2R38"Scope: 0.1 mM quinine hydrochloride, in the human study cited
Note: The concentration required and the matrix are essential to interpretation.
Show data
compound-quinine —has_perceived_bitterness_suppressed_by→ "Relatively high concentrations of sucrose, aspartame, and sodium chloride"Bitterness weakens sweetness back, too
Weakening does not always run only from sweetness to bitterness. Subjects who drank caffeine-containing coffee rated not only that coffee itself but also a separately served sucrose solution given afterward as less sweet. There was no difference in the ratings for bitterness, sourness, saltiness, or umami. That a pathway exists running from the bitter side back to weaken sweetness shows, within the scope of the abstract, that the balance of power between components is not one-directional.
Scope: A human sensory panel test comparing caffeine added to decaffeinated coffee against a quinine control condition (n=107)
Note: The adenosine-receptor-mediated mechanism is the authors' proposal; the test itself did not directly measure a receptor-level mechanism. No extrapolation is made to coffee and sugar confectionery in general.
Show data
concept-taste-mixture-suppression —reduces_taste_preference_for→ "A human sensory test (Choo et al., 2017; n=107) comparing decaffeinated coffee containing 200 mg of caffeine (equivalent to 1 cup of strong coffee) against a quinine condition adjusted to a matching level of bitterness. In the caffeine condition, both the sweetened coffee itself and a subsequently presented sucrose solution were rated as less sweet. This treatment, however, had no significant effect on ratings of bitterness, sourness, saltiness, or umami. The authors offer, as one possible explanation for this result, that caffeine's antagonism of adenosine receptors may alter taste response (a contested mechanistic hypothesis)."Some bitterness is not erased by salt
Sodium salt suppressed many bitter substances and their binary mixtures, but in the same test tetralone was not suppressed. Even though most combinations can be roughly explained by the average suppression rate for each component, an exception remains. This is a reservation that is convenient for seasoning design: because bitterness is not monolithic, there is no all-purpose recipe of simply adding salt.
Scope: A human sensory test in aqueous solution, with intensities matched to a weak bitterness level
Note: This claim itself retains a counterexample to any rule that sodium salt uniformly erases bitterness.
Show data
compound-sodium-chloride —can_reduce_perceived_bitterness_of→ "Several bitter substances alone and in binary mixtures. Tetralone and mixtures containing it, however, were not suppressed"Scope: The 5 weak bitter substances tested by Keast et al., with salt and sucrose
Note: An approximation within the tested range, not a law covering complex foods or every bitter substance.
Show data
concept-taste-mixture-suppression —is_partially_predicted_by→ "For binary bitter mixtures, the average of the suppression rate for each component"The same selectivity appears at the cellular level too. In cells expressing the human bitter receptor TAS2R16, the response to the bitter glycoside salicin weakened in the presence of sodium chloride, but the responses of TAS2R14, TAS2R43, and TAS2R46, tested at the same time, did not change. The sensory-test exception in which only tetralone was not suppressed, and the cell-test exception in which only one receptor was selected, happened in different experiments with different substances, but they take the same shape: "salt's suppression is not indiscriminate."
Scope: Layer: cells. TAS2R14, TAS2R16, TAS2R43, and TAS2R46 expressed in HEK293T-Gα16gust44 cells, combinations of each receptor's principal agonist with NaCl concentrations (0/22/43/65/87/108/130 mM), calcium imaging.
Note: Shows at the cellular level that NaCl suppresses receptor response specifically for the target compound (the TAS2R16-salicin pair only), supporting the existing item description 「苦味抑制効果は対象化合物と濃度に依存する」 (in translation: "the bitterness-suppressing effect depends on the target compound and concentration"). A sodium gluconate control confirms this is an effect of the sodium ion rather than the chloride ion.
Show data
compound-sodium-chloride —can_decrease→ "In cells expressing the human bitter receptor TAS2R16, the response to the bitter glycoside salicin was significantly reduced in the presence of NaCl, showing concentration-dependent suppression over the 0-43 mM range. By contrast, the responses of TAS2R14, TAS2R43, and TAS2R46, tested at the same time, showed no change with NaCl concentration."That suppression itself also seems to change shape with concentration. A review states that on the low side, at sodium concentrations of 10 mM (0.058% sodium chloride) or below, sodium stimulates the salt-taste receptor while at the same time suppressing the action of TAS2R16 and enhancing signaling by the sweet-taste receptor T1R2/T1R3. This reads as: even for the same sodium, how it acts on the salt-, bitter-, and sweet-taste receptors changes with where the concentration sits.
Scope: Layer: cell (receptor level). Literature (a description from a review).
Show data
sensation-salty —reduces→ receptor-tas2r16Raising concentration does not always strengthen the same direction
In a test of a substitute salt mixing KCl, NaCl, and L-arginine, raising the NaCl proportion from 70:20:10 to 55:35:10 lowered the bitterness score, but the significant difference appeared between the two lower ratios and the two higher ratios, not between each of the four steps individually. Here, concentration and the direction of suppression correspond straightforwardly.
Scope: Layer: solution. Aqueous solutions of KCl:NaCl:L-Arg mixing ratios 70:20:10/65:25:10/60:30:10/55:35:10 and a control (100% NaCl), at concentrations of 0.5/1.0/1.5% w/v, R-Index method (6 replicates) with a trained panel of 20.
Note: L-arginine (10%) is fixed across all conditions, so what can be isolated is the effect that accompanies changes in the KCl proportion. This does not show the effect of NaCl alone on bitter substances other than KCl.
Show data
compound-sodium-chloride —can_reduce_perceived_bitterness_of→ "In a KCl:NaCl:L-arginine mixed-salt solution, raising the NaCl proportion from the 70:20:10 and 65:25:10 groups to the 60:30:10 and 55:35:10 groups significantly lowered perceived bitterness intensity (mean rank score) by the R-Index method (the significant difference was shown between the two groups)."Yet the same operation of "raising the concentration" also has an example where it flips in the opposite direction. The experiment adding tannic acid to a quinine solution is a case where the sign of the multiplication reverses with concentration. While the astringent side's concentration is low, it acts to suppress bitterness; raising the concentration further undoes the suppression, and the bitterness score returns to a level above the control. The intuition that the more you increase the suppressing side, the more it works does not hold here.
Scope: The specific concentrations used in the cited human study.
Note: This is presented as an example of interaction between descriptors, not as proof that astringency is a form of bitterness.
Show data
compound-quinine —can_show_higher_bitterness_scores_with→ "Tannic acid suppresses quinine bitterness to about 61% of the control up to 0.05 (w/v)%, but at 0.15 (w/v)% this reverses and the bitterness score exceeds the control (the direction of the effect flips with dose).""Umami-bitter" splits into three layers
MSG, IMP, and L-theanine suppressed the response to a bitter ligand in cells expressing hTAS2R16. Meanwhile, in an actual mee soto broth, adding MSG or a nucleotide-containing flavor enhancer to a reduced-salt formulation increased umami and other qualities while lowering bitterness and sourness. The former is a single-receptor cell test; the latter is human sensory evaluation of a dish that also involves aroma and fat. Both are interesting. But they are not the same kind of evidence.
Scope: Calcium response to salicin and related ligands in HEK293T cells expressing hTAS2R16
Note: A single receptor in a cell system. Does not directly demonstrate umami-bitter in human sensory perception or in dishes as a whole.
Show data
compound-monosodium-glutamate —antagonizes→ receptor-tas2r16Scope: The same hTAS2R16-expressing cell system
Note: Reported as partial, non-surmountable antagonism. Not extrapolated to human sensory perception.
Show data
compound-inosine-monophosphate —antagonizes→ receptor-tas2r16Scope: The same hTAS2R16-expressing cell system
Note: Reported as partial, non-surmountable antagonism. Not extrapolated to the overall bitterness of tea or dishes.
Show data
compound-l-theanine —antagonizes→ receptor-tas2r16For L-theanine, the food-side observation is even more tangled. In a sensory and compositional analysis of green tea extracted in three successive infusions, some amino acids including L-theanine, together with catechins and caffeine, showed a positive correlation with the side of the first extract — the one rated as having the strongest astringency and bitterness. A substance that suppresses the receptor response in cells is, within the food matrix of tea, moving together with the strongly bitter and astringent side. Even for the same substance, the picture drawn changes depending on which layer you look at.
Scope: Layer: food. Green tea extracted in 3 successive infusions (GT1-GT3), five-taste sensory evaluation by electronic tongue, Pearson correlation analysis, and PLSR.
Show data
compound-l-theanine —has_perceived_bitterness_associated_with→ ingredient-green-teaScope: A difference-from-control test using Singaporean mee soto broth
Note: A specific dish in which decreased bitterness and sourness were reported together with increases in umami and other attributes. Not extended to home cooking formulations in general.
Show data
dish-mee-soto-broth —has_perceived_bitterness_suppressed_by→ "Adding MSG, or an MSG-nucleotide blend seasoning, to a reduced-salt formulation"Digging one level deeper into the mee soto broth experiment also reveals which side plays the leading role in the suppression. Formulations that reduced sodium chloride alone by 22% or by 31%, without adding a flavor enhancer, showed no significant difference in bitterness from the control. Bitterness dropped significantly only in formulations that reduced salt by 40% and also added MSG or a nucleotide-containing flavor enhancer. Here it becomes clear which comes first: reducing salt by itself did not move bitterness, adding umami did.
Scope: Layer: food. Difference-from-control test, a panel of 24 (mee soto broth panel), values from Table 7.
Show data
dish-mee-soto-broth —has_perceived_bitterness_unchanged_by→ "Against a bitterness sensory intensity of 5.94 for the control (blind, normal salt) mee soto broth, formulations with NaCl alone reduced by 22% (no flavor enhancer added) scored 5.60, and those reduced by 31% scored 5.63, neither significantly different from the control. Bitterness was significantly lower only for formulations with a 40% NaCl reduction plus 0.20% or 0.40% Ajiplus or MSG added (4.33-4.85, all P<0.05)."The tool of suppression is not always taste
Adding another flavor compound is not the only way to weaken quinine's bitterness. In an experiment measured with an electronic tongue, adding a sample of beef protein hydrolysate treated by the Maillard reaction to a quinine solution lowered the bitterness score by up to 38%. Here, what measures bitterness is not a human tongue but an instrument, the electronic tongue, and what is moving is not a new flavor compound but a component generated by a heating reaction.
Scope: Layer: solution. 5 mg/mL of the glycated product (AGE sample) was added to a quinine solution (the source does not state the concentration of quinine itself at the time of the electronic-tongue measurement). Electronic-tongue measurement using a partial least squares regression model for bitterness.
Note: An experimental system that used an electronic tongue to confirm a bitterness-modifying effect of the Maillard reaction (glycation).
Show data
compound-quinine —has_e_tongue_bitterness_reduced_by→ "Glycated beef protein hydrolysate (a Maillard reaction product formed by reacting Alcalase/chymotrypsin hydrolysates of beef protein with glucose at 120°C for 30 minutes; BPAH-AGEs/BPCH-AGEs)"One more example within the range of bitterness and masking: coating quinine sulfate pellets. Increasing the coating proportion to 10%, 20%, and 30% lowered the dissolution rate over the first 5 minutes to 9.2%, 5.9%, and 2.1% respectively, corresponding to a drop in the bitterness score from a taste sensor. This is not a story of covering the bitter substance itself with another taste, but of physically reducing the amount that touches the tongue. This example shows that the partner in the multiplication is not always "taste."
Scope: Layer: food. Quinine sulfate pellets by extrusion-spheronization, 3 levels of Eudragit EPO coating concentration, bitterness evaluation by the Astree electronic-tongue system.
Show data
process-pharmaceutical-taste-masking —reduces→ compound-quinineMultiplication becomes an experiment table, not an answer
Rather than stopping at "sweet-bitter," "salty-bitter," and "umami-bitter" as candidate product names, turn them into an experiment table that varies the bitter substance, the coexisting taste, concentration, temperature, order, food matrix, and the eater. What multiplication means here is not copy that links two words, but a method for increasing the number of conditions to explore the outline of bitterness.
Frequently asked
Does putting sugar in coffee really reduce bitterness?
In human tests using aqueous solutions, adding sucrose is reported to have lowered the bitterness ranking for children and adults across all 5 substances tested — urea, caffeine, denatonium benzoate, PROP, and quinine. Another human test also records that reducing the bitterness of a 0.1 millimolar quinine solution by about 80% required 800 millimolar of sucrose. These are experiments in aqueous solution, however, and this inventory does not include a source that directly measured how much sugar weakens bitterness within the food that is coffee. Conversely, an opposite result has also been reported: adding caffeine makes the sweetness of the coffee itself, and of sugar water drunk afterward, feel weaker.
If a dish is bitter, does adding salt make the bitterness disappear?
In experiments using aqueous solutions, sodium-containing salt suppressed bitterness in most cases, but the degree of suppression differed by the type of salt, and potassium chloride is reported not to have suppressed the bitterness of urea. In experiments using cells, the response of salicin at the bitter receptor TAS2R16 weakened according to the concentration of table salt, while the response of another receptor tested at the same time did not move even as the salt concentration was changed. In other words, whether salt works depends on the combination of bitter substance and receptor. How much this receptor-level result corresponds to a weakening of bitterness in an actual dish has not been confirmed within the scope of this inventory. When testing this in a dish, set aside a small portion before adding salt to the whole batch, and check separately whether the whole — saltiness included — becomes preferable even if the bitterness weakens.
What is not known
- Whether sweetness, saltiness, and umami act in the same non-monotonic way (reversing direction with concentration) on bitter substances other than quinine has not yet been confirmed.
- The mechanism by which tannic acid suppresses quinine's bitterness and then turns to enhancement has not been confirmed; the authors themselves state as unproven the possibility that it is a confusion with descriptions of astringency.
- How much the cell-level result that umami compounds inhibit TAS2R16 corresponds to an actual weakening of bitterness on a human tongue has not been confirmed within the scope of this inventory.
- No source has been found that measured, under the same concentration conditions at the same time, both the result that caffeine weakens sweetness and, conversely, the result that sucrose weakens caffeine's bitterness.
Related: "Bitter" Is Not a Single Taste / Bitterness Has a Clock / taste-mixture suppression / TAS2R16 / mee soto broth / Hypothesis Box
Continued: Bitter Crossings — a map and definitions page for eight names
How to cite this page
Aoki, M. (NIGAMI LAB) (2026). Does any seasoning remove bitterness? — Sugar and salt. https://nigamilab.com/en/notes/taste-is-not-addition/
Each claim card has a stable anchor (#clm-…). The dataset is CC BY 4.0; see the data page. Dataset DOI: 10.5281/zenodo.22739101