A Hard-to-Detect Ingredient Can Make a Blend Easier to Notice

Even a compound too dilute to detect readily on its own can make a mixture easier to notice when added to another odor. A 2008 study found that adding very dilute acetic or butyric acid to the aroma compound maple lactone increased detectability beyond the level predicted by simply combining independent detection probabilities.

What makes this interesting is that the study went beyond showing that adding something made an odor easier to detect. It accounted for the possibility of detecting the added compound itself, then tested for an additional benefit from mixing.

Too dilute to notice does not mean having no effect

An odor detection threshold is not a boundary below which sensation disappears completely. It is a reference concentration at which a specified proportion of responses is correct in a defined task. Odors can sometimes be detected below that threshold.

The researchers therefore reasoned that an increased detection rate after adding a weak component would not, by itself, demonstrate synergy. A mixture’s detection rate can rise simply because there are more opportunities to detect either component independently. Estimating that contribution from measured detection rates for the individual components was central to the study.

Seventeen participants selected the two odor presentations out of five

Participants were 17 healthy adults aged 23–47. They had participated in earlier experiments using the same method and also practiced with maple lactone to confirm that their sensitivity fell within the concentration range the apparatus could deliver.

The three compounds were maple lactone, acetic acid, and butyric acid. Maple lactone’s chemical name is 2-hydroxy-3-methyl-2-cyclopenten-1-one. An apparatus diluted each compound’s vapor with air, and concentrations delivered to the nose were calibrated using gas chromatography–mass spectrometry, or GC-MS. The researchers also checked that each compound’s concentration was the same in mixtures and single-compound presentations. The lowest concentrations, however, were below the analytical instrument’s sensitivity and were estimated by extrapolating calibration curves.

Each trial consisted of five 2.5-second presentations: two contained the odor at the same concentration and three contained clean air. Participants could resample if needed and selected the two presentations they thought contained odor. There were six concentration levels. Testing progressed from low concentrations upward, with breaks to limit adaptation. Each concentration of each compound or mixture was tested in 12 trials in total to establish the relationship between concentration and detection probability.

Acetic and butyric acid were first measured alone to determine two concentrations below each individual’s threshold. On average, these were about 34% and 15% of the threshold concentration for acetic acid, and 30% and 14% for butyric acid. These fixed acid concentrations were then added to each of the six maple lactone concentrations. Estimated detection probabilities for the acids alone, corrected for chance performance, were about 3–10%.

Detection improved even after accounting for the added component

Adding either acetic or butyric acid made the odor easier to detect on average than maple lactone alone. There was no clear difference in detection rate between the two acid concentrations selected for this study.

The researchers next calculated the detection rate expected if the two components were detected independently, without influencing each other. The idea is that noticing at least one component is enough to detect the mixture. For example, if one component has a 50% detection probability and the other 10%, the model predicts 55%. This is an illustration of the calculation, not an observed result from the study.

For both acetic and butyric acid, actual mixture detection rates significantly exceeded that prediction overall. The authors used “synergy” to describe the improvement beyond this prediction. They reported that another additive model gave essentially the same conclusion under these conditions.

The benefit was greatest within a particular concentration range

The additional benefit did not occur uniformly across all maple lactone concentrations. At the lowest concentration, detection did not exceed the prediction based on independence. The largest differences occurred at intermediate concentrations, while the difference narrowed at the highest concentration.

At high concentrations, detection is already quite accurate, leaving less room for improvement. Why synergy did not appear at the lowest concentration was unclear; the authors also noted the difficulty of measuring low detection rates. Rather than implying that adding a little always works, the findings suggest that the effect depends on the combination and concentration.

Studying compounds alone can miss how a blend works

The authors related these results to aroma analysis. Separating compounds and smelling them individually is a powerful way to identify substances that contribute to odor. But that approach alone may underestimate the importance of compounds that are barely noticeable individually yet aid detection when combined with others.

Possible sites of synergy discussed included receptor cells in the nose, central neural processing, and molecular interactions before or after entry into the nose. This experiment measured human responses and stimulus concentrations, not which mechanism was responsible.

The study also involved only 17 people and a limited set of combinations. It tested whether an odor could be detected, not sweetness, deliciousness, or pleasantness. Developing formulations that make foods or perfumes more appealing would require those additional assessments.

Conclusion

Adding very dilute acetic or butyric acid made mixtures containing maple lactone easier to detect within a particular concentration range. The improvement exceeded predictions based on independently detecting either component. Compounds that are hard to notice alone can therefore have effects that emerge in mixtures.

At the same time, synergy was absent at the lowest concentration, and the additional benefit diminished at high concentrations. Understanding a blend requires looking beyond component names and their individual odors to consider their partners and the concentrations at which they meet.

The study was supported by Ogawa & Co., Kraft Foods, the US NIH, and NIEHS. One author was affiliated with Ogawa & Co. at the time. No separate conflict-of-interest statement was found in the article.

Reference

Miyazawa T, Gallagher M, Preti G, Wise PM. Synergistic Mixture Interactions in Detection of Perithreshold Odors by Humans. Chemical Senses. 2008;33(4):363–369. doi:10.1093/chemse/bjn004