Taste Threshold

Abstract

Taste threshold is a type of sensory threshold: the minimum concentration of a dissolved substance that the gustatory system can detect or identify. Cognitive psychology distinguishes three limits that the single word threshold conceals: a detection threshold, the faintest concentration a taster can tell from plain water; a recognition threshold, the higher concentration at which the taste quality can be named; and a difference threshold, the smallest change in concentration that can be noticed. These limits are estimated with the same psychophysical methods used across the senses, and they vary by orders of magnitude between the five basic tastes and, strikingly, between individuals. This article covers how taste thresholds are defined and measured, how they differ across the basic tastes, and why genetics and age make one person's subthreshold another person's clear sensation.

Keywords: taste threshold, detection threshold, recognition threshold, gustatory psychophysics, supertaster

What a Taste Threshold Is

A taste threshold is the boundary between a concentration of a tastant that produces no reportable sensation and one that does. Because taste is a chemical sense, the stimulus is a concentration, usually expressed in moles per litre, and the threshold is the concentration at which a sensation first appears. The sensory system does not switch on sharply at a single value; detection rises gradually with concentration, so the threshold is defined statistically as the concentration detected on a fixed proportion of trials, conventionally half.

The single term threshold hides a distinction that matters for both theory and measurement, because tasting that something is present is not the same as tasting what it is. The detection threshold, sometimes called the absolute threshold, is the lowest concentration a person can reliably tell apart from plain water, without being able to say which taste it is. The recognition threshold is higher: it is the lowest concentration at which the quality can be correctly named as sweet, sour, salty, bitter, or umami. Above both sits the difference threshold, the smallest change in concentration a person can notice between two suprathreshold samples, which is governed by the same Weber-fraction logic that applies across the senses (Webb, Bolhuis, Cicerale, Hayes, & Keast, 2015). These are summarized in Table 1.

Taste thresholds are not fixed constants of the species but population statistics with enormous spread. A concentration of quinine that one person detects easily can be entirely tasteless to another, and the same holds for the bitter compounds propylthiouracil (PROP) and phenylthiocarbamide (PTC), whose thresholds split people into distinct groups (Bartoshuk, 2000). This variation is not noise around a true value; it is a real feature of the sense, and much of the modern study of taste thresholds is the study of why the spread is so large.

Measuring the Taste Threshold

The threshold cannot be read off a single sip, because any one judgment confounds sensitivity with the taster's willingness to say yes. Classical psychophysics solved this with procedures that present many concentrations across many trials and extract the threshold from the pattern of responses. The method of limits presents a series of ascending or descending concentrations and records the point at which the report switches between absent and present; averaging the switch points across several runs gives the threshold. Because an ascending run and a descending run bias the switch point in opposite directions, the two are alternated and averaged to cancel the bias.

The method of constant stimuli is more laborious but less biased: a fixed set of concentrations is presented many times each in random order, the proportion detected is plotted against concentration to form a psychometric function, and the threshold is read off as the concentration corresponding to the chosen criterion proportion. Adaptive staircases combine the efficiency of the method of limits with the rigor of forced choice by changing the concentration trial by trial in response to the taster's answers, stepping down after a correct detection and up after a miss, so that the series converges on the threshold and spends most of its trials near the value of interest. A two-alternative forced-choice version, in which the taster must say which of two samples carries the tastant rather than whether a taste is present, removes the criterion problem entirely, because guessing yields a known chance rate (Webb et al., 2015).

The difference between detection and recognition shows up directly in these functions. The psychometric curve for simply detecting that a sample differs from water rises at a lower concentration than the curve for correctly naming the quality, so there is a band of concentration in which a taster knows something is present but cannot yet say what, as shown in Figure 1 (Bartoshuk, 2000).

Figure 1

Detection and Recognition Psychometric Functions on a Concentration Axis

Thresholds Across the Basic Tastes

There are five well-established basic tastes, each served by its own receptor machinery, and their detection thresholds span several orders of magnitude. Bitter compounds are detected at the lowest concentrations, often in the micromolar range, because bitterness evolved as a warning system for potential toxins and a sensitive alarm is a useful one (Breslin, 2013). Sweet and umami, which signal calories and protein, are detected at much higher concentrations, in the millimolar range, while sour and salty fall in between. The ordering is not arbitrary: it tracks the biological cost of a miss, with the tastes that warn of danger set to trip at the faintest signal.

The molecular basis of these thresholds became clear over two decades of receptor biology. Sweet, umami, and bitter are transduced by G-protein-coupled receptors: the T1R family detects sugars and amino acids, and the roughly twenty-five T2R receptors detect a huge chemical range of bitter compounds (Chandrashekar, Hoon, Ryba, & Zuker, 2006). The identification of an amino-acid taste receptor assembled from T1R subunits gave umami a molecular identity and confirmed it as a basic taste rather than a flavor blend (Nelson et al., 2002). Umami itself had been proposed as a distinct quality nearly a century earlier, when its signature compound glutamate was isolated from kelp broth (Ikeda, 2002). The broader logic of how taste cells transduce chemicals into neural signals, and how those signals keep their quality labels on the way to the brain, was set out in reviews of taste transduction and of the taste receptor genes that specify sensitivity (Lindemann, 2001); (Bachmanov & Beauchamp, 2007). Each basic taste is carried by a dedicated population of receptor cells, a labeled-line arrangement conserved from insects to mammals (Yarmolinsky, Zuker, & Ryba, 2009), and the synaptic wiring of the taste bud that reads these cells out has since been mapped in detail (Roper & Chaudhari, 2017).

A candidate sixth taste illustrates how a threshold is used to adjudicate what counts as basic. Free fatty acids produce a detectable oral sensation at low concentrations, distinct from the texture of fat, and the demonstration that people have a genuine detection threshold for this taste, named oleogustus, was a central part of the case for treating fat perception as a taste rather than a mouthfeel (Running, Craig, & Mattes, 2015).

Individual Differences in Taste Thresholds

The most striking fact about taste thresholds is how much they differ between people. The textbook case is the bitter compound PTC, and its relative PROP, for which the population splits into non-tasters, who need high concentrations to detect any bitterness, and tasters, who detect it at far lower concentrations. Linda Bartoshuk refined this further by identifying supertasters, who experience intense bitterness from concentrations a non-taster cannot detect at all, and showed that these differences extend beyond bitterness to the perceived intensity of many tastes and oral sensations (Bartoshuk, 2000). Much of the PTC and PROP variation traces to the bitter receptor gene TAS2R38, though sensitivity is not fully explained by that single gene, and the phenotype depends on additional factors including the density of taste papillae on the tongue (Hayes & Duffy, 2007).

Comparing intensity across such different people created a measurement problem that shaped the whole field: if a supertaster and a non-taster both rate a solution as strong, they do not mean the same thing by the word. The labeled magnitude scale was developed to place taste intensity on a scale anchored to the strongest imaginable sensation of any kind, so that ratings could be compared validly across individuals with very different sensitivities (Green et al., 1996). The relationships among the various measures of taste function, detection thresholds, recognition thresholds, and suprathreshold intensity ratings, turn out to be modest, which means a low threshold does not guarantee intense suprathreshold experience and the measures capture partly separate aspects of the sense (Webb et al., 2015).

Beyond genetics, the clearest systematic influence on taste thresholds is age. Thresholds rise across the lifespan, but not uniformly: sensitivity to some tastes declines more than to others, so aging produces specific rather than global losses, with salt and bitter detection often more affected than sweet (Mojet, Christ-Hazelhof, & Heidema, 2001). A large study across the adult age range confirmed that detection of all five basic tastes declines with age, with the decline differing by sex and modulated by genetic variants, tying the aging effect back to the same receptor-gene variation that drives individual differences in the young (Barragan et al., 2018).

Worked Example: Reading a Threshold From a Psychometric Function

Suppose a laboratory estimates a detection threshold for sodium chloride using the method of constant stimuli. It presents five concentrations, each many times in random order against a water blank, and records the proportion of trials on which the taster correctly identifies the salt sample in a two-alternative forced choice. The proportions correct are 0.10 at 0.3 millimolar, 0.30 at 1 millimolar, 0.55 at 3 millimolar, 0.85 at 10 millimolar, and 0.98 at 30 millimolar.

The detection threshold is defined as the concentration yielding 50 percent correct above the guessing rate. Because the function is sigmoid on a logarithmic concentration axis, the threshold is found by interpolating between the two concentrations that bracket the criterion, here 1 millimolar at 0.30 and 3 millimolar at 0.55. On a base-ten log scale, log of 1 is 0 and log of 3 is 0.477. The fraction of the way from the lower to the upper proportion needed to reach 0.50 is the difference 0.50 minus 0.30, which is 0.20, divided by the difference 0.55 minus 0.30, which is 0.25, giving 0.80.

The log threshold is therefore 0 plus 0.80 times the log span of 0.477, which is 0.382, and the threshold concentration is ten raised to the 0.382 power, which is 2.41 millimolar. The recognition threshold, estimated the same way from a separate naming task, would fall at a higher concentration, reproducing the gap shaded in Figure 1. The MethodOfLimitsDemo runs the alternative staircase procedure on a simulated taster, and its reversal points converge on the same kind of value.

Discussion

The study of taste thresholds has moved from a search for the single number that would characterize human sensitivity to a recognition that there is no such number. The threshold for any tastant is a distribution across people, and the width of that distribution is as informative as its center. The three thresholds, detection, recognition, and difference, separate the questions of whether a taste is present, what it is, and how much it has changed, and keeping them distinct is what allows a clean psychophysics of a chemical sense that is otherwise easy to confound with smell and texture.

The field's durable achievement is methodological in two senses. First, it imported the forced-choice and scaling discipline of classical psychophysics into a sense where response bias is especially tempting, giving thresholds that can be compared across laboratories and across people. Second, by taking individual variation seriously rather than averaging it away, it connected a behavioral measure, the detection threshold, to a specific molecular cause, receptor-gene variation, more directly than almost any other perceptual threshold. The taste threshold is where a sip of quinine, a psychometric function, and a single nucleotide polymorphism meet.

Current Directions

The active frontier is genetic, and it is reshaping how thresholds are understood. Genome-wide and candidate-gene studies are mapping the receptor variants that set detection thresholds for each basic taste, and a systematic review of this literature documents both well-replicated associations, such as TAS2R38 with bitter detection, and the far larger number of weaker or unreplicated leads that remain to be sorted out (Dioszegi, Llanaj, & Adany, 2019). A recurring surprise is that genetic variation in one taste correlates with perception of another: a large twin study found that the covariation between sweet and bitter perception is substantially heritable, pointing to shared genetic influences on taste that cut across the classical quality boundaries (Hwang et al., 2016). Work on PROP taster status continues to show that it modulates perception of the other basic tastes and can be shifted by physiological manipulations, underlining that a threshold is a state of the whole system rather than a fixed property of one receptor (Melis & Tomassini Barbarossa, 2017). The practical payoff is a move toward individualized accounts of taste: once a person's receptor genotype and taster status are known, their thresholds for specific compounds become partly predictable, with implications for diet, food preference, and the specific losses that accompany aging.

Common Misconceptions

There is a single taste threshold for a substance.

There are at least three, and they differ. The detection threshold is where a sample is first told apart from water, the recognition threshold is the higher concentration where its quality can be named, and the difference threshold is the smallest noticeable change between suprathreshold samples (Webb, Bolhuis, Cicerale, Hayes, & Keast, 2015).

Everyone has roughly the same taste sensitivity.

Taste thresholds vary enormously between people. For the bitter compounds PTC and PROP the population splits into non-tasters, tasters, and supertasters, so a concentration that is clearly bitter to one person is tasteless to another (Bartoshuk, 2000).

A low detection threshold means intense taste experience.

The different measures of taste function are only modestly related, so a person with a low detection threshold does not necessarily rate suprathreshold solutions as especially strong; threshold and suprathreshold intensity capture partly separate aspects of the sense (Webb et al., 2015).

The basic tastes all have similar thresholds.

They differ by orders of magnitude. Bitter compounds are detected at the lowest concentrations because bitterness is a toxin-warning system, while sweet and umami, which signal nutrients, are detected only at much higher concentrations (Breslin, 2013).

Glossary

Absolute threshold.

Another name for the detection threshold: the lowest stimulus concentration that can be told apart from a blank on a fixed proportion of trials.

Adaptive staircase.

A threshold procedure that raises or lowers the concentration trial by trial in response to the taster's answers, converging efficiently on the threshold.

Detection threshold.

The lowest concentration of a tastant that can be reliably distinguished from water, without necessarily identifying the taste quality.

Difference threshold.

The smallest change in concentration that can be noticed between two suprathreshold samples; also called the just-noticeable difference.

Gustation.

The sense of taste, which detects chemicals dissolved in saliva through receptor cells clustered in taste buds.

Labeled line.

An arrangement in which each taste quality is carried by a dedicated population of receptor cells and neurons, preserving its identity to the brain.

Labeled magnitude scale.

A rating scale anchored to the strongest imaginable sensation of any kind, designed to let taste intensity be compared validly across people of different sensitivity.

Method of constant stimuli.

A procedure that presents a fixed set of concentrations many times in random order and reads the threshold off the resulting psychometric function.

Method of limits.

A procedure that presents ascending or descending concentration series and averages the points at which the report switches between absent and present.

Oleogustus.

The proposed taste of free fatty acids, supported by evidence of a distinct oral detection threshold separate from the texture of fat.

PROP.

6-n-propylthiouracil, a bitter compound whose detection threshold divides people into non-tasters, tasters, and supertasters.

Psychometric function.

The curve relating the proportion of correct detections to stimulus concentration, from which a threshold is estimated.

Recognition threshold.

The lowest concentration at which the quality of a taste can be correctly named, higher than the detection threshold.

Supertaster.

A person with an unusually low bitter threshold and intense suprathreshold taste experience, often with a high density of tongue papillae.

TAS2R38.

The bitter taste receptor gene whose variants largely account for the difference in PTC and PROP detection thresholds.

Umami.

The savory basic taste evoked by glutamate and related compounds, transduced by a receptor built from T1R subunits.

Key Researchers

Linda M. Bartoshuk

. University of Florida; discovered the supertaster phenotype and showed that taste thresholds and suprathreshold intensity vary enormously across individuals, driven in part by TAS2R38 sensitivity to PROP and PTC, and developed the cross-modal scaling methods that made valid between-person comparisons of taste intensity possible. ORCID - Wikipedia

Paul A. S. Breslin

. Rutgers University and the Monell Chemical Senses Center; studies the sensory psychophysics and genetics of human taste, including the evolutionary function of taste thresholds in nutrient detection and toxin avoidance and the heritability of sweet and bitter detection sensitivity. ORCID

Nirupa Chaudhari

. University of Miami Miller School of Medicine; characterized the taste-bud cell types and synaptic signalling that transduce tastants into neural signals, including work on the glutamate receptors that underlie umami, grounding the molecular basis of detection thresholds for the basic tastes. ORCID

Kikunae Ikeda

. Tokyo Imperial University; identified umami as a basic taste in 1908 by isolating glutamate from kelp broth, establishing that a distinct detection threshold exists for the savory quality and founding the research line that a century later located its receptor. Wikipedia

Danielle R. Reed

. Monell Chemical Senses Center; leads behavioral-genetic work on individual differences in taste perception, dissecting how variation in bitter and sweet receptor genes shapes detection thresholds and preferences, and co-authored the twin studies estimating the heritability of sweet and bitter sensitivity. ORCID

Charles S. Zuker

. Columbia University and the Howard Hughes Medical Institute; identified the receptors and labeled-line cells for sweet, bitter, sour, salty, and umami taste, establishing that each quality is detected by a dedicated receptor-cell population, the molecular architecture that determines what concentrations can be detected at threshold. Wikipedia

Frequently Asked Questions

What is a taste threshold?

A taste threshold is the minimum concentration of a dissolved substance that the gustatory system can detect or identify. Because the system does not switch on sharply, the threshold is defined statistically as the concentration detected on a fixed proportion of trials, conventionally half, rather than as a single all-or-none point.

What is the difference between a detection and a recognition threshold?

The detection threshold is the lowest concentration at which a sample can be told apart from plain water, without necessarily knowing what the taste is. The recognition threshold is higher: it is the lowest concentration at which the quality can be correctly named as sweet, sour, salty, bitter, or umami. There is a band of concentration where a taste is sensed but not yet identified.

How is a taste threshold measured?

With classical psychophysical procedures. The method of limits presents ascending and descending concentration series and averages the switch points; the method of constant stimuli fits a psychometric function to many fixed concentrations; and adaptive staircases converge on the threshold trial by trial. A two-alternative forced choice, asking which of two samples carries the tastant, removes the taster's response bias.

Why do bitter substances have the lowest thresholds?

Bitterness evolved as a warning system for potential toxins, and a sensitive alarm is biologically useful, so bitter compounds are detected at the lowest concentrations, often in the micromolar range. Sweet and umami, which signal calories and protein, are detected only at the much higher millimolar concentrations, because the cost of missing a nutrient is lower than the cost of swallowing a poison.

Why can some people taste bitterness that others cannot?

Much of the variation traces to the bitter receptor gene TAS2R38, whose variants set the detection threshold for the compounds PTC and PROP. The population splits into non-tasters, tasters, and supertasters, so a concentration that is intensely bitter to a supertaster can be completely tasteless to a non-taster. Papillae density and other factors also contribute.

What is a supertaster?

A supertaster is a person with an unusually low bitter threshold who experiences intense taste and oral sensations from concentrations others find faint. Linda Bartoshuk identified the group and showed that their heightened response extends beyond bitterness to many tastes, and that it is associated with a high density of taste papillae on the tongue.

Do taste thresholds change with age?

Yes. Detection thresholds rise across the lifespan, but not uniformly across the basic tastes, so aging produces specific rather than global losses, with salt and bitter detection often more affected than sweet. Large studies across the adult age range confirm the decline and show that it differs by sex and is modulated by the same receptor-gene variants that drive individual differences in the young.

Is fat a basic taste with its own threshold?

It is a strong candidate. Free fatty acids produce a detectable oral sensation at low concentrations, distinct from the texture of fat, and the demonstration of a genuine detection threshold for this sensation, named oleogustus, is a central part of the argument for treating fat perception as a taste rather than only a mouthfeel.

References