Content uploaded by Jos Mojet
Author content
All content in this area was uploaded by Jos Mojet on Jun 09, 2016
Content may be subject to copyright.
Chem. Senses 29: 671–681, 2004 doi:10.1093/chemse/bjh070
Chemical Senses vol. 29 no. 8 © Oxford University Press 2004; all rights reserved.
Correspondence to be sent to: J. Mojet, Agrotechnology and Food Innovations, Bornsesteeg 59, 6708 PD Wageningen, The Netherlands.
e-mail: jos.mojet@wur.nl
Abstract
An increase in concentration of one of the tastants in a ‘real food’ might affect not only the perception of the taste quality of
that manipulated tastant but also the other perceivable taste qualities. The influence of concentration increase of sodium or
potassium chloride in tomato soup, sucrose or aspartame in iced tea, acetic or citric acid in mayonnaise, caffeine or quinine
HCl in chocolate drink, monosodium glutamate (MSG) or inosine 5′-monophosphate (IMP) in broth on the other perceivable
taste qualities in these foods was studied in 21 young subjects (19–33 years) and 21 older subjects (60–75 years). The results
showed that for each of these tastants, except for the two acids, increasing the concentration provoked significant positive or
negative interaction effects on the perception of one or more other taste qualities of the product. Especially in the young, olfac-
tion plays a larger role in the assessment of taste intensity than has been hitherto assumed. The elderly are less able to discrim-
inate between the taste qualities in a product, whereas the young are more able to do so.
Key words: ageing, cross-modal intensity matching, food, olfactory deprivation, side-tastes
Introduction
Flavours in food consist of a mixture of tastes and odours
accompanied by a variety of oral sensations (Pangborn,
1960). Part of the research in this area is focused on
mixtures within one sensory modality such as taste
(Pangborn, 1960; Kamen et al., 1961; Bartoshuk, 1975;
Lawless, 1979, 1982; Gillan, 1983; Van der Heijden et al.,
1983; Frank and Archambo, 1986; Kroeze, 1989, 1990;
Calviño et al., 1990; Schifferstein and Frijters, 1990; Frank
et al., 1993; Schifferstein, 1995; Stevens, 1995; Stevens and
Traverzo, 1997; Kaneda et al., 2000; Prescott et al., 2001).
The excellent review of taste–taste interactions given by
Keast and Breslin (2002) led to the conclusion that in
general, suppression may be found with strong stimuli,
whereas cases of enhancement may be found mostly with
weak, near threshold stimuli.
In another approach the oral perception of mixtures of
cross-modal sensory stimuli such as odour and taste
mixtures was investigated (Lawless, 1977; Murphy et al.,
1977; Murphy and Cain, 1980; Rozin, 1982; Frank and
Byram, 1988; Calviño et al., 1990; Shaffer and Frank, 1990;
Frank et al., 1993; Stevenson et al., 1999; Kaneda et al.,
2000). The use of a specially developed instrument to
combine taste with orthonasal instead of retronasal smell
(Gillan, 1983; Hornung and Enns, 1984; Enns and Hornung,
1985, 1988) might not show an interaction between taste
and smell, since orthonasal sniffing has much weaker
effects than retronasal stimulation in the mouth
(Zoeteman, 1978). The general picture emerging from this
literature is that the interaction effect is strongly dependent
on the compounds used. Consonant odours and tastes
seem to enhance the perceived taste, dissonant odours and
tastes seem either to suppress the perceived taste or to show
no effect at all.
Only a few authors investigated age-related taste–taste
and odour–taste interaction effects. Hornung and Enns
(1984) reported no age-related effects, whereas Murphy
(1985), in a study on the ability to identify blended foods
while blindfolded, found that the average number of
correct responses on the first attempt to identify was signifi-
cantly higher in young than in old subjects. With feedback
and practice, performance improved, but the age effect
remained significant. When deprived of olfaction by
wearing a noseclip, the young lost their superiority over the
elderly.
Effect of Concentration on Taste–Taste Interactions in Foods for Elderly and
Young Subjects
Jos Mojet1,2, Johannes Heidema1 and Elly Christ-Hazelhof1
1Unilever Research & Development Vlaardingen, 3133 AT Vlaardingen, The Netherlands
2Present address: Wageningen University and Research, Department of Agrotechnology and Food
Sciences, 6700 AA Wageningen, The Netherlands
672 B. Mojet, J. Heidema and E. Christ-Hazelhof
The experimental questions addressed in the present study
are whether the increase of one taste compound in a food,
such as sodium chloride, influences the perception of the
other tastes of the food, such as the sweetness, sourness,
bitterness and umami taste. Furthermore, whether, when
such an influence is noticeable, it is different for elderly and
young or for men and women. Finally, in this experiment,
the role of olfaction in taste will be investigated.
This investigation is part of a larger study in which taste
perception with age has been studied with 10 tastants,
dissolved in water and in products and assessed by a fixed
group of elderly and young subjects wearing a noseclip or
not (Mojet et al., 2001, 2003, 2004a,b). The elderly had
higher thresholds than the young. While the relative percep-
tion (intensity discrimination) seemed to be remarkably
resistant to the effects of ageing, the absolute perception
(intensity rating) decreased with age for all tastants in water,
whereas in product this was only the case for the salty and
sweet tastants. Threshold sensitivity predicted neither supra-
threshold intensity perception, nor the optimally preferred
concentration of the tastants. The latter was not different for
the elderly and the young, whereas the former did differ with
age for a number of tastants. Apart from the intensities of
the taste qualities that were experimentally varied in the
food, the intensities of the non-manipulated concentrations
of other taste qualities were also assessed. Throughout the
present paper these taste qualities (other than the manipu-
lated taste quality in the product) will be referred to as side-
tastes.
Although previously we found a decrease in taste sensi-
tivity with age that was generic in nature, there were still
differences in taste losses for the different taste qualities—
see Table 1, based on results of Mojet et al. (2001, 2003).
Thus, when one taste quality is more affected than another
taste quality, the perceived interaction between these two
taste qualities and thus the integrated perception of the
product might be changed for the elderly. When thinking of
counteracting the effects of differential taste loss for the sake
of the well-being of the elderly, it is essential for the food
industry to gain insight in the taste–taste interactions (as
well as taste–odour interactions, which are not studied here)
for the elderly.
Materials and methods
Subjects
Twenty-one older subjects (age 60–75 years; 10 male, mean
= 66.0 years, SD = 3.6; 11 female, mean = 64.6 years, SD =
4.2) and 21 young subjects (age 19–33; 11 male, mean = 26.5
years, SD = 3.6; 10 female, mean = 23.2 years, SD = 3.3)
participated in the experiments. They had all taken part in
several series of experiments, one on threshold sensitivity
(Mojet et al., 2001), one on supra-threshold intensity (Mojet
et al., 2003) and one on pleasantness of the same taste
stimuli (Mojet et al., 2004a,b). All subjects were Caucasian
and met the following criteria: healthy, not on a therapeutic
diet, not living in a home for the elderly, not taking any
prescribed medicine, non-smoking, no heavy alcohol users,
non-pregnant or lactating, not subject to food allergies,
good dental hygiene and not wearing dentures (as it was very
difficult to recruit enough elderly persons without dentures,
subjects with partial dentures were admitted, but they were
not allowed to wear these during testing).
Furthermore, since hearing was used as a matching
modality for taste, it was made sure that all subjects
admitted to the experiments had normal hearing as tested at
750 Hz (for detailed selection criteria, see Mojet et al., 2001).
Subjects were recruited by advertisements in local newspa-
pers and on bulletin boards in senior citizen centres. At the
end of the experiments the subjects were paid for their
participation.
Stimuli
Five taste qualities were investigated: saltiness, sweetness,
sourness, bitterness and umami taste. For each taste quality,
two representative compounds were chosen, which were
administered at five concentrations in commercially avail-
able products. The compounds were grouped into two sets,
each containing one compound for each of the five taste
qualities. One set contained NaCl, sucrose, acetic acid,
caffeine and MSG, the other set consisted of KCl, aspar-
tame, citric acid, quinine HCl and IMP.
All tastants, which were of the highest grade of purity
available, were dissolved in foods. These food products were
iced tea to vary the sweet taste in, chocolate drink to vary the
Ta b l e 1 Variance components in% of the total variance in intensity perception to be ascribed to age
Age Age × taste
quality
Age × compound
(within taste
quality)
Age ×
concentration
Age × taste
quality ×
concentration
Age ×
concentration ×
compound (within
taste quality)
Error
Threshold 93.0 4.0 1.0 –––2.0
Water 96.4 1.1 0.3 1.0 0.8 00.4
Product 92.4 5.6 0.6 00.4 01.0
Product rated with
nose clipped
12.1 60.8 12.4 04.8 09.9
Effect of Concentration on Taste–Taste Interactions in Food 673
bitter taste in, mayonnaise to vary the sour taste in, tomato
soup to vary the salty taste in and bouillon to vary the
umami taste in. The products were versions of commercially
available products (all fromUnilever), which were varied by
the omission or addition of the tastants to be tested. Five
concentration levels of each tastant (ascending 0.2 log steps)
were used in the test. An exception in the concentration
levels was made for mayonnaise (ascending 0.1 log steps),
since the total concentration difference could not be >0.4 log
for technical reasons. Since the aim of the experiments was
to study age differences in the perception of tastants in
normal products rather than to investigate the effect of
different tastants on a product, the tastants were embedded
in products in which they do occur naturally in normal life
and were not all varied in the same product. Furthermore,
the second step in the range of five concentrations corres-
ponded to the usual concentration of the compound in each
selected product, except in the chocolate drink, where the
customary concentration was equal to the first step. Using
different products for different taste qualities, helped also to
better assess the generality of the age effects in a larger food
context.
The compounds were mixed with the dry product before-
hand. On the day of testing, the final products were
prepared. However, the mayonnaise was prepared before-
hand at the Unilever pilot plant at Vlaardingen. The subjects
received 20 ml of each stimulus in disposable 50 ml plastic
cups. All products were served at room temperature for
practical reasons.
The following concentration ranges in (g/l) of the tastants
were used: saltiness in tomato soup—sodium chloride
5.68–35.83 and potassium chloride 9.00–56.77; sweetness in
iced tea—sucrose 53.95–340.38 and aspartame 0.15–0.92;
umami taste in bouillon —MSG 1.58–9.95 and IMP
1.00–6.28; bitterness in chocolate drink—caffeine 0.63–3.98
and quinine HCl 0.01–0.03; sourness in mayonnaise—acetic
acid 0.27–0.67 and citric acid 0.02–0.05.
Five levels each of auditory (loudness), visual (size) and
kinaesthetic/tactile (weight) stimuli were included to be
eventually used as controls in cross-modal intensity
matching. As auditory stimuli, 1.5 s bursts of a narrow band
of noise centred at 750 Hz were recorded, with intensities
varying from 45 to 85 dB in 10 dB steps. These sounds were
delivered to the subjects through earphones. As kinaesthetic
stimuli, five weights were constructed varying in 0.2 log steps
from 33.7 to 212.6 g and hidden in small black (film)
containers of equal size which subjects lifted with the top of
the forefinger by means of a ring on a string connected to the
container. As visual stimuli, an irregular star figure was
multiplied in ascending 0.2 log size steps.
Procedure
The first week of this particular experiment, the subjects
assessed the stimuli while not wearing a noseclip on three
consecutive days, one session per day. The stimuli were
presented one after the other. The sip-and-spit method was
used, i.e. after tasting, the subjects rinsed their mouth with
distilled water and expectorated. For practical reasons sepa-
rate sessions were held for the elderly and the young. At the
start of a session and before each new trial the subject rinsed
with distilled water and expectorated. The subjects were
instructed to eat a piece of cream cracker at the end of a
series of samples of a given compound. In each session of an
experiment, all ten compounds were presented once at five
concentrations in five series, one for each taste quality.
These taste stimuli were alternated with samples of three
replications of auditory, visual and weight stimuli at five
levels. This alternation prevented monotony and adaptation
on the one hand and prolonged the inter-taste-stimulus
interval up to 2 min on the other hand, diminishing the
induction of fatigue. Before the experiment started the
subjects were familiarized with those stimuli by presenting
the lowest and highest in the range. Thus, the subjects
assessed 50 taste stimuli and 45 cross-modal stimuli in one
session. Each session lasted 2 h. Between two taste quality
series within a session there was a break of 5 min.
The order in which the taste qualities were presented
differed over the 3 days, as did the presentation order of the
stimuli within each series. In total, each taste stimulus was
assessed three times and each cross-modal stimulus was
assessed nine times per experiment. Since the major interest
was in unconfounded differences between the two age
groups, the presentation order of the compound concentra-
tions and the cross-modal stimuli was the same for all
subjects. Possible taste order effects were considered to be
strongly reduced by the interspersing of the taste stimuli
with the stimuli from the other sensory modalities.
For all taste samples intensity ratings were made of salt,
sweet, sour, bitter and umami and on liking, whereas the
cross-modal stimuli were rated on intensity only. Intensities
were marked on a nine-point scale with the anchors ‘very
weak’ to the left and ‘very strong’ to the right. For the taste
stimuli, liking for the product was assessed on a nine-point
pleasantness scale with the anchors ‘very little’ to the left and
‘very much’ to the right. Since in the Netherlands it is
considered inappropriate by many to admit to food dislike,
this scale was considered more appropriate than a like–
dislike scale. The second week was similar to the first, but
this time the subjects assessed the stimuli while wearing a
noseclip.
In this paper the effects of concentration of the tastants
that were experimentally varied on the other tastes (side-
tastes) of the products are reported.
Statistical analysis
Methods
The statistical analyses are conducted by means of SAS® and
SAS/STAT® with data arithmetically averaged over the
three replications. Since for the products different tastes
674 B. Mojet, J. Heidema and E. Christ-Hazelhof
were defined as side-taste (e.g. sweet, sour, bitter and umami
in tomato soup; sour, salty, bitter and umami in iced tea),
Repeated measures analysis is applied separately per
product and experimentally varied tastant (e.g. tomato soup
with NaCl and with KCl separately), with age, gender and
age by gender as between-subject factors and noseclip and
concentration as repeated within-subject factors, to investi-
gate the effect of these factors on the perceived side-tastes.
Since the sphericity test show that the patterns in the covari-
ance matrix of this experiment generally do not satisfy the
Huynh–Feldt condition, the multivariate test results instead
of the univariate test results are described, but for the
concentration effect only in those cases where the repeated
tests resulted in a significant linear or quadratic effect. An
interpretation of the results, in terms of higher polynomials,
would be rather meaningless.
Levels of significance
All effects that have a P-value of 0.05 or lower are reported
as ‘significant’. Power analysis shows that, with the number
of subjects in our study, an effect with a magnitude of 1.3
standard deviations and a P-value of 0.10 still has a power of
0.90. Therefore, a selection of the more interesting effects
with a P-value between 0.05 and 0.10 are reported addition-
ally. These effects will be denoted as ‘trends’.
Cross-modal intensity matching
All intensity and liking data reported here are based on
scores that were corrected by means of the cross-modal
intensity matching (CMIM) method, as described in a
previous paper from this group (Mojet et al., 2003). Audi-
tory stimulation was selected to correct the present data for
differences in scale usage since the sensitivity to low-
frequency sounds (∼750 Hz) is normally not impaired with
age and because an analysis of variance of the results for
each of the five sound levels did not show any systematic age
or gender variations.
The matrix of the CMIM data consisted of 42 subjects, 27
replications per individual and five different levels of loud-
ness per replication. On average, the standard deviation per
individual/level of loudness was fairly constant and of the
order of 0.9, indicating that the standard error of an indi-
vidual/level of loudness combination is in the order of 0.17
(0.9/√27). The correction steps were as follows. First, the
individual average and the age group average were deter-
mined for each sound level. Then, the (individual minus
group) averages (which are considered as one obsevation
each and thus have a standard deviation of ∼0.17 in conse-
quence) were regressed against the group averages using
polynomial functions of the latter, starting with a polyno-
mial of degree 0 (constant difference from group mean) and
ending with a polynomial of degree 4 (complete fit of indi-
vidual means). Subsequently, the lowest polynomial with a
residual standard deviation of ∼0.17 was selected as the
assessor’s correction formula for each subject. Finally, each
individual score on the scale was corrected by a value
obtained from the individual’s correction formula. All data
to be reported here are based on scores that were corrected
by means of this method.
Results
Overall age and gender differences
Before analysing the influence of concentration and noseclip
on the perception of the side-tastes, first the between-subject
effects are considered. This exploration shows a number of
significant between-subject effects [all Fs (1,38)]. The elderly
perceived both sourness and bitterness as side-tastes of NaCl
(F = 21.45, P < 0.0001 and F = 8.44, P < 0.007, respectively)
or KCl (F = 16.98, P < 0.0002 and F = 16.58, P < 0.0002) in
tomato soup and of MSG (F = 8.81, P < 0.006 and F = 4.10,
P < 0.05, respectively) or IMP (F = 7.89, P < 0.008 and F =
4.10, P < 0.05, respectively) in bouillon as stronger than the
young did. They also perceived bitterness as side-taste in
mayonnaise with either acetic acid (F = 4.99, P < 0.04) or
citric acid (F = 8.07, P < 0.008) as stronger than did the
young. However, they perceived sourness in caffeine-
flavoured chocolate drink as less intense (F = 4.36, P < 0.05).
Since many tests are conducted, the effects mentioned above
and below with a P-value between 0.01 and 0.05 will be
considered as indicating a tendency.
In general, no significant gender effects were found for the
side-tastes, with the exception of sweetness in KCl-flavoured
tomato soup, where men rated sweetness higher than women
did [F(1,38) = 4.31, P < 0.05]. An age by gender effect was
found for the bitterness perception of KCl-flavoured tomato
soup, where the elderly men and women perceived bitterness
as stronger than the young, but where the elderly women did
so to an extreme [F(1,38) = 5.95, P < 0.02]. A second age by
gender effect was found for the sourness perception of
caffeine-flavoured chocolate drink [F(1,38) = 5.71, P < 0.03].
Here, the young women perceived sourness most strongly,
followed by the young men, the older men and the older
women. Since only a few gender differences were found, the
mean intensities of all rated taste qualities are given in
Figures 1 and 2 per experimentally varied tastant for elderly
and young only.
Influence of concentration of the varied tastant on the
perception of the side-tastes
Is perception of side-tastes dependent on the concentration
of the experimentally varied taste and if so, do elderly and
young or men and women differ in this respect? Apart from
some differences, which will be discussed later, the effects of
the manipulation on the intensity of the side-tastes (see
Figures 1 and 2) show a strong similarity for both noseclip
conditions and are always in the same direction. Taken over
both conditions, increasing the concentration of both salts
NaCl and KCl, induced a significant [all Fs (4,35)] decrease
in sweetness of tomato soup (F = 9.31, P< 0.0001 and
Effect of Concentration on Taste–Taste Interactions in Food 675
Figure 1 Perceived intensities of all taste qualities for elderly and young, while not wearing a noseclip. Saltiness is represented by solid lines, sweetness by
lines of alternating stripes and dots, sourness by lines of stripes and double dots, bitterness by striped lines and the taste of umami by dotted lines.
676 B. Mojet, J. Heidema and E. Christ-Hazelhof
1
2
3
4
5
6
7
8
9
perceived intensity
1
2
3
4
5
6
7
8
9
perceived intensity
1
2
3
4
5
6
7
8
9
perceived intensity
1
2
3
4
5
6
7
8
9
perceived intensity
12345
0.2 log concentration step
12345
0.2 log concentration step
12345
0.1 log concentration step
12345
0.2 log concentration step
12345
0.2 log concentration step
NaCl - elder ly
with nose clip
NaCl - young
with nose clip
KCl - young
with nose clip
KCl - elderly
with nose clip
Sucrose - young
with nose clip
Aspartame - elderly
with nose clip
Aspartame - young
with nose clip
Sucrose - elderly
with nose clip
Acetic acid - elderly
with nose clip
Acetic acid - young
with nose clip
Citric acid - elderl y
with nose clip
Citric acid - young
with nose clip
Caffeine - elderly
with nose clip
Caffeine - young
with nose clip
Quinine HCl - elderly
with nose clip
Quinine HCl -young
with nose clip
MSG - elderly
with nose clip
MSG - young
with nose clip
IMP - elderly
with nose clip
IMP - young
with nose
salty sweet sour bitter umami
Figure 2 Perceived intensities of all taste qualities for elderly and young, while wearing a noseclip. Saltiness is represented by solid lines, sweetness by lines
of alternating stripes and dots, sourness by lines of stripes and double dots, bitterness by striped lines and the taste of umami by dotted lines.
Effect of Concentration on Taste–Taste Interactions in Food 677
F= 9.07, P < 0.0001, respectively), but significant increases
in bitterness (F = 3.10, P <0.03 and F = 6.91, P < 0.0003),
sourness (F = 8.70, P <0.0001 and F = 2.31, P < 0.08, trend
only) and umami taste (F = 2.69, P < 0.05, KCl only). The
increases in sourness with increasing NaCl and in bitterness
with increasing KCl are larger for the elderly than for the
young (F = 4.01, P <0.009 and F = 4.72, P < 0.004, respec-
tively).
The increase in sucrose and aspartame concentration in
iced tea resulted in significant [all Fs (4,35)] decreases in
bitterness (F = 3.92, P <0.01 and F = 6.59, P < 0.0005,
respectively), major decreases in sourness (F = 48.06,
P< 0.0001 and F = 30.67, P < 0.0001, respectively) and a
minor decrease in saltiness (sucrose only; F = 2.94, P < 0.04).
The decrease in bitterness with increasing aspartame was
larger for the young than for the elderly (F = 5.17, P <
0.003).
The increasing concentration of neither of the acids in
mayonnaise had a significant influence on the perception of
side-tastes and the slopes of the side-tastes also did not differ
for the elderly and young.
An increase in caffeine and quinine provoked a decrease in
sweetness of chocolate drink, which was large with caffeine
(F = 55.44, P < 0.0001) and less pronounced with quinine
(F= 3.93, P < 0.01).
An increase in MSG concentration in broth is accompa-
nied by an increase in saltiness (F = 16.86, P < 0.0001),
whereas an increase in IMP led to an increase in saltiness
(F= 4.18, P < 0.007) and in sweetness (F = 4.23, P < 0.007).
The elderly and young did not differ in the slopes of side-
tastes in mayonnaise, chocolate drink or broth. Men and
women did not show significant differences in their slopes
for any of the side-tastes in any of the products.
Deprivation of olfactory stimulation
To investigate whether in this experiment olfaction played a
role in taste perception, the noseclip conditions (off and on)
were compared. Significant differences were found in the
perception of side-tastes [all Fs (1,38)]. Generally, clipping
the nose resulted in a lower level of intensity ratings: lower
sourness and bitterness ratings of tomato soup with NaCl
(F= 10.02, P < 0.003 and F = 6.32, P < 0.02, respectively) or
of tomato soup with KCl (F = 9.71, P < 0.004 and F = 12.96,
P < 0.0009, respectively); lower ratings of saltiness, bitter-
ness and umami taste in iced tea with sucrose (F = 4.79, P <
0.04, F = 8.02, P < 0.008 and F = 6.64, P < 0.02, respectively)
or with aspartame (F = 6.20, P < 0.04, F = 12.80, P < 0.001
and F = 9.20, P < 0.005, respectively); lower ratings of salti-
ness in mayonnaise with acetic or with citric acid [F = 8.35,
P < 0.007 and F = 3.97, P < 0.06 (trend), respectively]; lower
ratings of sourness and saltiness in chocolate drink with
caffeine (F = 7.42, P < 0.01 and F = 9.82, P < 0.004, respec-
tively) or with quinine (F = 4.41, P < 0.05 and F = 10.76, P<
0.003, respectively); and lower saltiness ratings of broth with
MSG (F = 8.39, P < 0.007) or with IMP (F = 5.61, P < 0.03).
These reductions in perceived intensities are mainly due to
higher intensity ratings given by the young in the ‘noseclip
off’ condition than in the ‘noseclip on’ condition. t-Tests
showed that the differences between the ‘noseclip off’ and
the ‘noseclip on’ condition only deviated significantly from
zero for the elderly when tested over all tastants and all side-
tastes [T(20) = 2.18, P < 0.04]. This is also shown in the
scatter plots of Figure 3a and b. The ratings of the elderly
shown in Figure 3a are almost similar in the two noseclip
conditions (on and off), which is shown by the close prox-
imity of the ratings to the y = x line. The variance in the
group of elderly is much larger than the variance in the
group of young subjects, whose ratings are, compared to the
elderly, more scattered over the left upper side of the plot.
This indicates that the young make more use of their sense of
smell when assessing the intensities in the ‘noseclip off’
condition than the elderly. In all forty cases (10 tastants,
four side-tastes each, averaged over the five concentrations),
the young gave lower ratings when wearing a noseclip than
when not wearing a noseclip. Furthermore, in all cases but
one, these differences between the ‘noseclip on’ condition
and the ‘noseclip off’ condition, were larger than the differ-
ences found for the elderly (overall sign test: P < 0.0001).
The only exception was found for the bitterness of tomato
soup with KCl, where no significant difference between the
elderly and the young was found, but where the elderly
women perceived stronger bitterness when they wore no
noseclip than when they wore one. This difference between
the two conditions was the only difference found for one of
the age by gender groups that deviated significantly from
zero [T(10) = 3.15, P < 0.02].
Analysed per individual side-taste for each of the experi-
mentally varied tastants, the interaction between noseclip
and age reached significance [all Fs (1,38)] in 10 out of these
forty cases.
Thus, different effects of the noseclip condition on the
ratings of the elderly and young were found for bitterness of
iced tea, mayonnaise and bouillon, but not of tomato soup.
With the noseclip on, the extent to which the young gave
lower ratings to side-tastes than the elderly, was larger than
Figure 3 Scatter plots of the intensities of the side-tastes of the five
products, assessed by elderly and young while wearing a noseclip or not.
The intensities are averaged per person over all concentrations and all side-
tastes.
678 B. Mojet, J. Heidema and E. Christ-Hazelhof
when they had no noseclip on. This age by noseclip inter-
action on bitterness was found for broth with MSG (F =
7.26, P < 0.02), for mayonnaise with acetic acid (F = 5.47,
P< 0.03) or citric acid [F = 3.28, P < 0.08 (trend)], for iced
tea with aspartame (F = 4.97, P < 0.04), or sucrose (F = 5.64,
P< 0.03). An interaction effect was also found in the ratings
of sweetness of bouillon with MSG (F = 6.18, P < 0.02),
where, when wearing a noseclip, elderly subjects gave higher
ratings than young subjects. Furthermore, wearing a nose-
clip had also a different effect on elderly and young in the
assessment of saltiness of mayonnaise with acetic acid (F =
4.47, P < 0.05) or citric acid (F = 4.98, P < 0.04) and of choc-
olate drink with caffeine (F = 6.08, P < 0.02). With the nose-
clip off, the young gave higher ratings than the elderly,
whereas with the noseclip on the reverse was true. A similar
interaction was found in the assessment of sourness of choc-
olate drink with caffeine (F = 10.05, P < 0.03) or with
quinine (F = 5.84, P < 0.03) where with the noseclip on, the
young gave higher ratings than the elderly.
Deprivation of olfaction had no different effect on the
ratings of men and women, or on those of the four age by
gender groups.
In a few cases, wearing a noseclip had also an effect on the
slopes of the side-tastes [all Fs (1,38)]. Both, the slopes of the
sweet and the sour taste of tomato soup with increasing
NaCl were significantly flatter when subjects wore a noseclip
(sweet, F = 12.07, P < 0.002 and sour, F = 11.33, P < 0.002).
Furthermore, the sweetness slope of chocolate drink with
increasing caffeine was significantly flatter (F = 8.67, P <
0.006) when the assessment was made with the nose clipped.
Wearing a noseclip also led to a flatter slope (F = 3.08, P <
0.09, trend only) of sweetness in chocolate drink with
increasing quinine.
Discussion
Interactions
This first study over all taste qualities in the same group of
elderly and young people shows that increasing the concen-
tration of one dominant taste compound in a complex food
matrix has different effects on the perceived intensities of
side-tastes of the food. In addition, these effects were similar
for the two tastants within one taste quality. Since in this
experiment, foods were chosen on the basis of their repre-
sentativeness of salty, sweet, sour, bitter and umami foods in
‘real life’, different foods were used to investigate the effects
of the manipulated tastants on the side-tastes and as a result
no mutual effects between the taste qualities in one system
can be demonstrated.
Nevertheless, a generalization of the taste–taste interac-
tion effects over these different food types might help to
further the insight in these complex interactions. Figure 4,
which is a schematic overview of the results and is inspired
by Keast and Breslin (2002), summarizes this generalization.
They divide the taste–taste interaction effects found in the
literature into three schemata’s, one for low, one for medium
and one for high intensity/concentrations. The concentra-
tions of the varied tastants in the present experiment lie in
the same range as that which is found in normal every day
food. One could argue that experiments with much higher
concentrations bear little relevance for every day life and run
the risk of introducing artefacts caused by the strong devia-
tion from what subjects are used to.
The effects caused by the experimentally varied tastants in
the present study correspond mainly with the Keast and
Breslin (2002) medium concentration scheme. It should be
noted that in Figure 4 the arrows pointing away from a
given taste quality are directly comparable since they are
effects exerted in the same food matrix, whereas the arrows
coming towards a certain taste quality show how this quality
is affected by changes of other taste qualities in different
food matrices.
In contrast to the findings presented by Keast and Breslin
(2002), no effect of acids on any of the other taste qualities
was found in the present experiment. This lack of suppres-
sion or enhancement effect might be due to the relatively
limited concentration range in which the highest concentra-
tion was only 2.5 times larger than the lowest concentration.
Another difference exists with regard to the influence of salts
on bitterness. In their summary Keast and Breslin describe
this influence as suppressive, irrespective of the used concen-
tration, whereas in the present study an increase in bitterness
was found for both salts. Finally, in the present study,
neither an increase nor a decrease in sourness by bitter
tastants, as Keast and Breslin (2002) described for the
moderate and high concentrations respectively, was found.
This could be due to the fact that the sourness perceptibility
Figure 4 Pentagram of the taste–taste interactions in complex food
matrices. Outgoing arrows show the influence of one taste quality on the
other taste qualities in the same food. Incoming arrows represent
influences of different taste qualities from different foods. Triangles indicate
either an enhancement effect (when they point upwards) or a depressive
effect (when they point downwards). Circles indicate a nil effect. Blac
k
symbols represent both tastants within a taste quality, grey symbols
indicate that only one of the tastants exerts the effect.
Effect of Concentration on Taste–Taste Interactions in Food 679
in the chocolate drink was very low to begin with, or it might
be that the concentration range used here falls just between
the moderate and high intensities for which Keast and
Breslin (2002) describe opposite findings.
Umami perception was not influenced by any of the other
taste qualities, which supports the findings of Bartoshuk
(1975). She found that substances which show the least
compression when added to themselves, like the umami
tastants do (see Figures 1 and 2), also show the least suppres-
sion when other substances are added to them.
Slopes of the side-tastes with age
Taste–taste interaction effects have not been studied in
elderly to our knowledge, whereas odour–taste interactions
are reported by Enns and Hornung (1985) who found that,
when using odorant/tastant pairs, the magnitude of suppres-
sion [overall intensity/(smell intensity + taste intensity)] was
not affected by the age of the subjects. In the present experi-
ment only in 3 out of the possible 40 cases (sourness with
NaCl and MSG, bitterness with KCl) a steeper increase was
found for the elderly than for the young and in only one case
a more pronounced decrease in bitterness with increasing
aspartame was found for the young. This suggests that the
slopes of the perceived side-tastes are not systematically
affected by age.
Contribution of olfaction
No findings on the influence of the contribution of olfaction
on taste–taste interactions have been reported in the litera-
ture. However, Murphy (1985) reported that, when deprived
of olfaction the performance of young women, in a task
where they had to identify blended foods, fell to the same
level as that of elderly women. Obviously, when they did not
wear a noseclip, the young women took more advantage of
their sense of smell. Furthermore, it suggests that a decline
in olfactory sensitivity of the elderly women made them
perform less well than the young in this task with both olfac-
tory and taste cues.
Would the same phenomenon occur in the present experi-
ment where the side-taste intensities had to be assessed? In
line with a previous study (Mojet et al., 2003), the differences
between the ‘noseclip off’ and ‘noseclip on’ condition were
larger for the young than for the elderly in the present exper-
iment. Since the overall difference between the with and
without smell condition deviated not only significantly from
zero for the young but also (significantly, but to a minor
degree) for the elderly, it is clear that they too use their sense
of smell, if to a lesser degree. As was pointed out in a
previous paper (Mojet et al., 2003), there are two possible
ways in which smell could play a role. First, the tastants
themselves might have a smell. To the knowledge of the
present authors, this possibility, unlikely as it seems for most
tastants with the exception of acetic acid, has never been
tested seriously. In the second place, the presence of the
tastants might interact with the olfactory perception of the
medium in which they are presented, which is known as a
‘salting out’ effect (Segatin and Kiofutar, 2000). That this
latter possibility exists when the tastants are presented in
product is evident, since some of them (MSG and IMP) are
known flavour enhancers but might this also be true in the
case of water? It would suggest that water has a smell of its
own which is changed by tastants. To clarify this point,
further research has been carried out (Mojet and Köster,
2004) and its findings favour the idea that the tastants have
a smell that is perceptible when they are dissolved in water.
Age differences in taste quality discrimination ability
One interesting general finding has to be discussed here.
Over all products and tastants, the elderly gave higher
ratings to the perceived intensities of the side-tastes than did
the young, whether their olfactory input was blocked (37 out
of 40 cases) or was not blocked (29 out of 40 cases). This is
remarkable, since previously it was found (Mojet et al.,
2003) that the elderly rated the perceived intensities of the
experimentally varied taste lower than the young in 9 out of
10 cases (sign-test, P < 0.02) without wearing a noseclip and
in 6 out of 10 cases (n.s.) while wearing a noseclip.
To show this age difference in the perception of the side-
tastes graphically, the ratio between the scores for the domi-
nant taste and for the summated side-tastes are given in
Figure 5, in percentages of all scores. For example, the
percentage of all scores that is given to saltiness in tomato
soup with NaCl increases from 21.3% (concentration step 1)
to 45.1% (concentration step 5) for the elderly and from 33.3
to 54.4% for the young, whereas per definition, the
percentage given to the combined side-tastes decreases from
78.7 to 54.9% for the elderly and from 66.7 to 45.6% for the
young. It seems that the young are better able to discrimi-
nate between the different taste qualities than the elderly.
This is especially so in the case of tomato soup. Whether
this is because tomato soup is more complex than the
other products used, or because the taste quality varied
experimentally was salty, can not be concluded from this
experiment. That the elderly discriminate to a lesser degree
between the taste qualities was shown to some degree for all
products with the exception of chocolate drink with quinine,
where both the elderly and the young perceived the side-
tastes rather strongly and where the young consistently did
so to a higher degree. Previously, it was found that the
elderly have a less specified taste acuity than the young,
meaning that at threshold level the sensitivities for the 10
tastants were more strongly correlated for the elderly than
for the young (Mojet et al., 2001) and that at supra-
threshold level, the inter-correlation of all tastants dissolved
in water and in product was higher for the elderly than for
the young (Mojet et al., 2003).
For the phenomena described above, which both point in
the direction of a possible impairment in cognitive processes
(Essed and Eling, 1986), the noise hypothesis might provide
an explanation, either at a neural level, or at a psychological
680 B. Mojet, J. Heidema and E. Christ-Hazelhof
level. The neural noise hypothesis supposes that the signal to
noise ratio is lowered by a decrease in intensity of the signal,
by an increase in the level of spontaneously firing neurons,
or both and thus would make it more difficult for the elderly
to differentiate between the requested taste quality and the
other taste qualities. The perceptual noise hypothesis is
characterized by a decrease in the ability to neglect irrelevant
information. The irrelevant information could be considered
as a type of noise, but on a psychological/perceptual level
and not on a neural level (Stroop, 1935).
The present experiment reveals three points of interest for
further research. First, it can be concluded that increasing
the concentration of the dominant taste in foods provokes
significant positive or negative interaction effects on the
perception of one or more other taste qualities of the
product that are not age-related in most cases. Whether this
influence is tastant or product specific, or both, is a subject
for further research. Secondly, the fact that, especially in the
young, olfaction plays a larger role in the assessment of taste
intensity than has been hitherto assumed, also has to be
investigated further. Finally, the finding that the elderly are
less able to discriminate between the taste qualities in a
product, whereas the young are more able to do so, has not
been reported previously and should be pursued in future
research because of its possible gerontological implications.
Acknowledgements
Dr E.P. Köster and Dr J.H.A. Kroeze are greatly acknowledged
for their critical support in writing this paper.
References
Bartoshuk,L.M. (1975) Taste mixtures: is suppression related to compres-
sion? Physiol. Behav., 14, 643–649.
Figure 5 Percentages of the sum of intensity scores given to the dominant taste and to the combined side-tastes for elderly and young subjects. The
percentages of scores given by young subjects to the combined side-tastes are represented by the black areas, while the grey or striped areas (including the
black areas) show the percentages for the elderly (except for quinine, where the young attributed a higher percentage of their scores to the side-tastes than
the elderly). The white areas represent the percentage of intensity scores given to the dominant taste for the elderly. For the young the white and striped
area represent the percentages given to the dominant taste (reversed for quinine).
Effect of Concentration on Taste–Taste Interactions in Food 681
Calviño, A.M., García-Medina, M.R. and Cometto-Muñiz, J.E. (1990)
Interactions in caffeine–sucrose and coffee–sucrose mixtures: evidence
of taste and flavor suppression. Chem. Senses, 15, 505–520.
Enns,M.P. and Hornung, D.E. (1985) Contributions of smell and taste to
overall intensity. Chem. Senses, 10, 357–366.
Enns,M.P. and Hornung, D.E. (1988) Comparisons of the estimates of
smell, taste and overall intensity in young and elderly people.Chem.
Senses, 13, 131–139.
Essed,I. and Eling, P. (1986) Oorzaken van cognitieve verlangzaming bij
veroudering: ruishypothesen. Tijdsch. Gerontol. Geriatr., 17, 233–243.
Frank,R.A. and Archambo, G. (1986) Intensity and hedonic judgements
of taste mixtures: an integration analysis. Chem. Senses, 11, 427–438.
Frank,R.A. and Byram, J. (1988) Taste–smell interactions are tastant and
odorant dependent. Chem. Senses, 13, 445–455.
Frank, R.A., Van der Klauw, N.J. and Schifferstein, H.J. (1993) Both
perceptual and conceptual factors influence taste–odor and taste–taste
interactions. Percept. Psychophys., 54, 343–354.
Gillan,D.J. (1983) Taste–taste, odor–odor and taste–odor mixtures:
greater suppression within than between modalities. Percept. Psycho-
phys., 33, 183–185.
Hornung,D.E. and Enns, M.P. (1984) The independence and integration
of olfaction and taste. Chem. Senses, 9, 97–106.
Kamen, J.M., Pilgrim, F.J., Gutman, N.J. and Kroll, B.J. (1961) Inter-
actions of suprathreshold taste stimuli. J. Exp. Psychol., 62, 348–356.
Kaneda, H., Maeshima, K., Goto, N., Kobayakawa, T., Ayabe-
Kanamura, S. and Saito, S. (2000) Decline in taste and odor dis-
crimination abilities with age and relationship between gustation and
olfaction. Chem. Senses, 25, 331–337.
Keast,R.S.J. and Breslin, P.A.S. (2002) An overview of binary taste–taste
interactions. Food Qual. Pref., 14, 111–124.
Kroeze,J.H.A. (1989) Is taste mixture suppression a peripheral or central
event? In Laing, D.G., Cain, W.S., McBride, R.L. and Ache, B.W. (eds),
Perception of Complex Smells and Tastes. Academic Press, Sydney,
pp. 225–243.
Kroeze,J.H.A. (1990) The perception of complex taste stimuli. In McBride,
R.L. and MacFie, H.J.H. (eds), Psychological Basis of Sensory Evaluation,
Elsevier, London, pp. 41–68.
Lawless,H.T. (1977) The pleasantness of mixtures in taste and olfaction.
Sens. Processes, 1, 227–237.
Lawless,H.T. (1979) Evidence for neural inhibition in bittersweet taste
mixtures. J. Comp. Physiol. Psychol., 93, 538–547.
Lawless,H.T. (1982) Paradoxical adaptation to taste mixtures. Physiol.
Behav., 25, 149–152.
Mojet,J. and Köster, E.P. (2004) Sensory memory and food texture.Food
Qual. Pref., in press.
Mojet, J., Christ-Hazelhof, E. and Heidema, J. (2001) Taste perception
with age: generic or specific losses in threshold sensitivity to the five
basic tastes? Chem. Senses, 26, 845–860.
Mojet, J., Heidema, J. and Christ-Hazelhof, E. (2003) Taste perception
with age: generic or specific losses in supra-threshold intensities of five
taste qualities? Chem. Senses, 28, 397–413.
Mojet, J., Christ-Hazelhof, E. and Heidema, J. (2004a) Taste perception
with age: pleasantness and its relationships with threshold sensitivity
and supra-threshold intensity of five taste qualities. Food Qual. Pref., in
press.
Mojet, J., Köster, E.P. and Prinz, J.F. (2004b) Do tastants have a smell?
Chem. Senses, in press.
Murphy,C. (1985) Cognitive and chemosensory influences on age-related
changes in the ability to identify blended foods. J. Gerontol., 40, 47–52.
Murphy,C. and Cain, W.S. (1980) Taste and olfaction: Independence vs
interaction. Physiol. Behav., 24, 601–605.
Murphy, C., Cain, W.S. and Bartoshuk, L.M. (1977) Mutual action of
taste and olfaction. Sens. Processes, 1, 204–211.
Pangborn,R.M. (1960) Taste Interrelationships. Food Res., 25, 245–256.
Prescott, J., Ripandelli, N. and Wakeling, I. (2001) Binary taste mixture
interactions in Prop non-tasters, medium-tasters and super-tasters.
Chem. Senses, 26, 993–1003.
Rozin,P. (1982) ‘Taste–smell confusions’ and the duality of the olfactory
sense. Percept. Psychophys., 31, 397–401.
Schifferstein,H.J. (1995) Perception of taste mixtures. In Doty, R.L. (ed.),
Handbook of Olfaction and Gustation. Marcel Dekker, New York, pp.
689–714.
Schifferstein,H.N.J. and Frijters, J.E.R. (1990) Sensory integration in cit-
ric acid/sucrose mixtures. Chem. Senses, 15, 87–109.
Segatin,N. and Kiofutar, C. (2000) Salting-out of some alkyl acetates in
aquueous sodium chloride solutions. Monatsh. Chem., 131, 131–144.
Shaffer,G. and Frank, R.A. (1990) An investigation of taste–smell inter-
actions across four tastants and six odorants. Chem. Senses, 15, 638.
Stevens,J.C. (1995) Detection of heteroquality taste mixtures.Percept.
Psychophys., 57, 18–26.
Stevens,J.C. and Traverzo, A. (1997) Detection of a target taste in a
complex masker. Chem. Senses, 22, 529–534.
Stevenson, R.J., Prescott, J. and Boakes, R.A. (1999) Confusing tastes
and smells: how odours can influence the perception of sweet and sour
tastes. Chem. Senses, 24, 627–635.
Stroop,J.R. (1935) Studies of interference in serial verbal reactions. J. Exp.
Psychol., 18, 643–662.
Van der Heijden, A., Brussel, L.B.P., Heidema, J., Kosmeijer, J.G. and
Peer, H.G. (1983) Interrelationships among synergism, potentiation,
enhancement and expanded perceived intensity vs concentration.
J. Food Sci., 48, 1192–1196 and 1207.
Zoeteman,B.C.J. (1978) Sensory assessment and chemical composition of
drinking water. Thesis, Utrecht University, Utrecht.
Accepted July 29, 2004
Occupational Medicine 2004;54:000–000
DOI: 10.1093/occmed/kqhXXX
