Dysphagia
DOI 10.1007/s00455-009-9243-y
ORIGINAL ARTICLE
Sensory and Motor Responses of Normal Young Adults During
Swallowing of Foods with Different Properties and Volumes
Atsuko Igarashi Maiko Kawasaki
Shu-ichi Nomura Yuji Sakai Mayumi Ueno
Ichiro Ashida Yozo Miyaoka
Received: 16 December 2008 / Accepted: 21 July 2009
Springer Science+Business Media, LLC 2009
Abstract We examined the influence of rheological/tex-
tural properties and volumes of test foods on the sensory
and motor aspects of swallowing in healthy young adults.
Three test foods differing in thickening agent concentration
(0.0, 1.5, and 3.0%) were prepared and delivered in dif-
ferent volumes (*3, *5, and *7 ml) to subjects seated
on a chair. Viscosity analyses of the 1.5 and 3.0% test
foods revealed that they behaved as non-Newtonian fluids
and were thixotropic. The 1.5% test food differed from the
3.0% test food in its textural properties (hardness, cohe-
siveness, and adhesiveness). As determined by a linear
model equation method, the thickening agent concentration
affected the scores of all six sensory evaluation questions
that were answered by the subjects, which suggests that the
concentration affected the food properties being evaluated.
A. Igarashi
Department of Oral Health and Welfare, Niigata University
Faculty of Dentistry, Niigata 951-8514, Japan
M. Kawasaki
Department of Bio-Prosthodontics, Niigata University Graduate
School of Medical and Dental Sciences, Niigata 951-8514, Japan
S. Nomura
Division of Comprehensive Prosthodontics, Niigata University
Graduate School of Medical and Dental Sciences, Niigata
951-8514, Japan
Y. Sakai
M. Ueno
Product Development Center, Bourbon Corporation, Niigata
945-8611, Japan
I. Ashida
Y. Miyaoka (&)
Department of Health and Nutrition, Niigata University of
Health and Welfare School of Health Sciences, Niigata
950-3198, Japan
e-mail: miyaoka@nuhw.ac.jp
Consistent with previous reports, thickening agent con-
centration and test food volume also affected some dura-
tional parameters of laryngeal (recorded by a piezoelectric
sensor) and suprahyoid muscle (recorded on an electro-
myogram) motor activity. However, thickening agent
concentration and test food volume did not affect the single
amplitude parameter of the electromyogram that was
measured. The thixotropic property of foods can affect the
motor aspect of oropharyngeal swallowing as well as the
sensory aspect.
Keywords Rheological properties
Textural properties

Sensory evaluation
Motor activity
Oropharyngeal
swallowing
Normal humans
Deglutition

Deglutition disorders
Rheological and textural properties, such as viscosity and
hardness, of foods affect the duration and amplitude of
swallowing in humans [15]. Videofluoroscopic and man-
ometric studies [2, 3] have shown that as the density and
viscosity of a food preparation change from high to low,
statistically significant differences in durational parameters
of oropharyngeal swallowing are observed. These param-
eters include oral transit time, pharyngeal transit time,
pharyngeal clearance time, and upper esophageal sphincter
opening. Varying the viscosity of food preparations has
also been associated with significant changes in parameters
of pharyngeal swallowing in dysphagic patients with neu-
rological impairment [6]. Moreover, electromyographic
studies [1, 5] have shown by recording and analyzing
electromyograms (EMGs) of the suprahyoid (SH) muscles
that different rheological and textural properties of foods
affect oropharyngeal swallowing. For example, a food with
high density and viscosity was found to increase the active
123

durations and amplitudes of the SH EMG relative to the
effect of a food with low density and viscosity.
Liquid foods can be classified into two groups on the
basis of their flow properties, namely, Newtonian and non-
Newtonian fluids. For Newtonian fluids such as water and
watery beverages (tea, coffee, and beer), the shear rate is
directly proportional to the shear stress and viscosity is
independent of the shear rate [7]. However, most fluid and
semifluid foods fall into one of several classes of non-
Newtonian fluids [7]. One of these classes exhibits a unique
food property called thixotropy, in which the apparent
viscosity decreases with the time of shearing (i.e., the
viscosity exhibits time dependency [7]). Previous studies
have shown that aqueous solutions of gums that lack
thixotropy have a slimy mouthfeel, whereas gums with
thixotropy are not slimy [8, 9]. While one physicochemical
study has evaluated the thixotropy of maize-based and
maltodextrin-based thickeners [10], systematic analyses of
the effect of thixotropic foods on the sensory and motor
aspects of swallowing in humans have not been performed.
In this study we examined the effect of varying rheological
(viscosity) and textural properties of test foods on the
sensory and motor aspects (durational and amplitude
parameters) of oropharyngeal swallowing.
Materials and Methods
Subjects
Twelve healthy adults participated in this study. Four were
men [mean age standard deviation (SD) = 20.5 0.6
years] and eight were women (22.8 3.4 years). Although
all subjects were subjected to the sensory evaluation and
electrophysiology experiments described below, the elec-
trophysiology data collected from four subjects did not sat-
isfy our criterion of a good recording, namely, a signal-to-
noise ratio of an EMG that exceeded 3.0. Consequently, the
electrophysiology data from only eight subjects (four men,
20.5 0.6 years of age and four women, 24.8 4.5 years
of age) were analyzed. With regard to the sensory experi-
ments, the data from all 12 subjects were suitable and were
analyzed. All subjects participated in the study after pro-
viding informed consent. None complained of special
problems with regard to chewing and swallowing functions.
Preparation of Test Foods
Mixtures composed of a thickening agent, sucrose, citric
acid, and flavor served as the test foods (Table 1). The
thickening agent consisted of the following six materials:
guar gum, tara gum, carrageenan, xanthan gum, starch, and
dextrin. Three test food preparations that differed in
123
A. Igarashi et al.: Food Properties and Swallowing
Table 1 Raw materials constituting the test foods
Concentration of thicking agent Sucrose Citric acid Flavor
0.0 9.0 0.15 0.04
1.5 13.5 0.17 0.08
3.0 15.0 0.24 0.12
Values are percentages
percentage of the thickening agent (0.0, 1.5, and 3.0%)
were generated by dissolving the thickening agent in water
together with sucrose, citric acid, and flavor. The concen-
trations of sucrose, citric acid, and flavor in the three foods
varied on the basis of preliminary experiments so that the
resulting food preparations tasted and smelled the same.
The test foods were produced under sanitary conditions by
the Product Development Center of the Bourbon Corpo-
ration (Niigata, Japan) according to the following produc-
tion process: (1) the raw materials were measured precisely
(Table 1), (2) the powder materials were mixed, (3) the
powder materials were dissolved in water, (4) the dissolved
materials were heated, (5) the flavor was added to the
heated materials, and (6) the test foods were packed into
sealed containers. Thereafter, needleless syringes were
filled with three different weights (3, 5, and 7 g) of the
three test foods and kept at room temperature. The volumes
of the test foods differed slightly depending on the con-
centration of the thickening agent. Thus, for the 0.0, 1.5,
and 3.0% test foods, the 3-g syringes contained 3.2, 3.2,
and 3.4 ml, respectively, the 5-g syringes contained 5.3,
5.3, and 5.7 ml, respectively, and the 7-g syringes con-
tained 7.4, 7.5, and 8.0 ml, respectively. For convenience,
the 3, 5, and 7-g food preparations will be referred to as
small, medium, and large, respectively.
Measurements of Food Properties
The viscosity of the three test foods was measured by a
viscometer (TVB-10 M, Toki Sangyo Co., Ltd., Tokyo,
Japan) at room temperature (about 26C). Each sample was
placed into a glass beaker (500 ml) and subjected to rota-
tion by a suitable viscometer rotor (preliminary experi-
ments were used to determine which of the four rotors was
suitable for each experiment). The amount of each sample
in the beaker (about 500 ml) was determined by referring
to the manual of the viscometer. In one set of experiments,
viscosity was measured at four different rotation velocities
[6, 12, 30, and 60 revolutions per minute (rpm)] 60 s after
the rotor started. In another set of experiments, viscosity
was measured at 12 rpm at five time points (30, 60, 90,
120, and 150 s) after the start of rotation. Since the vis-
cosity of the samples was affected by the previous mea-
surement, each sample was measured only once (thus,
A. Igarashi et al.: Food Properties and Swallowing
different samples were used for each rotation velocity). For
each rotation velocity and time point, the measurements
were repeated five times. Thus, in total, 60 samples (3 test
foods 9 4 rotation velocities 9 5 repetitions) were sub-
jected to viscosity measurements. The viscosity measure-
ments were then expressed as mean SD.
Three textural properties (hardness, cohesiveness, and
adhesiveness) of the three test foods were measured by
using a texture profile unit (TPU-2S, Yamaden Inc., Tokyo,
Japan) at room temperature (about 26C). Thus, each
sample was placed in the stainless-steel container of the
texture profile unit and compressed twice at a speed of
5 mm/s with a clearance of 2.0 mm by a 20-mm-diameter
plunger. The same measurements were repeated ten times
for each sample and the values of the three textural prop-
erties were expressed as mean SD.
Sensory Evaluation Test
For the sensory evaluation experiment, a scoring method was
adopted. Each subject separately evaluated the three test
foods when they were delivered in three different volumes
(i.e., there were nine different tests). After swallowing, the
subject then scored the test foods by answering the six
questions listed in Table 2, which mainly aimed to determine
the texture and taste of the test foods. Scoring was done a
five-stage scale: very strong = ?2.0, strong = ?1.0, med-
ium = 0.0, weak = 1.0, and very weak = 2.0. In-
between scores such as -0.5 were not permitted.
Electrophysiological Recordings
To record surface electromyograms (EMGs), two pairs of
adhesive electrodes [EEG flat electrodes (MLAWBT9),
Bio Research Center, Nagoya, Japan] were attached to the
skin just above the suprahyoid (SH) and masseteric mus-
cles, respectively. The EMG signals were amplified, fil-
tered (using a bandwidth of 10-5000 Hz), fully rectified,
and integrated (s = 0.1 s) by using the PowerLab system
(PowerLab/8sp, ADInstruments Pty Ltd., Bella Vista,
Australia). The thin film (22 mm 9 171 mm 9 0.21 mm)
Table 2 Questions used for the sensory evaluation
Item # Questions
1 How hard was it to break the food down in the mouth?
2 How pleasant was the food in the mouth?
3 How tasty was the food?
4 How thin was the food in the mouth?
5 How easy was it to swallow the food?
6 How was the pharyngeal clearance?
of a piezoelectric sensor (LDT4-028 K/L, Tokyo Sensor
Co., Ltd., Tokyo, Japan) was then attached to the front of
the neck to record a laryngeal mechanogram during swal-
lowing. The upper edge of the sensor film was fitted to the
upper edge of the thyroid cartilage, and the whole of the
film was fixed onto the skin with surgical tape. The sensor
film was directly connected to the PowerLab system, which
could then record and analyze the electrical signals ema-
nating from the film.
Recording Procedure
Each subject was instructed to sit comfortably on a chair in
an electrically shielded room that was kept at a temperature
of about 25C. After the EMG electrodes and piezoelectric
sensor were attached, each subject was asked to open his/
her mouth to accept the contents of one of the nine syringes
that contained the three test foods at three different vol-
umes. The subject was also asked to swallow the test food
in a gulp without chewing, if possible, just after a vocal
command from the experimenter. The subject was then
given a questionnaire containing the questions listed in
Table 2 to record his/her sensory impressions of the test
food that he/she had just swallowed. The trials were sep-
arated by intervals of about 1 min and each experimental
session consisted of nine trials. In total, each subject per-
formed two experimental sessions on the same day.
Data Analysis
The electrophysiological data were analyzed only in terms
of durational parameters but the EMG data were analyzed in
terms of both durational and amplitude parameters. During
a swallow, the piezoelectric sensor usually detected large
(P1) and small positive deflections and a large negative (B1)
deflection (Fig. 1). Consequently, from each trajectory
recorded by the piezoelectric sensor, six durational
parameters were measured, namely, (1) from the start to the
end of each trajectory (on-off), (2) from the start to the large
peak (on-P1), (3) from the start to the bottom (on-B1), (4)
from the large peak to the bottom (P1-B1), (5) from the
large peak to the end (P1-off), and (6) from the bottom to
the end (B1-off). The start (or onset) was located by
observing when the trajectory clearly deviated from the
baseline, while the end (or termination) was located by
observing when the trajectory stably returned to the base-
line. The amplitude changes recorded by the piezoelectric
sensor were not analyzed, just as in previous studies [4, 11].
From the EMG data, three durational parameters and
one amplitude parameter were measured (Fig. 1; masse-
teric EMG data were not analyzed). The three durational
parameters were as follows: (1) from the start to the end of
the integrated SH EMG during swallowing (on-off), (2)
123

Fig. 1 Durational and amplitude parameters measured in the present
study. Piezo is the laryngeal trajectory that accompanies swallowing
and is recorded by a piezoelectric sensor (upward deflection, positive
polarity). P1 and P2 are the peaks of the first and second positive
deflections of the piezoelectric sensor. B1 is the bottom of a negative
deflection. SH EMG is the electromyogram that is recorded from the
suprahyoid muscles. intSH EMG is the integrated SH EMG. PE is the
peak of integrated SH EMG. The nine horizontal arrows indicate six
durations measured from the piezoelectric sensor and three durations
measured from the integrated SH EMG. A vertical arrow indicates the
peak amplitude measured from the integrated SH EMG. Filled and
open triangles indicate the start and the end of the laryngeal trajectory
and integrated SH EMG measurement, respectively. See text for
details
from the start to the peak (on-PE), and (3) from the peak to
the end (PE-off). As with the piezoelectric sensor trajec-
tories, the start (or onset) was located by observing when
the original and integrated EMGs deviated from the base-
line, while the end (or termination) was located by
observing when the EMGs stably returned to the baseline.
The net amplitude of the integrated SH EMG was calcu-
lated by subtracting the integrated value at the start of the
EMG from its peak value.
Statistical Analyses
The viscosity data were examined by using a general linear
model (GLM) method [12] with concentration, rotation
velocity, and measurement time serving as fixed factors.
This analysis was followed by Tukeys multiple-compari-
son tests for the detection of specific differences. Differ-
ences in the textural properties (hardness, cohesiveness,
123
A. Igarashi et al.: Food Properties and Swallowing
and adhesiveness) of the 1.5 and 3.0% test foods were
examined by Welchs modified t tests for unequal variance
data. The sensory evaluation data were examined by a
GLM method with concentration, volume, and subjects
serving as fixed factors followed by Tukeys multiple-
comparison tests for the detection of specific differences.
Differences in the durational and amplitude parameters
were examined by a GLM method with subject, concen-
tration, and volume of swallowing serving as fixed factors.
Marginal means of the levels of individual factors (con-
centration and volume) were also estimated and Tukeys
multiple-comparison tests were performed to detect spe-
cific differences between the levels of the factors. Signifi-
cance in all statistical analyses was determined to occur at
p \ 0.05.
Results
Food Properties
Figure 2 shows the data for the viscosities of the 1.5 and
3.0% test foods at the four different rotation velocities used
(which ranged from 6 to 60 rpm) 60 s after rotation started
(Fig. 2a) and at five time points (ranging from 30 to 150 s)
at a rotation velocity of 12 rpm (Fig. 2b, c). We did not
include the 0.0% test food in the statistical analysis since
the food showed very low viscosity (almost 0.0 Pa s)
regardless of the rotation velocity and the measurement
time. With regard to the effect of different rotation veloc-
ities on viscosity, we observed that the viscosity of the
1.5% test food was small (less than 10 Pa s at 6 rpm in
Fig. 2a) and decreased as the rotation velocity increased
from 12 to 60 rpm. In contrast, the 3.0% test food showed
much higher viscosity than the 1.5% food (more than
40 Pa s at 6 rpm in Fig. 2a) and its viscosity decreased as
the rotation velocity increased. With regard to the effect of
time on viscosity at a rotation velocity of 12 rpm, we
observed that the viscosity of the 1.5% test food 30 s after
the start of the rotor was less than 5 Pa s and decreased as
time progressed (150 s after starting, its viscosity was 93%
of the 30-s value; Fig. 2b). In addition, while the viscosity
of the 3.0% test food at 12 rpm was much higher than that
of the 1.5% food, its viscosity also decreased as time
progressed (Fig. 2c). The GLM method statistically con-
firmed that all three main effects (i.e., concentration of test
foods, rotation velocity, and measurement time) affected
significantly the viscosities of the test foods (p \ 0.01). In
addition, Tukeys multiple-comparison tests revealed that
30 s after starting the rotor, the viscosity of the 1.5 and
3.0% test foods was significantly higher than the viscosity
at the other time points (Fig. 2b, c).
A. Igarashi et al.: Food Properties and Swallowing
Fig. 2 Viscosity of the test foods. a Line graphs depict viscosity data
of the 1.5 and 3.0% test foods acquired at four rotation velocities (6,
12, 30, and 60 revolutions per minute). Both line graphs were
collected 60 s after the viscometer rotor was started. The open and
filled arrows indicate the viscosities of the 1.5 and 3.0% test foods,
respectively, that were measured at 12 revolutions per minute. b, c
Table 3 summarizes the textural properties (hardness,
cohesiveness, and adhesiveness) of the test foods. On
average, the 3.0% test food was about three times harder
than the 1.5% test food. Welchs modified t test showed
that this difference was statistically significant (p \ 0.01).
In addition, the 3.0% test food was on average at least
twice as cohesive and adhesive as the 1.5% test food
(Table 3). These differences were also statistically signif-
icant as determined by modified t tests (p \ 0.05 and
p \ 0.01, respectively).
Sensory Evaluation
Figure 3a depicts how varying the thickening agent con-
centration affected the texture and taste of the food, as
measured by the sensory evaluation experiment in which
each subject, after swallowing one the three test foods (0.0,
1.5, and 3.0%) [each of which was delivered in three dif-
ferent volumes (small, medium, and large), i.e., there were
nine tests in one experimental session], answered the six
questions listed in Table 2 by using a scoring system.
Table 3 Textural properties of test foods
Concentration Hardness (Pa) Cohesiveness Adhesiveness
(%) (J/m3)
1.5 464.2 67.0** 0.29 0.07* 203.5 59.4**
** *
3.0 1368.3 215.6 0.64 0.08 440.1 180.3**
Values are mean SD
* **
p \ 0.05, p \ 0.01
Line graphs depict viscosity data of the 1.5% (b) and 3.0% (c) test
foods at 12 revolutions per minute at five time points (30, 60, 90, 120,
and 150 s after the viscometer rotor was started). The open and filled
arrows indicate the viscosities of the 1.5 and 3.0% test foods,
respectively, that were measured 60 s after the start of the viscometer
rotor. See text for details
Analysis of the scores revealed that while the three test
foods varied in rank, the 0.0 and 3.0% test foods tended to
be in first and third place, respectively, for all questions
except question 1. Statistical analysis by the GLM method
revealed that for all six questions, the concentration of the
thickening agent significantly affected the scores (question
1, p \ 0.05; questions 2-6, p \ 0.01). Further analysis of
all possible concentration pairs with Tukeys tests detected
16 significant differences in total (Fig. 3a).
Figure 3b depicts how the different volumes of the test
foods affected the scores. Analysis of the scores generated
by answering the six questions revealed that the ranking of
the small, medium, and large volumes was inconsistent
(p [ 0.05). Moreover, significant differences were not
detected when Tukeys tests were performed (Fig. 3b).
Electrophysiological Data
Figure 4a summarizes the marginal means of six durational
parameters of the trajectories recorded by a piezoelectric
sensor during the swallowing of the three test foods (0.0,
1.5, and 3.0%). The overall durations (i.e., the on-off
parameter depicted in the first trajectory in Fig. 4a) tended
to increase as the concentration of the thickening agent
increased. However, changing the concentration did not
have a consistent effect on the other five durational
parameters. Use of the GLM method showed that the three
test foods differed significantly in three of the six dura-
tional parameters: (1) on-P (p \ 0.05), (2) on-B1
(p \ 0.01), and (3) B1-off (p \ 0.01). Further analysis of
test food pairs using Tukeys tests revealed significant
123

Fig. 3 The effect of different thickening agent concentrations and
test food volumes on the scores generated by answering six sensory
evaluation questions. a, b Point graphs summarize the effect of
thickening agent concentration (a) and test food volume (b) on the
marginal means and standard errors of means of the sensory
Fig. 4 Effect of thickening agent concentration on motor parameters.
The individual bars show the averages and standard errors of the data
collected from eight subjects. Table 1 gives the compositions of the
differences in three durations between four test food pairs:
(1) on-P1, 1.5% vs. 3.0% (p \ 0.05), (2) on-B1, 1.5% vs.
3.0% (p \ 0.01) and 0.0% vs. 3.0% (p \ 0.05), and (3) B1-
off, 1.5% vs. 3.0% (p \ 0.01) (Fig.4a).
Figure 4b summarizes the marginal means of three
durational parameters of the SH EMGs recorded after the
swallowing of the three test foods. The three durational
parameters tended to increase as the concentration of the
thickening agent increased. Use of the GLM method
revealed that the three test foods differed significantly in
two durational parameters: (1) on-off (p \ 0.05) and (2)
on-PE (p \ 0.01) (Fig. 4b). The three test foods did not
differ significantly in terms of the single amplitude
parameter that was measured (Fig. 4c).
123
A. Igarashi et al.: Food Properties and Swallowing
evaluation scores generated by testing 12 subjects. Table 2 lists the
six questions that were asked. The thin and thicker lines indicate
significant differences (p \ 0.05 and p \ 0.01, respectively). See text
for details
three test foods (0.0, 1.5, and 3.0%). See the legend to Fig. 1 for the
definitions of the parameter abbreviations. * p \ 0.05 and
** p \ 0.01. See text for details
Figure 5 summarizes the effect of changing the volume
of the test foods on the six durational piezoelectric sensor
trajectory parameters and the three durational parameters
and one amplitude SH EMG parameter. Except for P1-B1,
all durational and amplitude parameters tended to increase
as the volume increased. With regard to the piezoelectric
sensor trajectory parameters, use of the GLM method
revealed that the three volumes differed significantly in the
on-off parameter (p \ 0.05). Further analysis of volume
pairs using Tukeys tests detected a significant difference
in the on-off parameter between the small and large vol-
umes (p \ 0.05; Fig. 5a). With regard to the SH EMG
parameters, use of the GLM method revealed that the three
volumes differed significantly in the average on-off
A. Igarashi et al.: Food Properties and Swallowing
Fig. 5 Effect of test food volume on motor parameters. The
individual bars show the averages and standard errors of data
obtained by testing eight subjects. See the subsection Preparation of
Test Foods (in the Materials and Methods section) for a description
duration (p \ 0.01). Further analysis of volume pairs using
Tukeys tests detected a significant difference in the on-off
parameter between the small and large volumes (p \ 0.05;
Fig. 5b). The three volumes did not differ in terms of the
single amplitude parameter that was measured (Fig. 5c).
Discussion
An important feature of this study was the precise analysis
of the viscosity of the test foods that were used (Fig. 2).
This analysis initially revealed that while the 0.0% test food
acted very much like water, the viscosity of the 1.5 and
3.0% test foods decreased as the viscometer rotation
velocity increased (Fig. 2a). These observations suggested
that the latter two test foods are non-Newtonian fluids,
which are characterized by (1) flow properties that cannot
be described by a single constant viscosity and (2) a non-
linear relationship between shear stress and strain rate that
can be time-dependent [7]. Given these non-Newtonian
fluid properties, we then measured the viscosity of the 1.5
and 3.0% test foods at four rotation velocities 60 s after
starting the rotor (Fig. 2a) and over 150 s at 12 rpm rotation
(Fig. 2b, c). The 12-rpm rotation velocity used for the latter
experiment was chosen on the basis of food standards that
were developed especially for dysphagic patients in Japan.
We found that at 12 rpm, the viscosity of the 1.5 and 3.0%
test foods decreased as time progressed to 150 s (Fig. 2b, c).
This suggests that these two foods are thixotropic, i.e., their
apparent viscosity decreases with the time of shearing (i.e.,
their viscosity exhibits a time dependence [7]). To confirm
of the small, medium, and large test food volumes. See the legend to
Fig. 1 for the definitions of the abbreviations. * p \ 0.05. See text for
details
this suggestion from both the physiological and the clinical
perspective, we need to perform an experiment that enables
a comparison of two boluses of similar viscosity and other
sensory properties, one of which would behave like a
Newtonian fluid and the other of which would not.
For five of the six sensory evaluation questions
(Table 2), the scores obtained after swallowing the three
test foods can be explained by the food properties analyzed
in this study. First, question 1 asks about the textural
properties of the test foods, but it is difficult to explain the
result rationally from the instrumentally measured data in
this study (Table 3). Second, question 4 also asks about the
textural properties of the foods and indeed the resulting
score seems to reflect the differences in the adhesiveness
among the three foods (Table 3), although 0.0% food was
not an object of the instrumental measurement. Third,
questions 2, 5, and 6 ask about the rheological (Fig. 2) and
textural properties of the test foods, and indeed the scores
obtained after swallowing the three test foods seem to
reflect the differences in these properties between the test
foods. It should be noted that thickening agents have been
developed specifically to facilitate swallowing for dys-
phagic patients and to improve the safety of their swal-
lowing [13]; thus, while the dissolution of thickening
agents into fluids may improve the swallowing of dys-
phagic patients, this may not be the case for normal sub-
jects. However, we also observed that even the lowest
scores for the 3.0% test food for questions 2, 5, and 6
exceeded -0.5 (Fig. 3a), which were not the extremely low
scores. These two facts suggest that the thickening agent
used in the present study prevents the very low evaluation
of the test foods (1.5 and 3.0%) that would be expected
123

upon sensory evaluation by normal young adults. Finally,
question 3 asks about the impression of the taste of the test
foods and the scores of the question differ between thinner
(0.0 and 1.5%) and thick (3.0%) foods (Fig. 3a). In com-
parison to the thinner test foods, the 3.0% test food might
spread over the tongue insufficiently and stimulate the
lingual taste buds inadequately because of the high cohe-
siveness (Table 3) and high viscosity (Fig. 2) of the test
food.
In this study we found that thickening agent concen-
tration and test food volume both affected the durational
parameters of laryngeal movement that were recorded by a
piezoelectric sensor (Figs. 4 and 5) This is the first study to
examine the effect of thickening agent concentration on
both test food viscosity and laryngeal movement during
swallowing. Notably, a previous study that examined the
effect of test food volume on laryngeal movement [4]
showed that increasing the swallowing volume from 3 to
20 ml prolongs a duration parameter, namely, the 0-2
interval, which presumably corresponds to the pharyngeal
transit time [11]. It is not easy to match the wave com-
ponents measured in that study (which are labeled from 0
to 4 [11]) with ours (on, P1, B1, and off, Fig. 1), but it
appears that their 0 point is equivalent to our P1 point and
their 2 point is located just prior to our P2. If so, the
duration of the 0-2 period measured by the previous study
[11] is approximately equal to the duration from P1 to P2,
although we did not measure this duration for the present
study. Notably, the study described above shows that the
duration from the start (i.e., 0) to the end of the laryngeal
movement is prolonged as the swallowing volume
increases (Fig. 1 in [11]). Consistent with this, we also
observed that the on-off duration increased as the volume
increased (Fig. 5a). Thus, we confirm here that increasing
the bolus volume prolongs the duration of the laryngeal
elevation that accompanies the swallowing of liquids,
even when a small volume range from 3 to 8 ml is involved
[11].
Analysis of the EMG data we collected revealed that
thickening agent concentration (viscosity) significantly
affected the durational but not amplitude parameters that
were measured (Figs. 4 and 5). Specifically, we observed
that increasing the thickening agent concentration from 0.0
to 3.0% tended to prolong two durations of the integrated
SH EMG during swallowing, namely, the duration from the
start to the end and the duration from the start to the peak
(Fig. 4a). The effect of concentration on the whole duration
seems to be consistent with previous observations [1, 5],
although these studies used different test foods whose
rheological properties (e.g., Newtonian or non-Newtonian
fluids) are unclear (whereas the present study shows clearly
how viscosity and other food properties physiologically
affect SH EMG activity during swallowing). The previous
123
A. Igarashi et al.: Food Properties and Swallowing
reports also showed that increasing the viscosity of their
test foods augmented the SH (submental) EMG [1, 5],
whereas we did not observe that changing the viscosity
significantly changed the amplitude of the integrated SH
EMG (Figs. 4c and 5c). Rheological property differences
between the test foods in our study and those of the other
studies [1, 5] may be responsible for these disparate
observations regarding the effect of viscosity on SH EMG
amplitude.
In conclusion, this study examined the influence of
various food properties on both sensory and motor aspects
of swallowing in healthy young adults. We collected
sensory evaluation and electrophysiological data by using
three test foods containing different concentrations (0.0,
1.5, and 3.0%) of a thickening agent. The viscosity
measurements suggested that the 1.5 and 3.0% test foods
were non-Newtonian fluids with a thixotropic property.
We observed that thickening agent concentration (vis-
cosity) and test food volume significantly affected both
sensory and motor aspects of swallowing. While the
effects observed in this study are basically consistent with
previous findings, this is the first time that the rheological
properties of the test foods and their effects on sensory
and motor aspects of swallowing have been precisely
analyzed.
Acknowledgments We are grateful to Dr. Masako Fujiu-Kurachi
(Department of Eating Disorders and Dysphagia Rehabilitation,
University of Niigata Rehabilitation Graduate School) for her kind
help in preparing the manuscript. This study was supported in part
by Grants-in-Aid for Scientific Research from the Ministry of
Education, Science and Culture of Japan (No. 19592233 to AI and
No. 19500667 to YM) and from Niigata University of Health and
Welfare (to YM).
References
1. Dantas RO, Dodds WJ. Effect of bolus volume and consistency
on swallow-induced submental and infrahyoid electromyographic
activity. Braz J Med Biol Res. 1990;23:3744.
2. Dantas RO, Dodds WJ, Massey BT, Kern MK. The effect of
high- vs low-density barium preparations on the quantitative
features of swallowing. AJR Am J Roentgenol. 1989;153:
11915.
3. Dantas RO, Kern MK, Massey BT, Dodds WJ, Kahrilas PJ,
Brasseur JG, et al. Effect of swallowed bolus variables on oral
and pharyngeal phases of swallowing. Am J Physiol.
1990;258:G67581.
4. Ertekin C, Aydogdu I, Yuceyar N, Pehlivan M, Ertas M, Uludag
B, et al. Effects of bolus volume on oropharyngeal swallowing:
an electrophysiologic study in man. Am J Gastroenterol.
1997;92:204953.
5. Reimers-Neils L, Logemann J, Larson C. Viscosity effects
on EMG activity in normal swallow. Dysphagia. 1994;9:
1016.
6. Bisch EM, Logemann JA, Rademaker AW, Kahrilas PJ, Lazarus
CL. Pharyngeal effects of bolus volume, viscosity, and
A. Igarashi et al.: Food Properties and Swallowing

temperature in patients with dysphagia resulting from neurologic 13. Logemann JA. Evaluation and treatment of swallowing disorders.
impairment and in normal subjects. J Speech Hear Res. 2nd ed. Austin, TX: Pro-Ed; 1998.
1994;37:104159.
7. Bourne MC. Food texture and viscosity: concept and measure-
ment. 2nd ed. San Diego, CA: Academic Press; 2002. Atsuko Igarashi PhD
8. Stone H, Oliver S. Effect of viscosity on the detection of relative
sweetness intensity of sucrose solutions. J Food Sci. 1966;31: Maiko Kawasaki DDS
12934.
9. Szczesniak AS, Farkas E. Objective characterization of the Shu-ichi Nomura PhD
mouthfeel of the gum solutions. J Food Sci. 1962;27:3815.
10. Dewar RJ, Joyce MJ. The thixotropic and rheopectic behaviour of Yuji Sakai BAgr
maize starch and maltodextrin thickeners used in dysphagia
therapy. Carbohydrate Polymers. 2006;65:296305.
Mayumi Ueno BSc
11. Ertekin C, Pehlivan M, Aydogdu I, Ertas M, Uludag B, Celebi G,
et al. An electrophysiological investigation of deglutition in man.
Muscle Nerve. 1995;18:117786. Ichiro Ashida DAgr
12. Grafen A, Hails R. Modern statistics for the life sciences. Oxford,
UK: Oxford University Press; 2002. Yozo Miyaoka PhD

123