Journal of Oral Rehabilitation 2006 33; 2–7
Chewing-side determination of three food textures
J. PAPHANGKORAKIT*, N. THOTHONGKAM† & N. SUPANONT‡
Faculty of Dentistry, Khon Kaen University, Khon Kaen, Thailand, †Faculty of Dentistry, Naresuan University, Pitsanulok, Thailand and
‡
HRH Princess Maha Chakri Sirindhorn Medical Centre, Nakorn Nayok, Thailand
SUMMARY Food texture affects chewing movement
but it is not known if it also affects the chewing-side
pattern. This study determined the chewing sides of
three test foods with different textures during
habitual chewing. Twenty healthy dental students
(aged 20–24 years) chewed pieces of pork jerky,
fresh asparagus and almonds on two separate ses-
sions (1 week apart). In each session, each subject
chewed 30 food specimens, 10 of the same food type,
until swallowing while a video camera recorded the
displacement of the chin with respect to the other
two reference points vertically marked along the
facial midline. A slow-speed video playback was
used to identify the chewing side of each cycle. The
chewing-side pattern (right preference, left prefer-
ence, no preference) in each individual was deter-
mined statistically. The results showed that overall,
11 subjects did not have any side preference whereas
Introduction
Normal chewing is characterized by unilateral cycles
with periodic alternation of food between both sides of
the dentition although less unilateral cycles are
observed with homogenous and soft foods. Once the
food attains the consistency, homogeneity and cohe-
siveness ready to be swallowed, bilateral cycles can
occur (1, 2). The study of chewing-side pattern is useful
in understanding the neural control of chewing and the
design of dentures. Previous studies have shown that
most normal persons chew more on either the right or
left side, the so-called ‘preferred chewing side’ (3–5).
What determines the side preferably used during
chewing is not known. Studies have shown that the
preferred chewing side is not related to handedness
(6, 7) and not associated with the area of tooth contact
ª 2006 Blackwell Publishing Ltd
*Department of Oral Biology,
six and three subjects preferred to chew on right or
left sides respectively. The chewing-side pattern
remained unchanged between three food types in
about half of the subjects. When the same food was
compared between 2 days, the chewing-side pattern
of almonds was shown to be most reproducible (18
subjects). Unidentified cycles with little or no lateral
displacement, labelled as bilateral, were observed
more frequently near the end of the chewing
sequence with more occurrences in almonds and
jerky than asparagus (P < 0Æ01). It was suggested that
chewing-side preference is not a fixed characteristic.
Food texture seemed to influence the side prefer-
ence and also the occurrence of bilateral cycles.
KEYWORDS: preferred chewing side, food texture,
chewing pattern, jaw movement, mastication
Accepted for publication 3 May 2005
(8) or the chewing efficiency on that side (9). However,
Bourdiol and Mioche (10) found that the number of
cycles used during side-imposed chewing sequences of
frankfurters, cheese and toffee was less on the side with
larger occlusal contact area. Hannam et al. (11) sugges-
ted that the preferred chewing side was related to the
ability to move the jaw laterally towards that side. The
above variations could be due to the methods used to
verify the preferred chewing side. In some studies, the
chewing side was observed visually (6) while in others
more sophisticated tracking devices were used (5, 8,
12). The preferred chewing side was usually deter-
mined simply by the percentage of right- or left-side
chewing cycles without any statistical analysis except in
the study of Wilding and Lewin (4). Moreover, the
texture of test food also varies. Bread and toffee (3),
meat (5), gum (6), and wine-gums (4) have all been
2
 CHEWING-SIDE DETERMINATION OF THREE FOOD TEXTURES
used in previous studies. The texture of food can alter
some components of chewing cycle (2, 13–15) and
perhaps influence the preferred chewing side. The
present study was aimed to test if an individual’s
chewing-side pattern was affected by food texture and
if it differed between days.
Materials and methods
Experimental procedure
The study was approved by the Khon Kaen University
Ethical Committee. Twenty dental students, aged
20–24 years, having given informed consent, partici-
pated in this study. All subjects had normal masticatory
function with angle class I molar relationship and had
28 natural teeth (excluding third molars). The maxi-
mum lateral jaw excursion measured at the mid-
mandibular central incisors in each individual did not
show any apparent difference between right and left
sides. On the first day (day 1), each subject was asked to
chew three test foods, being pork jerky (1 cm long)*,
fresh asparagus (1 cm long selected from the middle
part of the stalk) and almond (approximately 2 cm
long, Blue Diamond)† representing tough, soft ductile
and brittle food respectively. As the physical properties
of test foods could be variable, the same foods used in
subjects were tested separately for their ultimate
strength and modulus of elasticity using a universal
testing machine‡ with a head speed of 5 mm min)1 as
follows. For almond, the whole seed was placed
horizontally and compressed between two metal plates.
For asparagus, a piece of 1 cm length selected from the
middle part, was placed horizontally and tested under
compression. For pork jerky, a 4-cm-long specimen was
cut with an orientation similar to that inserted into the
mouth and tested under tension (it was difficult to
recognize the orientation of muscle fibres in order to
test separately the specimens cut parallel and perpen-
dicular to the fibre orientation). Ten specimens of each
food were tested on two different days (7 days apart).
All foods tested were stored in the same way as in the
experimental sessions; almonds and jerky were kept in
a sealed container stored at room temperature whereas
the asparagus was newly bought from the supermarket
*Naem Lub Lae, Khon Kaen, Thailand.
†
Heritage Snacks and Beverages, Nakorn Pathom, Thailand.
‡ §
Lloyd, Lloyd Instruments, Hants, UK.
ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 2–7
3
(fresh supplies daily), wrapped in plastic and stored at
4
C not more than 2 days before use. During each
chewing session, subjects were seated upright in a
comfortable chair with a headrest and eyes aiming
horizontally while chewing the test food habitually
until they decided to swallow. In random order,
10 pieces of each of the three foods were given to each
subject, thus making a total of 30 food samples (trials)
being tested on each day. All subjects returned for the
second session 7 days later (day 2) and the entire series
was repeated. The technique used to observe the
chewing side is described as follows. Three pencilled
dots were marked on the subject’s facial skin, one at the
tip of the nose, one just below the center of the lower
lip and one on the chin. Subjects were asked to move
their head as little as they could during the recording
session so that the three marked dots were aligned
vertically during each chewing trial. A video camera
(Panasonic NV-VX3)§ was placed 1 m in front of the
subject and the picture zoomed in to cover the tip of the
nose and mouth region. The recorded tapes were played
off-line by another investigator using a slow-speed
playback mode of a video cassette player (Panasonic
NV-SR98)§ to determine the chewing side of each cycle.
A clear plastic ruler was placed over the subject’s mid-
facial line on the television screen to help distinguish
the path of chin movement. Using this method, most
cycles could be identified as right (R) or left (L) cycles.
Some of the cycles, however, could not be clearly
identified (U) because of indistinguishable transverse
movement. The validity of this visual method in
correctly identifying the right and left cycles was tested
by having three subjects’ chewing gum according to
any of four different preset sequences (12 cycles per
sequence). Ten chewing sequences (60 right and 60 left
cycles) were tested in total and the cycles were
identified by the same person as in the actual experi-
ments. The cross-tab analysis revealed agreement
cofficients (kappa) ranging from 0Æ70 to 0Æ83.
Data analysis
The number of right and left cycles was expressed as a
percentage of the total cycle number, i.e. %R and %L
respectively. On each day, the ‘chewing-side pattern’ of
a given subject chewing a given test food was deter-
mined by statistically comparing the %R and %L with
Matsushita Industrial, Osaka, Japan.
4 J . P A P H A N G K O R A K I T et al.

Mann–Whitney tests (n ¼ 10 trials for each food). Table 2. Side preference determination for each subject from two
A given subject was considered to have a ‘right’ sessions of 30 trials

chewing-side preference if %R was significantly

Subject %R %L Side preference*
(P < 0Æ05) greater than %L or a ‘left’ side preference if
%L was significantly (P < 0Æ05) greater than %R. Those 1 32Æ4 35Æ7 0Æ88
who did not show any side difference were classified as 2 40Æ6 37Æ5 0Æ06
3 39Æ3 44Æ3 0Æ88
‘no side preference’. The percentage of the unidentified
4 33Æ7 43Æ7 L (0Æ001)
(%U) cycles was not included in the comparison. Their 5 37Æ2 38Æ4 0Æ94
values were smaller than %R or %L and their inclusion 6 63Æ9 15Æ5 R (0Æ001)
would not have affected the determination of side 7 42Æ5 44Æ8 0Æ66
pattern. In order to further investigate the nature of the 8 46Æ2 40Æ9 0Æ23
9 38Æ1 43Æ0 0Æ24
unidentified cycles, the percentage of such cycles was
10 59Æ3 29Æ1 R (0Æ001)
calculated for each food type during the early, middle
11 38Æ9 47Æ3 0Æ32
and late (one-third) periods of the chewing sequence. 12 50Æ8 44Æ0 R (0Æ02)
The differences in the number of unidentified cycles for 13 39Æ2 47Æ8 L (0Æ03)
each period and food type were tested with two-way 14 50Æ5 34Æ7 R (0Æ001)
ANOVA. 15 46Æ5 39Æ2 0Æ42
16 43Æ3 28Æ6 R (0Æ001)
17 43Æ0 41Æ3 0Æ76
Results 18 38Æ9 50Æ0 L (0Æ02)
19 70Æ3 12Æ1 R (0Æ001)
The ultimate strength and the modulus of elasticity of 20 48Æ9 44Æ9 0Æ44
the test foods are shown in Table 1. The moduli of
%R and %L represent the percentage of cycles with side
elasticity were significantly different for the three food displacement to the right and left respectively (from 60 trials).
types, with pork jerky being the greatest, followed by *A given subject was considered to be predominantly right sided if
almond and asparagus (P < 0Æ01). The ultimate %R was significantly (P < 0Æ05) greater than %L or left sided if %L
strength, however, was greatest in almonds. There was significantly (P < 0Æ05) greater than %R. Unidentified or
presumed bilateral cycles were not taken into account (see Data
was no significant difference in both properties of the
Analysis). The numbers in brackets are P-values obtained from
same food type on two different days. All subjects used independent t-tests.
both sides to chew test foods and none of the subjects
chewed exclusively unilaterally. The numbers of cycles from all 60 trials were pooled, about half of the subjects
until swallowing (mean
SD) in chewing pork jerky, (11 subjects) did not reveal a preferred chewing side,
asparagus and almonds were 19Æ8
6Æ4, 10Æ9
3Æ7 and whereas six subjects preferred to chew on the right and
15Æ1
4Æ8 respectively and they were significantly three on the left (Table 2).
different (one-way ANOVA; P < 0Æ001). When the cycles Table 3 summarizes significance values for side pref-
erence in all trials for each day. Eleven of 20 subjects on
Table 1. Mechanical properties (mean
SD) of food samples day 1 and 10 of 20 subjects on day 2 changed their
(n ¼ 10 in each cell) chewing-side pattern for different food types. When
compared between 2 days, it was found that only seven
Day 1 Day 2 subjects consistently chewed the same food with the
Modulus of Ultimate Modulus of Ultimate
same side pattern (1, 4, 6, 8, 11, 16, 19). The chewing-
elasticity strength elasticity strength side pattern was most reproducible with almonds (18
(MPa) (MPa) (MPa) (MPa) subjects, excluding 5, 20), asparagus (12 subjects,
excluding 2, 7, 10, 12, 13, 14, 15, 17) and least
Almond 15Æ7
3Æ9 0Æ8
0Æ4 15Æ3
7Æ6 0Æ9
0Æ5
Pork jerky 23Æ1
10Æ0 0Æ40
0Æ16 20Æ4
10Æ4 0Æ43
0Æ19 reproducible with pork jerky (11 subjects, excluding 2,
Asparagus 1Æ1
0Æ2 0Æ3
0Æ1 1Æ3
0Æ4 0Æ3
0Æ04 3, 7, 9, 10, 13, 15, 17, 18). Six subjects (2, 7, 10, 13, 15,
17) changed their chewing-side pattern for both pork
Modulus of elasticity was calculated from the slope of the initial
linear part of the stress–strain curve. Ultimate strength was
jerky and asparagus on the second day.
determined by the highest stress at which pork jerky and almonds Only four subjects (subjects 4, 6, 10, 19) consistently
were broken apart and fresh asparagus yielded under force. showed a right- or left-side preference. 14 subjects did

ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 2–7

 CHEWING-SIDE DETERMINATION OF THREE FOOD TEXTURES 5

Table 3. Chewing-side determin-

Day 1 Day 2
ation established during each session
(days 1 and 2), with three different Subject Jerky Asparagus Almond Jerky Asparagus Almond
test foods
1 0Æ06 0Æ19 0Æ1 0Æ15 0Æ76 0Æ59
2 R(0.005) R(0Æ01) 0Æ06 L(0Æ001) L(0Æ02) 0Æ72
3 L(0.02) 0Æ58 0Æ9 0Æ91 0Æ18 0Æ9
4 0Æ06 0Æ25 L(0Æ001) 0Æ97 0Æ76 L(0Æ003)
5 0Æ22 0Æ91 0Æ24 0Æ4 0Æ73 L(0Æ01)
6 R(0Æ001) R(0Æ001) R(0Æ001) R(0Æ001) R(0Æ001) R(0Æ001)
7 L(0.03) 0Æ58 0Æ97 0Æ65 R(0Æ03) 0Æ06
8 0Æ91 0Æ68 0Æ35 0Æ34 0Æ28 0Æ49
9 L(0Æ04) 0Æ97 0Æ07 0Æ32 0Æ44 0Æ09
10 0Æ4 R(0Æ002) R(0Æ001) R(0Æ002) L(0Æ001) R(0Æ001)
11 0Æ22 0Æ53 0Æ25 0Æ72 0Æ35 0Æ64
12 0Æ12 0Æ53 0Æ08 0Æ08 R(0Æ03) 0Æ1
13 0Æ8 0Æ68 0Æ65 L(0Æ02) L(0Æ02) 0Æ56
14 R(0Æ01) 0Æ58 0Æ06 R(0Æ01) R(0Æ01) 1
15 R(0Æ05) R(0Æ02) 0Æ12 0Æ11 0Æ06 0Æ58
16 0Æ07 R(0Æ01) 0Æ06 0Æ06 R(0Æ03) 0Æ06
17 R(0Æ03) R(0Æ02) 0Æ64 0Æ67 0Æ94 0Æ2
18 0Æ07 0Æ44 0Æ07 L(0Æ001) 0Æ44 0Æ93
19 R(0Æ001) R(0Æ05) R(0Æ001) R(0Æ001) R(0Æ001) R(0Æ001)
20 0Æ4 0Æ28 R(0Æ01) 0Æ42 0Æ49 0Æ73

P-values for chewing-side preference obtained from Mann–Whitney tests are indicated for each
subject.
R ¼ right side preference; L ¼ left side preference; No label ¼ no side preference.

Table 4. Percentages (mean
SD)

Day 1 Day 2
of the unidentified cycles during
early, middle, and late periods of a Early Middle Late Early Middle Late
chewing sequence for each test food
(n ¼ 200 sequences in each cell) Jerky 6Æ0
10Æ6 8Æ4
11Æ9 12Æ8
14Æ5 2Æ4
6Æ5 5Æ1
10Æ3 9Æ5
12Æ4
Almond 4Æ8
8Æ5 7Æ7
10Æ1 12Æ2
12Æ2 2Æ0
5Æ9 4Æ6
8Æ8 9Æ4
12Æ0
Asparagus 4Æ0
8Æ1 7Æ3
10Æ3 8Æ8
8Æ7 1Æ5
6Æ2 3Æ2
6Æ9 5Æ1
7Æ7

Each period is obtained by dividing the chewing sequence into three equal intervals. Percentages
of unidentified cycles seen during late periods were, in general, higher than middle periods and
both were significantly higher than during early periods (P < 0Æ01). The percentages seen with
asparagus were always significantly lower than those with jerky and almond (P < 0Æ01).

not show a side preference when chewing almonds

(1, 2, 3, 7, 8, 9, 11, 12, 13, 14, 15, 16, 17, 18) whereas
with pork jerky and asparagus, more subjects showed a
side preference (subjects 2, 3, 6, 7, 9, 10, 12, 13, 14, 15,
16, 17, 18, 19), although with less consistency. The
percentages of unidentified cycles during early, middle
and late periods of the chewing sequence are shown in
Table 4. The percentage was increasingly larger during
the middle and late periods of the sequences (P < 0Æ01).
When compared between food types, the occurrence of
such cycles was significantly least with asparagus
(P < 0Æ01). The results were not appreciably different
between days 1 and 2.
Discussion
Although the method used to observe the chewing side
in this study was based on visual inspection, it showed a
fairly acceptable validity and had the advantage that it
did not use an intra-oral device that might interfere
with subjects’ natural jaw movements. The camera
might make subjects uneasy but its possible effect on
chewing behaviour was minimized by randomizing the
test foods. Movement of the chin as observed in this
study has been shown to correlate with that obtained
from a jaw-tracking device but provides a larger value
of lateral displacement (12) and this was in agreement

ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 2–7

6 J . P A P H A N G K O R A K I T et al.
with the present study. About the same number of our
subjects did or did not have a preferred chewing side.
This did not conform with some previous studies which
showed side preference in most subjects (4–6,16). The
difference could be due to the observation method,
number of trials, types of test food and the method used
in analysing the preferred chewing side. The determin-
ation of the preferred chewing side in this study was
similar to that used by Wilding and Lewin (4). They,
however, found that most subjects had a preferred
chewing side. The results of the present study suggest
that chewing-side preference is not a fixed character-
istic, especially when food texture or recording session
is different. More subjects (seven to nine subjects) had a
preferred chewing side when chewing tough jerky
compared to hard almonds (five subjects). Cutting
(grinding) tough food (like jerky) might require a more
comfortable chewing side whereas breaking almonds
might be indifferent between both sides of the jaw,
resulting in more frequent chewing-side alternation.
Interestingly, seven to nine subjects also showed a side
preference when chewing fresh asparagus. Although
tender foods are likely to be chewed indifferently on
either side of the jaw, fresh asparagus might be so
fibrous that it needed to be chewed on the preferred
side. Toughness of foods, therefore, might help explain
the choice of chewing side. Wilding and Lewin (4)
noticed a similarity of the chewing-side pattern
between two recording sessions in their subjects chew-
ing wine-gums. In the present study, almonds showed
the most reproducible chewing-side pattern (18 of
20 subjects), compared with asparagus and jerky. This
was probably because of the more regular texture of
wine-gums and almonds although the strength and
elasticity of pork jerky and asparagus tested on 2 days
was not significantly different. It was also possible that
the breakage of brittle foods like almonds might be
easily reproduced whereas that of jerky and asparagus
was less predictable. The unidentified cycles have never
been reported in previous preferred chewing-side
studies (4–6). These cycles could be true bilateral cycles
or those with such small lateral movements that the
current visual method was not able to detect them. This
type of cycle was more often observed during the later
parts of each chewing sequence (Table 4). The explan-
ation could be that as the chewing unfolded, the foods
became softer and more dissociated. Subjects tended
thus to use less lateral movement or even chew
bilaterally. Probably because of its fibrous texture, it
ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 2–7
was difficult to break fresh asparagus into pieces before
it could be chewed bilaterally. Dried pork jerky and
almonds, on the contrary, are more brittle and easily
broken into separate pieces, resulting in more bilateral
cycles. Dealing with natural food products, the present
study was limited by the difficulty to standardize the
rheological properties of the test foods, especially fresh
asparagus and pork jerky. Randomization of food
specimens might help to reduce the effect of such
variability. Some physical properties of the food sam-
ples were also tested in this study although more
comprehensive tests (such as plasticity and toughness
tests) might give a better picture. In conclusion, the
present study suggested that chewing-side pattern
depended on food texture and could change between
days. The determination of preferred chewing side
using natural foods of different textures, therefore, does
not seem to be a simple task. Among the three food
types tested, almond as brittle food seems to be the best
choice when food comminution (grinding effect) is
evaluated or if chewing-side reproducibility is needed
whereas tough or fibrous food may better reveal the
side preference when the cutting effect is at stake. It
should be added that the cohesiveness of the bolus
could play a role in the resultant side preference since
foods with great cohesion (e.g. fibrous food) are not
easily dissociated and tend to be chewed unilaterally.
Acknowledgments
The authors would like to thank Prof. J. W. Osborn for
reading and giving valuable advice on the original
version of this manuscript and Prof. P. W. Lucas who
advised on food property testing. This study was
supported by the Research Fund of Faculty of Dentistry,
Khon Kaen University.
References
1. Alexander RM. News of chews: the optimization of mastica-
tion. Nature. 1998;391:329.
2. Hiiemae K, Heath MR, Heath G, Kazazoglu E, Murray J,
Sapper D, Hamblett K. Natural bites, food consistency and
feeding behaviour in man. Arch Oral Biol. 1996;41:175–189.
3. Wictorin L, Hedegard B, Lundberg M. Cineradiographic
studies of bolus position during chewing. J Prosthet Dent.
1971;26:236–246.
4. Wilding RJ, Lewin A. A model for optimum functional human
jaw movements based on values associated with preferred
chewing patterns. Arch Oral Biol. 1991;36:519–523.
 CHEWING-SIDE DETERMINATION OF THREE FOOD TEXTURES
5. Stohler CS, Ash MM. Excitatory response of jaw elevators
associated with sudden discomfort during chewing. J Oral
Rehabil. 1986;13:225–233.
6. Christensen LV, Radue JT. Lateral preference in mastication: a
feasibility study. J Oral Rehabil. 1985;12:421–427.
7. Hoogmartens MJ, Stuyck J, Bockx E, Cheng N, Jacobs G,
Nijs J, Vanderfaeillie J, Vanhoof H, Wouters F. Stretch
sensitivity of primary spindle endings in masseter and tem-
poralis muscles: lateralization and association with unilateral
preference pattern of mastication. Electromyogr Clin Neuro-
physiol. 1980;20:369–381.
8. Wilding RJ, Adams LP, Lewin A. Absence of association
between a preferred chewing side and its area of functional
occlusal contact in the human dentition. Arch Oral Biol.
1992;37:423–428.
9. Wilding R. The association between chewing efficiency and
occlusal contact area in man. Arch Oral Biol. 1993;38:589–596.
10. Bourdiol P, Mioche L. Correlations between functional and
occlusal tooth-surface areas and food texture during natural
chewing sequences in humans. Arch Oral Biol. 2000;45:691–
699.
11. Hannam AG, De Cou RE, Scott JD, Wood WW. The relation-
ship between dental occlusion, muscle activity and associated
jaw movement in man. Arch Oral Biol. 1977;22:25–32.
ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 2–7
7
12. Gerstner GE, Fehrman J. Comparison of chin and jaw
movements during gum chewing. J Prosthet Dent.
1999;81:179–185.
13. Slagter AP, Bosman F, van der Glas HW, van der Bilt A.
Human jaw-elevator muscle activity and food comminution
in the dentate and edentulous state. Arch Oral Biol.
1993;38:195–205.
14. van der Bilt A, Weijnen FG, Ottenhoff FA, van der Glas HW,
Bosman F. The role of sensory information in the control of
rhythmic open-close movements in humans. J Dent Res.
1995;74:1658–1664.
15. Mioche L, Bourdiol P, Martin JF, Noel Y. Variations in human
masseter and temporalis muscle activity related to food
texture during free and side-imposed mastication. Arch Oral
Biol. 1999;44:1005–1012.
16. Kawazoe Y, Kotani H, Hamada T. Relation between integrated
electromyographic activity and biting force during voluntary
isometric contraction in human masticatory muscles. J Dent
Res. 1979;58:1440–1449.
Correspondence: Dr Jarin Paphangkorakit, Faculty of Dentistry, Khon
Kaen University, Khon Kaen 40002, Thailand.
E-mail: jarin@kku.ac.th