Joannis Grigoriadis Motor behavior during the first

Mats Trulsson
Krister G. Svensson
chewing cycle in subjects with fixed
tooth- or implant-supported prostheses

Authors’ affiliations: Key words: chewing, fixed implant-supported prostheses, fixed tooth-supported prostheses,
Joannis Grigoriadis, Mats Trulsson, Krister G. mastication, periodontal mechanoreceptor, sensory-motor control
Svensson, Department of Dental Medicine,
Karolinska Institutet, Huddinge, Sweden
Abstract
Corresponding author:
Joannis Grigoriadis
Objectives: Appropriate sensory information from periodontal mechanoreceptors (PMRs) is
Department of Dental Medicine important for optimizing the positioning of food and adjustment of force vectors during precision
Karolinska Institutet biting. This study was designed to describe motor behavior during the first cycle of a natural
P.O. Box 4064
SE-141 04 Huddinge, Sweden chewing task and to evaluate the role of such sensory input in this behavior.
Tel.: (+46) 8 524 88 055 Material and methods: While 10 subjects with natural dentition, 11 with bimaxillary fixed tooth-
Fax: (+46) 8 746 79 15 supported prostheses (TSP) and 10 with bimaxillary fixed implant-supported prostheses (ISP) (mean
e-mail: Joannis.Grigoriadis@ki.se
age 69 [range 61–83]) chewed a total of five hazelnuts, their vertical and lateral jaw movements
were recorded. Data obtained during the first chewing cycle of each hazelnut were analyzed.
Results: The amplitude of vertical and lateral mandibular movement and duration of jaw opening
did not differ between the groups, indicating similar behavior during this part of the chewing
cycle. However, only 30% of the subjects in the natural dentate group, but 82% of those in the
TSP and 70% in the ISP group exhibited slippage of the hazelnut during jaw closure in at least one
of five trials. The TSP and ISP groups also exhibited more irregular and narrower patterns of
motion (total lateral/vertical movement = 0.15 and 0.19, respectively, compared to 0.27 for the
natural group).
Conclusions: Subjects with fixed tooth- or implant-supported prostheses in both jaws show altered
behavior, including inadequate control of the hazelnut, during the first chewing cycle. We propose
that these differences are due to impairment or absence of sensory signaling from PMRs in these
individuals.

The masticatory process starts with food
intake and initial biting, followed by moving
the food to the posterior teeth for chewing,
which crushes and fragments the food and
presses in saliva to form a bolus that is trans-
ported further back in the mouth to be subse-
quently swallowed (Woda et al. 2006; Van der
Bilt 2011). Both chewing and biting involve
highly coordinated activities of several jaw
and facial muscles, together with the tongue.
Coordinated chewing is guided by rhythmic
signals produced by a “central pattern genera-
tor” (CPG) in the central nervous system
(CNS) and modified by other signals from
higher centers in the brain and from peripheral
sensory receptors situated in, for example, the
jaw muscles, temporomandibular joint, intra-
Date: oral mucosa, and periodontium (Dellow &
Accepted 31 December 2014
Lund 1971; Lund 1991). Appropriate informa-
To cite this article: tion from peripheral sensory receptors in the
Grigoriadis J, Trulsson M, Svensson KG. Motor behavior
during the first chewing cycle in subjects with fixed tooth- or orofacial area is also required to control the
implant-supported prostheses.
muscles involved in movement of the mandi-
Clin. Oral Impl. Res. 27, 2016, 473–480
doi: 10.1111/clr.12559 ble for precision biting (Lund 1991; Trulsson
& Johansson 1996a). In particular, appropriate
sensory information provided from periodon-
tal mechanoreceptors (PMRs) surrounding the
roots of natural teeth allows the nervous sys-
tem to optimize positioning of the food and
the force levels and vectors applied during pre-
cision biting (Trulsson & Johansson 1996b;
Trulsson 2006; Svensson & Trulsson 2009;
Svensson et al. 2013).
With fixed tooth-supported prostheses, con-
nection of several teeth in rigid constructions
will reduce the movement of individual teeth
and, consequently, the stimulation of PMRs
when forces are applied to these teeth
(Weinberg 1957a,b; Picton 1990; Nyman &
Lang 1994; Svensson et al. 2013). Moreover,
edentulous individuals with fixed implant-
supported prostheses in both jaws lack PMRs
and thus also the sensory information they
supply. These considerations explain the aber-
rant behavior of both of these groups in con-
nection with experimental and natural biting
involving holding, manipulating, and splitting

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 473
Grigoriadis et al
Motor behavior during the first chewing cycle
the food between the frontal teeth (Svensson
& Trulsson 2011; Svensson et al. 2013).
Inappropriate regulation of mandibular
movement can result in suboptimal occlusal
contact and difficulty in splitting food when
chewing. Indeed, subjects with implant-sup-
ported prostheses find it difficult to fragment
gelatinous food (Grigoriadis et al. 2011). Den-
tal status (e.g., the number of teeth remain-
ing, the presence of removable prostheses),
along with several other factors (e.g., tempo-
romandibular disorders, age, size of the food,
etc.) may influence also the adaptation of
mastication to the hardness of food (Peyron
et al. 2002; Woda et al. 2006). Earlier exami-
nation of chewing of standardized gelatinous
food of different degrees of hardness by sub-
jects with natural dentition or fixed implant-
supported prostheses demonstrated clearly
the importance of sensory information from
the PMRs for adaptation to the hardness (Gri-
goriadis et al. 2011).
In addition, undesirable occlusal contacts
during mastication, due to attenuated (fixed
tooth-supported prostheses) or no (fixed
implant-supported prostheses) input from the
PMRs, could cause unfavorable loading and
potential fracturing of prosthetic material.
Indeed, with fixed metal-ceramic partial den-
tures, both those supported by teeth and
implants, fracture of the ceramic material
occurs more frequently than with single-
tooth restorations (Lovgren et al. 2000; Link-
evicius et al. 2008; Wittneben et al. 2014).
As behavior and performance during biting,
including adaptation to the hardness of food,
by individuals with attenuated or no informa-
tion from the PMRs differs from that of natu-
rally dentated subjects, the behavior of such
patients during the initial stage of chewing,
when food is situated between the posterior
teeth, should also differ.
In the absence of information concerning
the direction of the force applied to the
tooth, the jaw muscles cannot direct the bite
force vector in an appropriate manner.
Accordingly, the aim of the present investiga-
tion was to describe and compare motor
behavior during the first cycle of a natural
chewing task in subjects with natural denti-
tion or fixed tooth- or implant-supported
prostheses in both jaws.
Material and methods
Subjects
Eleven subjects (six men and five women, 70
(61–83) years of age (mean (range))) with fixed
tooth-supported prostheses (the TSP group);
474 | Clin. Oral Impl. Res. 27, 2016 / 473–480
10 subjects (six men and four women, 72
(68–77) years of age) with fixed implant-sup-
ported prostheses (the ISP group); and 10 con-
trol subjects (six men and four women, 66
(61–72) years of age) with natural dentition
(the natural group) were included. The tooth-
supported prostheses (metal-ceramic type)
involved 9–14 (mean 11) units (including
abutment teeth and pontics) were supported
by 4–9 (mean 6.9) abutment teeth in each
jaw and had been in place for an average of
82 months (range 8–246 months). The per-
centage bone support (defined from the mar-
gin of the bridge to the apex of the root) was
assessed from available intraoral or pano-
ramic radiographs employing a Schei ruler
(Schei et al. 1959). A mean value for the
mesial and distal sides of each individual
tooth was determined and, thereafter, a mean
value for each subject calculated, giving an
overall mean bone support in the TSP group
of 80% (range 54–90%).
The fixed implant-supported prostheses
were of the metal-acrylic variety (except for
one metal-ceramic type in a maxilla);
extended to the premolar area; were sup-
ported by 4–6 (mean 5) (threaded machine- or
rough-surfaced) dental implants in each jaw;
and had been in place for an average of
81 months (range 1–240 months). All sub-
jects in the natural group had at least 24
occluding teeth and no known history of
periodontal disease.
The subjects in the TSP and ISP groups
were recruited from clinics specializing in
oral rehabilitation at the Department of Den-
tal Medicine at Karolinska Institutet and at
the Public Dental Service Clinics in Stock-
holm (Folktandv
arden Stockholms l€an AB,
Sweden), as well as from associated private
clinics in the greater Stockholm area. The
natural group consisted of present and former
staff members of Karolinska Institutet, as
well as members of a local senior citizens
organization.
All participants were in good general
health; visited their dentists on a regular
basis; had a normal intermaxillary relation-
ship; and exhibited no dental, oral or oro-fas-
cial symptoms or malfunction at the time of
testing. Some abutment teeth in the TSP sub-
jects had undergone a root canal procedure to
put a post and core in place to enable reten-
tion of the prostheses. Some premolar and
molar teeth of those with natural dentition
had been subjected to endodontic and/or
restorative treatment (e.g., received direct res-
torations or partial or fully covering crowns).
Written informed consent was provided by
all of the participants in accordance with the
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Declaration of Helsinki, and this study was
preapproved by the regional ethical review
board in Stockholm, Sweden (04-715/4).
Equipment
A device custom-built to monitor the man-
dibular movement in all three dimensions
(Ume
a University, Physiology Section, IMB,
Ume
a, Sweden) recorded changes in vertical
and horizontal positions of the incisor point
of the mandible with reference to the maxilla
(Grigoriadis et al. 2011; Svensson et al. 2013).
In briefly, the light-weight (220-g) head-worn
device, equipped with a total of eight mag-
netic sensors located beside the cheeks at
right angles to the horizontal Frankfurt plane,
tracked the position of a small
(10 9 5 9 5 mm) magnet attached (with
composite) to the labial surface of the man-
dibular central incisors, independently of the
posture of the head. This device interfered
minimally with oral function and allowed
free movement of the head (Fig. 1a).
EMG signals were recorded bilaterally from
the center of the masseter muscles by bipolar
surface electrodes (2 mm in diameter and
12 mm apart) attached with adhesive tape
and electrode gel after thorough cleaning of
the skin with aqueous alcohol.
Sounds related to the cracking of food were
recorded bilaterally using a custom-made
headset with microphones mounted inside
earplugs positioned at the entrance to the
external auditory meatus (Ume
a University,
Physiology Section, IMB, Ume
a, Sweden).
The microphones were calibrated for each
subject immediately to the experimental ses-
sions.
A detailed description of the devices
employed has been provided previously
(Svensson et al. 2013).
The task
A shelled, medium-sized hazelnut was placed
between the tongue and mid-section of the
hard palate by the subjects themselves, who
then closed their mouths and maintained
their teeth in an intercuspal position. There-
after, they were told to eat the hazelnut, with
no other instructions (unless they asked
which side they should chew on, in which
case the answer was whichever side they pre-
ferred).
The experimental procedure
The subjects were seated upright in a dental
chair, with their horizontal Frankfurter plane
approximately parallel to the floor, in a quiet
and electrically and magnetically shielded
room. After receiving verbal instructions, but
 Grigoriadis et al
Motor behavior during the first chewing cycle

(a) (b)

Fig. 1. (a) The device custom-built to monitor movement of the lower jaw relative to the upper jaw during chewing. Magnetic sensors (four on each side) located on arms pro-
jecting down from the frame track the position of a magnet attached to the labial surface of the mandibular incisors. EMG signals are recorded bilaterally from the centers of
the masseter muscles using bipolar surface electrodes. Sounds related to fracture of the hazelnut are recorded bilaterally by microphones located inside earplugs on a headset.
(b) Representative recordings from the first chewing cycle by a subject with natural dentition (subject no. 1 in the natural group, see Fig. 4). From top to bottom, these curves
illustrate the following: the vertical position, velocity, and acceleration of the mandible; the lateral position, velocity, and acceleration of the mandible; activity (the r.m.s.-pro-
cessed EMG signals) of the left and right masseter muscles; and sound recordings from the left and right microphones. The events of interest are as follows: the onset the jaw
opening (M1); end of the first jaw opening movement (M2); and end of the last jaw opening movement (M3) prior to the fracture of the hazelnut (M4). The fracture of the hazel-
nut was indicated by a clear sound signal ≥30% of the maximal sound recorded from the left or right microphone together with EMG activity in the left or right masseter mus-
cles associated with a positive vertical velocity (indicating closing of the mandible). In this particular chewing cycle, M2 and M3 were the same, that is, the hazelnut was
fractured during the first attempt.

no training, each performed the task five
times.
Data acquisition and analysis
Data collected during the first cycle of chew-
ing, defined as the period from the beginning
of jaw opening until initial fracture of the
hazelnut, in each trial were analyzed.
Mandibular movement (at 800 Hz), EMG
signals (3.2 kHz), and sound (25.6 kHz) were
all recorded and analyzed using a microcom-
puter-based data acquisition and analysis sys-
tem (WinSC/WinZOOM v1.54, Ume
a
University, Physiology Section, IMB, Ume
a,
Sweden). The velocity and acceleration of
mandibular movement were obtained by
symmetrical numerical time differentiation
(20 points) of the position and velocity,
respectively. The EMG signals were pro-
cessed as root-mean-squares (RMS) across a
moving time window containing 100
signals.
At several points of interest during the
individual trials, the time and positions were
recorded (Fig. 1b): The onset of jaw opening
(M1) was defined as the time-point at which
vertical acceleration at the beginning of man-
dible opening was maximal (peak negative
value) and the end of jaw opening (M2) when
the vertical velocity exceeded zero for the
first time thereafter. The fracture of the
hazelnut (M4) was defined as the time when
a clear sound signal ≥30% of the loudest sig-
nal from the left or right microphones was
obtained and the EMG activity in the left or
right masseter coincided with positive verti-
cal velocity (indicating closing of the mandi-
ble). The end of the last jaw opening (M3)
prior to the fracture of the hazelnut was
defined as the last time at which the vertical
velocity exceeded zero prior to M4. In cases
where the hazelnut was fractured during the
first attempt, M3 and M2 were the same,
while they were different if more than one
attempt was needed (see Fig. 2a). All of these
time-points were identified by the computer
software and checked thereafter manually for
error.
To quantify the range of motion, mandibu-
lar movement (M1 to M4) during the chew-
ing cycle was plotted from a frontal view
created by the analytical software (Win-
ZOOM) (Fig. 3a). The figures thus created
(one for every trial) were imported into
image-processing software (CorelDrawâ
Graphics Suite version 12.0, Corel Corp.,
Ottawa, ON, Canada) and the chewing cycle
“enclosed” by a line from the point of frac-
ture (corresponding to M4) to the start of jaw
opening (corresponding to M1) drawn with
the “Auto-closed curve” tool. The figure was
subsequently imported as a JPEG file into a
second graphic software program (Adobe
Photoshop CS4 version 11.0, Adobe Systems
Inc., San Jose, USA) and the area within the
peripheral lines filled in and the number of
pixels present determined (see Fig. 3b).
In an additional approach to quantifying
the lateral component of mandibular move-
ment, a line was drawn from start of jaw
opening (corresponding to M1) to the peak
vertical movement (corresponding to M2 or
M3) to produce a “cycle axis” and the longest
line that could be drawn perpendicular to the
first was added to generate a “cycle width”
(see Fig. 3c) (Piancino et al. 2005, 2008). The
ratio of cycle axis/cycle width was then cal-
culated for each chewing cycle.
For each subject, the data from all five tri-
als were combined to obtain mean values. In
the case of lateral movements, absolute

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 475 | Clin. Oral Impl. Res. 27, 2016 / 473–480
Grigoriadis et al
Motor behavior during the first chewing cycle

(a) (b)

Fig. 2. (a) Representative recordings from the first chewing cycle involving a slip (i.e., requirement of more than one attempt to crush the hazelnut; subject no. 1 in the ISP
group, see Fig. 4). From top to bottom, these curves illustrate the following: vertical mandibular movement and velocity; activity (the r.m.s.-processed EMG signals) of the left
and right masseter muscles; and sound recordings from the left and right microphones. The events of interest here are as follows: onset of jaw opening (M1); end of the first jaw
opening movement (M2); and end of the last jaw opening movement (M3) prior to fracture of the hazelnut (M4). The end of the last jaw opening movement (M3) was defined as
the last time-point at which the vertical velocity exceeded zero prior to the fracture (M4). (See the legend to Fig. 1b for additional definitions). (b) The occurrence of slips in at
least one of the five trials by naturally dentated subjects and those with fixed tooth-supported (TSP) or implant-supported (ISP) prostheses.

(a) (b) (c)

gender, group) were of primary interest, the
level for inclusion was set at P ≤ 0.10. In
subsequent pairwise comparisons, a P-value
≤ 0.05 was considered to be statistically
significant.

Results

General observations
All subjects began chewing by moving the
mandible downwards (jaw opening) to create
room for the hazelnut positioned (with the
assistance of the tongue) between the poster-
ior teeth. The mandible then moved
Fig. 3. (a) The figure created by plotting mandibular movements from the start of jaw opening (M1) to fracture of
upwards (jaw closing), and the hazelnut was
the hazelnut (M4) during a “first chewing cycle” from a frontal view. (b) Figure (a) imported into image-processing
software that “enclosed” the chewing cycle with a dashed line from the point of fracture (M4) to the start of jaw crushed between the teeth (Fig. 1b). The
opening (M1), filled in the enclosed area and counted the number of pixels present there. (c) Addition of a “cycle subjects then performed several additional
axis”, (i.e., a line connecting the start of jaw opening (M1) to the time-point of peak vertical movement (M2 or M3)) cycles of chewing to fragment the hazelnut
and a “cycle width” (i.e., the longest line that could be drawn perpendicular to the “cycle axis”). before swallowing it. However, only the first
chewing cycle, from the start of jaw opening
values were used. Group means and standard gender, group, trends during attempts and the until the nut was initially fractured, was
deviations (SD) for normally distributed data interaction between trend and group. The analyzed.
and medians (25–75 percentile) for trans- covariance structure utilized in all analyses A similar overall pattern of jaw opening
formed data were subsequently calculated as was unstructured and first-order autoregres- and closure was observed in all subjects in
specified in the Results. sive. The Akaike information criterion (AIC) the three groups (Fig. 4). However, several
and Bayes information criterion (BIC) were noteworthy differences indicating impaired
Statistical analyses checked (Weiss 2005). Residuals were motor control during this first chewing cycle
As several attempts to complete the task checked with respect to normality and were exhibited by the individuals with fixed
were made by the subjects, the data were approximately homogeneous variance. All tooth- or implant-supported prostheses.
analyzed employing a mixed-effects model analyses were performed using the SAS 9.2
for repeated measures. A fixed statistical statistical software (SAS Institutet INC., Performance
model centered for outcome mean and mean Cary, NC, USA). In initial statistical analyses In some cases, the hazelnut slipped and did
age was applied to analyze the effects of age, designed to determine which factors (age, not fracture during the first jaw closing, so

476 | Clin. Oral Impl. Res. 27, 2016 / 473–480 © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Fig. 4. The mandibular movements (plotted from a frontal view) from the start of jaw opening (M1) to fracture of
the hazelnut (M4) during a representative “first chewing cycle” by each subject with natural dentition (NAT) or a
fixed tooth- (TSP) or implant-supported prosthesis (ISP). All movements have been normalized in the vertical plane,
and some have been mirrored to facilitate comparison. The plots in gray are those illustrated in Figs 1b and 2a,
respectively. Note the wider and smoother mandibular movement of the subjects in the natural group.
that additional chewing was necessary
(Fig. 2a). The occurrence of slips differed sig-
nificantly between the groups (P = 0.031).
Fewer of the subjects with natural teeth
(30%) exhibited slips in at least one of their
five trials than those with tooth- (82%,
P = 0.038) or implant-supported (70%,
P = 0.006) prostheses (Fig. 2b).
Motor behavior
Duration
The time that elapsed from the start of jaw
opening until fracture of the hazelnut
(M1–M4) differed between the groups
(P = 0.049), with subjects with natural teeth
requiring 0.44 s (0.34–0.58) (median (25–75
percentile)), those with tooth-supported pros-
theses 0.57 s (0.39–0.74) (P = 0.109) and the
group with implant-supported prostheses
0.58 s (0.39–0.92) (P = 0.017). The average
durations of jaw opening (M1–M2) and of the
last jaw closing movement before fracture of
the hazelnut (M3–M4) for all three groups
were 0.29 s (0.07) (mean (SD)) and 0.2 s
(0.13–0.26) (median (25–75 percentile)),
respectively, with no difference between the
groups (P = 0.469 and 0.343).
Positions
The vertical and lateral positions of the man-
dible at the end of the first jaw opening
movement (M2) did not differ between the
groups (17.4 (0.21) (mean (SD)) and 1.4 mm
(1.2) for the natural group; 17.4 (0.28) and
1.2 mm (0.10) for the TSP group; and 17.8
(0.59) and 2.6 mm (0.40) for the ISP group;
P = 0.615 and 0.400, respectively).
Mandibular movements
Visual analysis of the range of motion of the
mandible during the first chewing cycle
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Grigoriadis et al
Motor behavior during the first chewing cycle
revealed a narrower pattern of movement for
the TSP and ISP groups (see Fig. 4).
The total number of pixels in the figures
created by plotting mandibular movement
(M1–M4) during the chewing cycle (see
Fig. 3b) differed between the groups
(P = 0.055), being 94.8 9 103 (38.7 9 103) for
the natural group, 68.6 9 103 (44.8 9 103) for
the TSP group (P = 0.049), and 63.2 9 103
(25.3 9 103) for the ISP group (P = 0.024),
clearly illustrating the wider range of mandib-
ular movement for those with natural teeth
compared to those with prostheses (Fig. 5a).
In addition, the ratio of cycle axis/cycle
width differed (P = 0.033), being 0.27 (0.13)
for the natural group, 0.15 (0.08) for the TSP
group (P = 0.009), and 0.19 (0.09) for the ISP
group (P = 0.084) (Fig. 5b).
In 66% of the subjects with natural denti-
tion, the largest part of the lateral displace-
ment of the mandible (as indicated by the
“cycle width”, see Fig. 3c) occurred during
jaw closure, whereas this occurred during jaw
opening in most of the subjects in the TSP
and ISP groups (67%, P = 0.00002 and 78%,
P = 0.00001, respectively) (test for simple
main effects: P = 0.001) (Fig. 5c).
The trajectory of the mandibular move-
ment was obviously smoother for the sub-
jects with natural dentition, while more
stuttering and probing behavior (regardless of
the presence or absence of slips) was apparent
for those with prostheses (see Fig. 4). To
quantify this difference, the number of times
the value for acceleration of the vertical
movement passed through “zero” in trials
without slips (i.e., when M2 = M3 see
Fig. 1b) was determined. Such a passage indi-
cates a switch from increasing to decreasing
velocity (or vice versa) of mandibular move-
ment. The natural group exhibited fewer pas-
sages (2.55 (2.25–3.15) (median (25–75
percentile))) than the TSP (4.0 (3.52–5.25),
P = 0.0004) and ISP (3.6 (2.4–4.5), P = 0.040)
groups (test for simple main effects:
P = 0.038). In addition, when examining all
trials, the highest number of passages
through “zero” for any single trial was 6 for
the natural group, 13 for the TSP group, and
20 for the ISP group, occurring in all cases
during trials when slips occurred (i.e.,
M26¼M3, see Fig. 2a).
Discussion
In this investigation, motor behavior during
the first cycle of chewing a hazelnut in indi-
viduals with differing dental status was
examined and compared. This is, to our
knowledge, the first time the first chewing
cycle, which is often disregarded when ana-
lyzing jaw kinematics during mastication,
has been analyzed in detail (Piancino et al.
2005, 2008; Farias Gomes et al. 2010; Grigor-
iadis et al. 2011; Molenaar et al. 2012). This
cycle is the initial stage of the chewing
sequence and involves a downward move-
ment of the mandible and proper positioning
of the hazelnut to be fractured between the
upper and lower posterior teeth, which due
to its irregular shape, requires a certain
degree of sensorimotor skill.
Performance
Successful fracture of the hazelnut during the
first closure of the jaw presumably indicates
proper positioning between the teeth. The
larger number of slips by the TSP (82%) and
ISP groups (70%) indicates inferior spatial
control in comparison with the natural group
(30%). This difference can be explained, at
least in part, by the impairment or the
absence of sensory input from the PMRs in
these individuals with prostheses.
Indeed, unimpaired sensory information
from the PMRs is of considerable significance
in connection with precision biting with the
anterior teeth, as well as for holding food and
biting with the posterior teeth (Johnsen et al.
2007; Svensson et al. 2013). The PMRs pro-
vide detailed information concerning spatial
location and the direction of forces applied to
teeth and the force load transmitted by a
fixed tooth-supported prosthesis to all abut-
ment teeth, and their periodontium in a com-
plex manner probably alters the pattern of
PMR signaling from the abutment teeth,
making it more difficult to crush the hazel-
nut successfully (Weinberg 1957a,b; Trulsson
et al. 1992; Trulsson 1993; Johnsen &
Trulsson 2003). Individuals with fixed
477 | Clin. Oral Impl. Res. 27, 2016 / 473–480
Grigoriadis et al
Motor behavior during the first chewing cycle

(a) (b) (c)

Fig. 5. (a) Total number of pixels in the figures created by plotting mandibular movement during the first chewing cycle (see Fig. 3b). (b) The ratio of lateral to vertical move-
ment of the mandible (calculated as cycle width/cycle axis) during the first chewing cycle (see Fig. 3c). (c) Relative lateral mandibular displacement associated with jaw opening
or closure during the first chewing cycle (see Fig. 4c), where the largest proportion of cycle width was associated with jaw closure. NAT = naturally dentated, TSP = fixed
tooth-supported prosthesis, and ISP= fixed implant-supported prosthesis.

implant-supported prostheses lack PMRs and
thereby also the sensory information they
can provide entirely. In such cases, other
types of mechanoreceptors in the oro-facial
tissues, with different sensitivities (e.g., pos-
sibly those located in muscle, joint, mucosal,
cutaneous and/or periosteal tissues) must
provide information about the magnitude and
direction of forces employed during biting
and chewing (Dellow & Lund 1971; Lund
1991; Trulsson & Johansson 2002; Luraschi
et al. 2012).
Differences in the surface structure of the
artificial teeth (i.e., for the subjects in the
TSP and ISP groups) might have influenced
the friction between the teeth and food.
Indeed, the friction coefficient for acrylic
resin is somewhat lower than for porcelain or
enamel (Tillitson et al. 1971; Koran et al.
1972; Schuh et al. 2005). However, we
believe that the impact of this factor on our
present findings is negligible.
Another possible explanation for the larger
number of slips by the TSP and ISP groups is
reduced occlusal area of their posterior teeth,
which complicates keeping the hazelnut
between the teeth during the first chewing
cycle. Narrower posterior teeth (in the bucco-
lingual direction) are often included in
implant-supported prosthesis to reduce the
area that transmits the bite force and thereby
the load on the implants; and the pontics of
tooth-supported prosthesis are sometimes
narrow for a similar reason, that is, to reduce
the load on abutment teeth (Douglas 2003;
Klineberg et al. 2007). Although for the pros-
theses involved here (including at least 10
units for the ISP and nine units for the TSP
group) no calculation of the bucco-lingual
width was performed, no pronounced devia-
tions from “normal” dimensions were
observed.
Motor behavior
All three groups moved their mandible down-
wards to a similar extent to provide a similar
amount of space between their teeth for the
hazelnut, employing similar amounts of time
for jaw opening (M1–M2) and closing (M3–
M4). The longer period from the start of jaw
opening until fracture of the hazelnut (M1–
M4) for the TSP and ISP groups can be
explained by their more frequent requirement
for more than one attempt to crush the nut
(due to more slips) (M26¼M3, Fig. 2b).
A natural chewing cycle in dentated sub-
jects includes vertical, lateral, and sagittal
movements of the mandible, the first two of
which were analyzed here. Our finding that
this lateral movement was approximately
27% of the vertical movement in the natural
group is in agreement with other values of
20–30% reported for masticatory movement
during free chewing by naturally dentated
healthy subjects (Pr€oschel & Hofmann 1988;
Shiga et al. 2003; Piancino et al. 2008). How-
ever, Pr€oschel & Hofmann (1988) and Shiga
et al. (2003) defined the lateral displacement
as the maximal lateral excursion, whereas
like Piancino et al. (2005, 2008), we mea-
sured the “cycle width” (Fig. 3c).
During a chewing cycle, the mandible
may initially move away from the working
side (where the food is to be bitten) before
swinging back toward that side (Gibbs et al.
1971; Miles et al. 2004), and this lateral
swing during closure may facilitate grasping
the food in the correct position between
the teeth. During the first cycle of chewing
the hazelnut, our TSP and ISP subjects
demonstrated a narrower pattern of move-
ment (28% and 33% fewer pixels) than the
natural group, as well as by a smaller lat-
eral displacement of the mandible (15% and
19% of the vertical movement, respectively,
versus 27% for the natural group). This
behavior might reflect differences in the
characteristics of the dental arches, in
which the posterior teeth were predomi-
nantly premolars in the TSP and ISP and
molars in the natural group. Indeed, the tra-
jectory of movement depends on the posi-
tioning of the food within the dental arch,
with narrower lateral movement (more
“chopping-like”) during chewing with first
premolars than first molars in dentate indi-
viduals (Hashii et al. 2009). Nevertheless,
we propose that the less pronounced lateral
movement observed here can be explained

478 | Clin. Oral Impl. Res. 27, 2016 / 473–480 © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

by a lack of appropriate information supplied
by the PMRs, so that the participants in the
TSP and ISP groups probably used the tongue
and cheek to keep the hazelnut between the
teeth and closed the mandible in a manner
more similar to the opening movement (Tak-
ada et al. 1996). Another possibility is that
the subjects with ISP continue to use the
strategy seen in edentulous subjects wearing
complete dentures, that is, more “chopping-
like” mandibular movements designed to sta-
bilize the dentures during chewing (Tallgren
et al. 1989; Posti
c et al. 1992).
Several studies on subjects with natural
dentition have revealed larger angles of
approach to the intercuspal position for the
mandible during jaw closure than opening
(Pr€oschel & Hofmann 1988; Wilding & Le-
win 1994; Shiga et al. 2003). This lateral
movement during jaw closure moves and
positions the morsel of food between the
teeth and, in addition, provides maximal
occlusal contact area with the food. The lat-
eral approach just before and during occlusal
contact allows the teeth to glide on a layer
of food to fractionate tough material for
maximal chewing efficiency (Suit et al.
1976; Yamashita et al. 1999; Rilo et al.
2009). Although we did not analyze jaw
kinematics following fracture of the hazel-
nut here, the wider range of mandibular
movement in general and relatively larger
lateral displacement during jaw closure for
the subjects with natural teeth indicate that
they employ a more lateral approach during
closure (Figs 4 and 5).
Spatial control
Kinesiographic as well as jerk-cost studies
have proposed that smooth jaw movements
during mastication are desirable (Mongini &
Tempia-Valenta 1984; Evans & Lewin 1986;
References
Birznieks, I., Macefield, V.G., Westling, G. &
Johansson, R.S. (2009) Slowly adapting mechano-
receptors in the borders of the human fingernail
encode fingertip forces. Journal of Neuroscience
22: 9370–9379.
Dellow, P.G. & Lund, J.P. (1971) Evidence for cen-
tral timing of rhythmical mastication. Journal of
Physiology 215: 1–13.
Douglas, R.D. (2003) Contemporary Fixed Prostho-
dontics. In: Rosenstiel, S.F., Land, M.F. & Fujim-
oto, J., eds. 4th edition, Pontic design, 617–648.
Philadelphia: Elsevier.
Evans, W.G. & Lewin, A. (1986) Some characteris-
tics of mandibular movement in a group of San.
Journal of the Dental Association of South Africa
41: 543–548.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Grigoriadis et al
Motor behavior during the first chewing cycle
Pr€oschel 1987; Yashiro et al. 1999). The more
irregular movements (and changes in vertical
acceleration) of the mandible, seen in our
subjects with prostheses, cannot readily be
attributed to lack of appropriate signaling
from the PMRs and are difficult to under-
stand. As the position of the food bolus (ante-
rior or posterior) during mastication does not
influence the smoothness of jaw movement
in subjects with natural dentition (Molenaar
et al. 2012), it appears reasonable to conclude
that the shorter dental arch, sometimes pres-
ent in individuals with fixed tooth- or
implant-supported prostheses, probably does
not per se influence the smoothness of their
jaw movements.
As biting and chewing are so rapid, regula-
tion of the direction and levels of forces must
occur in a predictive, feed-forward, sensory-
motor manner (Trulsson & Johansson 1996b;
Trulsson 2006). We propose that a stroke of
effective chewing demands awareness of the
positions of the maxillary and mandibular
teeth for optimizing muscular coordination
of the mandible to obtain efficient tooth-
food-tooth contact, an awareness that
requires appropriate input from the PMRs.
When this information is lacking, the “refer-
ence point” provided by the teeth appears to
be lost and the subjects must use another
strategy involving more probing with irregu-
lar movements, as seen here.
As in the case of the slowly adapting type
II mechanoreceptors (SA II) found in the skin,
PMRs have been proposed to be part of a gen-
eral proprioceptive system that provides a
representation of the body to the CNS (Bir-
znieks et al. 2009; Trulsson & Essick 2010).
Furthermore, an fMRI study revealed that
mechanical stimulation of teeth activates a
specific region of the cerebral cortex (the pos-
terior insula) involved in the phenomenon of
Farias Gomes, S.G., Custodio, W., Moura Jufer, J.S.,
Del Bel Cury, A.A. & Rodrigues Garcia, R.C.
(2010) Correlation of mastication and masticatory
movements and effect of chewing side preference.
Brazilian Dental Journal 21: 351–355.
Gibbs, C.H., Messerman, T., Reswick, J.B. & Der-
da, H.J. (1971) Functional movements of the
mandible. Journal of Prosthetic Dentistry 26:
604–620.
Grigoriadis, A., Johansson, R.S. & Trulsson, M.
(2011) Adaptability of mastication in people with
implant-supported bridges. Journal of Clinical
Periodontology 38: 395–404.
Hashii, K., Tomida, M. & Yamashita, S. (2009) Influ-
ence of changing the chewing region on mandibular
movement. Australian Dental Journal 54: 38–44.
“body ownership” (Karnath et al. 2005;
Trulsson et al. 2010). Therefore, the less
accurate spatial control of the mandible
observed for our subjects with fixed tooth- or
implant-supported prostheses may simply
reflect the fact that they do not know exactly
where their teeth are.
Conclusion
During the first chewing cycle, individuals
with fixed tooth- or implant-supported pros-
theses in both jaws exhibit less accurate con-
trol of mandibular movement (as demonstrated
by poorer performance and more slips) and
more probing behavior designed to establish
moment-to-moment control (demonstrated by
more irregular movements) than those with
natural dentition. We propose that these differ-
ences can be explained by the impairment or
the absence of sensory signaling from PMRs
associated with such prostheses.
Acknowledgements: We are grateful
to G€oran Westling and Anders B€ackstr€ om
(engineers at the Department of Integrative
Medical Biology, Ume
a University) for their
technical and software support and for the
assistance of Magnus Backheden (statistician
at the Department of Learning, Informatics,
Management and Ethics, LIME, Karolinska
Institutet). This study was supported
financially by grants from the Stockholm
County Council, King Gustaf V0 s and Queen
Victoria0 s Freemason Foundation, American
Dental Society of Sweden, Swedish Dental
Society, the Swedish National Graduate
School in Odontological Science, and
Karolinska Institutet. The authors declare
that they have no conflict of interests.
Johnsen, S.E., Svensson, K.G. & Trulsson, M. (2007)
Forces applied by anterior and posterior teeth and
roles of periodontal afferents during hold-and-split
tasks in human subjects. Experimental Brain
Research 178: 126–134.
Johnsen, S.E. & Trulsson, M. (2003) Receptive field
properties of human periodontal afferents
responding to loading of premolar and molar
teeth. Journal of Neurophysiology 89: 1478–1487.
Karnath, H.O., Baier, B. & N€agele, T. (2005) Aware-
ness of the functioning of one’s own limbs medi-
ated by the insular cortex? Journal of
Neuroscience 3: 7134–7138.
Klineberg, I., Kingston, D. & Murray, G. (2007) The
bases for using a particular occlusal design in
tooth and implant-borne reconstructions and
479 | Clin. Oral Impl. Res. 27, 2016 / 473–480
Grigoriadis et al
Motor behavior during the first chewing cycle
complete dentures. Clinical Oral Implants
Research 18: 151–167.
Koran, A., Craig, R.G. & Tillitson, E.W. (1972)
Coefficient of friction of prosthetic tooth materi-
als. Journal of Prosthetic Dentistry 27: 269–274.
Linkevicius, T., Vladimirovas, E., Grybauskas, S.,
Puisys, A. & Rutkunas, V. (2008) Veneer fracture
in implant-supported metal-ceramic restorations.
Part I: overall success rate and impact of occlusal
guidance. Stomatologija 10: 133–139.
Lovgren, R., Andersson, B., Carlsson, G.E. &
Odman, P. (2000) Prospective clinical 5-year
study of ceramic-veneered titanium restorations
with the Procera system. Journal of Prosthetic
Dentistry 84: 514–521.
Lund, J.P. (1991) Mastication and its control by the
brain stem. Critical Reviews in Oral Biology and
Medicine 2: 33–64.
Luraschi, J., Schimmel, M., Bernard, J.P., Gallucci,
G.O., Belser, U. & M€ uller, F. (2012) Mechanosen-
sation and maximum bite force in edentulous
patients rehabilitated with bimaxillary implant-
supported fixed dental prostheses. Clinical Oral
Implants Research 23: 577–583.
Miles, T.S., Nauntofte, B. & Svensson, P. (2004) Clin-
ical Oral Physiology. In: Miles, T.S., Nauntofte, B.
& Svensson, P., eds. Mastication, 1st edition,
219–243. Copenhagen: Quintessence Publishing.
Molenaar, W.N., Gezelle Meerburg, P.J., Luraschi,
J., Whittle, T., Schimmel, M., Lobbezoo, F., Peck,
C.C., Murray, G.M. & Minami, I. (2012) The
effect of food bolus location on jaw movement
smoothness and masticatory efficiency. Journal of
Oral Rehabilitation 39: 639–647.
Mongini, F. & Tempia-Valenta, G. (1984) A graphic
and statistical analysis of the chewing move-
ments in function and dysfunction. Journal of
Cranio-Mandibular Practice 2: 125–134.
Nyman, S.R. & Lang, N.P. (1994) Tooth mobility
and the biological rationale for splinting teeth.
Periodontology 2000 4: 15–22.
Peyron, M.A., Lassauzay, C. & Woda, A. (2002)
Effects of increased hardness on jaw movement
and muscle activity during chewing of visco-elas-
tic model foods. Experimental Brain Research
142: 41–51.
Piancino, M.G., Bracco, P., Vallelonga, T., Merlo,
A. & Farina, D. (2008) Effect of bolus hardness on
the chewing pattern and activation of masticatory
muscles in subjects with normal dental occlusion.
Journal of Electromyography and Kinesiology 18:
931–937.
Piancino, M.G., Farina, D., Talpone, F., Castroflo-
rio, T., Gassino, G., Margarino, V. & Bracco, P.
(2005) Surface EMG of jaw-elevator muscles and
chewing pattern in complete denture wearers.
Journal of Oral Rehabilitation 32: 832–870.
Picton, D.C. (1990) Tooth mobility - an update.
European Journal of Orthodontics 12: 109–115.
Postic, S.D., Krstic, M.S. & Teodosijevic, M.V.
(1992) A comparative study of the chewing cycles
480 | Clin. Oral Impl. Res. 27, 2016 / 473–480
of dentate and denture-wearing subjects. Interna-
tional Journal of Prosthodontics 5: 244–256.
Pr€
oschel, P. (1987) An extensive classification of
chewing patterns in the frontal plane. Journal of
Craniomandibular Practice 5: 55–63.
Pr€
oschel, P. & Hofmann, M. (1988) Frontal chewing
patterns of the incisor point and their dependence
on resistance of food and type of occlusion. Jour-
nal of Prosthetic Dentistry 59: 617–624.
Rilo, B., Fernandez-Formoso, N., Mora, M.J.,
Cadarso-Suarez, C. & Santana, U. (2009) Distance
of the contact glide in the closing masticatory
stroke during mastication of three types of food.
Journal of Oral Rehabilitation 36: 571–576.
Schei, O., Waerhaug, J., Lovdal, A. & Arno, A.
(1959) Alveolar bone loss as related to oral
hygiene and age. Journal of Periodontology 30:
7–16.
Schuh, C., Kinast, E.J., Mezzomo, E. & Kapczinski,
M.P. (2005) Effect of glazed and polished surface
finishes on the friction coefficient of two low-fus-
ing ceramics. Journal of Prosthetic Dentistry 93:
245–252.
Shiga, H., Kobayashi, Y., Arakawa, I. & Shonai, Y.
(2003) Selection of food and chewing side for eval-
uating masticatory path stability. Odontology 91:
26–30.
Suit, S.R., Gibbs, C.H. & Benz, S.T. (1976) Study of
gliding tooth contacts during mastication. Journal
of Periodontology 47: 331–334.
Svensson, K.G., Grigoriadis, J. & Trulsson, M.
(2013) Alterations in intraoral manipulation and
splitting of food by subjects with tooth- or
implant-supported fixed prostheses. Clinical Oral
Implants Research 24: 549–555.
Svensson, K.G. & Trulsson, M. (2009) Regulation of
bite force increase during splitting of food. Euro-
pean Journal of Oral Sciences 117: 704–710.
Svensson, K.G. & Trulsson, M. (2011) Impaired
force control during food holding and biting in
subjects with tooth- or implant-supported fixed
prostheses. Journal of Clinical Periodontology 38:
1137–1146.
Takada, K., Yashiro, K., Sorihashi, Y., Morimoto, T.
& Sakuda, M. (1996) Tongue, jaw, and lip muscle
activity and jaw movement during experimental
chewing efforts in man. Journal of Dental
Research 75: 1598–1606.
Tallgren, A., Mizutani, H. & Tryde, G. (1989) A
two-year kinesiographic study of mandibular
movement patterns in denture wearers. Journal of
Prosthetic Dentistry 62: 594–600.
Tillitson, E.W., Craig, R.G. & Peyton, F.A. (1971)
Friction and wear of restorative dental materials.
Journal of Dental Research 50: 149–154.
Trulsson, M. (1993) Multiple-tooth receptive fields
of single human periodontal mechanoreceptive
afferents. Journal of Neurophysiology 69: 474–481.
Trulsson, M. (2006) Sensory-motor function of
human periodontal mechanoreceptors. Journal of
Oral Rehabilitation 33: 262–273.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Trulsson, M. & Essick, G.K. (2010) Sensations
evoked by microstimulation of single mechanore-
ceptive afferents innervating the human face and
mouth. Journal of Neurophysiology 103:
1741–1747.
Trulsson, M., Francis, S.T., Bowtell, R. & McGlone,
F. (2010) Brain activations in response to vibrotac-
tile tooth stimulation: a psychophysical and fMRI
study. Journal of Neurophysiology 104:
2257–2265.
Trulsson, M. & Johansson, R.S. (1996a) Encoding of
tooth loads by human periodontal afferents and
their role in jaw motor control. Progress in Neu-
robiology 49: 267–284.
Trulsson, M. & Johansson, R.S. (1996b) Forces
applied by the incisors and roles of periodontal
afferents during food-holding and -biting tasks.
Experimental Brain Research 107: 486–496.
Trulsson, M. & Johansson, R.S. (2002) Orofacial
mechanoreceptors in humans: encoding charac-
teristics and responses during natural orofacial
behaviors. Behavioural Brain Research 135:
27–33.
Trulsson, M., Johansson, R.S. & Olsson, K.A. (1992)
Directional sensitivity of human periodontal
mechanoreceptive afferents to forces applied to
the teeth. Journal of Physiology 447: 373–389.
Van der Bilt, A. (2011) Assessment of mastication
with implications for oral rehabilitation: a
review. Journal of Oral Rehabilitation 38:
754–780.
Weinberg, L.A. (1957a) Force distribution in
splinted anterior teeth. Oral Surgery Oral Medi-
cine Oral Pathology 10: 484–494.
Weinberg, L.A. (1957b) Force distribution in
splinted posterior teeth. Oral Surgery Oral Medi-
cine Oral Pathology 10: 1268–1276.
Weiss, R.E. (2005) Modeling Longitudinal Data,
146–151. New York: Springer.
Wilding, R.J. & Lewin, A. (1994) The determination
of optimal human jaw movements based on their
association with chewing performance. Archives
of Oral Biology 39: 333–343.
Wittneben, J.G., Buser, D., Salvi, G.E., B€urgin, W.,
Hicklin, S. & Br€agger, U. (2014) Complication
and failure rates with implant-supported fixed
dental prostheses and single crowns: a 10-year
retrospective study. Clinical Implant Dentistry
and Related Research 16: 356–364.
Woda, A., Foster, K., Mishellany, A. & Peyron,
M.A. (2006) Adaptation of healthy mastication to
factors pertaining to the individual or to the food.
Physiology and Behavior 89: 28–35.
Yamashita, S., Hatch, J.P. & Rugh, J.D. (1999) Does
chewing performance depend upon a specific
masticatory pattern? Journal of Oral Rehabilita-
tion 26: 547–553.
Yashiro, K., Yamauchi, T., Fujii, M. & Takada, K.
(1999) Smoothness of human jaw movement dur-
ing chewing. Journal of Dental Research 78:
1662–1668.