Vol. 85, Issue 3, 927-934, September 1998
Muscle coordination in cycling: effect of surface incline and
posture
Li
Li and
Graham E.
Caldwell
Biomechanics Laboratory, Department of Exercise Science,
University of Massachusetts, Amherst, Amherst, Massachusetts
01003
 |
ABSTRACT |
The purpose of the present study was to examine the
neuromuscular modifications of cyclists to changes in grade and
posture. Eight subjects were tested on a computerized ergometer under
three conditions with the same work rate (250 W): pedaling on the level while seated, 8% uphill while seated, and 8% uphill while standing (ST). High-speed video was taken in conjunction with surface
electromyography (EMG) of six lower extremity muscles. Results showed
that rectus femoris, gluteus maximus (GM), and tibialis anterior had
greater EMG magnitude in the ST condition. GM, rectus femoris, and the vastus lateralis demonstrated activity over a greater portion of the
crank cycle in the ST condition. The muscle activities of gastrocnemius
and biceps femoris did not exhibit profound differences among
conditions. Overall, the change of cycling grade alone from 0 to 8%
did not induce a significant change in neuromuscular coordination. However, the postural change from seated to ST pedaling at 8% uphill
grade was accompanied by increased and/or prolonged muscle activity of hip and knee extensors. The observed EMG activity patterns
were discussed with respect to lower extremity joint moments.
Monoarticular extensor muscles (GM, vastus lateralis) demonstrated
greater modifications in activity patterns with the change in posture
compared with their biarticular counterparts. Furthermore, muscle
coordination among antagonist pairs of mono- and biarticular muscles
was altered in the ST condition; this finding provides support for the
notion that muscles within these antagonist pairs have different
functions.
coordination; muscle activity; biarticular muscles
| |
INTRODUCTION |
THE MAJORITY OF CYCLING STUDIES have examined questions
regarding riding on level surfaces (see Ref. 14 for review). Despite the wealth of knowledge concerning level cycling, there is a paucity of
information concerning cycling on uphill grades, with only a few
published papers that even mention incline riding (1, 12, 23). However,
cycling on graded surfaces is an important part of cycling competition.
The winner of single-day or multistage races is often a rider who
excels in the mountain stages of the course. Cycling on graded surfaces
also provides an opportunity to examine how the neuromuscular system
adapts to changes in the environment because the rider's orientation
to gravitational forces will change (4). An additional factor is an
alteration in posture, because cyclists often choose to stand on the
pedals during portions of an uphill climb. Such changes in posture are
not often observed during level riding because of the aerodynamic drag
associated with the standing position.
Neuromuscular patterns in standing cycling may be altered for a variety
of reasons, including changes in musculoskeletal geometry. For example,
in comparison to the seated posture, standing will change the range of
motion of the hip joint substantially during the crank revolution.
Similarly, the kinematics of the total body center of mass will be
modified by the standing posture, because the rider comes forward and
upward off the saddle. Such positional changes may lead to differences
in the pattern of forces applied to the pedal. Indeed, preliminary
reports from our laboratory confirmed that standing and seated postures
produce different pedal force, crank torque, and joint moment profiles
(6-8). Given the dynamic coupling between pedal forces and
joint torques, it seems reasonable to suggest that changes in muscle
activity patterns will occur also. A question of interest concerns the
manner in which such changes will take place in the highly redundant
neuromuscular system. Examination of the patterns of muscle
activity during level cycling and under different climbing strategies
will provide insight to this change process.
Electromyography (EMG) records have been used to study muscle activity
and neuromuscular coordination in cycling on level surfaces (13). One
of the first studies to examine EMG in cycling was conducted by Houtz
and Fisher (17), who investigated activity patterns in 14 lower
extremity muscles as a function of joint range of motion in three
subjects on a stationary bicycle. In a subsequent study, Despires
(11) recorded EMG from three subjects riding their own
bicycles on a treadmill while the height of the seat was varied. These
early studies laid the ground work for others that followed (13, 15,
16, 22). In studies of muscle coordination during cycling, eight lower
extremity muscles have been studied most often (12, 13, 16, 17):
gluteus maximus (GM), biceps femoris (BF), semimembranosus (SM), rectus
femoris (RF), vastus medialis (VM), vastus lateralis (VL),
gastrocnemius (GC), and tibialis anterior (TA). The selection of
muscles for any given study will depend on the exact question(s) being
addressed, although, in general, representative muscles controlling the
lower extremity joints in the sagittal plane are chosen.
Some EMG studies have sought to simplify matters by selecting one
muscle to represent the activity pattern of an anatomic group, using
the concept of a single equivalent muscle (2). In cycling, for example,
BF might be chosen from the hamstring muscles, whereas VL might
represent the quadriceps group. However, one reason for including more
than one muscle from a functional group is the recent interest in the
use of mono- and biarticular muscles (26). EMG studies of cycling have
demonstrated the degree of cocontraction of the muscles controlling the
knee joint and have shown the importance of two-joint muscles (15, 19,
22). Van Ingen Schenau and colleagues (24-26) have studied the
functional roles of mono- and biarticular muscle in human jumping and
cycling motions. They have proposed two unique roles for biarticular
muscles: 1) power produced
by a monoarticular muscle can be transported to adjacent joints by its
biarticular antagonist and 2) the
directional control of externally applied forces. Furthermore, the
coordination of related mono- and biarticular muscles is at least
partly under the influence of geometric constraints (24, 25). In
cycling, an alteration of grade and a change in posture result in
changes in the direction of the force applied to the pedal (7, 8) and
may alter the influence of geometric constraints in the leg. Thus the
relationship of activity patterns between monoarticular muscles and
their biarticular antagonists may well be modified under these
conditions.
Therefore, muscle activity patterns during uphill cycling in both
seated and standing postures are of interest. Changing grade will
modify the line of action of the gravitational force in relation to the
forces exerted on the pedal by the cyclist. Altering posture to
standing will not only change the relations of the forces concerned but
will also change the geometry of body segments during the motion. What
is the impact of these changing relationships of the forces on muscle
coordination? How does the neuromuscular system adapt to the change of
the body geometry? The purpose of the present study was to examine the
modification of lower extremity neuromuscular patterns to the change of
grade and posture during cycling.
| |
METHODS |
The subjects in this experiment were young, healthy, male university
graduate students with >2 yr of cycling experience. Informed consent
and medical clearance were obtained from each subject before the
experiment. Age, height, and body mass of the eight tested subjects
were 24 ± 2 (mean ± SD) yr, 1.76 ± 0.06 m, and 71 ± 2 kg, respectively.
Subjects were tested on their own road-racing bicycles mounted on a
computerized Velodyne cycling ergometer (Schwinn). The bicycle was
mounted to the ergometer, with the front fork in a fixed position and
the rear wheel free to spin without lateral motion. A roller in contact
with the rear tire provided resistance to pedaling. The amount of
resistance was dictated by a computer-controlled electric motor. The
subjects were examined under three conditions: pedaling on a level
surface while seated (LS), 8% uphill grade while seated (US), and 8%
uphill grade while standing (ST). To emphasize the alteration
associated with grade and posture, the work rate for all three
conditions was kept constant at 250 W. Compared with having subjects
actually riding on the road with these conditions, this
setup allowed us to investigate the influence of posture and grade
without other confounding factors (such as lateral sway in standing,
wind resistance, and so forth). After a 10-min warm-up, subjects
pedaled with a self-selected gear ratio that was the same for all three
conditions. Subjects were instructed to find a comfortable speed and
keep it constant for each condition throughout the testing session. In
all three conditions, subjects gripped the handlebars on the upper part
near the brake hoods. Testing order for the three conditions was
randomized, with data from five consecutive crank cycles collected
within the last minute of a 3-min riding session for each condition.
High-speed video (200 Hz) of the left sagittal view of each trial was
taken in conjunction with EMG, with a trial defined as one complete
crank revolution. Retro-reflective markers were attached to the
subjects at appropriate anatomic locations. The shoulder joint was
defined by a marker over the humeral head, marking the rotation center
of the upper arm. The upper pelvis location was indicated by a marker
on the anterior superior iliac spine. Ankle, knee, and hip joints were
defined by markers on the lateral malleolus, a point 1 cm above the
superior margin of the lateral tibia and the greater trochanter,
respectively. The lateral side of the head of the fifth metatarsal, the
pedal spindle, and the crank center were also marked by
retro-reflective tape. A marker on a stiff fin at the back of the pedal
was used with the pedal spindle marker to determine the pedal angle.
Video data from the motion of these retro-reflective markers were
processed by using standard planar calibration techniques to determine
the sagittal plane kinematic motion. Raw positional data were filtered
with a fourth-order, zero-lag, low-pass Butterworth digital filter with
a cutoff frequency of 5 Hz. The data were expressed as a function of
the crank arm angle (
c) as it
rotated from the highest pedal position [0°or top dead center
(TDC)] to the lowest (180° or bottom dead center) and back to
TDC to complete a 360° crank cycle.
EMG data from GM, BF, RF, VL, GC, and TA of the left leg were collected
by using Ag-AgCl surface electrodes. Preamplified electrode pairs were
placed on each muscle belly along the longitudinal line of muscle
fibers, after the sites were shaved and cleaned with alcohol. Each
electrode pair was attached to the skin with an adhesive pad by using
electrolytic gel to improve conductivity. A common ground electrode was
placed on the distal end of the left radius. After appropriate
amplification (amplifier frequency response: 20-4,000 Hz; input
impedance: >25 M
at DC; common reject ratio: 87 dB at 60 Hz), the
EMG data were collected with a 12-bit analog-to-digital (A/D) converter
at 1,000 samples/s. Different gains were used for each muscle to
optimize the resolution of the digital signals without saturation or
clipping. A synchronizing signal was collected by the A/D board and
used to illuminate a light-emitting diode in view of the video camera
during each trial. The kinematic data were then synchronized with the
EMG to locate the start and end point of each cycle within the EMG
data.
A low-pass Butterworth filter (cutoff frequency 22 Hz) was applied to
the rectified raw EMG data to produce linear envelopes for each muscle
activity pattern. The rather high cutoff frequency of 22 Hz was chosen
to minimize filter effects in quantifying muscle activity onsets yet
still provide a relatively smooth representation of muscle activity
changes throughout the crank cycle. To quantify the muscle activity
pattern, a series of variables were calculated from the linear envelope
records of each trial. The overall activity level of each muscle was
identified by the mean EMG magnitude for one cycle
(Meancycle), defined as the
integrated EMG (IntEMG)
over one crank
cycle divided by 360°. Peak EMG was the maximum value from the EMG
linear envelope during each trial. Periods of higher muscle activity
were defined by the period when the signal was above a threshold of
25% of the peak EMG of each trial. The 25% threshold value was chosen
in concert with the choice of linear envelope filter cutoff as an
appropriate level to indicate the beginning and end of a period of
muscle exertion. A burst of muscle exertion was defined as the muscle
activity between the starting crank angle of a higher activity phase
(SMA) and the end of this phase (EMA). The duration of the muscle
activity (DMA) was calculated by the amount of crank rotation between
SMA and EMA (Fig. 1). The IntEMG was
calculated as the area under the linear envelope curve within the
duration of higher muscle activity. Peak EMG,
Meancycle, and IntEMG are all
reported in terms of activity level at the electrode-skin interface.
For display purposes, ensemble average curves of the linear envelopes
from the five trials of each condition and subject were also
calculated. The differences in SMA, EMA, DMA,
Meancycle, peak EMG, and IntEMG among the different conditions were tested by using a repeated-measures ANOVA. The statistical significance level was set at
= 0.05. As another measure of relative change in
muscle activity between cycling conditions, the similarity of linear
envelope patterns was assessed by using a cross-correlation technique.
The following equation was used to calculate the correlation
coefficient (9)
where
x
and y represent the two EMG time
series of interest, N is the total
number of data points in each time series, and
k is the number of data points shifted
in the calculation (k = 0 was selected
here). The correlations of EMG ensemble curves between the LS vs. US,
and between the US vs. ST conditions, were examined using the
correlation coefficient
(rxy).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 1.
Illustration of start muscle activity (SMA), end muscle activity (EMA),
and duration of muscle activity (bar) measurements made by using a
sample electromyogram (EMG) ensemble curve. Actual values were
calculated from linear envelopes of individual trials.
|
|
| |
RESULTS AND DISCUSSION |
A relatively constant pedaling frequency was maintained by each subject
in each condition to achieve the work rate of 250 W. The pedaling
frequency was different for each subject, ranging from 60 to 85 rpm.
Muscle activity patterns from the three cycling conditions are
represented with ensemble linear envelopes (5 trials × 8 subjects
per condition) of the EMG data (Fig. 2).
These data are presented by using the same arbitrary scale within each
panel, although the scales used in different panels may be different. Therefore the EMG amplitudes should only be compared within each panel.
Overall, there is excellent agreement between our results in the seated
conditions and the results found in the literature (12, 13, 16, 17,
22).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 2.
Ensemble curves of linear envelope EMG for 6 muscles for all
conditions. LS, level seated; US, uphill seated; ST, uphill standing.
All curves in 1 panel here used same arbitrary units on vertical axes.
Scales of vertical axes used in different panels may be different.
Horizontal axes are labeled by corresponding crank angle (in degrees).
One complete cycle = top dead center (TDC) from 0° to TDC
360°.
|
|
Muscle activity level.
Among the six muscles tested, only GM and TA displayed significant
differences in peak EMG between conditions (Fig.
3). The mean peak EMG of GM in ST (1.04 mV)
was nearly 50% higher than in LS and US conditions (0.72 and 0.67 mV,
respectively). Meancycle of GM
increased by ~65% in the ST condition [from 0.14 (LS) and 0.15 (US)
to 0.24 mV], whereas the IntEMG of GM increased by
>80%, reaching 72 mV · degrees for ST compared
with 41 and 38 mV · degrees for LS and US. The higher
IntTEMG of GM in ST is caused by not only the higher activity level
during the burst but also by the burst duration's lasting over a
greater crank angle (Fig. 4). For all
measurements of EMG magnitude, the EMG activity of GM was higher
in the ST condition than in either seated condition.
View larger version (40K):
[in this window]
[in a new window]
|
Fig. 3.
Mean value of selected measures of muscle activities for 6 muscles
in 3 conditions for all subjects. GM, gluteus maximus; BF,
biceps femoris; RF, rectus femoris; VL, vastus lateralis; GC,
gastrocnemius; TA, tibialis anterior; peak EMG, maximum value from EMG
linear envelope for each trial;
meancycle, mean EMG magnitude for
1 cycle; IntEMG, integrated EMG over 1 crank
cycle divided by 360°. P
values displayed underneath bars representing
each muscle indicate between-condition comparisons.
* P  0.05. a and
b Homogeneous groups
tested by least significant difference (LSD) method with
= 0.05.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4.
Higher levels of muscle activation indicated by bars, displayed as a
function of crank position. Symbols ,  , A, B, a, and b,
homogeneous groups (LSD method with = 0.05) for start
(bottom of bars), end
(top of bars), and duration of
activation, respectively. Open bars, LS; hatched bars, US; solid bars,
ST.
|
|
For TA, the mean peak EMG in the ST condition (2.15 mV) was >40%
higher than in the two seated conditions (1.44 and 1.59 mV for LS and
US, respectively). The higher peak EMG activity of TA in ST was offset
by a relatively narrow peak (Fig. 2), so that Meancycle and IntEMG for TA were
the same in all three conditions. If we assume bilateral symmetry of
the leg motion during the crank cycle, the higher TA activity in late
recovery occurred at the same time as higher GM activity for the
contralateral leg in its late downstroke.
Although GM and TA were the only muscles displaying alterations in peak
EMG, the biarticular RF exhibited changes in the variables Meancycle and IntEMG.
Meancycle for RF in ST (0.56 mV)
was higher than in the two seated conditions (0.35 and 0.32 mV, LS and
US, respectively; see Fig. 3). This indicates more overall
EMG activity, which coincides with the higher IntEMG for RF in ST (181 mV · degrees, compared with 101 and 87 mV · degrees in LS and US, respectively). These EMG
increases for RF may be associated with the greater range of crank
angles over which it is active (crank angle of 219° vs. 135°
and 136° for the LS and US, respectively; see Fig. 4), because no
difference in peak EMG was observed between conditions. The longer duration of RF activation in the ST condition is caused by
both earlier beginning in the upward recovery phase and delayed ending
during the subsequent downstroke (Fig. 4).
The EMG activity measures for BF, VL, and GC did not display
significant alterations with the change of grade and/or
posture. However, in the case of the knee extensor VL, a change in its overall pattern of activity is evident from Fig. 2. Qualitatively, it
appears that VL is activated earlier in the upward recovery phase, and
the activity lasted longer into the subsequent downward power phase for
the standing condition. However, Fig. 4 illustrates that only the
duration was significantly greater in the ST condition, whereas SMA and
EMA values are statistically equivalent for all conditions. The longer
duration of VL activity did not produce a significantly higher IntEMG
of VL (Fig. 3), possibly because of a lower mean activation within the
burst (Fig. 2).
A better understanding of these activity changes can be gained by
placing them within the context of joint moment changes associated with
alterations in grade and posture. During the downstroke of seated
cycling, extensor moments are needed at all three lower extremity
joints, except for the knee joint moment that changes from extensor to
flexor in the middle of the downstroke (14, 21). Examples of joint
moment data from one subject are presented in Fig.
5. These moment profiles were calculated by
using techniques described in detail elsewhere (6, 7). Briefly,
measured pedal forces (3) were collected (100 Hz, 12-bit A/D) and
smoothed (10 Hz), followed by synchronization with the kinematic data. These data were used to calculate pedal kinetics in a global reference system by using the equations reported in Coyle et al. (10) and were
used as input to a standard inverse dynamics model that calculated
moments at the ankle, knee, and hip joints (28).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 5.
Exemplar joint moment patterns (ensemble of 5 trials) for 1 subject,
displayed as function of crank angle. One complete cycle = TDC from
0° to 360°. Joint moments were calculated by using an inverse
dynamics model with measured pedal forces and lower extremity
kinematics as input data (7).
|
|
EMG and joint moment data in the two seated conditions were comparable,
although there was a distinct change in the pelvic angle in the uphill
condition (Fig. 6). However, much larger
changes in muscle activities and joint moments were seen with the
alteration in posture. From the seated to standing posture, the peak
EMG value of hip extensor GM increased dramatically. If other mechanics of hip joint motion were unchanged, the higher GM activity by itself
would indicate an increase in hip extensor moment.
However, in the ST, the hip joint was further forward in relation to
the crank spindle (Fig. 6). This more forward position reduced the horizontal distance between the hip joint and the point of force application on the pedal. Therefore, the moment arm of the vertical pedal reaction force in relation to the hip joint axis was reduced. Previous data have shown this vertical force to be the major component of the pedal reaction force while standing (7, 8). Therefore, when
considered in relation to the pedal force-vector direction and
musculoskeletal geometry associated with the standing position, increased GM activity during the downstroke does not necessarily result
in a greater hip extensor moment (6). In fact, the hip moment is seen
to decrease for some subjects (Fig. 5). The increased GM activity in
the ST condition may be associated with an increase in hip joint
stiffness or greater stabilization of the pelvis, which, in the ST
condition, unlike in seated cycling, is not supported by the bike
saddle.
View larger version (44K):
[in this window]
[in a new window]
|
Fig. 6.
Stick figure representation for 3 cycling conditions of 1 subject (1 pedaling cycle of each condition). Segments were linked by markers at
shoulder, anterior superior iliac spine, hip, knee, ankle, pedal
spindle, and crank center. Note change of orientation from LS to US
condition and change of posture from seated to ST condition.
|
|
The muscle activity of the one-joint knee extensor (VL) displayed no
significant alteration related to grade alone and only increased its
duration of activity with the change in posture. A similar increase in
DMA was seen for the biarticular RF. No differences in VL
peak EMG, Meancycle, and IntEMG
would indicate a similar level of muscle activity, and, therefore,
potentially the same magnitude of knee moment in the early part of the
downstroke, if it is assumed that the activity level of other knee
muscles also remained the same. This scenario coincides with the knee joint-moment measurement (Fig. 5). As with the hip, the moment arm of
the vertical pedal force with respect to the knee joint axis may be
altered during the downstroke, because the knee also moved forward from
the seated to ST condition (Fig. 6). However, the effect is not as
dramatic as with the hip, because the knee joint center is physically
closer to the pedal, with only one intervening joint (the ankle). This
lessens the degree of forward knee translation compared with the amount
of hip translation. Figure 5 indicates that the knee-extensor moment is
prolonged during the downward stroke in ST, consistent with the greater duration of VL and RF activity, but that the peak magnitude of knee-extensor moment is similar between conditions.
The ankle plantar flexor (GC) did not exhibit any significant change in
EMG activity or duration (Figs. 2 and 3) with condition. This is
surprising, because the kinetic data indicate that the peak plantar
flexor moment increased from seated to ST conditions (Fig. 5). This may
be caused by the biarticular nature of GC, as it also serves as a knee
flexor. With the extended period of knee-extensor torque, increased GC
activity would be contraindicated. The single-joint plantar flexor
soleus may play a more important role in the increase of the ankle
joint plantar flexor moment, because its activity is mostly
concentrated in the downstroke phase of level cycling (14, 26).
Muscle coordination.
Although the previous section described changes in activity levels of
individual muscles associated with the three cycling conditions, the
coordination of muscle activity was not addressed. To examine
coordination among these muscles, important variables of interest are
the starting and ending crank angles of the activity bursts and the
correlation of activity patterns between conditions.
Overall, Figs. 2-4 illustrate that the two seated conditions (LS
and US) had similar muscle activity patterns that differed from the ST.
The muscle activity of GM started just before TDC for all conditions.
However, GM displayed activity over a greater portion of the crank
cycle in ST, with activity well into the later part of the downstroke
(to ~160°). RF, which acts as both a hip flexor and knee
extensor, also was active for a longer duration in ST. This increased
duration has two components, because the muscle activity started
earlier before TDC and continued later into the power stroke. The
single-joint knee extensor VL also displayed a greater duration of
muscle activity in ST, although the differences in SMA and EMA were not
significant between conditions (Fig. 4). The remaining three muscles
(BF, GC, and TA) had similar values in onset time and burst durations
in the three conditions. The fact that three muscles had consistent
patterns across conditions, whereas three others showed altered ST
profiles, is indicative of a change in muscle coordination during the
ST condition.
The correlation scores give a quantitative indication of similarity in
muscle activity patterns between conditions. High correlation of muscle
activity between conditions would indicate a similar pattern of usage
in any two conditions. The correlation coefficients (Table
1) showed that the correlation between
the two seated conditions
(r1, LS and US)
was high for all muscles, ranging from 0.93 to 0.99. These correlations
dropped for all muscles in the comparison between the two uphill
conditions (r2,
ST and US), ranging from 0.73 to 0.95. The greatest change in these
correlation scores was observed for the monoarticular hip and knee
extensors (GM and VL). Correlation of the ensemble curves of GM
activity decreased from 0.97 for the two seated conditions to 0.73 for the two uphill conditions. The correlation of the ensemble curves of VL
activity was also much higher (0.99) for the two seated conditions than
for the two uphill conditions (0.77). The differences in variation
calculation shows ~40% less common variation in the ST and US
conditions than in the two seated conditions for both GM and VL
activity patterns.
View this table:
[in this window]
[in a new window]
|
Table 1.
Correlation coefficient between ensemble EMG curves of each muscle for
seated and standing uphill conditions and for level and uphill seated
conditions
|
|
It is interesting that the activity patterns of single joint VL and GM
muscles display greater responses to the postural change (Table 1,
r22 = 0.59 and 0.53, for
seated and standing uphill conditions respectively) than do the
biarticular RF and BF muscles
(r22 = 0.90 and 0.87, respectively). This result concurs with other studies that report
functional differences between mono- and biarticular muscles (26).
These authors speculate that monoarticular muscles contribute to
positive work, whereas biarticular muscles control the direction of
pedal force and transportation of the power produced by the
monoarticular muscles to adjacent joints (24, 26). One necessary
condition for a biarticular muscle to transport power generated by its
monoarticular antagonist is that both muscles are active
simultaneously. Table 2 shows the overlap
in activity of two such pairs, GM-RF and VL-GC, for each condition. The
simultaneous muscle activity for both antagonist pairs was equivalent
in LS and US but increased in the ST condition. The changes in activity
in the ST condition are likely caused not only by power transport but
also by control of the direction of the force vector applied to the
pedal; these changes are associated with the removal of the support of
the saddle. Profound differences in the pedal force
profiles have been observed in the ST condition (7).
The relative roles of the mono- and biarticular muscles in cycling may
shape the changes that occur in response to the postural change. Future
studies should investigate the underlying reason(s) why the GM and VL
activity patterns changed more than the RF and BF patterns and why
there was increased overlap in GM-RF and VL-GC activities. These
activity alterations may be linked to pedal force production, power
transport needs, and/or changes in muscle kinematics. For
example, if a muscle is working to produce power, it will be
contracting concentrically; in the case of the GM and VL, this would
indicate a more protracted period of energy generation. For two-joint
muscles, transporting power to adjacent joints will be associated with
much lower muscle-shortening velocities. Investigation of muscle
kinematics and power transport with musculoskeletal models (5, 20, 27)
will provide more insight to the changes in muscle activities reported
here.
In cycling, monoarticular muscles have relatively clear roles in terms
of their function. For example, GM functions mainly as a hip extensor,
although, because of the mechanical coupling of body segments, its
effect is seen throughout the lower extremity (18). The function of
two-joint muscles may not be as clear; BF can act as a hip extensor
and/or a knee flexor. In fact, various cyclists use the muscle
differently, even in the same task. Figure 7 shows muscle activity of BF during ST for
two subjects. Subject 2 exhibited a
pattern similar to that seen in Fig. 2, with BF activity associated
entirely with hip and knee extension during the downstroke. In
contrast, subject 4 had BF activity
starting well before TDC but ending earlier in the crank cycle than
activity for subject 2. In this case,
BF activity was associated with hip and knee flexion in late recovery
before TDC, rather than with hip and knee extension in the early
downstroke. Others have reported similar subject-specific data for
level cycling (22).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 7.
muscle activity patterns of biceps femoralis for 2 subjects
[subjects 2 and
4 (S2 and S4, respectively)] in
ST condition. One complete cycle = TDC from 0° to 360°. Both
curves use same arbitrary unit for vertical axis.
|
|
The different usage of BF may be related to subject-specific pedaling
techniques. For example, the use of a relatively fixed ankle joint
throughout the crank cycle requires different thigh kinematics than
does a style that permits greater ankle range of motion. The GC
activity pattern may change with these different techniques, which in
turn may result in different muscle coordination at the knee joint
caused by the biarticular influence of GC. The different usage of BF
may also be related to overall neuromuscular coordination, such as
different usage of biarticular muscles for power transport or pedal
force control (25, 26). Relative muscle strength is another factor, as
stronger one-joint hip extensors may be associated with increased RF
activity for power transportation, whereas weaker one-joint hip
extensors may need help from BF to forcefully extend the hip joint.
Because the primary goal of this study was to investigate the change of
muscle activity patterns between conditions, overall estimations of EMG
activity were presented, even though such large
individual differences were observed. Individuality of EMG activity is
an important facet of understanding cycling mechanics, especially for
issues related to performance of individual cyclists, and individual
differences should be addressed in the future.
Clearly, more changes in muscle activity patterns were observed with
the change of posture rather than grade alone (Table 1), especially for
the patterns of the monoarticular hip (GM) and knee (VL) extensors.
Greater differences also were observed in the coordination between the
monoarticular muscles and their biarticular antagonists between seated
and ST conditions (Table 2). Further investigation will focus on the
influence of the anatomic position on the contractile properties of
individual muscles or muscle groups to examine more details about the
mechanisms of activity pattern alteration with posture.
Summary.
The change of cycling grade from 0 to 8% did not
induce a significant change in lower extremity neuromuscular
coordination in cycling. However, a postural change from seated to ST
pedaling at 8% uphill grade was accompanied by increased muscle
activity of some hip and knee extensors. Among all the muscles tested, GM and RF exhibited the most significant increase of muscle activity as
demonstrated by the mean and IntEMG. Compared with the biarticular muscles, monoarticular muscles (GM and VL, particularly) demonstrated larger modifications in activity patterns with a change in posture. In
the standing posture, both the activity patterns of individual muscles
and the coordination between antagonist pairs were altered. These
changes may be related to the different functional roles of one- and
two-joint muscles.
| |
ACKNOWLEDGEMENTS |
This work was supported by the Office of Research Affairs at the
University of Massachusetts through Faculty Research Grant 1-03451 (to.
G. E. Caldwell).
| |
FOOTNOTES |
Address for reprint requests: L. Li, 112 Long Fieldhouse, Dept. of
Kinesiology, Louisiana State University, Baton Rouge, LA
70803.
Received 11 February 1997; accepted in final form 21 April
1998.
| |
REFERENCES |
1.
Álvarez, G.,
and
J. Vinyolas.
A new bicycle pedal design for on-road measurements of cycling forces.
J. Appl. Biomech.
12:
130-142,
1996.
2.
Bouisset, S.
EMG and muscle force in normal motor activities.
In: New Development in Electromyography and Clinical Neurophysiology, edited by J. E. Desmedt. Basel: Karger, 1973, p. 547-583.
3.
Broker, J. P.,
and
R. J. Gregor.
A dual piezoelectric force pedal for kinetic analysis of cycling.
Int. J. Sports Biomech.
6:
394-403,
1990.
4.
Brown, D. A.,
S. A. Kautz,
and
C. A. Dairaghi.
Muscle activity patterns altered during pedaling at different body orientations.
J. Biomech.
29:
1349-1356,
1996[Medline].
5.
Caldwell, G. E.,
and
A. E. Chapman.
The general distribution problem: a physiological solution which includes antagonism.
Hum. Mov. Sci.
10:
355-392,
1991.
6.
Caldwell, G. E.,
J. M. Hagberg,
S. D. McCole,
and
L. Li.
Lower extremity joint moments during uphill cycling.
In: Proceedings of the 9th Canadian Society of Biomechanics Conference, edited by A. Hoffer. Vancouver, BC, Canada: Canadian Society of Biomechanics, 1996, p. 182-183.
7.
Caldwell, G. E., L. Li, S. D. McCole, and
J. M. Hagberg. Pedal and crank kinetics in uphill
cycling. J. Appl. Biomech. In press.
8.
Caldwell, G. E.,
S. D. McCole,
and
J. M. Hagberg.
Pedal force profiles during uphill cycling.
In: Proceedings of the 8th Canadian Society of Biomechanics Conference, edited by W. Herzog,
and B. M. Nigg. Calgary, Alberta, Canada: Canadian Society of Biomechanics, 1994, p. 58-59.
9.
Chatfield, C.
The Analysis of Time Series. New York: Chapman & Hall, 1984.
10.
Coyle, E. F.,
M. E. Feltner,
S. A. Kautz,
M. T. Hamilton,
S. J. Montain,
A. M. Baylor,
L. D. Abraham,
and
G. W. Petrek.
Physiological and biomechanical factors associated with elite endurance cycling performance.
Med. Sci. Sports Exerc.
23:
93-107,
1991[Medline].
11.
De Groot, G.,
E. Welbergen,
L. Clijsen,
J. Clarijs,
J. Cabri,
and
J. Antonis.
Power, muscular work and external forces in cycling.
Ergonomics
37:
31-42,
1994[Medline].
12.
Despires, M.
An electromyographic study of competitive road cycling conditions simulated on a treadmill.
In: Biomechanics IV. Baltimore, MD: University Park, 1974, p. 349-355.
13.
Faria, I. E.,
and
P. R. Cavanagh.
The Physiology and Biomechanics of Cycling. New York: Wiley, 1978.
14.
Gregor, R. J.,
J. P. Broker,
and
M. M. Ryan.
The biomechanics of cycling.
In: Exercise and Sport Science Review, edited by J. O. Holloszy. Baltimore, MD: Williams & Wilkins, 1991, p. 127-169.
15.
Gregor, R. J.,
P. R. Cavanagh,
and
M. LaFortune.
Knee flexor moments during propusion in cycling: a creative solution to Lombard's paradox.
J. Biomech.
18:
307-316,
1985[Medline].
16.
Gregor, R. J.,
D. Green,
and
J. J. Garhammer.
An electromyographic analysis of selected muscle activity in elite competitive cyclists.
In: Biomechanics VII. Baltimore, MD: University Park, 1982, p. 537-541.
17.
Houtz, S. J.,
and
F. J. Fisher.
An analysis of muscle action and joint excursion on stationary bicycle.
J. Bone Joint Surg. Am.
41A:
123-131,
1959[Abstract/Free Full Text].
18.
Hoy, M. G.,
and
R. F. Zernicke.
The role of intersegmental dynamics during rapid limb oscillations.
J. Biomech.
19:
867-877,
1986[Medline].
19.
Jorge, M.,
and
M. L. Hull.
Analysis of EMG measurements during bicycling pedaling.
J. Biomech.
19:
683-694,
1986[Medline].
20.
Pandy, M. G.,
F. E. Zajac,
E. Sim,
and
W. S. Levine.
An optimal control model for maximum-height human jumping.
J. Biomech.
23:
1185-1198,
1990[Medline].
21.
Redfield, R.,
and
M. L. Hull.
On the relation between joint moments and pedaling rates at constant power in bicycling.
J. Biomech.
19:
317-329,
1986[Medline].
22.
Ryan, M. M.,
and
R. J. Gregor.
EMG profiles of lower extremity muscles during cycling at constant workload and cadence.
J. Electromyogr. Kinesiol.
2:
69-80,
1992.
23.
Stone, C.,
and
M. L. Hull.
Rider/bicycle interaction loads during standing treadmill cycling.
J. Appl. Biomech.
9:
202-218,
1993.
24.
Van Ingen Schenau, G. J.
From rotation to translation: constraints on the multi-joint movements and the unique action of bi-articular muscles.
Hum. Mov. Sci.
8:
301-337,
1989.
25.
Van Ingen Schenau, G. J.,
M. F. Bobbert,
and
R. H. Rozendal.
The unique action of bi-articular muscles in complex moments.
J. Anat.
155:
1-5,
1987[Medline].
26.
Van Ingen Schenau, G. J.,
P. J. M. Boots,
G. de Groot,
R. J. Snackers,
and
W. W. L. M. Woenzel.
The constrained control of force and position in multi-joint movements.
Neuroscience
46:
197-207,
1992[Medline].
27.
Van Soest, A. J.,
A. L. Schwab,
M. F. Bobbert,
and
G. J. van Ingen Schenau.
The influence of the bi-articularity of the gastrocnemius on vertical jumping.
J. Biomech.
26:
1-8,
1993[Medline].
28.
Winter, D. A.
Biomechanics and Motor Ccontrol of Human Movement. New York: Wiley Interscience, 1990.
J APPL PHYSIOL 85(3):927-934
8570-7587/98 $5.00
Copyright © 1998 the American Physiological Society
Top
Abstract
Introduction
