Journal of Applied Biomechanics, 2013, 29, 1-11
© 2013 Human Kinetics, Inc.
An Official Journal of ISB
www.JAB-Journal.com
ORIGINAL RESEARCH
The Roles of Whole Body Balance, Shoe-Floor Friction,
and Joint Strength During Maximum Exertions:
Searching for the “Weakest Link”
Steven L. Fischer,1 Bryan R. Picco,2 Richard P. Wells,2 and Clark R. Dickerson2
1Queen’s
University; 2University of Waterloo
Exerting manual forces is critical during occupational performance. Therefore, being able to estimate maximum
force capacity is particularly useful for determining how these manual exertion demands relate to available
capacity. To facilitate this type of prediction requires a complete understanding of how maximum force capacity
is governed biomechanically. This research focused on identifying how factors including joint moment strength,
balance and shoe-floor friction affected hand force capacity during pulling, pressing downward and pushing
medially. To elucidate potential limiting factors, joint moments were calculated and contrasted with reported
joint strength capacities, the balancing point within the shoe-floor interface was calculated and expressed
relative to the area defined by the shoe-floor interface, and the net applied horizontal forces were compared
with the available friction. Each of these variables were calculated as participants exerted forces in a series of
conditions designed to systematically control or restrict certain factors from limiting hand force capacity. The
results demonstrated that hand force capacity, in all tested directions, was affected by the experimental conditions (up to 300%). Concurrently, biomechanical measures reached or surpassed reported criterion thresholds
inferring specific biomechanical limitations. Downward exertions were limited by elbow strength, whereas
pulling exertions were often limited by balance along the anterior-posterior axis. No specific limitations were
identified for medial exertions.
Keywords: biomechanics, hand force, ergonomics, force capability
Incorporating human force producing capability
into job design can be an effective method to match job
demands with workers’ functional capacity. Since the
nineteenth century, researchers have measured force
production in an effort to match individual capability
with anticipated performance demands (Sargent, 1897) to
optimize performance and minimize injury risk. However,
measuring individual worker capabilities to facilitate this
matching is time and cost intensive. Alternatively, models
designed to predict force producing capability could
enable designers to more readily match job demands
with prospective worker capabilities. However, there have
been few attempts to develop comprehensive predictive
models (Grieve, 1979a; 1979b; Kerk et al., 1994), and
none designed to incorporate three-dimensional tasks.
This may be due to an incomplete understanding about
how force capacity is limited.
Steven L. Fischer (Corresponding Author) is with the School of
Kinesiology and Health Studies, Queen’s University, Kingston,
ON, Canada. Bryan R. Picco, Richard P. Wells, and Clark R.
Dickerson are with the Department of Kinesiology, University
of Waterloo, Waterloo, ON, Canada.
Several factors, both extrinsic and intrinsic to the
worker, can limit force producing capability during
manual material handling exertions. Limiting factors
include whole body balance (Kerk et al., 1998; Holbein
& Chaffin, 1997), shoe-floor friction (Kroemer, 1974),
hand-handle friction (Seo et al., 2010), and individual
joint moment strength (Chaffin, 1997).
Whole body balance, as a factor limiting occupational performance, has been operationally defined within
the paradigm of postural stability. Postural stability is
achieved when static moment equilibrium is met (Grieve,
1979a; 1979b; Kerk et al., 1994). In this paradigm the
hand force and the force of gravity acting on the center of
mass (COM) each contribute moments about a balancing
point or fulcrum within the area defined by the shoe-floor
interface and must balance to achieve static equilibrium.
Maximum pulling force for example (Figure 1), could
therefore be achieved by shifting the balancing point as far
forward as possible while concurrently shifting the COM
posteriorly. This strategy serves to increase the moment
arm and corresponding moment contribution from the
COM, while reducing the moment arm of the applied
hand force, allowing the maximum applied hand force to
be obtained such that the applied moment does not exceed
the available counterbalancing moment of the COM.
1
2 Fischer et al.
Figure 1 — An example of a pulling exertion with annotations to conceptualize static postural stability. The force of
gravity acting at the center of mass ( ) creates a moment
about the balance point (where FCOM is the force and rCOM is the
moment arm to the balance point) which must counterbalance
the moment from the reaction force at the hand (where Fhand is
the reaction force and rhand is the moment arm to the balance
point) to achieve equilibrium or postural stability. However
equilibrium must be achieved while satisfying the constraint
that the balance point must remain with the area defined by the
shoe-floor interface (thin white lines marking the anterior and
posterior limits of the feet).
This biomechanical rationale has also been supported empirically. Hoffman (2008) reported that participants increased the distance between their pelvis and
the anterior limit of their shoe-floor interface as they
pulled maximally. This finding coincides with shifting
the COM posteriorly and the balancing point anteriorly
toward the front edge of the shoe-floor interface to
maximize force production. The combination of theoretical and empirical support underscores the hypothesis
that balance can be limiting; however, it remains unclear
if this limitation is reached before any other possible
limitations during maximum exertions, such that it truly
limits performance.
Joint strength is also commonly identified as a
fundamental limiting factor to hand force production
capability. Operationally, joint strength is defined as the
net muscle moment required at a given joint to counter­
balance the reaction moment from the applied hand
force satisfying static equilibrium. Through decades
of research, Chaffin and colleagues conceptualized
the use of joint strength as a constraint on capability
in the workplace (Chaffin, 1969; Chaffin & Baker,
1970; Garg & Chaffin, 1975; Chaffin et al., 2006) in
the development of the Michigan Three-dimensional
Static Strength Prediction Program (3DSSPP). The
strength constraints in the 3DSSPP are based on work by
Schanne (1972) and Stobbe (1982) who developed population scalable strength estimates to predict maximum
joint strength at each joint, depending on the posture
and exertion type. Although these estimates of joint
strength are widely cited, a comparison of measured
joint strengths and estimated maximum joint strengths
is rare (Nussbaum & Lang, 2005; Langenderfer et al.,
2006).
Shoe-floor friction and hand-handle friction can
also be limiting factors, though they may be less likely
to limit performance under most conditions. Grieve
(1983) concluded that friction is most likely the limiting
factor only during conditions where the coefficient of
friction is greatly reduced. Similarly, Seo et al., (2010)
demonstrated that hand-handle friction is also most
limiting only when it is greatly reduced, and further, it
can be eliminated as a constraint if the handle is perpendicularly to the hand.
The purpose of this research was to investigate
how maximum unilateral hand force capability changed
during pulling, pressing down, and medially pushing
under different experimental conditions designed to
systematically eliminate and control prospective limiting factors (whole-body balance, shoe-floor friction,
and joint strength). We aimed to provide improved
clarification to understand what specific performance
limiters are most likely to impact unilateral hand force
capability as the direction of force application was
manipulated.
Methods
Participants
A convenience sample of fourteen right hand dominant male university students (stature: 1.78 ± 0.08 m;
body mass: 80.3 ± 10.9kg) participated. According
to an a priori power analysis, a minimum of fourteen
Body Balance, Shoe-Floor Friction, and Joint Strength 3
participants was necessary to provide >80% power
to detect significant differences between mean hand
forces with an effect size greater than 0.4 when using
an alpha level of 1%. The study exclusion criteria were
self-reported upper extremity or low back disorders or
pain within the previous year. The research protocol
was approved by the university research ethics review
board.
Instrumentation
Three-dimensional motion was tracked using an
8-camera Vicon MX20 System (Vicon, Oxford, UK).
Thirty-eight individual markers were placed over anatomical landmarks including the C7 and L5 vertebrae,
over the suprasternal notch, xiphoid process, and bilaterally over the 2nd and 5th metacarpals, radial and ulnar
styloids, medial and lateral epicondyles, the acromion,
ear, anterior superior iliac spine, greater trochanter,
medial and lateral condyles of the knee, medial and
lateral malleolus, the tip of the 1st and 5th metatarsals,
and at the posterior border of the calcaneus. Additional
marker clusters secured on rigid plates, were positioned
over the sternum, and bilaterally over the forearm,
upper arm, leg, shank, and over the top of the foot
(shown in Figure 2). The marker clusters were used to
track segment movement during experimental testing.
A static calibration frame established the relationship
between the clusters and the calibration markers over
the anatomical landmarks, and subsequently joint centers and segment coordinate systems were described
(Kingma et al., 1996). Force was measured using an
AMTI 6 degree-of-freedom transducer (MC3A, AMTI
MA, USA), rigidly fixed to a custom handle to allow
the participant to exert force in the specified directions
(described below). Force was sampled synchronously
with motion data at 50 Hz using VICON Nexus 1.2
software (Oxford, UK).
Experimental Protocol
Participants completed unilateral maximal exertions
using their right arm, in the medial (pushing right to left,
across the body), horizontal pull (inwards toward the
body), and downward press (toward the floor) directions
against a handle using a power grip. Each exertion was
performed twice within each of the five experimental
conditions. Participants completed each exertion within
a five second window and were asked to ramp up to their
maximum force during the first 1–2 s, and then sustain
that maximum until the end of the five seconds (Chaffin,
1975). A minimum of two minutes of rest was provided in
Figure 2 — The experimental conditions tested. Participants completed downward presses, pulls, and medial pushes in five conditions: (1) shoulder width foot placement (“SWFP” seen in frame A), (2) free foot placement (“FFP” seen in frame B), (3) high
friction (“HF” seen in frame A, with feet taped to floor), (4) lower body braced (“LBB” seen in frame C, without upper body chest
strap), and upper body braced (“UBB” seen in frame C).
4 Fischer et al.
between exertions (Chaffin, 1975). The five experimental
conditions were as follows.
the body, at a distance of approximately 80% of limb
length for all exertions.
1.Shoulder Width Foot Posture (SWFP): Each
participant stood with their feet shoulder width apart
(Figure 2-A). This condition represented a generic
working position.
2.Free Foot Posture (FFP): Without altering the
position of the torso and shoulder relative to the
handle, participants were given up to five minutes
to test different foot placements to determine a
position that would allow them to produce the most
force in the required direction (Figure 2-B). This
condition represented a generic work posture where
participants could choose their own preferred posture
to maximize hand force.
3. High Friction (HF): Participants stood with their feet
shoulder width apart, similar to Figure 2-A with the
soles of their shoes taped to the floor using double
sided indoor Scotch carpet tape (3M, Minnesota,
USA) . This condition was intended to eliminate
friction as a possible limiting factor.
4.Lower Body Braced (LBB): The upper legs were
braced with a rigid fixture (Figure 2-C, with only
the legs braced) with the feet positioned and still
taped similar to the HF condition. This condition was
intended to eliminate balance as a possible limiting
factor.
5.Upper Body Braced (UBB): Both the lower
and upper body were braced (Figure 2-C). This
condition was intended to eliminate the influence
of trunk strength as a possible limiting factor during
downward exertions, and to further reduce balance
as a possible constraint.
Data Analysis
Exposure to experimental conditions was block randomized, whereby participants completed exertions within the
blocks of SWFP, FFP, and the group of HF, LBB, and
UBB in a random order. The exertion direction was randomized within each block. For all exertions the handle
was positioned at shoulder height, along the midline of
Three types of information, corresponding to each limiting factor category, were extracted from the data to
determine which biomechanical factor was most likely
limiting performance during each exertion, where specific
evaluation criteria are detailed below and summarized
in Table 1.
1.Strength limitations: Elbow and shoulder strength
limitations were determined by calculating net joint
moments at the elbow and shoulder and expressing
those moments as a percentage of the angle corrected
maximum predicted threshold strength values for
a 50th percentile male, obtained using equations
provided in the literature (Chaffin et al., 2006).
If the normalized moment reached or exceeded
100% of the predicted maximum threshold it was
deemed limiting. Shoulder and elbow moments
were calculated using a 3D static linked segment
model (adapted from Dickerson et al., 2007) based
on the acquired force and posture information. For
this study peak hand force was determined as the
peak value resulting from a 500 ms moving window
average over the raw force trace. The corresponding
postural data were also extracted and averaged over
the same 500 ms window.
2.Balance limitations: The location of the balance
point, or fulcrum, was calculated as the point along
the shoe-floor interface that satisfied equations
of static moment equilibrium about the anterior /
posterior and medial / lateral axes. This is consistent
with an approached used previously by Kerk et al.,
(1994). The derivation of the balance point along the
A/P axis is given below with reference to Figure 1,
where the Y direction is in the plane of progression
and the Z direction is in line with gravity:
rCOM × FCOM = rhand × Fhand (1)
Table 1 Seven biomechanical criteria that were used to evaluate potential biomechanical limiting
factors during the maximum hand force exertions in the tested directions and experimental
conditions
Limiter to Evaluate
Measured / Calculated Variable
Threshold
Elbow Strength
Shoulder Strength
Shoulder Strength
Shoulder Strength
Friction
Balance (ML)
Balance (AP)
Elbow flex/ext moment
Shoulder rotation moment
Shoulder ab/adduction moment
Shoulder flex/ext moment
Fhorizontal
X position of the balance point
Y position of the balance point
Chaffin et al. (2006) population strength
Chaffin et al. (2006) population strength
Chaffin et al. (2006) population strength
Chaffin et al. (2006) population strength
μFnormal
ML boundaries of the shoe-floor interface
AP boundaries of the shoe-floor interface
Body Balance, Shoe-Floor Friction, and Joint Strength 5
Simplifying Equation 1 into a 2D case (solving about
the A/P axis) yields:
dCOMY × FCOMZ = d handY × FhandZ + d handZ × FhandY
(2)
where, FCOMZ, FhandZ, FhandY are the respective Y and Z
force components acting at the whole body COM and
hand (known values), dhandZ is the moment arm component along the Z axis between the grip center and the
floor (also a known value), while dCOMY, and dhandY are
the respective moment arms along the Y axis, and remain
unknown. However, they both depend on the location
of the balance point. Therefore dCOMY, and dhandY can be
expressed as:
dCOMY = BalancePoint y − WBcog y
d handY = Gripy − BalancePoint y
(3)
(4)
where the location of the whole body center of mass
(WBcogy) and grip center (Gripy) are known. Then by
substituting Equations 3 and 4 into Equations 2, the
remaining unknown (BalancePointy) can be solved.
This approach is underscored by the assumption
that ground reaction forces can be model to act through
a single center of pressure (or balance point) between
the two feet, a common assumption used in similar work
(Holbein & Chaffin, 1997; Holbein & Redfern, 1997; Lee
& Lee, 2003; Holbein-Jenny et al., 2007). However, this
assumption does not take into account the force coupling
between the feet as they act independently to counter
the axial moments about the body. This is a recognized
limitation in the current work and affects the interpretation of biomechanical limitations when the applied hand
force imposes axial moments about the balance point.
This limitation is more completely presented in the
discussion section.
3.Friction limitations: The maximum available shoefloor friction force was also calculated as the product
of the normal force and a coefficient of friction (μ) of
0.525 (average coefficient of friction during pushing
and pulling on standard floors from Boocock et al.,
2006). The total horizontal hand force was then
normalized to the maximum available friction force.
During unconstrained exertions, the net horizontal
hand force should theoretically remain below the
available friction force; however if it exceeded that
limit during corresponding high friction or braced
exertions, and hand force increased, the nonbraced
exertion was considered to be friction limited.
Statistical Analysis
Repeated measures analyses of variance (ANOVA) were
used to examine the effect of condition (SWFP, FFP,
HF, LBB, UBB) on the peak hand force for each force
direction. Nonparametric Wilcoxon signed ranks tests
were conducted on each normalized dependent variable
(corresponding to a possible limiting factor) as measured in each condition. The mean grouped response for
each dependent measure was compared with the 100%
criterion. If the measure was not significantly less than
100% threshold criterion (i.e., the null hypothesis was
retained), it was inferred to be potentially limiting. To
balance between the increased risk of type 1 error due to
the number of comparisons, and the increased risk of type
2 error by applying a conservative Bonferroni correction,
significance was set at p < .01 for all comparisons. All
statistical processing was completed using SPSS software
(SPSS INC., Chicago, IL, USA).
Results
The experimental conditions significantly affected peak
hand force output in each of the tested directions. Downward force differed between experimental conditions (p
< .001) where the highest average force was reached in
the FFP condition (220 ±51 N) and the lowest in the UBB
condition (156 ± 36 N). Pairwise differences in downward
forces between each condition are highlighted in Figure
3. Medial force differed between experimental conditions (p < .001), where the highest force was produced
in the LBB condition (151 ± 45 N), and the lowest in
the SWFP condition (95 ± 27 N). Pairwise differences
in medially directed hand forces between conditions are
shown in Figure 4. Lastly, pulling force also differed
between conditions (p < .001), where the peak pull force
occurred in the UBB condition (563 ± 160 N), and the
lowest force occurred in the SWFP condition (191 ± 91
N). Pairwise differences between conditions for pulling
forces are indicated in Figure 5.
Downward Exertions—Limiting Factors
Maximum downward exertions were likely strength
limited, where the elbow extensor moment statistically
reached the 100% strength criterion during each of the
experimental conditions (Figure 3). The shoulder internal
rotation moment and shoulder extension moment were
also statistically equal to or greater than the 100% strength
criterion for all conditions except UBB. The AP balance
boundary criterion was also exceeded during downward
exertions, but only during the LBB and UBB conditions.
Medial Exertions—Limiting Factors
A clear limiting factor for maximum medial exertions did
not emerge for any condition except the LBB condition
where the shoulder internal rotation moment statistically
equaled the 100% strength criterion (Figure 4). The
ML balance 100% criterion was statistically reached
or exceeded during the LBB and UBB conditions, and
the AP balance boundary was additionally statistically
reached during the UBB condition; however, due to the
bracing in these conditions balance could not functionally
limit performance.
6 Fischer et al.
Figure 3 — The maximum downward hand force capability (N) and standard deviation for each experimental condition. Bars
with different letters are significantly different (p < .001) (Top graph). The table beneath the figure shows the corresponding mean
(and standard deviation in parentheses) for each of the biomechanical responses normalized relative to their maximum thresholds.
Highlighted boxes indicated where the null hypothesis was retained (inferring that the variable was a potential biomechanical limiter). SWFP = shoulder width foot placement; FFP = free foot placements; HF = high friction; LBB = lower body braced; UBB =
upper body braced.
Pulling Exertions—Limiting Factors
Maximal pulling exertions were differentially limited.
During the braced conditions (LBB and UBB) the
shoulder internal rotation moment statistically reached
or surpassed the 100% strength criterion in both the LBB
and UBB conditions (Figure 5). During the SWFP and HF
conditions, the AP boundary was reached or exceeded.
None of the assessed thresholds were reached when pulling in the FFP condition.
Interpreting the Statistical Results—
Detectable Effect Sizes
This study was designed to ensure that the ANOVA model
was adequately powered to detect differences in hand
forces with effect sizes greater than 0.4. However, many
comparisons were also conducted using the Wilcoxon
signed ranks. Therefore a post hoc analysis of the detectable effect size (Hoenig & Heisey, 2001) was conducted
to determine how small of an effect could be detected at
a power of 0.8. The result of this analysis indicated that
this study was sufficiently powered to detect differences
at the 0.01 alpha level (when using the Wilcoxon signed
ranks test) with an effect size equal to or greater than 1.0.
To aid in interpretation in the context of this project, for
the null hypothesis to be retained (a weakest link variable
reaching or exceeding its criterion threshold) the standard
deviation for each measure had to be less than the difference between 100% and the mean normalized value for
that measure. The effect sizes for each comparison are
presented in Table 2. As the effect size approaches 1.0
the risk of type II error increases, therefore comparisons
where the null hypothesis was retained (the variable
was inferred to be limiting) and where the corresponding effect size is approaching 1.0 should be interpreted
with caution.
Discussion
The main results of this study demonstrated that the
downward exertions were likely limited by elbow or
shoulder strength, where the pulling exertions were likely
limited by AP balance. Although these insights may not
be surprising, this work provides context to help understand the impact of biomechanical limiters as factors
affecting occupational performance. For example, AP
balance is a likely factor to affect pulling force capability; but it is important to quantify its specific influence
on capacity; pulling forces were nearly three times
greater when balance was eliminated as a biomechanical
Figure 4 — The maximum medial hand force capability (N) and standard deviation for each experimental condition. Bars with different letters are significantly different (p < .001) (Top graph). The table beneath the figure shows the corresponding mean (and standard
deviation in parentheses) for each of the biomechanical responses normalized relative to their maximum thresholds. Highlighted
boxes indicated where the null hypothesis was retained (inferring that the variable was a potential biomechanical limiter). SWFP =
shoulder width foot placement; FFP = free foot placements; HF = high friction; LBB = lower body braced; UBB = upper body braced.
Figure 5 — The maximum pulling hand force capability (N) and standard deviation for each experimental condition. Bars with different letters are significantly different (p < .001) (Top graph). The table beneath the figure shows the corresponding mean (and standard
deviation in parentheses) for each of the biomechanical responses normalized relative to their maximum thresholds. Highlighted
boxes indicated where the null hypothesis was retained (inferring that the variable was a potential biomechanical limiter). SWFP =
shoulder width foot placement; FFP = free foot placements; HF = high friction; LBB = lower body braced; UBB = upper body braced.
7
8 Fischer et al.
Table 2 The effect size for each comparison made using the Wilcoxon signed ranks test. As the
effect size increases from 1.0, variables were increasingly likely not to be found limiting based on
the statistical analysis. Conversely, as effect sizes decreased from 1.0, variables were increasingly
likely to be limiting based on the statistical analysis.
Pulling Exertions
Variable
Elbow flex/ext strength
Shoulder rotation strength
Shoulder ab/adduction strength
Shoulder flex/ext strength
Frictional capacity
ML balance point boundary
AP balance point boundary
SWFP
FFP
HF
LBB
UBB
3.81
1.89
3.89
2.48
3.50
4.40
0.11
4.91
2.98
5.65
3.20
2.91
2.90
1.42
4.50
1.29
2.83
2.14
2.77
2.49
–0.30
1.88
–0.26
3.54
2.04
–1.14
1.37
–3.23
1.83
0.26
6.22
2.53
–1.59
1.16
–3.86
SWFP
FFP
HF
LBB
UBB
7.45
2.89
6.96
3.03
16.54
2.72
9.63
6.13
1.73
2.90
1.71
10.17
4.86
6.13
5.35
1.27
3.19
2.46
9.52
1.20
13.10
5.03
0.55
1.93
2.16
6.48
–0.55
1.39
4.22
1.08
1.70
2.87
4.36
–0.55
0.57
Medial Exertions
Elbow flex/ext strength
Shoulder rotation strength
Shoulder ab/adduction strength
Shoulder flex/ext strength
Frictional capacity
ML balance point boundary
AP balance point boundary
Downward Exertions
Elbow flex/ext strength
Shoulder rotation strength
Shoulder ab/adduction strength
Shoulder flex/ext strength
Frictional capacity
ML balance point boundary
AP balance point boundary
SWFP
FFP
HF
LBB
UBB
–0.74
0.17
1.59
0.50
7.60
6.77
3.14
–0.40
–0.21
1.44
0.56
6.66
4.29
3.11
–0.88
0.28
1.44
0.26
7.53
11.11
3.00
–0.55
0.62
1.47
0.82
6.25
6.14
–0.18
–0.23
1.72
1.55
1.45
6.45
3.96
0.09
Note. Highlighted boxes indicated where the null hypothesis was retained (inferring that the variable was a potential biomechanical limiter). SWFP
= shoulder width foot placement; FFP = free foot placements; HF = high friction; LBB = lower body braced; UBB = upper body braced.
constraint. Other unique findings include the lack of an
identifiable limitation for medial exertions and the presence of a shoulder internal rotation strength limitation
within all exertion directions.
Perhaps the most intriguing and impactful finding
was the lack of an identifiable limitation for the medial
exertions. Medial hand force was significantly affected
between the SWFP-FFP and the SWFP-LBB conditions
(Figure 4), with none of the dependent measures reaching or exceeding threshold criteria. This suggests that
the biomechanical limitation affecting this capacity was
not likely measured in the current study design. Since a
large medially directed hand force (at an extended reach
distance) imparts a large axial moment about the body
it is plausible that internal shoulder rotation strength or
trunk axial rotation strength may limit performance to
counterbalance the large external moment. However,
in general, shoulder rotation strength did not reach or
exceed its expected capacity (except during the LBB
condition) suggesting that trunk axial rotation strength
capacity may be a stronger possibility, especially in the
UBB condition, as balance related factors were experimentally eliminated.
However, as indicated in Figure 4, medially directed
hand forces differed between the SWFP and FPP conditions, inferring a balance related limitation. Recalling
from the balance limitations section in the methodology,
the modeling process did not account for the force coupling that occurs between the feet as they work independently to produce axial moments counterbalancing the
axial moments on the body from the applied hand force.
Since this force coupling was not accounted for in the
model, yet the data indicate a balance related challenge,
we attribute the increase in force between the SWFP and
Body Balance, Shoe-Floor Friction, and Joint Strength 9
FFP to an increased ability to optimize the force coupling
between the feet in the FFP condition.
Alternately, the static body could have been modeled using two contact points with the ground, a left foot
contact point and a right foot contact point. This would
enable a more complete understanding of the force
coupling occurring between the feet, especially during
medial exertions. Although straightforward conceptually, this issue has not yet been explored in the hand
force prediction literature. The central challenge lies in
determining the distribution of forces between each foot.
Experimentally this could be achieved using two force
plates; operationally in the field this issue remains more
challenging, where only posture and anthropometric
information are easily available.
Joint strength was expected to limit all braced exertions in addition to all downwardly directed exertions.
Indeed, a plausible joint strength limitation emerged
for all downward exertions and braced (UBB or LBB)
medial or pulling exertions, where a trunk rotation
strength limitation was inferred during braced medially
exertions, as indicated above. Two findings in this regard
emerge as being particularly relevant to understanding
and modeling maximum hand force capability. Firstly,
during downward exertions multiple strength thresholds
were exceeded. From a biomechanical perspective it is
unlikely that two joints would reach their upper boundary at the same time. The reason for reaching multiple
thresholds during the same exertion is likely a reflection
of the data used to describe the upper boundaries (Chaffin
et al., 2006). It is well documented in the seminal work
describing the development of these limits that they represent conservative estimates of strength (Schanne, 1972).
Therefore it may be more plausible that elbow extension
strength is the true limiter during downward exertions as
it further exceeded its conservative threshold. In terms of
modeling hand force capacity the conservative strength
thresholds may have a significant impact on the upper
limit of a predicted hand force capacity.
The second joint strength based finding that is
particularly interesting is the presence of a shoulder
rotation strength limitation in at least one condition for
each direction. It would be expected that shoulder joint
adduction would be a likely limiter during medial exertions, whereas downward exertions could hypothetically
be limited by extension strength. Conversely, for pulling,
due to the force vector being so closely aligned with the
shoulder joint center, it would not likely be suspected
as a limiter at all. The presence of a shoulder rotation
limitation may be related to the off-axis hand force
contributions, which commonly occur during maximum
exertions in a specified direction (Hoffman et al., 2011).
For example when participants pulled, they often tended
to produce off-axis downward forces as well, likely
through a combination of shoulder internal rotation,
adduction and extension.
The suggestion that the off-axis forces acted to
increase joint moments is not consistent with the current
state of literature searching for principles to define why
off-axis forces are produced. Most research to date has
investigated pushing and pulling exertions, suggesting
that off axis forces are created to help minimize joint
moments, specifically at the joint that was thought to limit
performance (de Looze et al., 2000; Hoozemans et al.,
2004; Hoffman et al., 2007). However, previous research
has not explored the possibility that balance may have
limited those exertions, not joint strength, and the off-axis
forces were generated to help protect against reaching a
balance limitation. Based on the results here, the authors
hypothesize that the off-axis forces may rather be used
to help maintain balance. Consider Figure 1, and picture
the addition of an off-axis downward force, which would
cause a reaction force aimed upwards on the figure. The
moment contribution from this reaction force would aid
the moment created by the COM in counteracting the
moment produced from the reaction force due to pulling. In this example, the off-axis forces were not likely
protecting against a joint strength limitation, but against
the factor most likely limiting hand force capacity, in
this case balance. These insights may be useful to further
guide model development aimed at predicting maximum
hand force capability by demonstrating how maximum
force is produced, possibly by optimizing off-axis forces
in a way that helps to protect against reaching a limiting
threshold. Future work modeling hand force capacity
should aim to include estimates of off-axis forces as they
seem particular relevant as evidenced by the high shoulder
moments that resulted during pulling, although the force
vector in the desired direction closely aligned with the
shoulder joint center.
This study had inherent limitations and assumptions.
Postanalysis, it was evident that a primary limitation was
the lack of use of force plates to more readily detect how
medial exertion strength was limited. The presence of
this limitation invites future research advancing on this
methodology to more completely measure and document
force sharing and balance point distribution between the
feet when performing medial exertions. A second limitation or assumption that is important to highlight is the
presumption of a balance limitation. When participants
were braced to eliminate the potential to lose balance;
ankle, knee and hip strengths were also constrained,
making it impossible to partition out balance as a limiter.
However, ankle strength remains unlikely as a limiting
factor. Ankle plantar flexor strengths, normalized to body
weight and height are reportedly between 5.94–8.06 N·m
in older adults (Pavol et al., 2002; Grabiner et al., 2005),
which would provide sufficient ankle strength capacity
to meet the ankle moment demands calculated in the
present work. In addition, errors arising from the postural
data collection, such as intertrial skin motion and marker
placement accuracy, and the use of population-based
anthropometric tables to estimate segment masses, may
have caused some uncertainly in the balance point calculation and the linked segment modeling outcomes. These
effects were mitigated through the use of rigid clusters
to help reduce potential artifacts in the motion capture
data (Kingma et al., 1996) and the error from the use of
anthropometric tables should be randomly distributed
across participants.
10 Fischer et al.
This study applied a novel methodology to deduce
when specific factors may limit hand force capacity, and
further, how they correspond to the underlying biomechanics of the system. It is evident from this work that
the prediction of maximum hand force capacity depends
on the primary and off-axis hand forces, the posture and
the distribution of forces between the feet. These insights
may provide a framework to help understand how posture
could be manipulated or controlled within this paradigm
to optimize force production through the upper extremity.
Acknowledgments
Steven Fischer and Bryan Picco were supported through
NSERC graduate student awards. Additional project support
came from and NSERC discovery grant held by Clark Dickerson. The authors would also like to thank Alan Cudlip for his
assistance with data collection.
References
Boocock, M.G., Haslam, R.A., Lemon, P., & Thorpe, S. (2006).
Initial force and postural adaptations when pushing and
pulling on floor surfaces with good and reduced resistance to slipping. Ergonomics, 49, 801–821. PubMed
doi:10.1080/00140130600562876
Chaffin, D.B. (1969). A computerized biomechanical
model: Development and use in studying gross body
actions. Journal of Biomechanics, 2, 429–441. PubMed
doi:10.1016/0021-9290(69)90018-9
Chaffin, D.B. (1975). Ergonomics guide for the assessment of human static strength. American Industrial
Hygiene Association Journal, 36, 505–511. PubMed
doi:10.1080/0002889758507283
Chaffin, D.B. (1997). Development of computerized human
static strength simulation model for job design. Human
Factors and Ergonomics in Manufacturing, 7, 305–322.
doi:10.1002/(SICI)1520-6564(199723)7:4<305::AIDHFM3>3.0.CO;2-7
Chaffin, D.B., Anderson, G.B.J., & Martin, B.J. (2006). Occupational Biomechanics (4th ed.). Hoboken, NJ: John Wiley
and Sons, Inc.
Chaffin, D.B., & Baker, W.H. (1970). Biomechanical model
for analysis of symmetric sagittal plane lifting. American
Institute of Industrial Engineers Transactions, 2, 16–27.
De Looze, M.P., Van Greuningen, K., Rebel, J., Kingma, I., &
Kuijer, P.P.F.M. (2000). Force direction and physical load
in dynamic pushing and pulling. Ergonomics, 43, 377–390.
PubMed doi:10.1080/001401300184477
Dickerson, C.R., Chaffin, D.B., & Hughes, R.E. (2007).
A mathematical musculoskeletal model for proactive
ergonomic analysis. Computer Methods in Biomechanics and Biomedical Engineering, 10, 389–400. PubMed
doi:10.1080/10255840701592727
Garg, A., & Chaffin, D.B. (1975). A biomechanical computerized simulation of human strength. American Institute of
Industrial Engineers Transactions, 7, 1–15.
Grabiner, M.D., Owings, T.M., & Pavol, M.J. (2005). Lower
extremity strength plays only a small role in determining
the maximum recoverable lean angle in older adults. Journal of Gerontology, 66A, 1447–1450. PubMed
Grieve, D.W. (1983). Slipping due to manual exertion. Ergonomics, 26, 61–72. PubMed doi:10.1080/00140138308963313
Grieve, D.W. (1979a). The postural stability diagram
(PSD): Personal constraints on the static exertion
of force. Ergonomics, 22, 1155–1164. PubMed
doi:10.1080/00140137908924690
Grieve, D.W. (1979b). Environmental constraints on the static
exertion of force: PSD analysis in task design. Ergonomics,
22, 1165–1175. PubMed doi:10.1080/00140137908924691
Hoenig, J.M., & Heisey, D.M. (2001). The abuse of power:
The pervasive fallacy of power calculations for data
analysis. The American Statistician, 55, 19–24.
doi:10.1198/000313001300339897
Hoffman, S.G., Reed, M.P., & Chaffin, D.B. (2011). A study
of the differences between nominal and actual hand
forces in two-handed sagittal plane whole-body exertions.
Ergonomics, 54, 47–59. PubMed doi:10.1080/00140139
.2010.535021
Hoffman, S.G. (2008). Whole-body postures during standing
hand-force exertions: Development of a 3D biomechanical
posture prediction model (Doctoral dissertation). Retrieved
from ProQuest Dissertations & Theses database. (UMI
No. 3304986).
Hoffman, S.G., Reed, M.P., & Chaffin, D.B. (2007). The relationship between hand force direction and posture during
two-handed pushing tasks. Proceedings of the Human
Factors and Ergonomics Society 51st annual meeting,
928-932.
Holbein, M.A., & Chaffin, D.B. (1997). Stability limits in
extreme postures: Effects of load positioning, foot placement, and strength. Human Factors, 39, 456–468. PubMed
doi:10.1518/001872097778827160
Holbein, M.A., & Redfern, M.S. (1997). Functional stability
limits while holding loads in various positions. International Journal of Industrial Ergonomics, 19, 387–395.
PubMed doi:10.1016/S0169-8141(96)00023-6
Holbein-Jenny, M-A., McDermott, K., Shaw, C., & Demchak,
J. (2007). Validity of functional stability limits as a measure of balance in adults aged 23-73 years. Ergonomics,
50, 631–646. PubMed doi:10.1080/00140130601154814
Hoozemans, M.J.M., Kuijer, P.F.M., Kingma, I., van Dieën,
J.H., de Vries, W.H.K., van der Woude, L.H.V., . . . FringsDresen, M.H.W. (2004). Mechanical loading of the low
back and shoulders during pushing and pulling activities.
Ergonomics, 47, 1–18. PubMed doi:10.1080/001401303
10001593577
Kerk, C.J., Chaffin, D.B., & Keyserling, W.M. (1998). Stability as a constraint in sagittal plane human force exertion
modeling. Occupational Ergonomics, 1, 23–39.
Kerk, C.J., Chaffin, D.B., Page, G.B., & Hughes, R.E. (1994).
A comprehensive biomechanical model using strength,
stability, and COF constraints to predict hand force exertion capability under sagittally symmetric static conditions.
American Institute of Industrial Engineers Transactions,
26, 57–67.
Kingma, I., de Looze, M., Toussaint, H.M., Klijnsma, J.G., &
Bruijnen, T.B.M. (1996). Validation of a full body 3-D
dynamic linked segment model. Human Movement Science, 15, 833–860. doi:10.1016/S0167-9457(96)00034-6
Kroemer, K.H. (1974). Horizontal push and pull forces
exertable while standing in working positions on various surfaces. Applied Ergonomics, 5, 94–102. PubMed
doi:10.1016/0003-6870(74)90085-4
Body Balance, Shoe-Floor Friction, and Joint Strength 11
Langenderfer, J.E., Carpenter, J.E., Johnson, M.E., An, K., &
Hughes, R.E. (2006). A probabilistic model of glenohumeral external rotation strength for healthy normal’s and
rotator cuff tear cases. Annals of Biomedical Engineering,
34, 465–476. PubMed doi:10.1007/s10439-005-9045-9
Lee, T.H., & Lee, Y.H. (2003). An investigation of stability
limits while holding a load. Ergonomics, 46, 446–454.
PubMed doi:10.1080/0014013021000039583
Nussbaum, M.A., & Lang, A. (2005). Relationships between
static load acceptability, ratings of perceived exertion,
and biomechanical demands. International Journal of
Industrial Ergonomics, 35, 547–557. doi:10.1016/j.
ergon.2005.01.005
Pavol, M.J., Owings, T.M., Foley, K.T., & Grabiner, M.D.
(2002). Influence of lower extremity strength in healthy
older adults on the outcome of an induced trip. Journal
of the American Geriatrics Society, 50, 256–262. PubMed
doi:10.1046/j.1532-5415.2002.50056.x
Sargent, D. (1897). Strength tests and the strong men of Harvard. Boston Society of Medical Sciences, 1, 7–18.
Schanne, F.A. (1972). A three-dimensional hand force capability model for the seated operator. (Doctoral dissertation).
Retrieved from ProQuest Dissertations & Theses database.
(UMI No. 7229190).
Seo, N-J., Armstrong, T.J., & Young, J.G. (2010). Effects of
handle orientation, gloves, handle friction and elbow posture
on maximum horizontal pull and push forces. Ergonomics,
53, 92–101. PubMed doi:10.1080/00140130903389035
Stobbe, T. (1982). The development of a practical strength
testing program for industry. (Doctoral dissertation).
Retrieved from ProQuest Dissertations & Theses database.
(UMI No. 8215093).