J Exerc Sci Fit •Vol 6 •No 2 •2008 67
THE SIZE PRINCIPLE AND A CRITICAL ANALYSIS OF
THE UNSUBSTANTIATED HEAVIER-IS-BETTER
RECOMMENDATION FOR RESISTANCE TRAINING
Ralph N. Carpinelli
Human Performance Laboratory, Adelphi University, Garden City, New York, USA
The size principle states that motor units are recruited in an orderly manner from the smaller (lower threshold)
to the larger (higher threshold) motor units, and that the recruitment is dependent on the effort of the activity.
Greater recruitment produces higher muscular force. However, the pervasive faulty assumption that maximal
or near maximal force (very heavy resistance) is required for recruitment of the higher-threshold motor units
and optimal strength gains is not supported by the size principle, motor unit activation studies, or resistance
training studies. This flawed premise has resulted in the unsubstantiated heavier-is-better recommendation
for resistance training.
Keywords: effort, motor units, recruitment, strength training
The purpose of this review is to show that the strongly
supported size principle of motor unit recruitment has
been incorrectly applied in the resistance training litera-
ture. Because greater motor unit activity produces a great-
er force output, it is mistakenly believed that a greater
force (very heavy resistance) is required for maximal
motor unit activation. This commonly held belief is an in-
valid reverse inference of the size principle. The belief is
apparently based on a faulty premise, which was estab-
lished without any supporting evidence (resistance train-
ing studies). This review also shows that the references
cited by the authors, who claim that a heavier resistance
produces greater strength gains, do not substantiate
their training recommendations and in many instances
are irrelevant to their claims. Although the section enti-
tled Misapplication of the Size Principle is lengthy, it is
important to show specifically how each of the authors’
references failed to support their claims and recommen-
dations. It is not sufficient simply to cite the references
without noting exactly what the authors of those studies
and reviews reported.
Despite the lack of supporting evidence, the heavier-
is-better belief continues. Perhaps because these authors
are considered by many to be the leading authorities in
resistance training, their unsupported heavier-is-better
training philosophy has gone unchallenged for decades.
The Size Principle
A motor unit consists of the neuron, its axon, and all
the muscle fibers it innervates. A single motor unit may
innervate a few to several hundred muscle fibers, and
a specific muscle may contain a few to several hundred
motor units. Smaller motor units are more easily excitable
and innervate fewer muscle fibers than larger motor
units (Guyton 1991). Although the size and specific force
capability of each muscle fiber in a motor unit con-
tribute to the total force capacity of a specific muscle,
the variation in the number of innervated fibers is the
Ralph N. Carpinelli, P.O. Box 241, Miller Place,
NY 11764, USA.
Tel: (1) 516 877 4276
Fax: (1) 516 877 4274
predominant factor in differences in motor unit force
and therefore total force capacity. The size principle states
that when the central nervous system recruits motor
units for a specific activity, it begins with the smallest,
more easily excited, least powerful motor units and pro-
gresses to the larger, more difficult to excite, most pow-
erful motor units to maintain or increase force. Gradation
of force is a result of the activation of a number of active
motor units (recruitment) and their rate of discharge
(rate coding). This orderly recruitment of motor units
provides a smooth gradation of force (Guyton 1991).
A motor neuron pool consists of all the motor units
available for a specific activity. If the input to a specific
pool of motor units exceeds the excitation threshold
of the motor neuron, the neuron will generate action
potentials (nerve impulses) to activate the innervated
muscle fibers (Kossev & Christova 1998; Senn et al. 1997;
Fuglevand et al. 1993; Guyton 1991). Thus, the motor
neuron’s recruitment threshold is dependent on the
interaction between the organization and strength of
the synaptic inputs and the motor neuron’s intrinsic
responsiveness to that input (Cope & Pinter 1995).
Maximal recruitment occurs between 30% and 90% of
maximal voluntary contraction (MVC), depending on the
specific muscle (Enoka & Fuglevand 2001; DeLuca et al.
1996; Erim et al. 1996; Kukulka & Clamann 1981), with
the threshold for maximal recruitment relatively lower in
dynamic compared with static muscle actions (Linnamo
et al. 2003). Smaller muscles, where smaller gradations
in force are important, recruit all the motor units in the
pool at relatively low force levels, and rate coding (firing
more frequently) then becomes the primary means of
modulating any increase in force (Fuglevand et al. 1993).
Recruitment plays a greater role in force development
in larger muscles, whereas rate coding is the predomi-
nant force-generating factor in smaller muscles (Seki &
Origin of the size principle
Denny-Brown and Pennybacker (1938), who were inves-
tigating the occurrence of fibrillation in people with amy-
otrophic lateral sclerosis, stated that a specific voluntary
movement always begins with the discharge of the
same motor units. They noted that more intense contrac-
tions are accomplished by the addition of a greater
number of the increasingly larger motor units, which are
recruited in a specific orderly sequence. The motor units
recruited first were always the smaller motor units, and
the sequence of recruitment was always followed by
recruitment of the larger more powerful motor units,
each controlling many more muscle fibers. In addition,
they observed that in muscles that were affected by the
loss of motor neurons, the larger motor units became
activated without sufficient previous recruitment of the
smaller motor units. The result of this disease process
was an uneven gradation of force.
Henneman (1957) tested Denny-Brown and
Pennybacker’s (1938) observations on cats with tran-
sected spinal cords. He reported that recruitment of pro-
gressively larger motor units required progressively
greater increases in the intensity of the stimulus.
Henneman concluded that the susceptibility of neurons
to discharge is a function of their size; hence, the size
principle. Henneman (1968) later noted: “The smaller
a motor neuron is the more easily it can be fired; the
larger it is the greater is the amount of excitatory input
required to discharge it. Since the size of a motor neuron
and the size of a motor unit are directly related, it follows
that the participation of a motor unit in graded motor
activity is dictated by its size” (p. 625).
The reliability for the orderly recruitment of motor
units, which is the cornerstone of movement control,
has been substantiated over the last 40 years in ani-
mals (Wakeling et al. 2002; Cope & Sokoloff 1999; Cope
et al. 1997; Bawa et al. 1984; Clamann & Henneman
1976; Clamann et al. 1974; Henneman et al. 1974;
Henneman & Olson 1965; Henneman et al. 1965a,
1965b) and humans (Adam & DeLuca 2003; Houtman
et al. 2003; Akaboshi et al. 2000; Scutter & Turker 1998;
Feiereisen et al. 1997; Ivanova et al. 1997; Schmied et al.
1997; Jabre & Spellman 1996; Masakado et al. 1995;
Riek & Bawa 1992; Knaflitz et al. 1990; Moritani &
Muro 1987; Thomas et al. 1987; Thomas et al. 1986;
DeLuca et al. 1982; Desmedt & Godaux 1981; Maton
1980; Desmedt & Godaux 1979; Goldberg & Derfler
1977; Freund et al. 1975; Milner-Brown et al. 1973).
Thus, the size principle of motor unit recruitment is per-
haps the most supported principle in neurophysiology.
The size principle determines the recruitment order
(from smaller, low-threshold motor units to larger, higher-
threshold motor units) during isometric, slow and fast
muscle actions (Enoka & Fuglevand 2001; Ivanova et al.
1997; Masakado et al. 1995; Desmedt & Godaux 1977).
For example, Masakado and colleagues reported the
recruitment pattern during two consecutive isometric
contractions of the first dorsal interosseous (index finger
abduction) in five healthy males (age, 26–37 years). One
smooth isometric contraction was followed by a contrac-
tion performed as fast as possible. Based on the recorded
motor unit action potentials, the authors stated that
there was no evidence of high-threshold motor units
being selectively engaged during the fast contractions
68 J Exerc Sci Fit •Vol 6 •No 2 •2008
(Masakado et al. 1995). That is, there were no observ-
able violations of the size principle. In a similar investi-
gation, Desmedt and Godaux (1977) noted that their
data definitively excluded the hypothesis of preferential
recruitment of fast-twitch higher-threshold motor units.
The authors concluded that they did not observe a sin-
gle instance when the faster (larger, higher threshold)
motor units were recruited before the slower (smaller,
lower threshold) motor units during either a slow con-
traction or a brisk ballistic contraction (Desmedt &
Thus, motor units are recruited in a regular, nonran-
dom sequence (Cope & Pinter 1995), and the predeter-
mined order of recruitment reduces the time required
to select an appropriate combination of active motor
units because the input is distributed uniformly to the
motor units (Senn et al. 1997). The size principle also
ensures that the most frequently used and most easily
excited motor units are those that are most resistant to
fatigue, thereby enhancing energy efficiency (Cope &
Pinter 1995). Edstrom and Grimby (1986) noted that
the weaker, slower contracting, fatigue resistant motor
units recruited before the stronger, rapidly contracting,
fatigable motor units makes great functional sense for
graded movements and prolonged exercise. Cope and
Pinter (1995) concluded that the preponderance of avail-
able evidence suggests that there are no functionally
meaningful violations of the size principle.
Practical application of the size principle
It is important to recognize that at the conception of
the size principle, Denny-Brown and Pennybacker (1938)
and Henneman (1957) specifically stated that the
independent variable was the intensity of the stimulus
and the dependent variable was the recruitment of
higher-threshold (larger, more difficult to excite) motor
units. Force was not the prerequisite for recruitment;
force was the result of a more intense stimulus. The level
of effort in voluntary muscle actions determines the de-
gree of motor unit activity (see section on Interpolated
twitch technique below). The fundamental neuromus-
cular concept is that all voluntary muscle actions initiate
in the brain and action potentials are generated, which
is analogous to the external electrical stimulus gener-
ated in Denny-Brown and Pennybacker’s (1938) and
Henneman’s (1957) studies: the greater the stimulus
(internal action potentials generated by the brain or
external stimulus generated in vitro), the greater recruit-
ment of motor units through the size principle. Each
individual’s effort determines motor neuron activation
Intensity of Effort and Motor Unit Activation
Interpolated twitch technique
The interpolated twitch technique (ITT) is an indirect
measure of the level of motor unit activation during an
MVC. During an MVC, a supramaximal electrical stim-
ulus (typically a single, double or multiple stimuli) is
superimposed (either manually or computer generated)
with surface electrodes onto a muscle or its nerve. When
the superimposed twitch technique is applied prop-
erly, the electrical stimulus fully activates all the motor
units in the pool. If all the motor units have been re-
cruited and are firing at optimal frequencies, no addi-
tional force will be detected. However, if some motor
units are silent or firing at a low frequency, the inter-
polated stimulus will produce a twitch on top of the
voluntary force. The evoked interpolated twitch is usu-
ally measured in Newtons (N) or Newton·meters (N·m).
ITT cannot distinguish between incomplete recruitment
and insufficient motor unit firing rate responsible for
any increment in force as a result of the superimposed
stimulus. Therefore, the term motor unit activation level
(AL) is preferred (Folland & Williams 2006; Oskouei et al.
2003; Herbert & Gandevia 1999; Behm et al. 1996;
Brown et al. 1990).
When there is an increment in external muscle force
(the interpolated twitch) as a result of the superimposed
stimulus during the MVC, the magnitude of the incre-
ment represents that portion of the muscle mass not
activated by the voluntary effort. The interpolated twitch
is typically normalized to the resting twitch force; that is,
the force evoked by the same stimulus applied to the
resting muscle (pre- and/or post-MVC). The formula for
estimating the motor unit activation level is AL (%) =[1 –
Evoked Force (superimposed on the voluntary force)/
Evoked Response (control twitch force)] ×100 (Suter &
Herzog 2001; Allen et al. 1998; Suter et al. 1996).
The amplitude of the interpolated twitch declines
with increasing contraction intensity. A large interpolated
twitch indicates a low level of voluntary activation,
whereas a small or nonexistent interpolated twitch indi-
cates near-maximal or truly maximal voluntary motor
unit activation. Complete motor unit activation is indi-
cated by no increment in force and implies that all motor
units in a specific pool have been recruited and are firing
at rates sufficient to produce a maximal force (Gandevia
et al. 1998; Kent-Braun & Le Blanc 1996; Allen et al.
Although there are some questions concerning the
resolution of force, number of control stimuli, and the
sensitivity of twitch interpolation (Shield & Zhou 2004),
J Exerc Sci Fit •Vol 6 •No 2 •2008 69
a single stimulus, compared to a double or multiple stim-
uli, has been shown to provide a high enough level of
sensitivity for detecting even slight decrements in vol-
untary drive (motor unit inactivity) and are appropriate
for estimating AL. ITT is capable of detecting decrements
in voluntary activation of less than 1% (Allen et al. 1998;
Allen et al. 1995). One should expect that in any specific
comparison study (see examples below), all methodolog-
ical aspects such as sensitivity, variability, validity and
reliability would be similar between the comparison
Comparisons of activation levels
Several studies have reported the force (N) or torque
(N·m) values for MVC, which is an isometric muscle
action in most cases, and the AL in younger and older
groups of previously untrained but healthy, physically
active subjects (males and females). Values for specific
muscles such as the knee extensors (Knight & Kamen
2001; Roos et al. 1999; Hurley et al. 1998), elbow flexors
and extensors (Jakobi & Rice 2002; Klein et al. 2001;
De Serres & Enoka 1998) and ankle plantarflexors and
dorsiflexors (Klass et al. 2005; Connelly et al. 1999;
Kent-Braun & Ng 1999; Kent-Braun & Le Blanc 1996;
Vandervoort & McComas 1986) were reported for sub-
jects in very different age groups (∼a half century).
For example, Roos and colleagues (1999) tested 13
younger (age ∼26 years) and 12 older (age ∼ 80 years)
moderately active males for MVC and AL of the quadri-
ceps. The younger group’s force was 758 N compared
with the older group’s 396N. However, there was no sig-
nificant difference in motor unit AL between the younger
(93.6%) and older groups (95.5%). Roos and colleagues
concluded that the MVC of the older subjects was ∼48%
less than that of the younger subjects. Yet, despite the
older group’s significantly lower maximal force, ITT indi-
cated that there was no significant difference in the abil-
ity of young and old subjects to activate their muscles
(Roos et al. 1999).
Connelly and colleagues (1999) reported the tibialis
anterior MVC and AL in six younger (age ∼21 years)
and six older (age ∼82 years) healthy, recreationally
active males. The MVC in the younger group (44 N·m)
was significantly greater than that in the older group
(32 N·m). However, there was no significant difference
in AL between the younger (99.3%) and older (99.1%)
subjects. They concluded that although the mean max-
imal torque was only 74% of that produced by the
younger subjects, the older subjects were able to acti-
vate the tibialis anterior to a similar high level of ∼99%
(Connelly et al. 1999).
Jakobi and Rice (2002) reported the MVC and AL
for the elbow flexors and extensors in six younger (age
∼24 years) and six older (age ∼83 years) males. MVC
was significantly greater in the younger group (367 N)
compared to that in the older group (208N), but there
was no significant difference in AL (98% and 96%,
younger and older groups, respectively) for the elbow
flexors. There was also a significant difference in elbow
extensor MVC between the younger (318 N) and older
(169 N) groups, with no significant difference in AL (98%
and 99%, respectively). The authors concluded that
although elbow flexion and extension MVC was signifi-
cantly lower (43% and 47%, elbow flexors and exten-
sors, respectively) in the older group, the maximal
voluntary muscle activation was similar for the two
groups (Jakobi & Rice 2002).
Klass and colleagues (2005) tested 20 younger (age
∼26 years) and 19 older (age ∼77 years) males and
females to determine MVC and AL for isometric, concen-
tric and eccentric muscle actions of the tibialis anterior.
Although the mean MVC in the older group was 20%
lower than that in the younger group, the AL (100%) was
not significantly different between the younger and older
groups, or between males and females (Klass et al. 2005).
Kent-Braun and Le Blanc (1996) reported the MVC
and AL in a healthy group and a group with amyotrophic
lateral sclerosis (ALS). The healthy group generated a
higher MVC (152 N) than the ALS group (114 N), but there
was no significant difference in motor unit AL follow-
ing a single stimulus (100% and 96%, respectively) or
double stimuli (100% and 96%, respectively). The only
significant difference was following the superimposed
train of multiple stimuli where the healthy group had
a greater AL (96%) than the ALS group (85%).
Vandervoort and McComas (1986) reported the MVC
and AL in five age groups (age ∼20–100 years) of males
and females for the dorsiflexors and plantarflexors. The
males were consistently stronger than the females across
the age groups (31–39% and 22–43%, dorsiflexors and
plantarflexors, respectively), and the average MVC for
the oldest groups (males and females) was ∼55% of the
youngest groups. However, males and females in all the
age groups, including the youngest (20–32 years) and
the oldest (80–100 years), were able to voluntarily fully
activate (AL ∼100%) their specific pool of motor units
(Vandervoort & McComas 1986).
Practical application of ITT studies
These motor unit activation studies (Klass et al. 2005;
Jakobi & Rice 2002; Klein et al. 2001; Knight & Kamen
2001; Connelly et al. 1999; Kent-Braun & Ng 1999; Roos
70 J Exerc Sci Fit •Vol 6 •No 2 •2008
et al. 1999; De Serres & Enoka 1998; Hurley et al. 1998;
Kent-Braun & Le Blanc 1996; Vandervoort & McComas
1986), which compared different age groups, males and
females, healthy and ALS patients, showed that the dif-
ferences in the ability to generate force did not affect
the ability to voluntarily activate specific motor units. The
studies strongly support the previously discussed neu-
rophysiological concept put forth by Denny-Brown and
Pennybacker (1938) and Henneman (1957).
It is the intensity of the effort (maximal in the
aforementioned ITT studies) that determines the AL of
motor units and the resultant force output. A greater
effort produces greater motor unit activation. Maximal
effort produces maximal, or near maximal, activation of
motor units. The resultant force, which is the dependent
variable—not the independent variable—is a maximal
force produced in a specific individual for a specific
exercise. It is entirely dependent on the intensity of
effort. However, it is important to recognize that none
of the authors of the aforementioned ITT studies spec-
ulated on a minimal recruitment threshold for strength
The motor unit AL required for an optimal stimulus
to increase muscular strength is unknown, and may be
considerably less than the 85–100% AL reported in the
ITT studies. Therefore, a maximal or near maximal effort
may not be required for optimal strength gains. A max-
imal effort only ensures maximal voluntary motor unit
Perhaps the most relevant interpolated twitch study
with the greatest practical application to resistance train-
ing was conducted by Behm and colleagues (2002). They
measured recruitment properties by testing 14 resist-
ance trained males (age ∼ 21 years) for voluntary and
evoked contractile properties before and after perform-
ing dynamic 5, 10 and 20RM (repetition maximum)
elbow flexion exercise (dumbbell curls) at recovery inter-
vals of 30 seconds, 1, 2 and 3 minutes. Repetition dura-
tion was consistent across trials at 3 seconds concentric,
1 second isometric, and 3 seconds eccentric, with a re-
sultant time-under-tension of 35, 70 and 140 seconds
for the 5, 10 and 20RM, respectively. The 12 sessions
were randomly allocated and separated by at least 48
hours. The force amplitudes of superimposed stimulation
compared with post-contraction stimulation was used
to estimate the extent of activation during the MVC. The
superimposed stimulation activated those motor units
left inactivated by the maximal voluntary muscle action.
There was no significant difference in voluntary motor
unit activation following 5RM (95.5%), 10 RM (93.5%)
and 20 RM (95.1%). The three different loads (amount of
resistance) in the 5, 10 and 20 RM protocols, and the very
different time-under-tension (35, 70 and 140 s, respec-
tively), elicited similar AL of motor units (93.5–95.5%).
Behm and colleagues (2002) noted that the most impor-
tant result of their study was that the 5 RM protocol did
not produce greater motor unit activation than the 10 RM
or 20 RM protocols. They concluded: “The commonly
repeated suggestion that maximal strength methods
[resistance heavier than a 6RM] produce greater neural
adaptations or increases in neural drive was not substan-
tiated in this study” (p. 213). In fact, their study unequiv-
ocally demonstrated the direct relationship between
intensity of effort—not the amount of resistance or time-
under-tension—and voluntary motor unit activation.
Misapplication of the Size Principle
Although the size principle is described reasonably
accurately in the resistance training literature, it is often
followed by a misunderstanding of the underlying neu-
rophysiological concept and its practical application to
resistance training. Many resistance training authors state
that in order to produce maximal muscular force, the
larger and more difficult to excite motor units must be
stimulated through a size principle continuum. That
statement is supported by the size principle (Adam &
DeLuca 2003; Houtman et al. 2003; Akaboshi et al. 2000;
Scutter & Turker 1998; Feiereisen et al. 1997; Ivanova
et al. 1997; Schmied et al. 1997; Jabre & Spellman 1996;
Masakado et al. 1995; Riek & Bawa 1992; Knaflitz et al.
1990; Moritani & Muro 1987; Thomas et al. 1987;
Thomas et al. 1986; DeLuca et al. 1982; Desmedt &
Godaux 1981; Maton 1980; Desmedt & Godaux 1979;
Goldberg & Derfler 1977; Freund et al. 1975; Milner-
Brown et al. 1973). However, many authors (see below)
also claim that in order to recruit the larger motor units,
and more importantly to maximize strength gains, a
maximal—or near maximal—resistive force is required.
The latter statement is an invalid reverse inference of
the size principle.
Science places the entire burden of proof on those
who recommend heavier-is-better resistance training to
support their claims and recommendations with resist-
ance training studies and sound neurophysiological prin-
ciples. However, the size principle, interpolated twitch
studies, and resistance training studies do not support
the contention that individuals must train with maximal
or near maximal resistance to achieve optimal strength
gains. The following are specific examples of misappli-
cation of the size principle.
J Exerc Sci Fit •Vol 6 •No 2 •2008 71
Fleck and Kraemer (1997) noted that based on in-
creasing demands of a specific activity, higher-threshold
motor units are progressively recruited after lower-
threshold motor units. That statement is supported by
the size principle. However, they also stated: “Heavier
resistances (e.g., 3 to 5 RM) require the recruitment of
higher-threshold motor units than lighter resistances
(e.g., 12 to 15RM)” (p. 61). It should be clearly under-
stood that a 3–5 RM range does not merely describe
the execution of 3, 4 or 5 repetitions. A 3–5 RM range
specifically refers to the termination of the set because
of the inability to perform another repetition. A 3RM
designates the completion of the 3rd repetition, with the
inability to execute a 4th repetition; a 4 RM designates
the completion of the 4th repetition, with the inability
to execute a 5th repetition; and a 5 RM designates
the completion of the 5th repetition, with the inability
to execute a 6th repetition. With each specific RM, the
completion of an additional repetition in good form
is not possible. Similarly, a 12–15 RM range describes
a set where the 12th, 13th, 14th or 15th repetition
is a maximal effort. Regardless of the specific amount
of resistance used (e.g., 3 RM, 4 RM, 5 RM, 12 RM,
13 RM, 14 RM or 15 RM), the progressive recruitment
of lower- to higher-threshold motor units is similar
because the effort is similar at the completion of the
set—maximal. Therefore, the statement by Fleck and
Kraemer that recruitment of the higher-threshold mo-
tor units is possible only with the heavier resistance
(3–5 RM) and not the lighter resistance (12–15 RM)
appears to be based on their misapplication of the size
Fleck and Kraemer (1997) concluded that when per-
forming a set of repetitions, the higher-threshold motor
units are recruited as the required force increases. How-
ever, higher-threshold motor units are recruited as the
effort increases throughout the set, not because of in-
creased force. If the exercise form and repetition dura-
tion remain constant as one progresses through a set of
repetitions, the force requirement at any specific point
in the range of motion does not increase. Because of
fatiguing motor units—not increased force require-
ment—recruitment of larger motor units and greater
rate coding activity are required to maintain the force
required to complete the set. Therefore, the degree of
effort—not force—increases with each repetition.
Another important distinction is that the completion
of an RM set (e.g., attempting the 7th repetition with a
6 RM) is an MVC, but does not involve a maximal load
(force). Relatively speaking, it is a maximal force out-
put for the fatigued muscle.
The simplest example of increasing effort and a con-
stant force is during an isometric muscle action. If a
person is holding a 20 kg dumbbell at an elbow angle
of 90 degrees, the internal (muscular) torque of the elbow
flexor muscles is equal to the external (resistance) torque.
The first 10 seconds may feel relatively easy. As the task
becomes increasingly more difficult because of the grad-
ual fatigue of the recruited motor units, the brain in-
creases recruitment (through the size principle) and rate
coding in order to maintain the required muscular force.
If after about 60 seconds—despite a maximal effort
to hold the resistance—the external torque exceeds the
internal torque, the person will no longer be able to hold
the 20 kg mass. Despite the increasing effort throughout
the 60 seconds duration, the muscular force remained
constant until it decreased at 60 seconds when the
individual was no longer capable of producing a sub-
stantial muscular force to maintain the required internal
torque. At the point of maximal effort (∼60 seconds),
all the motor units in the pool were recruited (including
the larger motor units) for that specific isometric muscle
Fleck and Kraemer (1997) stated: “The factor that
determines whether to recruit high- or low-threshold
motor units is the total amount of force necessary to
perform the muscular action” (p. 62), and “Typically Type
II motor units have a high twitch force and so are not
recruited unless high forces are needed” (p. 138). These
are more examples of the misapplication of the size prin-
ciple. That is, it does not require a maximal or near max-
imal resistance (external) force to recruit the larger motor
units. According to the size principle, it simply requires
a maximal or near maximal effort, which occurs near
or at the end of any commonly used RM performance
(e.g., 3–5RM, 8–10 RM, 10–12 RM, 12–15 RM).
Kraemer and colleagues (1988) noted that the
amount of resistance is probably one of the most im-
portant aspects of resistance training and that 6RM
loads result in the largest strength gains. They cited four
references (Fleck & Kraemer 1987; Anderson &
Kearney 1982; Atha 1981; Clarke 1973) in an attempt to
support those statements. Anderson and Kearney (1982)
compared 6–8 RM, 30–40 RM and 100–150 RM proto-
cols. Because they did not compare any other commonly
used and tested protocols (e.g., 3–5 RM, 8–10 RM,
10–12RM, 12–15 RM), which may have produced sim-
ilar increases in strength, the claim by Kraemer and
colleagues lacks support. In the review by Atha (1981),
he noted that there were several attempts to show that
training with the heaviest loads produces the greatest
gains in strength and he cited studies by Berger (1963)
72 J Exerc Sci Fit •Vol 6 •No 2 •2008
and O’Shea (1966). As described in a critical analysis
(Carpinelli et al. 2004) of these and other resistance
training studies, Berger (1963) reported no significant
difference in strength gains for the bench press exer-
cise as a result of training with a 2RM (17%), 6 RM (21%)
or 10RM (20%); O’Shea (1966) reported no significant
difference in strength gains for the squat exercise as
a result of training with 2–3 RM (22%), 5–6RM (27%),
or 9–10RM (20%). Atha concluded: “From these stud-
ies, one begins to believe that the importance of load
magnitude may have been exaggerated” (p. 13). Then,
without any explanation, Atha antithetically claimed:
“Tension, not fatigue, is the strengthening stimulus…”
(p. 15). Atha claimed that because an RM load such as
a 10RM does not result in an absolutely greater tension
than a 5 RM, all the motor units available for the spe-
cific exercise would not be activated. That statement
was probably the result of an incorrect application of
the size principle by Atha.
The third reference cited by Kraemer and colleagues
(1988) is a review by Clarke (1973), who cited two other
studies by Berger (1962a, 1962b). Berger (1962a) trained
199 male college students three times a week for 12
weeks. All the participants performed the free weight
bench press for 1 set of 2, 4, 6, 8, 10 or 12 RM. Berger
and Clarke claimed that the gains in 1 RM bench press
in the 4, 6 and 8 RM groups were significantly greater
than in the 2, 10 and 12RM groups (neither the pre-
training data nor the percent gains were reported).
According to Berger’s Table 2 (p. 337), strength gains
as a result of 4, 6 and 8 RM were significantly greater
than 2 RM, and strength gains using the 8 RM protocol
were significantly greater than 2, 10 and 12 RM. How-
ever, there was no significant difference in strength gains
between the 2 RM group and the 10 or 12 RM groups,
which indicated that there was no significant difference
in strength gains as a result of training with the lightest
(12 RM) and the heaviest (2 RM) loads. And, contrary
to the statement by Clarke and in Berger’s narrative,
Berger’s Table 2 (p. 337) showed no significant difference
between the 6 RM group and the 4, 8, 10 and 12 RM
In Berger’s other study (1962b), there was no
significant difference in strength gains (1 RM bench
press) between groups using the lightest load (10 repe-
titions) and the heaviest load (2 repetitions) for either
the lower volume 1-set protocol (12.2 and 11.3 kg, 10 and
2 repetition protocol, respectively) or the higher volume
3-set protocol (13.0 and 13.3 kg, 10 and 2 repetition
protocol, respectively). That is, the heavier resistance did
not elicit a significantly greater strength gain. The fourth
reference cited by Kraemer and colleagues (1988) is the
first edition (Fleck & Kraemer 1987) of the previously
discussed book by the same authors (Fleck & Kraemer
1997). The four references cited by Kraemer and col-
leagues (1988) do not support their claims.
Fleck and Kraemer (1988) also recommended
2–6 RM loads for increasing strength (Figure 1, p. 165)
and stated that loads lighter than 6 RM (7 RM or lighter)
result in progressively smaller strength gains. For exam-
ple, they claimed that strength gains as a result of train-
ing with a 6 RM are 50% greater than with a 10RM.
However, they did not cite any evidence to support their
statements or recommendation.
Kraemer and Bush (1998) described the size principle
as the orderly recruitment of motor units, which is de-
pendent on the recruitment threshold, and that maximal
force production requires the recruitment of the high-
threshold motor units at a sufficiently high activation
rate. However, they recommended 2–3 sets of 3–5 rep-
etitions for increasing strength. They also claimed that
advanced weightlifters may not require the orderly re-
cruitment stipulated by the size principle because these
advanced trainees can inhibit the lower-threshold motor
units and preferentially activate the higher-threshold
motor units. Kraemer and Bush did not cite any refer-
ences to support their training recommendation, their
incorrect application of the size principle, or violations
of the size principle.
Kraemer and colleagues (1998) noted that motor unit
activation is governed by the size principle, and they
described the process: “Specifically, motor units are re-
cruited according to their size and recruitment thresh-
olds” (p. 111). However, they did not cite any references
to support their statement that a resistance equal to or
greater than a 6 RM has the greatest effect on strength
gains (p. 114).
Kraemer and Newton (2000) stated that according
to the size principle, the smaller low-threshold motor
units are recruited first and the progressively higher-
threshold motor units are recruited with increasing
demands of the activity. However, their statement is
followed by four misapplications of the size principle:
1. “Heavier resistances (e.g., 3–5 RM) require the recruit-
ment of higher-threshold motor units than a lighter
resistance (e.g., 12–15 RM)” (p. 363). As previously
noted, by their description of RM loads, the demands
(effort) with both the heavier and lighter resistance
are similar at the end of the set.
2. “Lifting heavier resistances according to the size
principle, however, starts with the recruitment of low-
threshold motor units (Type I). The high-threshold
J Exerc Sci Fit •Vol 6 •No 2 •2008 73
motor units (Type II) needed to produce greater force
are recruited as the needed force increases” (p. 363).
As previously discussed, the force does not increase
throughout a set of repetitions with a specific resist-
ance such as 3–5 RM or 12–15 RM.
3. “Exercises performed by weightlifters such as the
clean and jerk lift may have the potential for recruit-
ment patterns not adhering to the size principle to
enhance power production. Rather than starting with
the recruitment of low-threshold motor units, high-
threshold motor units are recruited first. This means
that the low-threshold motor units are not recruited
in the activity but are skipped over to recruit the high-
threshold motor units first” (p. 363). There was no
reference cited to support their opinion that the size
principle is violated.
4. “The determining factor of whether to recruit high-
or low-threshold motor units is the total amount of
force necessary to perform the muscular action” (p.
363). As previously discussed, their statement is not
supported by the size principle or interpolated twitch
In a book edited by Kraemer and Hakkinen, Kraemer
claimed that a 6 RM load or heavier is required for
strength gains (Kraemer 2002); Fleck (2002) recom-
mended 2–6 repetitions to increase strength; Hasegawa
and colleagues (2002) recommended 1–5 repetitions.
None of these authors cited any references to substan-
tiate their heavier-is-better training philosophy.
Kraemer and Gomez (2001) noted that the size
principle is one of the primary concepts in neuromus-
cular activation, and that lower-threshold motor units
are recruited before the higher-threshold units. They also
claimed that there is documentation that shows inhi-
bition of lower-threshold motor units, which supposedly
allows the brain to more quickly activate higher-
threshold motor units. However, they did not cite any
evidence to support their belief that the size principle
Kraemer (2003) cited the previously discussed review
by Atha (1981), and claimed that heavier resistance is
associated with greater strength gains than lighter resist-
ance. Kraemer (1983) expressed a similar opinion when
he claimed that changing from a 10RM to a 5 RM proto-
col would dramatically affect strength training outcomes,
and that a 10RM builds strength at a slower rate than a
5 RM. The only reference he cited was the previously dis-
cussed training study by Anderson and Kearney (1982)
who compared 6–8 RM, 30–40 RM and 100–150 RM
protocols. They did not use a 5 RM or 10 RM protocol,
which was incorrectly reported by Kraemer (1983).
Kraemer and Ratamess (2004) noted that motor unit
activity increases during the last few repetitions of a set.
They stated: “Maximizing strength, power, and hypertro-
phy may only be accomplished when the maximal num-
bers of motor units are recruited. Thus, heavy loading in
experienced individuals is needed to recruit the high-
threshold motor units that may not be activated during
light-to-moderate lifting” (p. 677). There is no reference
cited to support either of their statements.
Kraemer and Fragala (2006) stated that by exercising
larger muscles first, a superior training stimulus is applied
to the involved muscles and that this would enhance the
ability to lift a greater resistance. The authors claimed
that a heavier resistance creates an optimal stimulus for
strength gains. They also recommended a resistance
heavier than 6 RM or greater than 85% 1 RM for optimal
strength gains. However, they did not cite any evidence
to support their heavier-is-better training philosophy
(Kraemer & Fragala 2006).
In describing the size principle, Kraemer and Vingren
(2007) stated that motor units with progressively higher
thresholds are recruited based on increasing demands
of the activity. They claimed that a 3–5 RM requires the
recruitment of higher-threshold motor units than a
12–15 RM. “If only a low resistance (e.g., 12–15 RM) is
used, then the largest motor units will not be recruited
and thus will not benefit. Therefore, to maximize strength
gains from a resistance training program, high resistance
(1–5 RM) must also be used to stimulate adaptation in the
largest motor units” (p. 19–20). Their Figure 1.10 (p. 20)
depicts a high motor unit activation threshold for the
1 RM and 5 RM, and a low activation threshold for 15 RM
and 20 RM. Kraemer and Vingren also claimed that the
high activation threshold for the larger motor units is not
attained unless high levels of force or power are pro-
duced, that there are specialized high-power motor units
that are recruited only during high-power muscle contrac-
tions, and that during high-velocity movements, smaller
motor units are skipped over (inhibited activation) so
that larger motor units can be recruited first. They did
not cite any evidence to support their heavier-is-better
opinions, their invalid reverse inference of the size prin-
ciple, or their claim for violations of the size principle.
Kraemer and colleagues (2007) noted: “The amount
of resistance used for a specific exercise is one of the
key factors in any resistance training program. It is the
major stimulus related to changes in strength…” (p. 49).
They claimed that if the goal is to maximize strength
gains, then heavy loads with a few repetitions are requir-
ed. The authors did not cite any references to support
74 J Exerc Sci Fit •Vol 6 •No 2 •2008
Brown and colleagues (2007) stated that improving
maximal strength is most successful when heavy loads
are used; that is, 2–4 repetitions with loads close to the
1 RM. They did not cite any evidence to support their
claim. Brown and colleagues also stated that loads be-
tween 85% and 95% 1 RM create the overload required
for maximal strength gains and they cited a meta-
analysis (Peterson et al. 2004) to support their recom-
mendation. The credibility of the meta-analysis by
Peterson and colleagues has been previously challenged
(Otto & Carpinelli 2006). For example, Peterson and
colleagues claimed that there was a trend for increased
strength gains with a greater percent 1RM up to 85%
1 RM. However, their own data actually failed to support
that claim. Their effect sizes for training with 70%, 75%,
80% and 85% 1 RM were 0.07, 0.73, 0.57 and 1.12,
respectively. The reported effect size was 10 times greater
for training with 75% 1 RM compared with 70% 1 RM,
decreased for 80% 1 RM, and then doubled for 85%
1 RM. Their data also implied the unlikely scenario
that training with 70% 1 RM to muscular fatigue had no
effect on strength gains. More importantly, there is no
known physiological hypothesis to explain why a 5%
difference in resistance, performed for one or two fewer
or greater repetitions with a similar effort (RM), would
result in such large differences in outcomes. In addition,
many studies in the meta-analysis had no control group,
which required the use of a pooled standard deviation
rather than the pre-training standard deviation employed
by Peterson and colleagues. Most of the studies included
in their meta-analysis did not compare different RMs
or percent 1 RM; they simply reported the effects of one
specific protocol (e.g., with or without dietary supple-
ments) on different outcomes. The data from this meta-
analysis do not support the claims by Peterson and
colleagues (2004) or Brown and colleagues (2007).
Stone (1993) noted that the size principle describes
the recruitment process from the smaller to the larger
motor units and the evidence suggests that this orderly
sequence is similar for gradual or explosive voluntary
and reflex muscle actions. However, he also stated that
the larger motor units are only recruited when high
force or high power outputs are required, and that in
order to activate the larger motor units, very high force-
ful contractions must be employed. The former state-
ment is valid. The latter statement is inconsistent with
the size principle, which has resulted in unsubstanti-
ated resistance training recommendations. According
to the size principle, there is no reason to believe that
non-explosive, lower-power repetitions performed with
a moderate resistance and a reasonable effort—rather
than explosive high force or high power repetitions—
are any less effective for the recruitment of the larger
Stone (1982) also stated that purposefully slow
movements reduce the training effect because the lower
forces require the recruitment of fewer motor units;
therefore, the larger motor units will not be effectively
trained. He did not cite any training studies to support
his opinion that the size principle is violated. As previ-
ously discussed, the size principle dictates that motor
unit recruitment is determined by the degree of effort,
which can be maximal or near maximal at the end of a
set of repetitions, regardless of repetition duration or
absolute muscular force.
Stone and O’Bryant (1984) stated: “Tension, not
fatigue, is the major factor in developing maximum
strength” (p. 148). Because the only reference to support
their questionable application of the size principle was
the previously discussed review by Atha (1981), their
statement remains unsubstantiated.
Garhammer (1987) described the size principle as
the recruitment of smaller motor units first and larger
motor units last. He also claimed that performing 3
sets of 5 repetitions for the bench press and squat exer-
cises was the most productive way to train, but recom-
mended 10–15 repetitions for other exercises such as
the thigh curl. No rationale was presented for his appar-
ent incorrect interpretation of the size principle (heavier
weights to recruit the larger motor units), or why the
hamstring and quadriceps would require a different
range of repetitions.
Berger (1982) noted: “The actual force of a muscle
depends on the number of stimulated motor units and
their frequency of firing or MUI [motor unit involve-
ment]” (p. 15). However, he then stated: “To increase
force capacity of a muscle, relatively heavy loads must
be lifted” (p. 30). Berger’s statement is surprising be-
cause his own studies (Berger 1963, 1962a, 1962b) do
not support his claim, and because he specifically dis-
tinguished between force and effort. That is, Berger cor-
rectly stated that if a specific load (resistance) is held
until fatigue occurs, the force of contraction remains
constant as motor unit involvement increases (Berger
Palmieri (1983) noted that there is an orderly
recruitment of motor units based on the size principle,
with the smaller motor units recruited first followed by
the larger motor units. However, his claim that the resist-
ance (load) must be high because moderate loads do
not require the recruitment of larger motor units is
inconsistent with the size principle.
J Exerc Sci Fit •Vol 6 •No 2 •2008 75
Haff and Potteiger (2001) described the size principle,
but they then claimed that “… larger more powerful
motor units are recruited only when high force or high
power outputs are demanded by the activity. Thus, in
order to activate the larger motor units, explosive exer-
cises—which generally require high force and high
power outputs—are needed” (p. 14). The authors did
not cite references to support their statement, which
was an apparent misapplication of the size principle.
Hoffman (2002) described the size principle as
the orderly recruitment of motor units from the smaller,
lower-threshold motor units to the larger, higher-
threshold motor units. However, he also stated that with
high-velocity, high-power muscle actions, the larger,
higher-threshold motor units may be recruited first. He
defined exercise intensity as the percent of an individ-
ual’s RM for a specific exercise, and noted: “To maxi-
mize muscular strength gains, the muscle needs to be
stimulated with a resistance of relatively high intensity”
(p. 72); and “It appears that RM loads of 6 or fewer
have the greatest effect on maximal strength or power
output” (p. 80). Hoffman also recommended specific
ranges of repetitions for different outcomes. Interestingly,
he recommended 6–8 RM for maximal strength, which
contradicted his previous claim of 6 or fewer repeti-
tions, and 10–12 RM for muscular hypertrophy. He cited
no references to support his recommendations or his
erroneous interpretation of the size principle. In fact,
there is very little evidence to suggest that his recom-
mended differences in the range of repetitions elicit
different outcomes (Carpinelli et al. 2004).
In an extensive review of resistance exercise inten-
sity, Fry (2004) operationally defined intensity as the
percent of maximal strength (%1 RM) for a specific ex-
ercise. He noted that exercise intensity is one of the
most important training variables, and that the greatest
percent increase in maximal strength occurs with loads
approaching 100% 1 RM. He claimed that his Figure 3
“clearly illustrates” (p. 670) this specificity of training.
His Figure 3 shows that resistance training with 95%
1 RM increased 1 RM by approximately 60% compared
with training loads of 80%, 60% and 40% 1 RM (∼36%,
34% and 36% increases, respectively) for the leg press
exercise. However, the same figure also shows that the
gains in 1 RM for the knee extension exercise were simi-
lar when using loads of 95%, 80% and 60% 1 RM (∼58%,
51% and 52% increases, respectively). That is, Fry
claimed that training the leg press with 95% 1 RM pro-
duced almost twice the strength gain as training with
80% 1 RM, but using 95% 1 RM and 80% 1 RM with
the knee extension exercise produced similar strength
gains (58% and 51%, respectively). Fry did not address
the disparity in outcomes between these two exercises;
that is, he did not explain what factors could possibly
contribute to such large differences in the strength gains
between these muscle groups.
A much more revealing issue is the examination of
the four studies (Green et al. 1998; Hakkinen & Pakarinen
1993; Hakkinen et al. 1990; Thorstensson et al. 1976)
that Fry (2004) cited to support his clearly illustrated
data in Figure 3 (p. 670). Green and colleagues (1998)
trained six previously untrained males (age, 19.2 years)
three times a week for 12 weeks. All the participants
performed 3 sets of 6–8 RM in the squat, leg press and
knee extension exercises. Green and colleagues reported
on histochemical analyses of the quadriceps muscles
(fiber area, fiber-type distribution, enzyme activity, and
capillarization). They reported the resistance (6–8 RM)
used at the beginning and the end of the study. However,
contrary to what Fry stated, there was no report of pre-
or post-training 1 RM. In addition, Fry did not explain
how he was able to determine the percent 1RM (training
intensity, according to Fry) from the range of repeti-
tions. It has been shown that the number of repetitions
performed to fatigue differs among individuals as well
as among different exercises within an individual at a
specific percent of the 1 RM (Hoeger et al. 1990, 1987).
Hakkinen and Pakarinen (1993) reported the acute
hormonal responses in 10 male strength athletes (age,
29.7 years) to specific resistance training protocols.
In one session, the athletes performed 20 sets of 1 RM
squats with 3 minutes of rest between sets. At another
session (separated by 48 hours), the subjects per-
formed 10 sets of 10 RM squats with 3 minutes of rest
between sets. Because of fatigue, the resistance was
adjusted during the sessions so that the subjects were
able to complete the designated number of repetitions
in each of the 20 sets of 1 RM and each of the 10 sets
of 10RM. Resistance was reduced 10.3% during the
1 RM session and 24.6% in the 10RM session. This
was not a longitudinal training study. Therefore, the
study does not support what Fry (2004) claimed in his
Figure 3 (p. 670).
Hakkinen and colleagues (1990) reported neuro-
muscular adaptations, hormonal concentrations, and
force production in seven non-competitive physically
active females (age, 24.7 years) after 16 weeks of various
body-weight jumping exercises. Strength training during
the last 12 weeks consisted of several sets of 5–6 repe-
titions with 40–80% 1 RM for the leg press or squat
exercise. Neither pre- nor post-training 1 RM was re-
ported. Therefore, Fry’s use of this study to support his
76 J Exerc Sci Fit •Vol 6 •No 2 •2008
claim for changes in 1 RM as a result of training with
a specific percent of the 1 RM (40%, 60%, 80% or 95%
1 RM) is highly questionable (Fry 2004).
Thorstensson and colleagues (1976) trained 14
physical education students (age, 19–31 years) with 3
sets of 6 RM squats, three times a week for 8 weeks.
The 1 RM squat increased 67%. However, they did not
report the percent 1 RM used in training, and as previ-
ously discussed, the percent of the 1 RM for a specific
exercise in a group of subjects cannot be accurately
estimated from the number of repetitions performed.
There was insufficient information reported in the four
studies (Green et al. 1998; Hakkinen & Pakarinen 1993;
Hakkinen et al. 1990; Thorstensson et al. 1976) cited
by Fry (2004) to support his attempt to clearly illustrate
a specificity (percent 1 RM) of training, and none of
the studies he cited actually compared the effects of
training with different percents of the 1RM. Therefore,
it is unclear how Fry derived the data shown in his
Fry (2004) also stated that it has been theorized
that his training philosophy (heavier resistance produces
better results) is more important for advanced trainees.
The four references he cited to support his claim are
books—not resistance training studies. The misappli-
cation of the size principle in the books by Fleck and
Kraemer (1997) and Stone and O’Bryant (1984) has been
previously discussed. In their book entitled Periodization
Breakthrough!, Fleck and Kraemer (1996) claimed that
to increase maximal strength and power, 1–6 repeti-
tions are required. They specifically noted that training
with a 5 RM load produces mostly gains in maximal
strength and power and some muscular size, while train-
ing with a 7RM load elicits smaller gains in strength
and power and greater muscular hypertrophy. They did
not cite any training studies to support their opinion that
training with 5 repetitions per set versus 7 repetitions
per set would produce differential outcomes—either
statistically or clinically significant—in muscular strength,
power or hypertrophy.
In the book by Zatsiorsky (1995), the author reported
his direct observations of the USSR weightlifting team
during the year prior to the 1988 Olympic Games. Only
7% of all the lifts employed loads heavier than 90%
maximum. These elite competitive athletes performed
as many lifts (8%) with loads lighter than 60% of maxi-
mum. The majority of all the lifts (85%) were with
loads between 60% and 90% maximum (Figure 4.4,
p. 97). Zatsiorsky noted that the largest proportion of
weights lifted was between 70% and 80% maximum
resistance, and that these levels should be “thoughtfully
implemented” (p. 107). Fry’s heavier-is-better training
philosophy (Fry 2004) is not supported by Zatsiorsky’s
Fry (2004) concluded: “The bottom line is that rela-
tively heavy loads must be utilized if maximal strength
is to be increased and/or maintained” (p. 671). However,
neither the size principle nor the references (Green et al.
1998; Fleck & Kraemer 1997; Fleck & Kraemer 1996;
Zatsiorsky 1995; Hakkinen & Pakarinen 1993; Hakkinen
et al. 1990; Stone & O’Bryant 1984; Thorstensson et al.
1976) cited by Fry supports his conclusion.
Several other claims related to the size principle
and motor unit recruitment in the aforementioned book
by Zatsiorsky (1995), which was cited by Fry (2004)
and also cited five times in the American College of
Sports Medicine’s Position Stand (2002) on resistance
training, warrant examination. After he described the
size principle, Zatsiorsky expressed opinions that lack
any scientific support. For example, Zatsiorsky claimed
in his Figure 4.7 (p. 103) that a 1RM recruits all the
motor units within the pool, with slower and interme-
diate motor units recruited but not exhausted; and the
larger motor units recruited and exhausted (he did not
define exhausted). He claimed in his narrative that
these larger exhausted motor units are the only motor
units subjected to a training stimulus. However, there
is no neurophysiological evidence to suggest that during
a maximal effort (e.g., 1RM, 5 RM or 10 RM), an optimal
stimulus requires exhaustion of the motor units. Further-
more, he did not cite studies that support any of his
Zatsiorsky (1995) noted that maximal muscular
tension is achieved by lifting a maximal resistance (1 RM)
or lifting a submaximal resistance to failure (e.g., 5 RM).
Although a 1 RM could elicit high muscular tension, this
is not the case with the submaximal resistance. The effort
is the same (maximal), but as previously discussed,
the tension created within the muscle is less with the
submaximal load (assuming similar repetition duration).
The tension remains at that level throughout a set of
repetitions until fatigue causes a reduction in the ability
to generate tension.
Zatsiorsky (1995) stated: “If an athlete can lift a bar-
bell 12 times but lifts only 10, the exercise set is worth-
less” (p. 105). He noted that the smaller motor units
are recruited but not exhausted, and the larger motor
units are not recruited at all (Figure 4.7, p. 103). He
claimed that if motor units are not fatigued (he did not
define fatigued), then they are not trained. Because
there is no universally accepted definition for the con-
cept of exhausted or fatigued muscles, these words are
J Exerc Sci Fit •Vol 6 •No 2 •2008 77
extremely ambiguous unless specifically defined by the
author. Zatsiorsky did not cite any references to support
his opinion that not training to failure is worthless or to
substantiate his speculation on exhausted or fatigued
Bird and colleagues (2005) recommended 8–12 RM
for maximal strength gains and their only reference to
support that protocol was a meta-analysis by Rhea and
colleagues (2003). Rhea and colleagues claimed that
previously untrained individuals experience maximal
strength gains training with a 12 RM and that training
with an 8 RM produces the greatest strength gains in
previously trained (not defined) individuals. However,
there are a few very important examples to show that
the meta-analysis by Rhea and colleagues is illogical and
impractical. In their Table 1 (p. 458), they reported an
effect size of 2.8 as a result of training with an average
60% 1 RM and an effect size of 1.2 as a result of 70%
1 RM training. Assuming that the effort at the end of
each set in the previously untrained subjects was similar
in both situations, the effect size as a result of training
with 60% 1 RM was 2.3 times greater than training
with 70% 1 RM (a difference of 1.6 standard deviations).
There is no known physiological mechanism that would
explain their reported pattern of outcomes.
Rhea and colleagues’ (2003) data for advanced
trainees are also illogical from a practical aspect. They
claimed that training with an average 80% 1 RM resulted
in an effect size (1.8) that was almost three times greater
than using 85% 1 RM (0.65). In other words, their data
suggest that if the average 1RM is 100 kg for a specific
exercise, and training involves an 8 RM with 80 kg, which
was estimated by Rhea and colleagues to be 80% 1 RM
(e.g., with a variation of 70–90% 1 RM), the effect on
strength gains is 2.8 times greater (more than 1 stan-
dard deviation difference) than performing that spe-
cific exercise with an average of 85 kg (85% 1 RM or
approximately 6–7 RM). There is no known physiologi-
cal mechanism that would explain this large difference
in outcomes as a result of a relatively small change (5%)
in resistance. In addition, the practical application of an
average percent 1RM is also questionable and was not
addressed by Rhea and colleagues. For example, training
three times a week with an average 80% 1 RM could
refer to 1 set for a specific muscle group at each of the
three sessions using 75%, 80% and 85% 1 RM (average =
80%1 RM); or 3 sets for the muscle group with 65%, 80%
and 95% 1 RM (average=80% 1 RM); or some combi-
nation of these examples. They did not explain how
they determined the average percent 1 RM or its prac-
tical application to resistance training. The conclusions
regarding training intensity (percent 1RM) reported by
Rhea and colleagues are illogical, without foundation,
and have no reasonable practical application to resist-
ance training. Their own data do not support their own
claims (Rhea et al. 2003) or the training protocol rec-
ommended by Bird and colleagues (2005).
In a lengthy review of adaptations to resistance
training, Crewther and colleagues (2005) stated: “Heavy
loads would seem fundamental to strength develop-
ment, as high forces are associated with maximal motor
unit recruitment according to the ‘size principle’, with
these units also firing at higher frequencies” (p. 975).
They also noted: “Given the relative importance of high
forces for adaptation, the use of heavy training loads
would appear to provide the superior stimulus for max-
imal strength development” (p. 975). Crewther and
colleagues did not cite any references to support either
of their apparent misapplications of the size principle.
Crewther and colleagues (2005) stated that because
of the maximal efficiency of eccentric muscle actions,
supramaximal loads (>100% concentric 1 RM) may facil-
itate the development of “high-load forces” (p. 977).
They claimed that a study by Murphy and colleagues
(1994) supported the use of supramaximal eccentric
loading to increase force production. However, Murphy
and colleagues merely attempted to develop a so-called
iso-inertial force-mass relationship with loads ranging
from 30% to 150% concentric 1 RM bench press, deter-
mine its relationship to dynamic physical performance
and isometric tests, and report the reliability of isometric
and iso-inertial tests. This was not a training study.
Furthermore, Murphy and colleagues did not suggest
that training with higher supramaximal loads would
elicit superior outcomes. Actually, they reported (Figure 2,
p. 253) no significant difference in force measurements
(∼1150, 1160, 1170 and 1190 N, respectively) with loads
(mass) of 100% concentric 1 RM (concentric-only mus-
cle action), 100%, 130% and 150% concentric 1 RM
(eccentric-only muscle action). This was antithetical to
the claim by Crewther and colleagues that the force
output increased “… with the heavier masses in the
eccentric conditions” (p. 977).
Crewther and colleagues (2005) also stated that larger
high-threshold motor units may be preferentially
recruited during lengthening muscle actions with
resistance greater than 100% concentric 1 RM, and that
this supramaximal loading would seem ideal for maxi-
mal strength development. They cited a review by Behm
(1995). In fact, Behm merely stated that Nardone and
colleagues (1989) reported some alterations in motor
unit recruitment during lengthening muscle actions,
78 J Exerc Sci Fit •Vol 6 •No 2 •2008
with some high-threshold motor units firing prior to low-
threshold motor units. However, Nardone and colleagues
reported electromyographic activity in five subjects who
performed concentric, isometric and eccentric plantar
flexion muscle actions with a torque that corresponded
to 15–20% of the torque produced during a maximal
voluntary muscle action. Neither the review by Behm
nor the study by Nardone and colleagues supports the
claim by Crewther and colleagues because the forces
were less than 20% maximal and not supramaximal,
as mistakenly reported by Crewther and colleagues.
Furthermore, Crewther and colleagues (2005) noted
that ballistic training (described by them as throwing
and catching the resistance) may enhance the training
stimulus. They speculated that near maximal forces are
required to elicit maximal strength gains. They did not
cite any references to support their recommendation
for throwing and catching the resistance.
Crewther and colleagues (2005) claimed that another
benefit of supramaximal loading is the higher incidence
of microscopic muscle injury, which provides an ideal
stimulus for morphological adaptation to occur according
to the protein degradation-synthesis response of muscle.
However, the studies they cited (Nosaka & Newton 2002;
Jamurtas et al. 2000) did not measure or report micro-
scopic muscle injury. Nosaka and Newton reported the
acute effects of performing 3 sets of 10 maximal eccen-
tric arm curl muscle actions in one limb compared with
3 sets of 10 submaximal (50% maximal isometric force
at 90 degrees) eccentric muscle actions in the contralat-
eral limb of eight previously untrained males. Plasma cre-
atine kinase concentration significantly increased after
both sessions, with peak levels occurring 4–5 days post-
exercise. These changes in creatine kinase were signifi-
cantly greater following the session of maximal eccentric
loading compared with submaximal loading, and the dif-
ferences became greater after 3–5 days. The authors sug-
gested that the difference in load affected the recovery
process several days after the exercise session, which was
significantly more than the changes immediately post-
exercise. Nosaka and Newton specifically noted: “… CK
[creatine kinase] activity may not be a quantitative re-
flection of muscle damage” (p. 207). However, they did
advise against using maximal or near-maximal resistance
for eccentric muscle actions in potential non-compliant
trainees, and recommended the use of submaximal re-
sistance and machines with eccentrically shaped cams
(e.g., Nautilus machines) that vary the external torque.
Contrary to the statement by Crewther and colleagues,
Nosaka and Newton did not measure microscopic mus-
cle damage. More importantly, they did not suggest that
greater muscle damage would produce an ideal stimulus,
nor did they speculate on how muscle damage would
stimulate any strength-related adaptations.
In the other reference cited by Crewther and col-
leagues (2005), Jamurtas and colleagues (2000) reported
the acute responses to 6 sets of knee extension and
calf raise exercises performed to volitional exhaustion
in 24 previously untrained males. Two sessions were
separated by 6 weeks. The three groups of participants
performed the exercises using 70% of their 1 RM con-
centric-only machine exercise, eccentric-only machine
exercise, or plyometric jumps (concentric and eccentric
muscle actions). Creatine kinase concentration signifi-
cantly increased post-exercise, peaked at 24 hours, and
remained elevated for 72 hours. The changes were not
significantly different among the groups (concentric,
eccentric and plyometric) at any time (24, 48 and 72
hours post-exercise). Muscle soreness was significantly
lower in the concentric group compared with the other
two groups. Jamurtas and colleagues speculated that
the eccentric phase of the plyometric exercises may
have produced more microscopic damage to the muscle
fibers, as indicated by the greater degree of soreness.
Because of the reported tenderness around the muscu-
lotendinus junction, they hypothesized that damage to
connective tissue may have contributed to the perceived
soreness. They did not measure microscopic muscle
damage, nor did they speculate on the effect of muscle
damage as a stimulus for adaptation, as erroneously
reported by Crewther and colleagues.
Interestingly, four training studies compared con-
ventional concentric/eccentric training with concentric/
accentuated-eccentric training (also known as nega-
tive accentuated training). Brandenburg and Docherty
(2002) randomly assigned 23 young males to one of
two training groups: 4 sets of ∼10 concentric/eccentric
repetitions to concentric failure with ∼ 75% of the con-
centric 1 RM, or 3 sets of ∼10 concentric/accentuated-
eccentric repetitions to concentric failure using ∼75%
of the 1 RM for the concentric muscle actions and
110–120% of the concentric 1 RM for the eccentric mus-
cle actions. Both groups followed a 2-second concentric
2-second eccentric protocol for elbow flexion and elbow
extension exercises for 9 weeks. They reported a sig-
nificant increase in elbow flexor 1RM (11% and 15%),
elbow extensor 1RM (9% and 24%), and specific tension
(concentric 1 RM divided by the cross-sectional area)
for the biceps (9% and 9%) and triceps (13% and 22%)
in the traditional concentric/eccentric and concentric/
accentuated-eccentric groups, respectively. There was
no significant difference between groups except for
J Exerc Sci Fit •Vol 6 •No 2 •2008 79
1 RM triceps strength, which was significantly greater in
the accentuated-eccentric group. Muscle cross-sectional
area of the biceps and the triceps as measured with mag-
netic resonance imaging did not significantly increase
in either group. Out of the six reported dependent vari-
ables, only one (triceps 1RM) showed a significant advan-
tage to supramaximal loading. Notably, one third (4 of
12) of the subjects in the accentuated-eccentric (supra-
maximal) group withdrew from the study; two because
of injury and two for non-compliance (Brandenburg &
Barstow and colleagues (2003) randomly assigned
39 previously trained males and females to perform
traditional resistance elbow flexor exercise (60% 1 RM
for concentric and eccentric muscle actions) or negative
accentuated exercise (60% 1 RM concentric and 100%
1 RM eccentric muscle actions) for 3 sets of 6–10 repe-
titions two times a week for 12 weeks. The traditional
group significantly increased strength (concentric 1RM)
by 13.8% and the negative accentuated group increased
15.5%, with no significant difference between groups.
Barstow and colleagues concluded: “The results of this
study suggest that although dynamic training load in-
creased approximately 27% for both training groups,
the C1RM [concentric 1 RM] analysis did not support
the superiority of enhanced-eccentric training for improv-
ing isotonic elbow flexor strength in a group of trained
subjects” (p. 66).
Two other studies reported no significant difference
in strength gains or anthropometric outcomes (Godard
et al. 1998; Ben-Siri et al. 1995), which strongly sug-
gested that there was no additional benefit to training
with supramaximal loads (refer to Carpinelli et al. 2004
for details). These four studies (Barstow et al. 2003;
Brandenburg & Docherty 2002; Godard et al. 1998;
Ben-Siri et al. 1995) are curiously missing from the
review by Crewther and colleagues (2005). Moreover,
Crewther and colleagues did not cite a single study to
support their claim that there are superior outcomes
as a result of training with supramaximal eccentric loads
compared with concentric and eccentric training with
a similar moderate resistance.
Crewther and colleagues (2005) stated: “…near
maximal forces are necessary to induce maximal stre-
ngth through neural mediated adaptation” (p. 984–
5). One of the two references cited (Bloomer & Ives
2000; Schmidtbleicher 1992) is a chapter in a book.
Schmidtbleicher (1992) claimed that compared with
the requirement for the initial rate of force develop-
ment for very light loads and maximal rate of force
development for moderate loads, “maximal strength
predominates” (p. 381) with very heavy loads.
Schmidtbleicher did not cite any reference to support
his opinion, which was reported by Crewther and col-
leagues. The other reference cited by Crewther and
colleagues is a review by Bloomer and Ives (2000) who
described the size principle, but then noted: “…if
muscle fibers are recruited but not fatigued, they are
not trained” (p. 31). One of the two references cited by
Bloomer and Ives is the previously discussed book
by Zatsiorsky (1995). The other reference is a review by
Sale (1987). However, Sale noted: “… adaptations are
induced only in those MUs [motor units] that are active
in the training exercise” (p. 141). Sale specifically re-
ferred to the activation of motor units. He did not claim
that motor units had to be fatigued. These references
(Bloomer & Ives 2000; Zatsiorsky 1995; Schmidtbleicher
1992; Sale 1987) do not support the heavier-is-better
claims by Crewther and colleagues (2005).
In a lengthy review of molecular and cellular adap-
tations to resistance training, Toigo and Boutellier (2006)
expressed conflicting opinions. They noted that the size
principle predicts that motor unit recruitment is deter-
mined by the force requirement and larger motor units
are recruited only when the force requirement is high.
However, their proposed model of motor unit recruit-
ment states that as long as an exercise is performed to
volitional muscular fatigue, different load magnitudes
such as 60%, 75% and 90% 1 RM will result in similar
complete motor unit recruitment. Obviously, their mis-
application of the size principle and their proposed model
of recruitment are antithetical.
In addition to the infiltration of the heavier-is-better
training philosophy in resistance training guidelines,
reviews and books, it has also affected the reporting of
resistance training studies. For example, Hatfield and
colleagues (2006) recently reported the acute responses
in nine resistance-trained males to two different repe-
tition durations (incorrectly noted by Hatfield and col-
leagues as speed of movement and movement velocity)
after performing as many repetitions as possible with
60% and 80% 1 RM for the squat and shoulder press
exercises. The longer repetition duration protocol (10 sec-
onds concentric, 10 seconds eccentric muscle actions)
resulted in significantly fewer repetitions, lower force
levels, power and volume of work for both exercises
compared with the self-selected repetition duration (not
specified). There was no significant difference in the rat-
ing of perceived exertion between protocols or exercises
(15.8 and 16.2 for the shoulder press using 60% 1 RM,
15.0 and 15.4 with 80% 1 RM, 15.4 and 16.6% for the
squat at 60% 1 RM, and 15.9 and 14.0% at 80% 1 RM,
80 J Exerc Sci Fit •Vol 6 •No 2 •2008
longer and self-selected repetition durations, respec-
tively). Their data showed that there was a similar effort
at the termination of each of the eight exercise protocols.
According to the size principle, a similar number of
motor units were activated at the end of each exercise—
regardless of the amount of resistance, number of rep-
etitions, or repetition duration. This important point was
apparently missed by Hatfield and colleagues.
Hatfield and colleagues (2006) stated in their
Discussion section that the longer repetition duration
would not allow one to train with an appropriate amount
of resistance, and that maximum or near maximum
force is required for optimal increases in muscular
strength and hypertrophy. In an attempt to support their
conclusion, they cited four studies (Shoepe et al. 2003;
Siegel et al. 2002; Schlumberger et al. 2001; Sogaard
et al. 1996). Schlumberger and colleagues compared
single-set and multiple-set protocols in females who
used similar amounts of resistance (6–9RM). Shoepe and
colleagues compared functional properties of muscle
fibers in two groups of males (sedentary versus resist-
ance trained). Siegel and colleagues simply reported the
acute response (power output) to exercise with 30%,
40%, 50%, 60%, 70%, 80% and 90% 1 RM. Sogaard and
colleagues reported motor unit recruitment patterns
in six females during concentric and eccentric muscle
actions of the elbow flexors with a 2 kg load. None of
these studies (Shoepe et al. 2003; Siegel et al. 2002;
Schlumberger et al. 2001; Sogaard et al. 1996) claimed or
even suggested that high forces are required for opti-
mal strength gains. Nor did any of them support Hatfield
and colleagues’ (2006) apparently incorrect application
of the size principle.
Hatfield and colleagues (2006) also concluded that
the highest possible force was required to optimize
muscular performance and tissue adaptations. They
cited only one reference to support that claim, which
they noted as reference number 32. However, because
they listed only 31 references, their statement remains
In another study, Shimano and colleagues (2006)
concluded that performing as many repetitions as pos-
sible in the free weight squat, bench press and arm
curl exercises using 60%, 80% and 90% 1 RM resulted
in similar levels of perceived exertion in previously un-
trained males and resistance trained males. The practical
application of these results and those of the aforemen-
tioned study by Hatfield and colleagues (2006) is that
the perceived effort and therefore the recruitment of
available motor units is similar when a set of repe-
titions is continued to the point of maximal effort
(RM)—regardless of the number of repetitions, repetition
duration, or amount of resistance. However, Shimano
and colleagues apparently did not recognize the impli-
cation of their results and claimed that a resistance
load of at least 90% 1 RM should be used for optimal
strength gains in previously untrained and resistance
trained individuals. They cited no evidence to support
that claim or their misapplication of the size principle.
Minimal Support for Heavier-is-Better
Although not cited by any of the aforementioned authors
who endorsed the heavier-is-better training philosophy,
there is one study by Campos and colleagues (2002) who
reported some advantage to a lower repetition protocol.
They randomly assigned 32 previously untrained males
(age, ∼22 years) to one of three training groups (4 sets
of 3–5 RM, 3 sets of 9–11 RM, or 2 sets of 20–28 RM)
or a control group. The three training groups performed
the leg press, squat and knee extension exercises 2 days
a week for the first 4 weeks and three times a week for
the next 4 weeks. Although the authors stated that they
selected three practical ranges of repetitions, 20–28 RM
is not commonly employed in most resistance training
protocols. Therefore, comparisons are justified only be-
tween the 3–5 RM and 9–11 RM groups. The strength
gain (1 RM) for the squat (∼61%) and leg press (∼100%)
was significantly greater in the 3–5 RM group compared
with the 9–11 RM group (∼31% and ∼80%, squat and
leg press, respectively). There was no significant differ-
ence between the groups for the gains in knee exten-
sion strength (∼67% and ∼56%, 3–5 RM and 9–11 RM
groups, respectively). Exercise-induced fiber transforma-
tion (from Type IIB to IIA) was similar for both groups,
and muscle hypertrophy of the three fiber types (Types
I, IIA and IIB) significantly increased in both groups, with
no significant difference between groups. The authors
did not speculate on why the heavier resistance (3–5RM)
produced more favorable strength gains in the leg press
and squat exercises and not in the knee extension
exercise, and their only comment regarding the simi-
lar hypertrophic responses in the 3–5RM and 9–11 RM
groups was that it was interesting. Campos and coll-
eagues (2002) concluded: “It has often been accepted
that improved strength/power results from high inten-
sity/low volume training, whereas low intensity/high
volume training maximizes muscle hypertrophy. Based
on data from the present investigation, this may not be
entirely true. Indeed, data from the present investiga-
tion suggest low and intermediate RM training induces
J Exerc Sci Fit •Vol 6 •No 2 •2008 81
similar muscular adaptations, at least after short-term
training in previously untrained subjects” (p. 58).
Authors who have asserted that very heavy or near-
maximal resistance is required for optimal strength
gains did not cite any training studies that actually sup-
port their apparent incorrect application of the size
principle (Table 1).
Studies That Do Not Support Heavier-is-Better
Studies that reported the effects of training with differ-
ent amounts of resistance (Harris et al. 2004; Vincent
et al. 2002; Hortobagyi et al. 2001; Bemben et al.
2000; Chestnut & Docherty 1999; Faigenbaum et al.
1999; Graves et al. 1999; Masuda et al. 1999; Weiss
et al. 1999; Hisaeda et al. 1996; Kerr et al. 1996; Taaffe
et al. 1996; Pruitt et al. 1995; Stone & Coulter 1994;
Schmidtbleicher & Haralambie 1981; Withers 1970;
O’Shea 1966; Berger 1963, 1962a, 1962b) strongly sup-
port the previously discussed acute neurophysiological
responses reported by Behm and colleagues (2002).
For example, there are studies that compared
2 RM, 6RM and 10 RM free-weight bench press (Berger
1963), 2 RM, 10RM and 12 RM bench press (Berger
1962a), 2 RM and 10RM bench press (Berger 1962b),
2–3 RM, 5–6 RM and 9–10 RM free-weight squats
(O’Shea 1966), 3–5 RM and 13–15 RM barbell squats
(Weiss 1999), 4 RM and 10 RM for seven upper-body
exercises (Bemben et al. 2000), 7–10 RM and 15–20 RM
knee extension exercise (Graves et al. 1999), 4–6 RM
and 15–20 RM knee extension (Hisaeda et al. 1996),
3 RM, 5 RM and 7 RM for three upper- and lower-body
free-weight exercises (Withers 1970), and 6 RM, 9 RM
and 15 RM upper-body and lower-body free-weight and
machine exercises (Harris et al. 2004). They all reported
no significant difference in the strength gains among
the various groups using different amounts of resistance
(RMs) for training (Table 2).
The study by Graves and colleagues (1999) is note-
worthy because the 10 pairs of identical twins were
the quintessentially matched groups. After training two
times a week for 10 weeks, increases in isometric strength
(averaged for the eight positions tested) were signifi-
cant in both groups. However, there was no significant
difference in the strength gains as a result of training
with 7–10RM (13.2%) or 15–20 RM (12.8%).
The misapplication of the size principle is an error in
reasoning. Recommendations to train with very heavy
resistance (loads heavier than 6RM), because they
purportedly result in superior strength gains, are based
on a faulty premise (an invalid reverse inference of the
82 J Exerc Sci Fit •Vol 6 •No 2 •2008
Table 1. Summary of authors who recommended heavier-is-
better resistance training based on an incorrect application
of the size principle
Brown et al. 2007 Kraemer & Gomez 2001
Kraemer & Vingren 2007 Kraemer & Newton 2000
Kraemer et al. 2007 Kraemer & Bush 1998
Hatfield et al. 2006 Kraemer et al. 1998
Kraemer & Fragala 2006 Fleck & Kraemer 1997
Shimano et al. 2006 Fleck & Kraemer 1996
Toigo & Boutellier 2006 Zatsiorsky 1995
Bird et al. 2005 Stone 1993
Crewther et al. 2005 Schmidtbleicher 1992
Fry 2004 Fleck & Kraemer 1988
Kraemer & Ratamess 2004 Kraemer et al. 1988
Peterson et al. 2004 Fleck & Kraemer 1987
Kraemer 2003 Garhammer 1987
Rhea et al. 2003 Stone & O’Bryant 1984
Fleck 2002 Kraemer 1983
Hasegawa et al. 2002 Palmieri 1983
Hoffman 2002 Berger 1982
Kraemer 2002 Stone 1982
Haff & Potteiger 2001 Atha 1981
Table 2. Summary of studies that reported resistance training
protocols using different amounts of resistance and showed no
significant difference in strength gains
Reference Training protocol
Harris et al. 2004 6 RM, 9 RM and 15 RM
Vincent et al. 2002 50%1 RM and 80% 1 RM
Hortobagyi et al. 2001 40%1 RM and 80% 1 RM
Bemben et al. 2000 40% 1 RM and 80% 1 RM
Chestnut & Docherty 1999 4 RM and 10 RM
Faigenbaum et al. 1999 6–8 RM and 13–15 RM
Graves et al. 1999 7–10RM and 15–20 RM
Masuda et al. 1999 40–80% 1 RM and 90% 1 RM
Weiss et al. 1999 3–5 RM and 13–15 RM
Hisaeda et al. 1996 4–6 RM and 15–20 RM
Kerr et al. 1996 8–10RM and 20–25 RM
Taaffe et al. 1996 40% 1 RM and 80% 1 RM
Pruitt et al. 1995 40% 1 RM and 80% 1 RM
Stone & Coulter 1994 6–8 RM, 15–20 RM and
Schmidtbleicher & 30% 1 RM and 90–100% 1 RM
Withers 1970 3 RM, 5 RM and 7RM
O’Shea 1966 2–3RM, 5–6 RM and 9–10 RM
Berger 1963 2RM, 6 RM and 10 RM
Berger 1962a 2 RM, 10 RM and 12 RM
Berger 1962b 2 RM and 10 RM
size principle) and have very little supporting evidence.
The question is whether the incorrect application was
unintentionally created and perpetuated because of a
misunderstanding, or whether it was intentionally cre-
ated and perpetrated in order to support a preconceived
opinion. Whatever the reason, the grossly distorted
heavier-is-better training philosophy has deeply infil-
trated the resistance training literature.
The correct interpretation of the size principle and its
practical application should help dedicated trainees
understand what constitutes a proper stimulus for resist-
ance training and how to apply that stimulus. That is,
the size principle does not support the popular resistance
training recommendation to use a maximal or near
maximal resistance. The size principle and interpolated
twitch studies support the contention that if maximal
motor unit activation is desired, a maximal or near
maximal effort at the end of a set of repetitions—
regardless of the amount of external resistance—will
elicit maximal motor unit activity. Effective resistance
training does not require the use of a maximal or near
maximal force to stimulate the available motor units
and produce significant increases in muscular strength.
The preponderance of studies strongly suggest that
effective resistance training simply requires the selec-
tion of a desired range of repetitions (e.g., 3–5, 6–9,
10–12RM), which is based on a personal preference
rather than a specific goal, and a progression of the
resistance to stay within the desired range of repetitions
(Carpinelli et al. 2004). Very high RMs (e.g., loads lighter
than 20 RM) or an extensive time under load (e.g., longer
than 2–3 minutes) may involve mechanisms of fatigue
that are not conducive to stimulate optimal increases
in muscular strength. Despite the plethora of opinions
in the resistance training literature, the specific mecha-
nisms of fatigue and exactly what constitutes an optimal
stimulus for strength gains are unknown. If a maximal—
or near maximal—effort is applied at the end of a set of
repetitions, the evidence strongly suggests that the dif-
ferent external forces produced with different amounts
of resistance elicit similar outcomes.
If the heavier-is-better training philosophy is truly
valid, as claimed by the majority of resistance training
experts, the training studies would overwhelmingly
support that concept. In fact, the preponderance of resist-
ance studies reported no significant difference in strength
gains as a result of training with heavier loads (Figure).
If the size principle was correctly applied, effective
resistance training may appeal to a larger proportion
of the population. This would include competitive and
recreational athletes as well as those in the general
population who perceive resistance exercise as the lift-
ing of very heavy weights and therefore potentially
dangerous. Because some people may have a fear of
injury—that need not exist—the heavier-is-better percep-
tion may actually be a deterrent to resistance training,
which deprives those most in need of health-related
benefits. These potential health-related benefits include
the prevention of osteoporosis, falls, fractures and dis-
ability, changes in risk factors associated with cardio-
vascular disease, some cancers, diabetes (improvements
in glucose tolerance and insulin resistance), enhanced
lipid profiles, elevated resting metabolic rate, decreased
resting blood pressure, reduced back pain and subse-
quent disabilities, and greater functional ability (Winett
& Carpinelli 2001)—all in addition to muscular hypertro-
phy and strength gains.
Special thanks to Sandee Jungblut, M.S., Arty Conliffe,
B.A., Richard Winett, Ph.D., and Robert Otto, Ph.D. for
their critical conceptual and editorial feedback on the
preparation of this manuscript—and to Rocco Carpinelli
for his inspiration.
J Exerc Sci Fit •Vol 6 •No 2 •2008 83
Fig. Twenty resistance training studies reported no significant
difference in strength gains compared with only one study
that reported a significant difference as a result of training with
a heavier resistance.
Adam A, DeLuca CJ (2003). Recruitment order of motor units in human
vastus lateralis muscle is maintained during fatiguing contractions.
J Neurophysiol 90:2919–27.
Akaboshi K, Masakado Y, Chino N (2000). Quantitative EMG and motor
unit recruitment threshold using a concentric needle with quadrifilar
electrode. Muscle Nerve 23:361–7.
Allen GM, Gandevia SC, McKenzie DK (1995). Reliability of measure-
ments of muscle strength and voluntary activation using twitch
interpolation. Muscle Nerve 18:593–600.
Allen GM, McKenzie DK, Gandevia SC (1998). Twitch interpolation of the
elbow flexor muscles at high forces. Muscle Nerve i21:318–28.
American College of Sports Medicine (2002). Kraemer WJ, Writing Group
Chairman. Position Stand: Progression models in resistance training
for healthy adults. Med Sci Sports Exerc 34:364–80.
Anderson T, Kearney JT (1982). Effects of three resistance training pro-
grams on muscular strength and absolute and relative endurance.
Res Q Exerc Sport 53:1–7.
Atha J (1981). Strengthening muscle. Exerc Sports Sci Rev 9:1–73.
Barstow IK, Bishop MD, Kaminski TW (2003). Is enhanced-eccentric
resistance training superior to traditional training for increasing
elbow flexor strength? J Sports Sci Med 2:62–9.
Bawa P, Binder MD, Ruenzel P, Henneman E (1984). Recruitment order
of motoneurons in stretch reflexes is highly correlated with their
axonal conduction velocity. J Neurophysiol 52:410–20.
Behm DG (1995). Neuromuscular implications and applications of resist-
ance training. J Strength Cond Res 9:264–74.
Behm DG, Reardon G, Fitzgerald J, Drinkwater E (2002). The effect of 5,
10, and 20 repetition maximums on the recovery of voluntary and
evoked contractile properties. J Strength Cond Res 16:209–18.
Behm DG, St-Pierre DMM, Perez D (1996). Muscle inactivation: assess-
ment of interpolated twitch technique. J Appl Physiol 81:2267–73.
Bemben DA, Fetters NL, Bemben MG, Nabavi N, Koh ET (2000). Muscu-
loskeletal responses to high- and low-intensity resistance training in
early postmenopausal women. Med Sci Sports Exerc 32:1949–57.
Ben-Siri DA, Ayalon A, Tavi M (1995). The effect of different types of
strength training on concentric strength in women. J Strength Cond Res
Berger RA (1962a). Optimum repetitions for the development of strength.
Res Q 33:334–8.
Berger RA (1962b). Effect of varied weight training programs on strength.
Res Q 33:168–81.
Berger RA (1963). Comparative effects of three weight training programs.
Res Q 34:396–8.
Berger RA (1982). Applied Exercise Physiology. Lea & Febiger,
Bird SP, Tarpenning KM, Marino FE (2005). Designing resistance training
programmes to enhance muscular fitness. A review of the acute pro-
gramme variables. Sports Med 35:841–51.
Bloomer RJ, Ives JC (2000). Varying neural and hypertrophic influences
in a strength program. Strength Cond J 22:30–5.
Brandenburg JP, Docherty D (2002). The effects of accentuated eccentric
loading on strength, muscle hypertrophy, and neural adaptations in
trained individuals. J Strength Cond Res 16:25–32.
Brown AB, McCartney N, Sale DG (1990). Positive adaptations to weight-
lifting training in the elderly. J Appl Physiol 69:1725–33.
Brown LE, Findley BW, Murray DP, Bera SG (2007). Workout schedule and
rest. In: Brown LE (ed). Strength Training. Human Kinetics, Chicago, IL,
Campos GER, Luecke TJ, Wendeln HK, Toma K, Hagerman FC, Murray
TF, Ragg KE, Ratamess NA, Kraemer WJ, Staron RS (2002). Muscular
adaptations in response to three different resistance-training regi-
mens: specificity of repetition maximum training zones. Eur J Appl
Carpinelli RN, Otto RM, Winett RA (2004). A critical analysis of the ACSM
position stand on resistance training: insufficient evidence to sup-
port recommended training protocols. JEPonline 7:1–64.
Chestnut JL, Docherty D (1999). The effects of 4 and 10 repetition maxi-
mum weight-training protocols on neuromuscular adaptations in
untrained men. J Strength Cond Res 13:353–9.
Clamann PH, Gillies JD, Skinner RD, Henneman E (1974). Quantitative
measures of output of a motoneuron pool during monosynaptic
reflexes. J Neurophysiol 37:1328–37.
Clamann PH, Henneman E (1976). Electrical measurement of axon
diameter and its use in relating motoneuron size to critical firing
level. J Neurophysiol 39:844–51.
Clarke DH (1973). Adaptations in strength and muscular endurance
resulting from exercise. Exerc Sports Sci Rev 1:73–102.
Connelly DM, Rice CL, Roos MR, Vandervoort AA (1999). Motor unit fir-
ing rates and contractile properties in tibialis anterior of young and
old men. J Appl Physiol 87:843–52.
Cope TC, Pinter MJ (1995). The size principle: still working after all these
years. Physiology 10:280–6.
Cope TC, Sokoloff AJ (1999). Orderly recruitment among motoneurons
supplying different muscles. J Physiol (Paris) 93:81–5.
Cope TC, Sokoloff AJ, Dacko SM, Huot R, Feingold E (1997). Stability of
motor-unit force thresholds in the decerebrate cat. J Neurophysiol
Crewther B, Cronin J, Keogh J (2005). Possible stimuli for strength and power
adaptation. Acute mechanical responses. Sports Med 35:967–89.
DeLuca CJ, Foley PJ, Erim Z (1996). Motor unit control properties in
constant-force isometric contractions. J Neurophysiol 76:1503–16.
DeLuca CJ, LeFever RS, McCue MP, Xenakis AP (1982). Control scheme
governing concurrently active human motor units during voluntary
contractions. J Physiol 329:129–42.
Denny-Brown D, Pennybacker JB (1938). Fibrillation and fasciculation in
voluntary muscle. Brain 61:311–34.
De Serres SJ, Enoka RM (1998). Older adults can maximally activate the
biceps brachii muscle by voluntary command. J Appl Physiol 84:
Desmedt JE, Godaux E (1977). Fast motor units are not preferentially
activated in rapid voluntary contractions in man. Nature 267:717–9.
Desmedt JE, Godaux E (1979). Voluntary motor commands in human
ballistic movements. Ann Neurol 5:415–21.
Desmedt JE, Godaux E (1981). Spinal motoneuron recruitment in man:
rank deordering with direction but not with speed of voluntary
movement. Science 214:933–5.
Edstrom L, Grimby L (1986). Effect of exercise on the motor unit. Muscle
Enoka RM, Fuglevand AJ (2001). Motor unit physiology: some unresolved
issues. Muscle Nerve 24:4–17.
Erim Z, DeLuca CJ, Mineo K, Aoki T (1996). Rank-ordered regulation of
motor units. Muscle Nerve 19:563–73.
Faigenbaum AD, Westcott WL, Loud RL, Long C (1999). The effects of dif-
ferent resistance training protocols on muscular strength and
endurance development in children. Pediatrics 104 e5:1–7.
Feiereisen P, Duchateau J, Hainaut K (1997). Motor unit recruitment
order during voluntary and electrically induced contractions in the
tibialis anterior. Exp Brain Res 114:117–23.
Fleck SJ (2002). Periodization of training. In: Kraemer WJ, Hakkinen K
(eds). Strength Training for Sport. Blackwell Science, Oxford, 55–68.
Fleck SJ, Kraemer WJ (1987). Designing Resistance Training Programs.
Human Kinetics, Champaign, IL.
Fleck SJ, Kraemer WJ (1988). Resistance training: basic principles (part 1
of 4). Physician Sportsmed 16:160–71.
Fleck SJ, Kraemer WJ (1996). Periodization Breakthrough! Advanced
Research Press, New York.
Fleck SJ, Kraemer WJ (1997). Designing Resistance Training Programs
(2nd ed). Human Kinetics, Champaign, IL.
Folland JP, Williams AG (2006). Methodological issues with the interpo-
lated twitch technique. J Electromyography Kinesiol 17:317–27.
Freund HJ, Budingen HJ, Dietz V (1975). Activity of single motor units
from human forearm muscles during voluntary isometric contrac-
tions. J Neurophysiol 38:933–46.
Fry AC (2004). The role of resistance exercise intensity on muscle fibre
adaptations. Sports Med 34:663–79.
84 J Exerc Sci Fit •Vol 6 •No 2 •2008
Fuglevand AJ, Winter DA, Patla AE (1993). Models of recruitment and rate
coding organization in motor-unit pools. J Neurophysiol 70:2470–88.
Gandevia SC, Herbert RD, Leeper JB (1998). Voluntary activation of human
elbow flexor muscles during maximal concentric contractions. J Physiol
Garhammer J (1987). Sports Illustrated Strength Training. Time Inc., New York.
Godard MP, Wygand JW, Carpinelli RN, Catalano S, Otto RM (1998).
Effects of accentuated eccentric resistance training on concentric
knee extensor strength. J Strength Cond Res 12:26–9.
Goldberg LJ, Derfler B (1977). Relationship among recruitment order,
spike amplitude, and twitch tension of single motor units in human
masseter muscle. J Neurophysiol 40:879–90.
Graves JE, Pollock ML, Jones AE, Jones WE, Colvin A (1999). Number of
repetitions does not influence the initial response to resistance train-
ing in identical twins [abstract]. Med Sci Sports Exerc 26 Suppl 5:S74.
Green H, Goreham C, Ouyang J, Ball-Burnett M, Ranney D (1998). Regula-
tion of fiber size, oxidative potential, and capillarization in human
muscle by resistance exercise. Am J Physiol 276:R591–6.
Guyton AC (1991). Basic Neuroscience. Anatomy and Physiology (2nd ed).
WB Saunders, Philadelphia, PA.
Haff GG, Potteiger JA (2001). A brief review: explosive exercises and
sports performance. Strength Cond J 23:13–20.
Hakkinen K, Pakarinen A (1993). Acute hormonal responses to two dif-
ferent fatiguing heavy-resistance protocols in male athletes. J Appl
Hakkinen K, Pakarinen A, Kyrolainen H, Cheng S, Kim DH, Komi PV
(1990). Neuromuscular adaptations and serum hormones in females
during prolonged power training. Int J Sports Med 11:91–8.
Harris C, DeBeliso MA, Spitzer-Gibson TA, Adams KJ (2004). The effect of
resistance-training intensity on strength-gain response in the older
adult. J Strength Cond Res 18:833–8.
Hasegawa H, Dziados J, Newton RU, Fry AC, Kraemer WJ, Hakkinen K
(2002). Periodized training programmes for athletes. In: Kraemer
WJ, Hakkinen K (eds). Strength Training for Sport. Blackwell Science,
Hatfield DL, Kraemer WJ, Spiering BA, Hakkinen K, Volek JS, Shimano T,
Spreuwenberg LPB, Silvestre R, Vingren JL, Fragala MS, Gomez AL,
Fleck SJ, Newton RU, Maresh CM (2006). The impact of velocity of
movement on performance factors in resistance exercise. J Strength
Cond Res 20:760–6.
Henneman E (1957). Relation between size of neurons and their suscep-
tibility to discharge. Science 126:1345–7.
Henneman E (1968). Peripheral mechanisms involved in the control of
muscle. In: Mountcastle VB (ed). Medical Physiology (12th ed). CV
Mosby, St. Louis, MO.
Henneman E, Clamann PH, Gillies JD, Skinner RD (1974). Rank order of
motoneurons within a pool: law of combination. J Neurophysiol
Henneman E, Olson CB (1965). Relations between structure and func-
tion in the design of skeletal muscles. J Neurophysiol 28:581–98.
Henneman E, Somjen G, Carpenter DO (1965a). Functional significance
of cell size in spinal motoneurons. J Neurophysiol 28:560–80.
Henneman E, Somjen G, Carpenter DO (1965b). Excitability and inhibil-
ity of motoneurons of different sizes. J Neurophyiol 28:599–620.
Herbert RD, Gandevia SC (1999). Twitch interpolation in human mus-
cles: mechanisms and implications for measurement of voluntary
activation. J Neurophysiol 82:2271–83.
Hisaeda H, Miyagawa K, Kuno S, Fukunaga T, Muraoka I (1996).
Influence of two different modes of resistance training in female
subjects. Ergonomics 39:842–52.
Hoeger WWK, Barette SL, Hale DF, Hopkins DR (1987). Relationship
between repetitions and selected percentages of one repetition max-
imum. J Appl Sport Sci Res 1:11–3.
Hoeger WWK, Hopkins DR, Barette SL, Hale DF (1990). Relationship
between repetitions and selected percentages of one repetition max-
imum: a comparison between untrained and trained males and
females. J Appl Sport Sci Res 4:47–54.
Hoffman J (2002). Physiological Aspects of Sport Training and Performance.
Human Kinetics, Champaign IL.
Hortobagyi T, Tunnel D, Moody J, Beam S, DeVita P (2001). Low- or high-
intensity strength training partially restores impaired quadriceps
force accuracy and steadiness in aged adults. J Gerontol Biol Sci
Houtman CJ, Stegeman DF, Van Dijk JP, Zwarts MJ (2003). Changes in
muscle fiber conduction velocity indicate recruitment of distinct
motor unit populations. J Appl Physiol 95:1045–54.
Hurley MV, Rees J, Newham DJ (1998). Quadriceps function, propriocep-
tive acuity and functional performance in healthy young, middle-
aged and elderly subjects. Age Aging 27:55–62.
Ivanova T, Garland SJ, Miller KJ (1997). Motor unit recruitment and dis-
charge behavior in movements and isometric contractions. Muscle
Jabre JF, Spellman NT (1996). The demonstration of the size principle in
humans using macro electromyography and precision decomposi-
tion. Muscle Nerve 19:338–41.
Jakobi JM, Rice CL (2002). Voluntary muscle activation varies with age
and muscle group. J Appl Physiol 93:457–62.
Jamurtas AZ, Fatouros IG, Buckenmeyer P, Kokkinidis E, Taxildaris K,
Kambas A, Kyriazis G (2000). Effects of plyometric exercise on mus-
cle soreness and plasma creatine kinase levels and its comparison
with eccentric and concentric exercise. J Strength Cond Res 14:68–74.
Kent-Braun JA, Le Blanc R (1996). Quantitation of central activation fail-
ure during maximal voluntary contractions in humans. Muscle Nerve
Kent-Braun JA, Ng AV (1999). Specific strength and voluntary muscle acti-
vation in young and elderly women and men. J Appl Physiol 87:22–9.
Kerr D, Morton A, Dick I, Prince R (1996). Exercise effects on bone mass
in postmenopausal women are site-specific and load-dependent.
J Bone Min Res 11:218–25.
Klass M, Baudry S, Duchateau J (2005). Aging does not affect voluntary
activation of the ankle dorsiflexors during isometric, concentric, and
eccentric contractions. J Appl Physiol 99:31–8.
Klein CS, Rice CL, Marsh GD (2001). Normalized force, activation, and
coactivation in the arm muscles of young and old men. J Appl
Knaflitz M, Merletti R, De Luca CJ (1990). Inference of motor unit recruit-
ment order in voluntary and electrically elicited contractions. J Appl
Knight CA, Kamen G (2001). Adaptations in muscular activation of the
knee extensor muscles with strength training in young and older
adults. J Electromyography Kinesiol 11:405–12.
Kossev A, Christova P (1998). Discharge pattern of human motor units
during dynamic concentric and eccentric contractions. Electroen-
cephalography Clin Neurophysiol 109:245–55.
Kraemer WJ (1983). Exercise prescription in weight training: manipulat-
ing program variables. NSCA J 5:58–9.
Kraemer WJ (2002). Developing a strength training workout. In: Kraemer
WJ, Hakkinen K (eds). Strength Training for Sport. Blackwell Science,
Kraemer WJ (2003). Strength training basics. Physician Sportsmed 31:39–43.
Kraemer WJ, Bush JA (1998). Factors affecting the acute neuromuscular
responses to resistance exercise. In: LaFontaine T (senior ed).
ACSM’s Resource Manual for Guidelines for Exercise Testing and
Prescription (3rd ed). Williams & Wilkins, Baltimore, MD, 164–73.
Kraemer WJ, Duncan ND, Volek JS (1998). Resistance training and elite
athletes: adaptations and program considerations. J Orthop Sports
Phys Ther 28:110–9.
Kraemer WJ, Fleck SJ, Deschenes M (1988). A review: factors in exercise
prescription of resistance training. NSCA J 10:36–41.
Kraemer WJ, Fragala MS (2006). Personalize it: program design in resist-
ance training. ACSM Health Fitness J 10:7–17.
Kraemer WJ, Gomez AL (2001). Establishing a solid fitness base. In:
Foran B (ed). High-performance Sports Conditioning. Human Kinetics,
Champaign, IL, 3–17.
Kraemer WJ, Hatfield DL, Fleck SJ (2007). Types of muscle training. In:
Brown LE (ed). Strength Training. Human Kinetics, Chicago, IL, 45–72.
Kraemer WJ, Newton RU (2000). Training for muscular power. Phys Med
Rehabil Clin N Am 11:341–68.
J Exerc Sci Fit •Vol 6 •No 2 •2008 85
Kraemer WJ, Ratamess NA (2004). Fundamentals of resistance training:
progression and exercise prescription. Med Sci Sports Exerc 36:674–88.
Kraemer WJ, Vingren JL (2007). Muscle anatomy 101. In: Brown LE (ed).
Strength Training. Human Kinetics, Chicago, IL, 3–28.
Kukulka CG, Clamann PH (1981). Comparison of the recruitment and
discharge properties of motor units in human brachial biceps and
adductor pollicis during isometric contractions. Brain Res 219:45–55.
Linnamo V, Moritani T, Nicol C, Komi PV (2003). Motor unit activation
patterns during isometric, concentric and eccentric actions at differ-
ent force levels. J Electromyography Kinesiol 13:93–101.
Masakado Y, Akaboshi K, Nagata M, Kimura A, Chino N (1995). Motor
unit firing behavior in slow and fast contractions of the first dorsal
interosseous muscle of healthy men. Electroencephalography Clin
Masuda K, Choi JY, Shimojo H, Katsuta S (1999). Maintenance of myoglo-
bin concentration in human skeletal muscle after heavy resistance
training. Eur J Appl Physiol 79:347–52.
Maton B (1980). Fast and slow motor units: their recruitment for tonic
and phasic contraction in normal man. Eur J Appl Physiol 43:45–55.
Milner-Brown HS, Stein RB, Yemm R (1973). The orderly recruitment of
human motor units during voluntary isometric contractions. J Physiol
Moritani T, Muro M (1987). Motor unit activity and surface electromyo-
gram power spectrum during increasing force of contraction. Eur J
Appl Physiol 56:260–5.
Murphy AJ, Wilson GJ, Pryor JF (1994). Use of iso-inertial force mass rela-
tionship in the prediction of dynamic human performance. Eur J
Appl Physiol 69:250–7.
Nardone A, Romano C, Schieppati M (1989). Selective recruitment of
high-threshold human motor units during voluntary isotonic length-
ening of active muscles. J Physiol 409:451–71.
Nosaka K, Newton M (2002). Difference in the magnitude of muscle
damage between maximal and submaximal eccentric loading.
J Strength Cond Res 16:202–8.
O’Shea P (1966). Effects of selected weight training programs on the
development of strength and muscle hypertrophy. Res Q 37:95–102.
Oskouei MAE, van Mazijk BCF, Schuiling MHC, Herzog W (2003).
Variability in the interpolated twitch torque for maximal and sub-
maximal voluntary contractions. J Appl Physiol 95:1648–55.
Otto RM, Carpinelli RN (2006). A critical analysis of the single versus
multiple set debate. JEPonline 9:32–48.
Palmieri GA (1983). The principles of muscle fiber recruitment applied to
strength training. NSCA J 5:22–4, 63.
Peterson MD, Rhea MR, Alvar BA (2004). Maximizing strength develop-
ment in athletes: a meta-analysis to determine the dose-response
relationship. J Strength Cond Res 18:377–82.
Pruitt LA, Taaffe DR, Marcus R (1995). Effects of a one-year high-intensity
versus low-intensity resistance training program on bone mineral
density in older women. J Bone Min Res 10:1788–95.
Rhea MR, Alvar BA, Burkett LN, Ball SD (2003). A meta-analysis to deter-
mine the dose response for strength development. Med Sci Sports
Riek S, Bawa P (1992). Recruitment of motor units in human forearm
extensors. J Neurophysiol 68:100–8.
Roos MR, Rice CL, Connelly DM, Vandervoort AA (1999). Quadriceps
muscle strength, contractile properties, and motor unit firing rates in
young and old men. Muscle Nerve 22:1094–1103.
Sale DG (1987). Influence of exercise and training on motor unit activa-
tion. Exerc Sport Sci Rev 15:95–151.
Schlumberger A, Stec J, Schmidtbleicher D (2001). Single- vs. multiple-
set strength training in women. J Strength Cond Res 15:284–9.
Schmidtbleicher D (1992). Training for power events. In: Komi PV (ed).
Strength and Power in Sport. Blackwell Scientific Publications, Boston,
Schmidtbleicher D, Haralambie G (1981). Changes in contractile properties
of muscle after strength training in man. Eur J Appl Physiol 46:221–8.
Schmied A, Morin D, Vedel JP, Pagni S (1997). The “size principle” and
synaptic effectiveness of muscle afferent projections to human exten-
sor carpi radialis motoneurons during wrist extension. Exp Brain Res
Scutter SD, Turker KS (1998). Recruitment stability in masseter motor
units during isometric voluntary contractions. Muscle Nerve 21:1290–8.
Seki K, Narusawa M (1996). Firing rate modulation of human motor
units in different muscles during isometric contraction with various
forces. Brain Res 719:1–7.
Senn W, Wyler K, Clamann HP, Kleinle J, Luscher HR, Muller L (1997).
Size principle and information theory. Biol Cybern 76:11–22.
Shield A, Zhou S (2004). Assessing voluntary muscle activation with the
twitch interpolation technique. Sports Med 34:253–67.
Shimano T, Kraemer WJ, Spiering BA, Volek JS, Hatfield DL, Silvestre R,
Vingren JL, Fragala MS, Maresh CM, Fleck SJ, Newton RU,
Spreuwenberg LPB, Hakkinen K (2006). Relationship between the
number of repetitions and selected percentages of one repetition
maximum in free weight exercises in trained and untrained men.
J Strength Cond Res 20:819–23.
Shoepe TC, Stelzer JE, Garner DP, Widrick JJ (2003). Functional adaptabil-
ity of muscle fibers to long-term resistance exercise. Med Sci Sports
Siegel JA, Gilders RM, Staron RS, Hagerman FC (2002). Human muscle
power output during upper- and lower-body exercises. J Strength
Cond Res 16:173–8.
Sogaard K, Christensen H, Jensen BR, Finsen L, Sjogaard G (1996). Motor
control and kinetics during low level concentric and eccentric contrac-
tions in man. Electroencephalography Clin Neurophysiol 101:453–60.
Stone MH (1982). Considerations in gaining a strength-power training
effect (machines vs. free weights). NSCA J 9:22–4, 54.
Stone MH (1993). Literature review: explosive exercises and training.
NSCA J 15:7–15.
Stone MH, O’Bryant HS (1984). Weight Training: A Scientific Approach.
Burgess Publishing, Minneapolis MN.
Stone WJ, Coulter SP (1994). Strength/endurance effects from three resist-
ance training protocols with women. J Strength Cond Res 8:231–4.
Suter E, Herzog W (2001). Effect of number of stimuli and timing of
twitch application on variability in interpolated twitch. J Appl Physiol
Suter E, Herzog W, Huber A (1996). Extent of motor unit activation in the
quadriceps muscles of healthy subjects. Muscle Nerve 19:1046–8.
Taaffe DR, Pruitt L, Pyka G, Guido D, Marcus R (1996). Comparative
effects of high- and low-intensity resistance training on thigh muscle
strength, fiber area, and tissue composition in elderly women. Clin
Thomas CK, Ross BH, Calancie B (1987). Human motor-unit recruitment
during isometric contractions and repeated dynamic movements.
J Neurophysiol 57:311–24.
Thomas CK, Ross BH, Stein RB (1986). Motor-unit recruitment in human
first dorsal interosseous muscle for static contractions in three differ-
ent directions. J Neurophysiol 55:1017–29.
Thorstensson A, Hulten B, Von Dobelin W, Karlsson J (1976). Effect of
strength training on enzyme activities and fibre characteristics in
human skeletal muscle. Acta Physiol Scand 96:392–8.
Toigo M, Boutellier U (2006). New fundamental resistance exercise deter-
minants of molecular and cellular muscle adaptations. Eur J Appl
Vandervoort AA, McComas AJ (1986). Contractile changes in opposing
muscles of the human ankle joint with aging. J Appl Physiol 61:361–7.
Vincent KR, Braith RW, Feldman RA, Magyari PM, Culter RB, Persin SA,
Lennon SL, Gabr AH, Lowenthal DT (2002). Resistance exercise and
physical performance in adults aged 60–83. J Am Geriatr Soc 50:1100–7.
Wakeling JM, Kaya M, Temple GK, Johnston IA, Herzog W (2002).
Determining patterns of motor recruitment during locomotion. J Exp
Weiss LW, Coney HD, Clark FC (1999). Differential functional adaptations
to short-term low-, moderate-, and high-repetition weight training.
J Strength Cond Res 13:236–41.
Winett RA, Carpinelli RN (2001). Potential health-related benefits of
resistance training. Prev Med 33:503–13.
Withers RT (1970). Effect of varied weight-training loads on the strength
of university freshmen. Res Q 41:110–4.
Zatsiorsky VM (1995). Science and Practice of Strength Training. Human
Kinetics, Champaign, IL.
86 J Exerc Sci Fit •Vol 6 •No 2 •2008