Pattern of Pulses That Maximize Force Output From Single Human
Thenar Motor Units
C. K. THOMAS,1 R. S. JOHANSSON,2 AND B. BIGLAND-RITCHIE3
1
The Miami Project to Cure Paralysis, Departments of Neurological Surgery and Physiology and Biophysics, University of
Miami School of Medicine, Miami, Florida 33136; 2Department of Physiology, University of Umeå, SE-90187 Umeå,
Sweden; and 3Department of Pediatrics, Yale University Medical School, New Haven, Connecticut 06520
INTRODUCTION
When a human motor unit is recruited in a voluntary
contraction it often fires with two closely spaced impulses,
termed a “doublet”, followed by longer interpulse intervals
that maintain the contraction (Bawa and Calancie 1983;
Desmedt and Godaux 1977; Kudina and Alexeeva 1992).
Impulse patterns of one or two short interpulse intervals
followed by longer intervals maximize the force-time integral from triceps surae motor units of the cat (Burke et al.
1976; Parmiggiani and Stein 1981; Zajac and Young
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
3188
1980a,b). Likewise, whole mammalian muscles show similar response behaviors when their peripheral nerves are
stimulated (Binder-Macleod and Barker 1991; Cooper and
Eccles 1930; Stein and Parmiggiani 1981). Thus the pattern
by which motoneurons fire when initially recruited strongly
influences the force motor units subsequently generate. The
first aim of the present study was to determine the sequence
of pulses that maximize force from individual human thenar
motor units when stimulated intraneurally. These data provide the information needed to effectively stimulate muscle
that cannot be driven voluntarily, thereby providing the
possibility of restoring limb movement artificially. The second aim of this study was to examine the magnitude of
twitch force potentiation that resulted from the delivery of
these stimuli. Twitch potentiation following muscle activation is common in human, cat, and rat motor units (Bagust
et al. 1974; Burke et al. 1973; Kernell et al. 1975; Nordstrom and Miles 1990; Olson and Swett 1971; Stephens and
Usherwood 1977; Thomas et al. 1990; Young and Mayer
1981). Force changes due to potentiation and different patterns of stimulation could then be compared. Some of the
present results have been published as an abstract (BiglandRitchie et al. 1997).
METHODS
Subjects and experimental setup
Four men and three women with no history of any neurological
disorder were chosen as research subjects. Each subject gave their
informed consent to participate in the experiments, all of which were
approved by the ethics committee of Umeå University, Sweden.
The experimental setup shown in Fig. 1A has been described
previously (Westling et al. 1990). Briefly, the subject reclined in a
dental chair and the subject’s right arm was extended, the hand
supinated, and the forearm immobilized in a vacuum cast. The dorsal
aspects of the hand and fingers rested in, and were firmly stabilized
by, molded modeling clay. The fingers were also constrained by
U-shaped clamps anchored in the clay. The extended thumb rested
against a two-dimensional force transducer placed against the interphalangeal joint. The thumb position was adjusted until the force
transducer registered 0.5 N resting tension in the directions of both
abduction and flexion. Thumb flexion and abduction forces were
measured simultaneously at right angles by the use of the twodimensional force transducer.
Single motor axons of the median nerve that innervate thenar motor
units were stimulated electrically through a tungsten microelectrode
(Vallbo et al. 1979) inserted into the median nerve ;10 cm proximal
to the elbow. After an appropriate nerve fascicle was impaled, the
0022-3077/99 $5.00 Copyright © 1999 The American Physiological Society
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
Thomas, C. K., R. S. Johansson, and B. Bigland-Ritchie. Pattern of
pulses that maximize force output from single human thenar motor
units. J. Neurophysiol. 82: 3188-3195, 1999. We assessed the sequence of nerve impulses that maximize force output from individual
human thenar motor units. When these motor units were stimulated
intraneurally by a variable sequence of seven pulses, the pattern of
pulses that elicited maximum force always started with a short (5–15
ms) interpulse interval termed a “doublet.” The twitch force summation caused by this “doublet” elicited, on average, 48 6 13% (SD) of
the maximum tetanic force. The peak amplitude of “doublet” forces
was 3.5 times that of the initial twitches, and twitch potentiation
appeared to have little influence on twitch force summation elicited by
the “doublets.” For some units, the second optimal interpulse interval
was also short. Peak forces elicited by the third to sixth interpulse
intervals did not change substantially when the last interpulse interval
was varied between 5 to 55 ms, so maximum force could not be
attributed to any unique interpulse interval. Each successive pulse
contributed a smaller force increment. When five to seven pulses were
delivered in an optimal sequence, the evoked force was close to that
recorded during maximal tetanic stimulation. In contrast, maximal
force-time integral was evoked with one short interpulse interval
(5–15 ms) then substantially longer interpulse intervals (.100 ms).
Maximum force and force-time integrals were therefore elicited by
different patterns of stimuli. We conclude that a brief initial interpulse
interval (5–15 ms) is required to elicit maximum “doublet” force from
human thenar motor units and that near-maximal tetanic forces can be
elicited by only five or six additional post-“doublet” pulses if appropriately spaced in time. However, the rate at which these post“doublet” stimuli must be provided is fairly uncritical. In contrast,
maximum post-“doublet” force-time integrals were obtained at intervals corresponding to motoneuronal firing rates of ;7 Hz, rates close
to that typically used to recruit motor units and to maintain weak
voluntary contractions.
IMPULSE PATTERNS THAT MAXIMIZE MUSCLE FORCE
3189
infrared blood pressure pulse detector attached to the subject’s finger
(Westling et al. 1990). Thus all stimuli or pulse trains were separated
by ;0.8 s intervals, depending on the heart rate.
Protocol
Data collection and analysis
FIG. 1. A: experimental setup. A and F, forces of abduction and flexion,
respectively. B: typical sequence of pulses in run 2.
tungsten microelectrode was moved in minute steps until a weak
stimulus current evoked all-or-none electromyogram (EMG) and
force in a thenar motor unit. Strict criteria were applied to ensure that
the force and EMG responses arose from only a single motor unit
(Westling et al. 1990). Distal and proximal unitary thenar surface
EMG signals were recorded using three flexible electrodes, each made
from braided strands (0.004 mm2) of silver-coated copper wire (Fig.
1A). A common electrode lay transversely across the mid-muscle
belly, a distal electrode across the metacarpal joint, and a proximal
electrode over the base of the thenar eminence. The resultant force
elicited from each motor unit was calculated continuously from the
recorded abduction and flexion components. Force baseline fluctuations caused by cardio-respiratory processes were minimized by resetting the force electronically to a preset zero just before the delivery
of each stimulus or package of stimuli. All stimuli were triggered in
fixed relation to the pulse pressure cycle, which was assessed from an
Motor unit EMG and force data were sampled on-line at 3,200 Hz
and 400 Hz, respectively, using SC/Zoom software (Department of
Physiology, Umeå University, Sweden). Event markers and stimulus
pulses were recorded at 0.3 ms accuracy (Westling et al. 1990). Peak
force, force-time integral, time to peak force (contraction time, CT),
and time for the force to fall from peak to half-peak force (halfrelaxation time) were measured from averages of 5 consecutive
twitches recorded before and after the interval test, and from 1 to 3
single twitches obtained within the interval test. For each unit, we
normalized the values obtained within and after the interval test to the
corresponding initial twitch parameters to assess twitch force and
force-time integral potentiation and contractile speed changes. We
also normalized twitch force values to the maximum tetanic force to
determine twitch/tetanus force ratios.
Contractile responses to multiple pulses (n 5 2–7) were measured
as described above for twitches (peak force, force-time integral from
force start to end, and time to peak force, CT; see Fig. 2, inset) except
that no measures of half-relaxation time were made. For each response
in a run, the peak force and force-time integral of the initial fixed pulse
response (e.g., two pulses in run 2) were subtracted from the values
obtained for all pulses delivered (e.g., three pulses in run 2) to
determine which last interpulse interval evoked the greatest peak force
and force-time integral increment.
Statistics
Mean values 6 SD are given. “Doublet” data from all 21 units are
provided. Analysis of twitch force potentiation during the interval test
was based on the 14 units that completed all runs of the interval test
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
For each motor unit, we determined the pattern of seven pulses
which maximized force output, a process we termed an “interval
test.” First, two stimuli were delivered to the motor axon at
progressively shorter intervals until they were 5 ms apart. Starting
with an interpulse interval of 500 ms, this was reduced in steps of
100 ms to 200 ms (4 pulse pairs), then in 20 ms steps from 180 ms
to 80 ms (6 pulse pairs), and in 5 ms steps from 75 ms to 5 ms (15
pulse pairs). This sequence, a total of 25 pulse-pairs, is referred to
as “run 1” of the interval test. The interpulse interval that elicited
the strongest force was termed the first optimal interval and was the
first interpulse interval in all subsequent runs. In run 2, the second
optimal interpulse interval was determined by delivering a third
pulse at progressively shorter intervals in relation to the preceding
force-optimized pulse-pair (Fig. 1B). We repeated this process in
runs 3– 6 until six optimum interpulse intervals were defined. Thus
each pulse train in runs 2– 6 involved delivery of 3, 4, 5, 6, and 7
pulses, respectively. Before and after the interval test, each unit
was stimulated with 5 pulses at 1 Hz to elicit single twitches
(Thomas et al. 1990). Single twitches were also recorded within the
interval test when the interpulse intervals were longer than those at
which any twitch fusion occurred.
After the interval test, the relation between motor unit force and
stimulus frequency was recorded by stimulating each unit for 1–2 s at
rates from 5 Hz to 100 Hz as described in Thomas et al. (1991a). This
provided values of maximum tetanic force and the frequency required
to elicit any fraction thereof. The units were then fatigued with 13
pulses at 40 Hz each second for two minutes and force fatigue indices
were calculated (final force/initial force) (Burke et al. 1973; Thomas
et al. 1991b).
3190
C. K. THOMAS, R. S. JOHANSSON, AND B. BIGLAND-RITCHIE
RESULTS
Force responses to a pair of stimuli
FIG. 2. Force summation. Representative force and proximal and distal
EMG signals recorded from a typical motor unit in run 1. Two pulses were
delivered to its motor axon at progressively shorter intervals until 5 ms
apart. Only 13 of 25 total records are shown (interpulse intervals of 400,
200, 160, 120, 80, 70, 60, 50, 40, 30, 20, 10, and 5 ms). Peak force,
contraction time (CT), and force-time integral were measured as shown in
inset.
using repeated measures analysis of variance with run number as the
fixed effect (7 levels; prerun and run 1– 6). Least squares linear
regression was used to assess correlations between different parameters. Statistical significance was set at P , 0.05.
Figure 2 shows how the twitch forces of a representative unit
summed as pairs of stimulus pulses were delivered at progressively closer interpulse intervals. Peak force and force-time
integral were greatest when the pulses were 5 ms apart, and
were respectively 3.4 and 5.0 times greater than the values for
single twitches (Fig. 3, A and B). Peak force was 39% of the
maximum tetanic force (Fig. 3C). Two shocks applied 5 ms
apart also elicited the most rapid rise in force and the shortest
time to peak force (CT, Fig. 3D). EMG potentials produced by
each stimulus were strikingly similar at all interpulse intervals
.5 ms (Fig. 2). With a 5 ms interpulse interval, the two EMG
potentials of most units began to fuse, although force continued
to rise or was similar to that produced by a 10 ms interval. At
interpulse intervals ,5 ms, tested prior to the formal protocol,
FIG. 3. Values recorded from the unit
shown in Fig. 2 as a pair of pulses were
delivered at progressively shorter intervals.
A: peak force; B: force-time integral; C: peak
force normalized to tetanic force; D: contraction time (time to peak force) in relation to
interpulse interval. - - -, initial twitch values.
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
Twenty-one thenar motor axons were stimulated intraneurally to determine the temporal pattern of seven pulses that
maximized motor unit force. The twitch forces of these units,
measured before the interval test, ranged from 3.0 to 44.5 mN
(mean 16.1 6 13.0 mN), contraction times ranged from 37.5 to
91.9 ms (57.9 6 16.1 ms), and half-relaxation times ranged
from 32.5 to 130.0 ms (66.5 6 21.5 ms), all unimodally
distributed. Their tetanic forces in response to 50 Hz stimulation ranged from 28 to 193 mN (99.3 6 50.6 mN), force
fatigue indices ranged from 0.33 to 1.14 (0.83 6 0.25), and
twitch/tetanus force ratios ranged from 0.03 to 0.31 (0.18 6
0.09). These values and their distributions were similar to those
found for the larger population of 45 units studied for other
purposes (Thomas et al. 1990, 1991a,b; Westling et al. 1990).
The first optimal interpulse interval was determined for all
21 units, but successful recording from some units was subsequently lost. The number of units from which complete sets of
measurements could be made declined to 14 units in run 6.
When mean values are shown for the entire interval test (runs
1– 6), they are based on data from 14 units. However, the
behavior of these units was representative of the initial 21
units.
IMPULSE PATTERNS THAT MAXIMIZE MUSCLE FORCE
3191
FIG. 4. Mean ““doublet”” values for
different thenar motor units. A, B: force and
force-time integral summation in relation to
initial twitch values; C: force summation in
relation to twitch/tetanic force ratios; D:
time to “doublet” force in relation to twitch
CT. All correlations showed slope coefficients different from zero (P , 0.05).
Figure 5B shows, for the same unit, alterations in peak force of
the last pulse in runs 1– 6. At long intervals, the last pulse in
each run evoked only twitches. The force from the last pulse in
each run only increased as the twitches began to fuse with the
Force generated by pulses that followed the initial
“doublet”
Figure 5A shows, for a representative unit, how force summated as an additional pulse was delivered at progressively
shorter intervals in run 4. With an interpulse interval of 120 ms,
the last pulse added substantially to the force-time integral but
did not contribute to the peak force. Peak force only increased
when the last interpulse interval was reduced to 70 ms or less.
FIG. 5. Multi-pulse force summation for a representative unit. A: force
responses when the last interpulse interval in run 4 was 120, 50, and 5 ms (z z z,
- - -, and —, respectively). B: peak force of the last pulse during runs 1– 6. C:
force-time integral measures for each contraction in runs 1– 6 as a function of
the interval between the last pair of pulses.
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
the force was depressed relative to that evoked by a 5 ms or 10
ms interpulse interval (see Burke et al. 1976).
All but one unit generated maximal force when two pulses
were delivered either 5 ms or 10 ms apart (15 and 5 units,
respectively). The force from only one unit was greatest with
an interpulse interval of 15 ms. The force-time integrals also
reached maximum values at 5, 10, and 15 ms for 12, 7, and 2
units, respectively. We termed the short interpulse intervals
that generated maximum force “doublet’s” because they resembled the closely-spaced muscle potentials that are often
recorded at the onset of voluntary contractions that are termed
doublet’s (Bawa and Calancie 1983; Desmedt and Godaux
1977; Kudina and Alexeeva 1992).
Both the peak force and force-time integral elicited by a
“doublet” varied considerably between units. For all 21 units,
the “doublet” forces averaged 3.5 6 1.6 times the initial twitch
force and were 48 6 13% of maximum tetanic force. The
force-time integrals of the “doublets” were 5.6 6 3.6 times
those of the initial twitches. The greatest summation (“doublet”
force/twitch force) occurred in units with weak twitch forces,
low twitch/tetanic force ratios (Fig. 4, A and C), and small
twitch force-time integrals (not shown). Units with small
twitch force-time integrals also showed the greatest force-time
integral summation (Fig. 4B). No correlation was found between force summation and tetanic force however. The mean
time to peak “doublet force” (CT) was 1.4 6 0.4 times the
initial twitch contraction time. The ratio of “doublet” CT to
initial twitch CT was greatest for fast units (Fig. 4D). Force
summation (“doublet” force/twitch force) was not correlated to
twitch CT.
3192
C. K. THOMAS, R. S. JOHANSSON, AND B. BIGLAND-RITCHIE
FIG. 6. Multi-pulse forces and increments
in force and force-time integrals. A: maximum
force elicited from one unit by 2, 3, 4, 5, 6, and
7 pulses in runs 1– 6, compared with the force
from constant rate stimulation at 50 Hz (- - -).
B: peak forces of individual units (1) and
means across all units (●) in runs 1– 6, relative
to maximal tetanic force. C: mean 1 SD force
increment contributed by the last pulse (●) and
mean interpulse interval for maximum force
(E) in runs 1– 6 (n 5 14 units). D: mean 2 SD
force-time integral increment contributed by
the last pulse (●) and mean interpulse interval
for maximum force-time integral (E) in runs
1– 6.
Figure 6A shows the maximal forces elicited from one motor
unit in runs 1– 6, recorded when interpulse intervals of 5, 15,
25, 30, 35, and 35 ms, respectively, were delivered. These
forces are compared with the force evoked by a constant
frequency train of stimuli at 50 Hz (- - -). The maximal force
increased during runs 1– 4. The largest force increment, relative to the twitch, occurred in response to the “doublet” (run 1).
The force increment elicited by each additional pulse became
progressively smaller until, in runs 5 and 6, no further force
increase occurred.
Figure 6B shows the increase in force output as we gradually
added interpulse intervals that maximized the peak force output
in runs 1– 6. Mean unit force increased from 51 6 12% of
maximal tetanic force in run 1 (“doublet” force) to 92 6 11%
in run 6. It is interesting to note that while a “doublet” elicited
approximately half-maximal tetanic force, it took four or five
additional pulses to elicit a further 41% maximum tetanic
force. Nevertheless, when an appropriate pulse pattern was
FIG. 7. Twitch potentiation. A: superimposed twitch forces from a typical motor unit
recorded initially and after runs 1– 6, showing gradual force increase. B: twitch force
from each unit (1) and mean values (●)
before run 1 (Pre) and after runs 1– 6 (n 5
14). C, D: twitch force after run 6 normalized to initial twitch force vs. initial twitch
force and twitch/tetanus force ratios, respectively (both are P , 0.05).
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
preceding fixed pulse response. At intervals shorter than 45 ms,
the force contribution from the last pulse was indistinguishable
from the fixed pulse response. It was also independent of
interpulse interval during runs 3– 6. For this unit, the interpulse
interval in run 2 had to be reduced to 5 ms to elicit maximum
force, as in run 1. However, for other units, peak forces elicited
by pulses in runs 2– 6 did not change substantially when the
last interpulse interval was varied between 5 and 55 ms. Thus
in runs 3– 6 and sometimes run 2, the peak force elicited could
not be attributed to any unique interpulse interval.
Figure 5C shows, for the same motor unit, the effects of
interpulse interval on the force-time integral for the whole
contraction. For this and other units, the force-time integrals
increased as more pulses were added; the increase was roughly
proportional to the number of pulses delivered. This was in
contrast to the increase in peak force, which was largest in run
1, declined progressively in runs 2– 4, and was negligible
thereafter (Fig. 5B).
IMPULSE PATTERNS THAT MAXIMIZE MUSCLE FORCE
applied, almost full tetanic force was achieved with only six or
seven pulses.
The mean values for the force and force-time integral increment contributed by the last pulse in each run are shown in Fig.
6, C and D, respectively (F). Although the maximum additional force per pulse was greatest for “doublets” and then
declined progressively in runs 2– 6, the mean increments in
force-time integrals per pulse were similar for each additional
pulse. Maximal force output in runs 1– 6 was obtained with
mean interpulse intervals of 9 6 8, 19 6 11, 28 6 11, 33 6 10,
36 6 9, and 29 6 7 ms, respectively (Fig. 6C, E). Using these
interpulse intervals, maximal force-time integral in runs 2– 6
occurred when the last interpulse interval was 104 6 41, 143 6
36, 145 6 52, 134 6 41, and 144 6 30 ms, respectively (Fig.
6D, E). The force-time integral in run 1 was greatest when the
two pulses were 7 6 4 ms apart, on average.
The twitch forces increased significantly during the course
of the interval test (Fig. 7, A and B). On average, twitch forces
increased from 17.4 6 11.5 mN, measured before the test, to
23.9 6 11.5 mN at the end of run 6. That is, after 675 pulses
had been applied over approximately 4 min, twitch force was
1.5 6 0.6 times initial values. Force-time integral values were
1.7 6 1.1 times initial values after run 6. In contrast to twitch
forces, “doublet” forces did not show any signs of potentiation.
At the end of runs 1 and 2, mean “doublet” forces were 46.2 6
27.1 and 46.1 6 28.0 mN, respectively. “Doublet” forces were
not recorded in runs 3– 6 because these runs involved 4 –7
pulses.
Figure 7, C and D, shows that twitch force potentiation was
significantly greater for units with weak twitch forces and low
twitch/tetanic force ratios. Potentiation was not accompanied
by significant changes in twitch contraction and half-relaxation
times. However, units with long initial twitch CT (slow units)
became somewhat faster, whereas those with short initial
twitch CT tended to slow down in run 1, with little further
change thereafter. After runs 1 and 6, twitch CT was 1.04 6
0.22 and 1.05 6 0.23 times initial, respectively. Neither potentiation of twitch force nor of twitch force-time integral was
correlated to initial CT or half-relaxation.
DISCUSSION
The pulse sequence that elicited the greatest force from a
human thenar motor unit always started with a short 5–15 ms
interval between the first two pulses. For all 21 units, this
“doublet” caused the force to increase by more than the linear
sum of two twitches elicited separately (mean, 3.43 twitch
force). Force summation was greatest for weak units with small
twitch force-time integrals and with low twitch/tetanus ratios.
Some units needed a second short interpulse interval to maximize force. Additional pulses caused a further increase in
force, but the extra force elicited became progressively smaller
and was less sensitive to changes in interpulse interval. After
seven pulses were delivered, the force reached 92 6 11%
maximum unit tetanic force measured by pulse trains at 50 Hz.
The pulse pattern that elicited peak force and force-time integral did not coincide after the first short interpulse interval.
Longer interpulse intervals maximized the force-time integral
whereas shorter intervals maximized peak force.
This behavior of human thenar motor units is similar to that
reported previously by Burke et al. (1976), Parmiggiani and
Stein (1981), and Zajac and Young (1980a) when single units
and/or whole muscles were stimulated in a similar way in cat.
Again, all units exerted maximum force-time integral when
stimulated by a train of pulses commencing with a 5–10 ms
“doublet” followed by longer intervals. Zajac and Young
(1980a) measured changes in force-time integrals for isometric
contractions of cat medial gastrocnemius motor units. They
found that fast units with low twitch/tetanus force ratios often
needed two, rather than one, short (transitional) intervals before high force-time integrals could be sustained by subsequent
longer intervals, as found here for some human thenar motor
units (see Fig. 5). They suggested that two transitional intervals
were needed when unit force was less potentiated. The protocol
used in the present study did not allow these relationships to be
explored in detail. They also found that maximal values required a steady state post-“doublet” interpulse interval of about
1.83 CT for each unit. For human thenar units, the maximum
force-time integral during runs 3– 6 usually occurred at approximately 140 ms intervals (2.43 mean twitch contraction time).
This difference in post-“doublet” interval between the results
of Zajac and Young (1980a) and ours probably arises because
we selected the interpulse interval that maximized the peak
force rather than the force-time integral. Differences may also
relate to the lack of fast fatigable type units found in human
thenar muscles (Thomas et al. 1991b) and to the absence of any
significant correlations between unit force and speed observed
in most human motor unit studies (Bigland-Ritchie et al. 1998).
“Doublet” force summation
That the “doublet” force should be substantially greater than
the summed force of two twitches elicited separately is explained in part by muscle mechanics. Hill (1949, 1953) demonstrated that, for frog muscle, the muscle series elastic component primarily determines the twitch/tetanus force ratio. If a
twitch is preceded by a quick stretch such that the muscle series
elastic elements become fully prestretched, the twitch force is
then equal to that of the tetanus. Thus the increased amplitude
of a “doublet,” compared with a twitch, would result directly
from the stretch provided by the contractile response of the first
pulse that takes up slack and stretches the unit’s series components. A second pulse following closely thereafter will then
elicit a force response substantially greater than that of the sum
of two twitches elicited separately. Thus the “doublet” response approaches, to a greater or lesser extent, the force from
maximal tetanic stimulation.
In the current experiments, “doublet” forces in run 1 averaged 48 6 13% maximum tetanic force, and twitch forces
averaged 17.8 6 8.8%. Although “doublet” force-time integrals were 6.8 6 6.7 times those of single twitches, the increases were mainly because of the increase in force amplitude
(3.5 6 1.6 times initial twitch values). The force-time integral
also increased because of the increase in force duration (e.g.,
CT was 1.4 6 0.4 times initial). Differences in twitch summation between units, when two pulses were delivered, may
depend on differences in series compliance. This predicts that
“doublet” force enhancement would be greatest for units with
low twitch/tetanus force ratios, as was found here for thenar
units. A low twitch/tetanus force ratio would be expected for
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
Twitch force and force-time integral potentiation
3193
3194
C. K. THOMAS, R. S. JOHANSSON, AND B. BIGLAND-RITCHIE
Potentiation
The potentiation of twitch forces observed during these
experiments was similar to that reported previously for human
thenar motor units (Thomas et al. 1990), and motor units in
other human muscles (Nordstrom and Miles 1990; Stephens
and Usherwood 1977; Young and Mayer 1981). Units with
weak twitch forces and low twitch/tetanus force ratios potentiated most and were those from which a “doublet” elicited the
greatest twitch force summation. Because the sequence in
which stimuli were delivered was not randomized, we assume
that the influence of potentiation on the magnitude of twitch
summation was cumulative. However, it seems unlikely that
twitch potentiation played a major role in twitch force summation induced by “doublets.” The twitch forces gradually
increased during the entire interval test but not by more than
1.5 times the initial values, whereas the mean “doublet” force
was already 3.5 times single twitch forces in run 1. Furthermore, the “doublet” force did not show any signs of potentiation between runs 1 and 2.
Optimal patterns of stimulation following the “doublets”
Our results show that an initial “doublet” increased the force
three to four times that of the single twitch. After the subsequent five optimal interpulse intervals, the force generated was
at near-maximum levels (92 6 11% maximal tetanic force, Fig.
6B). In contrast, if shocks were applied to human thenar motor
units at a constant rate (nonoptimal intervals), it may take up to
20 pulses delivered at 30 –100 Hz before maximum tetanic
force is reached (Thomas et al. 1991a). Thus when a unit is
recruited with a “doublet” it generates force more quickly,
primarily by reducing the effect of series compliance (Fig. 6A).
Furthermore, that the interpulse periods of the post-“doublet”
stimuli required to provide near-maximal force output were
rather uncritical (5–55 ms) suggests that the irregular rates at
which motor units often fire during voluntary contractions do
not necessarily compromise force generation. Irregular motor
unit firing rates are common during fatigue, for example,
(Griffin et al. 1998) or after spinal cord injury (Thomas and
Kozhina 1999).
Numerous studies have noted the occurrence of doublet’s at
the onset of human voluntary contractions (Bawa and Calancie
1983; Denslow 1948; Desmedt and Godaux 1977; Kudina and
Alexeeva 1992), in animal muscles contracting in response to
reflex drive (Cordo and Rymer 1982), during decerebrate cat
locomotion (Hoffer et al. 1981; Zajac and Young 1980b),
under conditions of both normal and pathological tremor (Elek
et al. 1991; Freund 1983), and during spasms of paralyzed
muscle (Thomas and Ross 1997). Macefield et al. (1996)
assessed the influence of an initial 10 ms “doublet” on the
force-frequency relationship for single human toe extensor
units when trains of pulses were applied at different constant
frequencies (2–100 Hz). They found that such “doublets” effectively elevated the force at low stimulation rates (e.g., 2 to
10 Hz) but caused little force change when delivered during
pulse-trains of 20 Hz or above. Thus a “doublet” fired by a
motoneuron probably has little effect on force when it occurs
during the course of long-lasting, relatively strong contractions. It is most effective at contraction onset when firing rates
are low or when the motor unit is engaged in brief actions.
Indeed, even at the lowest stimulus rates, the force increment
generated by a “doublet” typically disappears within 1–2 s
(Macefield et al. 1996).
General considerations
The results reported here demonstrate that a brief initial
interpulse interval (5–15 ms) is required to elicit maximum
“doublet” force from human thenar motor units and that nearmaximal tetanic forces can be elicited by only five or six
additional post-“doublet” pulses if appropriately spaced in
time. Furthermore, the rate at which these post-“doublet” stimuli must be provided to human thenar motor units for near
maximal force is fairly uncritical. Hence, if trains of pulses that
start with a “doublet” were used in functional electrical stimulation of impaired or paralyzed muscle, rather than the more
usual constant frequency trains, this may markedly reduce the
risk of fatigue due to transmission failure at the neuromuscular
junction and/or within the t-tubular system (Binder-Macleod
and Barker 1991; Binder-Macleod et al. 1997). Minimizing the
number of pulses would also reduce the overall metabolic load
of the neuromuscular system during motor actions.
It is noteworthy that maximum force-time integrals were
recorded when most post-“doublet” interpulse intervals were
about 140 ms (i.e., runs 3– 6), which corresponds to a motoneuronal firing rate of 7 Hz. This rate is close to the minimum rate
at which motor units are recruited and it is similar to the rate
used to maintain weak voluntary contractions (Monster and
Chan 1977; Tanji and Kato 1973a,b). Thus, if preceded by a
“doublet,” these low rates would make the human thenar
muscles contract with maximum efficiency even when acti-
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
units that are “on slack,” i.e., at short length in relation to their
range of motion. The ten-fold range of twitch/tetanus force
ratios reported here, 0.03– 0.31, suggests that different units
have a correspondingly wide range of functional intrinsic series
compliance with the hand in the position adopted in the present
experiments.
Differences in resting tension among individual motor units
may have contributed to the wide range of twitch/tetanus force
ratios, despite the precautions to make all measurements at a
standard overall muscle length (see METHODS). The weaker
units may thus have been “slacker” than those generating
stronger forces. Thomas et al. (1991b) found that thenar units
with weaker tetanic forces tended to exert their force preferentially in the direction of flexion.
In addition to stretch of passive series compliance as discussed above, “doublet” force enhancement may also be explained by increased Ca21 release from the sarcoplasmic reticulum because, in cat muscles, dantrolene sodium abolished
part of the short interval force enhancement (Parmiggiani and
Stein 1981). Indeed, twitch force and its time integral are also
determined by the duration and amount of actomyosin crossbridge binding (“active state”; Hill 1949), which in turn are
related to the amount of calcium released from the sarcoplasmic reticulum per impulse and the time course of its reuptake.
The possibility that the additional force increment came
from recruitment of inactive muscle fibers is unlikely because,
for intervals of 10 ms or longer, both EMG potentials had
similar shapes. For 5 ms intervals, unit EMG potentials began
to merge but without force decrements.
IMPULSE PATTERNS THAT MAXIMIZE MUSCLE FORCE
vated at their minimum rates. Moreover, the similarity between
the optimal interpulse patterns described here, in animal studies, and those recorded from muscles or nerves during naturally
elicited contractions, suggests that the nervous system indeed
uses these advantageous pulse-patterns.
The authors thank G. Westling and L. Bäckström for help during the
experiments.
This work was supported by National Institute of Neurological Disorders
and Stroke Grants NS-30226 to C. K. Thomas and NS-14756 to B. BiglandRitchie, and by Medical Research Council of Sweden Project 08667 to R. S.
Johansson.
Address for reprint requests: C. K. Thomas, The Miami Project to Cure
Paralysis, University of Miami School of Medicine, P.O. Box 016960 (R48),
1600 NW 10th Ave., R-48, Miami, FL 33101-9844.
Received 18 May 1999; accepted in final form 30 August 1999.
REFERENCES
HOFFER, J. A., O’DONOVAN, M. J., PRATT, C. A., AND LOEB, G. E. Discharge
patterns of hindlimb motoneurons during normal cat locomotion. Science
213: 466 – 468, 1981.
KERNELL, D., DUCATI, A., AND SJÖHOLM, H. Properties of motor units in the
first deep lumbrical muscle of the cat’s foot. Brain Res. 98: 37–55, 1975.
KUDINA, L. P. AND ALEXEEVA, N. L. Repetitive “doublet”s of human motoneurones: analysis of interspike intervals and recruitment pattern. Electroencephalogr. Clin. Neurophysiol. 85: 243–247, 1992.
MACEFIELD, V. G., FUGLEVAND, A. J., AND BIGLAND-RITCHIE, B. Contractile
properties of single motor units in human toe extensors assessed by intraneural motor axon stimulation. J. Neurophysiol. 75: 2509 –2519, 1996.
MONSTER, A. W. AND CHAN, H. Isometric force production by motor units of
extensor digitorum communis muscle in man. J. Neurophysiol. 40: 1432–
1443, 1977.
NORDSTROM, M. A. AND MILES, T. S. Fatigue of single motor units in human
masseter. J. Appl. Physiol. 68: 26 –34, 1990.
OLSON, C. B. AND SWETT, C. P., JR. Effect of prior activity on properties of
different types of motor units. J. Neurophysiol. 34: 1–16, 1971.
PARMIGGIANI, F. AND STEIN, R. B. Nonlinear summation of contractions in cat
muscles. II. Later facilitation and stiffness changes. J. Gen. Physiol. 78:
295–311, 1981.
STEIN, R. B. AND PARMIGGIANI, F. Nonlinear summation of contractions in cat
muscles. I. Early depression. J. Gen. Physiol. 78: 277–293, 1981.
STEPHENS, J. A. AND USHERWOOD, T. P. The mechanical properties of human
motor units with special reference to their fatiguability and recruitment
threshold. Brain Res. 125: 91–97, 1977.
TANJI, J. AND KATO, M. Recruitment of motor units in voluntary contraction of
a finger muscle in man. Exp. Neurol. 40: 759 –770, 1973a.
TANJI, J. AND KATO, M. Firing rate of individual motor units in voluntary
contraction of abductor digiti minimi muscle in man. Exp Neurol. 40:
771–783, 1973b.
THOMAS, C. K., BIGLAND-RITCHIE, B., AND JOHANSSON, R. S. Force-frequency
relationships of human thenar motor units. J. Neurophysiol. 65: 1509 –1516,
1991a.
THOMAS, C. K., JOHANSSON, R. S., AND BIGLAND-RITCHIE, B. Attempts to
physiologically classify human thenar motor units. J. Neurophysiol. 65:
1501–1508, 1991b.
THOMAS, C. K., JOHANSSON, R. S., WESTLING, G., AND BIGLAND-RITCHIE, B.
Twitch properties of human thenar motor units measured in response to
intraneural motor-axon stimulation. J. Neurophysiol. 64: 1339 –1346, 1990.
THOMAS, C. K. AND KOZHINA, G. Motor unit firing rates during voluntary
contractions. International Society of Biomechanics XVIIth Congress, 1999,
p. 242.
THOMAS, C. K. AND ROSS, B. H. Distinct patterns of motor unit behavior during
muscle spasms in spinal cord injured subjects. J. Neurophysiol. 77: 2847–
2850, 1997.
VALLBO, Å. B., HAGBARTH, K.-E., TOREBJÖRK, H. E., AND WALLIN, B. G.
Somatosensory, proprioceptive, and sympathetic activity in human peripheral nerves. Physiol. Rev. 59: 919 –956, 1979.
WESTLING, G., JOHANSSON, R. S., THOMAS, C. K., AND BIGLAND-RITCHIE, B.
Measurement of contractile and electrical properties of single human thenar
motor units in response to intraneural motor-axon stimulation. J. Neurophysiol. 64: 1331–1338, 1990.
YOUNG, J. L. AND MAYER, R. F. Physiological properties and classification of
single motor units activated by intramuscular microstimulation in the first
dorsal interosseous muscle in man. In: Motor Unit Types, Recruitment and
Plasticity in Health and Disease, edited by J. E. Desmedt. Basel: Karger,
1981, vol. 9, p. 17–25.
ZAJAC, F. E. AND YOUNG, J. L. Properties of stimulus trains producing maximum tension-time area per pulse from single motor units in medial gastrocnemius muscle of the cat. J. Neurophysiol. 43: 1206 –1220, 1980a.
ZAJAC, F. E. AND YOUNG, J. L. Discharge properties of hindlimb motoneurons
in decerebrate cats during locomotion induced by mesencephalic stimulation. J. Neurophysiol. 43: 1221–1235, 1980b.
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on June 16, 2017
BAGUST, J., LEWIS, D. M., AND LUCK, J. C. Post-tetanic effects in motor units
of fast and slow twitch muscle of the cat. J. Physiol. (Lond.) 237: 115–121,
1974.
BAWA, P. AND CALANCIE, B. Repetitive “doublet”s in human flexor carpi
radialis muscle. J. Physiol. (Lond.) 339: 123–132, 1983.
BIGLAND-RITCHIE, B., FUGELVAND, A. J., AND THOMAS, C. K. Motor units and
their behavior. Man is not a cat. Neuroscientist 4: 240 –249, 1998.
BIGLAND-RITCHIE, B., THOMAS, C. K., STITT, J. T., AND JOHANSSON, R. S.
Excitation rates that optimize force output from human thenar motor units.
Neurosci. Abs. 23: 1048, 1997.
BINDER-MACLEOD, S. A. AND BARKER, C.B.I. Use of a catchlike property of
human skeletal muscle to reduce fatigue. Muscle Nerve 14: 850 – 857, 1991.
BINDER-MACLEOD, S. A., LEE, S.C.K., AND BAADTE, S. A. Reduction of the
fatigue-induced force decline in human skeletal muscle by optimized stimulation trains. Arch. Phys. Med. Rehabil. 78: 1129 –1137, 1997.
BURKE, R. E., LEVINE, D. N., TSAIRIS, P., AND ZAJAC III, F. E. Physiological
types and histochemical profiles in motor units of the cat gastrocnemius.
J. Physiol. (Lond.) 234: 723–748, 1973.
BURKE, R. E., RUDOMIN, P., AND ZAJAC III, F. E. The effect of activation
history on tension production by individual muscle units. Brain Res. 109:
515–529, 1976.
COOPER, S. AND ECCLES, J. C. The isometric responses of mammalian muscles.
J. Physiol. (Lond.) 69: 377–385, 1930.
CORDO, P. J. AND RYMER, W. Z. Contributions of motor-unit recruitment and
rate modulation of compensation for muscle yielding. J. Neurophysiol. 47:
797– 809, 1982.
DENSLOW, J. S. Double discharges in human motor units. J. Neurophysiol. 11:
209 –215, 1948.
DESMEDT, J. E. AND GODAUX, E. Ballistic contractions in man: characteristic
recruitment pattern of single motor units of the tibialis anterior muscle.
J. Physiol. (Lond.) 264: 673– 693, 1977.
ELEK, J. M., DENGLER, R., KONSTANZER, A., HESSE, S., AND WOLF, W. Mechanical
implications of paired motor unit discharges in pathological and voluntary
tremor. Electroencephalogr. Clin. Neurophysiol. 81: 279 –283, 1991.
FREUND, H.-J. Motor unit and muscle activity in voluntary motor control.
Physiol. Rev. 63: 387– 436, 1983.
GRIFFIN, L., GARLAND, S. J., AND IVANOVA, T. Discharge patterns in human
motor units during fatiguing arm movements. J. Appl. Physiol. 85: 1684 –
1692, 1998.
HILL, A. V. The abrupt transition from rest to activity in muscle. Proc. Soc.
Lond. B 136: 399 – 420, 1949.
HILL, A. V. The plateau of full activity during a muscle twitch. Proc. Soc.
Lond. B 141: 498 –503, 1953.
3195