Core and skin temperature influences on the surface electromyographic responses to an isometric force and position task
Core and skin temperature influences on the surface electromyographic responses to an isometric force and position task
Nico A. Coletta 0 1
Matthew M. Mallette 0 1
David A. Gabriel 0 1
Christopher J. Tyler 1
Stephen S. Cheung 0 1
☯ These authors contributed equally to this work. 1
0 Department of Kinesiology, Brock University , St. Catharines, Ontario , Canada , 2 Department of Life Sciences, University of Roehampton , London , United Kingdom
1 Editor: Antoine Nordez, Universite de Nantes , FRANCE
The large body of work demonstrating hyperthermic impairment of neuromuscular function has utilized maximal isometric contractions, but extrapolating these findings to whole-body exercise and submaximal, dynamic contractions may be problematic. We isolated and compared core and skin temperature influences on an isometric force task versus a position task requiring dynamic maintenance of joint angle. Surface electromyography (sEMG) was measured on the flexor carpi radialis at 60% of baseline maximal voluntary contraction while either pushing against a rigid restraint (force task) or while maintaining a constant wrist angle and supporting an equivalent inertial load (position task). Twenty participants performed each task at 0.5ÊC rectal temperature (Tre) intervals while being passively heated from 37.1±0.3ÊC to 1.5ÊC Tre and then cooled to 37.8±0.3ÊC, permitting separate analyses of core versus skin temperature influences. During a 3-s contraction, trend analysis revealed a quadratic trend that peaked during hyperthermia for root-mean-square (RMS) amplitude during the force task. In contrast, RMS amplitude during the position task remained stable with passive heating, then rapidly increased with the initial decrease in skin temperature at the onset of passive cooling (p = 0.010). Combined hot core and hot skin elicited shifts toward higher frequencies in the sEMG signal during the force task (p = 0.003), whereas inconsistent changes in the frequency spectra occurred for the position task. Based on the patterns of RMS amplitude in response to thermal stress, we conclude that core temperature was the primary thermal afferent influencing neuromuscular response during a submaximal force task, with minimal input from skin temperature. However, skin temperature was the primary thermal afferent during a position task with minimal core temperature influence. Therefore, temperature has a task-dependent impact on neuromuscular responses.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
and analysis, decision to publish, or preparation of
High ambient temperatures impair whole-body exercise tolerance [
] and reduce self-paced
exercise intensity [
]. Multiple mechanisms have been advanced for this premature fatigue and
reduced exercise capacity, including that sustained heat storage and hyperthermia directly
causes a decrement in neuromuscular function [
]. Prior work, predominantly performed
using isometric, maximal single-joint muscle contractions has endorsed a direct hyperthermic
impairment on central nervous system activation [4±7]. In these studies, voluntary activation
progressively decreased with higher core temperature and then progressively increased back
towards baseline levels only upon subsequent core cooling, with these changes in voluntary
activation being independent of changes in skin [
] and muscle temperature [
However, extrapolating core temperature impairment during isometric and maximal
contractions to whole-body exercise may be problematic [
]. While maximal isometric
contractions provide an understanding of the force producing capacity of the neuromuscular system
], neural activation patterns of maximal isometric contractions are poorly related to dynamic
movements such as sprinting and vertical jump test performance [
]. Rather, isotonic
contractions±which inherently demonstrate lower electromyographic activity [
], lower motor
unit discharge rate [
], recruitment of additional motor units during prolonged contractions
], and greater rate of increase in central neural activity [
] compared to isometric
contractions±may have greater neural and mechanical similarities to dynamic movements
typical of whole-body exercise [
]. Dynamically maintaining position of a joint against a
specified load with a consistent muscle tension±often termed an iso-inertial contraction±is
neurologically similar to an isotonic contraction, and may be a useful analog for studying
dynamic muscle movements.
Differences in how core and skin thermal stimuli affect different neuromuscular tasks have
been previously reported [4,6,8,17±21]. Contrary to observations during isometric
contractions, maximal dynamic isokinetic contractions appear to be unaffected by rising core
temperature; rather, lowering skin temperature immediately impaired force production prior to any
reduction in core temperature, with this impairment sustained throughout the period of core
temperature decreasing back to baseline [
]. These data suggest that neuromuscular function
during hyperthermia is dependent on both the source of thermal stimuli and the type of task
being performed. Surface electromyograpy (sEMG) is a non-invasive method to objectively
quantify neuromuscular activation of muscle [
], and could be a valuable technique to
investigate the physiological mechanisms responsible for these differences. Additionally, sEMG has
been used extensively to investigate various neuromuscular disorders [
], muscular fatigue
[9,25±27], and the effects of temperature on muscle properties [
]. Yet, none of the
previous works have directly compared the effects of core and skin temperature on different
neuromuscular tasks, nor investigated the contribution of thermal afferents while using sEMG.
Therefore, the purpose of this experiment was to compare the effect of high core versus
skin temperature on the neuromuscular response to static and dynamic muscle contractions.
Submaximal (60% of maximal voluntary contraction [MVC]) isometric force and position
tasks were used as a surrogate for static and dynamic muscle contractions, respectively,
because they are biomechanically comparable with respect to changes in muscle length and
moment arm versus a freely movable limb while maintaining a constant muscle tension. The
force (isometric) task consisted of wrist flexion against an immovable brace, while the position
(isotonic) task±where the force requirements are the same, and the limb is free to moveÐ
required dynamically maintaining the wrist in a stable position against 60% of baseline
maximal force. Testing was performed at core temperature intervals of 0.5ÊC during heating
to 1.5ÊC above baseline core temperature, and then during cooling back to baseline in order
2 / 17
to test the relative contributions of core versus skin temperatures on neural drive. It was
hypothesized that core temperature would be the primary influence on sEMG magnitude and
frequency changes during the isometric force task, while skin temperature would be the
primary influence on these variables during the isometric position task.
Twenty healthy individuals (13 males and 7 females) were recruited. The mean (± SD) age,
height, mass, and body fat percentage was 23.8 ± 2.1 years, 175.3 ± 8.5 cm, 70.4 ± 9.2 kg, and
14.8 ± 6.0%, respectively. The study was approved by the Bioscience Research Ethics Board of
Brock University (REB 15±076) and conformed to the standards set by the Declaration of
Helsinki. All participants were screened for cardiovascular and neuromuscular health using the
Physical Activity Readiness Questionnaire developed by the Canadian Society for Exercise
Physiology, then were informed of the experimental protocol and associated risks prior to
providing verbal and written informed consent.
Participants completed a familiarization session to acquaint themselves with the apparatus and
to practice the experimental protocol. During this session, height (cm) and mass (kg) were
determined using standard laboratory equipment. Body fat percentage (%) was determined
through a 7-site (triceps, sub-scapula, abdomen, supra-iliac crest, mid-axilla, thigh, and
pectoralis major) skinfold thickness [
] with manual calipers (Harpenden, Bay International,
West Sussex, UK).
The study was a repeated measures design occurring over a single session, with neuromuscular
testing taking place during passive heating from baseline to voluntary tolerance, which was an
increase of at least 1.5ÊC rectal temperature (Tre), followed by passive cooling back towards
baseline Tre. Female participants were in the early follicular phase of their menstrual cycle as
determined through self-report to control for differences in initial Tre. Prior to the
experimental session, participants were asked to abstain from strenuous exercise and the consumption of
caffeine for 12 h and alcohol for 24 h. Upon arrival to the laboratory (~22ÊC, ~43% relative
humidity) between 0800±1300 h for the experimental session, participants voided their bladder
and euhydration was confirmed using a refractometer (PAL-10S, Atago, USA) and a threshold
urine specific gravity 1.020 [
Participants self-inserted a thermistor (Mon-A-Therm Core, Mallinkrodt Medical,
St. Louis, USA) 12 cm beyond the anal sphincter to measure Tre. Thermocouples
(PVC-T-24190, Omega Environmental Inc. Laval, QC, Canada) were taped at four sites to calculate mean
skin temperature (Tsk) using a weighted average of 0.3(chest) + 0.3(arm) + 0.2(thigh) + 0.2
Participants were dressed in a three-piece liquid-conditioning garment (BCS 4 Cooling
System, Allen Vanguard, Ottawa, CAN) consisting of 1/8º Tygon tubing sewn into a stretchable
suit. The liquid-conditioning garment covered the arms, upper and lower legs, and torso; the
face, head, neck, hands, and feet were left uncovered. Males wore shorts, while females wore
shorts and a sports bra. To minimize evaporative heat loss, an impermeable polyvinyl rain suit
was worn overtop the liquid-conditioning garment. Passive hyperthermia±target of a
minimum 1.5ÊC ΔTre above baseline±was performed by adding 49.5ÊC water maintained by a
3 / 17
Fig 1. Arm brace setup. Brace isolating the flexor carpi radialis muscle during wrist flexion for both the isometric force and position tasks, illustrating the skin
temperature thermistor for local forearm temperature (Tloc) and sEMG electrodes. Resistance for the position task provided by weights suspended on a pulley at the end
of the wooden arm.
temperature controller (Model 5202, Polyscience, Niles, IL, USA) and pumped (MED-ENG,
Pembroke, CAN) at a flow rate of ~1.5 L min-1. This was then followed by passive cooling,
with 10ÊC water flowing through the liquid-conditioning garment until Tre decreased to 1.0ÊC
above baseline or lower.
To isolate wrist flexion and the flexor carpi radialis, participants were positioned in a
semirecumbent position on an examination table with their right forearm supported at
approximately 135Ê elbow extension in a custom-made apparatus (Fig 1). For the force task, the fingers
were extended and the carpals to distal phalanges of the right hand were placed between two
fixed aluminum plates, with the right forearm positioned such that the styloid process was
aligned with the axis of rotation. These plates were secured to a calibrated load cell and
potentiometer to measure torque and angular displacement of the wrist joint, respectively. For
calibration, a known weight was attached to the load cell and the voltage output (torque acquired
through A-D converter @ 2500Hz) viewed; voltages outputs for weight below expected force
to above expected force were used to build a linear regression equation.
Participants were tasked with performing wrist flexion against the fixed plates, and viewed
a monitor that displayed the target torque. For the position task, the same elbow, forearm, and
4 / 17
wrist support occurred, but the aluminum plates bracing the hand were not fixed, such that
wrist flexion and extension were enabled. Participants were tasked with dynamically
maintaining the wrist in a neutral position against a load that forced wrist extension, while the monitor
provided information on angular displacement.
The heating and cooling protocol, along with timing of neuromuscular testing, is outlined in
Fig 2. Neuromuscular testing before (PRE) and following (POST) the passive heating and
cooling protocol consisted of eliciting five maximal M-wave (Mmax) and five H-reflex responses
through electrical stimulation of the median nerve, each separated by 15 s. Subsequently,
participants performed five maximal voluntary isometric contractions (MVC); each contraction
was 5 s in duration separated by 2 min. These were performed to test for whether the
contractions performed over the course of the experiment, or the heating and cooling protocol itself,
induced significant fatigue. The highest MVC pre-heating was used to calculate the 60% MVC
force output required during the subsequent submaximal force and position tasks.
During the passive heating and cooling protocol, participants performed a set of 60% MVC
force and position tasks at each 0.5ÊC Tre interval; each task was 3 s in duration and consisted
of two submaximal isometric force tasks followed by two position tasks (the order of
contractions was counterbalanced across individuals). Participants also performed a single
submaximal (60% of MVC) isometric force and position tasks for 1 min at each of four distinct
temperature states: initial Tre and Tsk (BASE); hot Tre, hot Tsk (H-H) after passive heating of
Tre 1.5ÊC above baseline; hot Tre, cool Tsk (H-C) shortly after cold (10ÊC) water began
Fig 2. Schematic of the experimental timeline, outlining the neuromuscular test battery along with the heating and cooling protocol. BASE: testing at baseline
prior to passive heating; H-H: Hot core, hot skin, taken at the highest point of Tre rise tolerated (>1.5ÊC Tre for all participants); H-C: Hot core, cool skin, taken <5 min
after initiation of cooling; POST: Neutral core, cool skin, taken at the end of cooling after Tre returned to <+1.5ÊC from BASE.
5 / 17
circulating through the liquid-conditioning garment; and the end of the protocol where Tre
returned to near baseline and Tsk was cool (POST). A 1 min, 60% MVC was used in this study
because this was the highest submaximal contraction that could be sustained and repeated
without evoking fatigue. This was demonstrated through no changes in root-mean-square
(RMS) amplitude in the sEMG signal before and after the heating protocol.
Pilot testing was completed to determine the effect of time, boredom, and potential fatigue on
neuromuscular function. Three participants completed a control trial during which they
performed the experimental protocol without any thermal manipulations, with one participant
fully instrumented with sEMG. The control trial lasted 2 h, with participants performing
submaximal contractions every 20 min; no differences in torque output or EMG parameters were
Previous work [
] has shown that stabilization of the force and sEMG trace is achieved
within 10 contractions. Therefore, at the familiarization session for all participants, maximal
contractions were practiced until a plateau was achieved, and the 60% force and positions
tasks were practiced 10 times. In order to reduce the likelihood of fatigue occurring during the
familiarization session, we used shorter 10±20 s practice trials.
For the experimental session for all participants, MVCs and evoked potentials before (PRE)
and following (POST) the passive heating and cooling protocol served as further control, to
determine that any neuromuscular changes that occurred during the experimental protocol
were a result of thermal manipulations rather than the protocol itself.
Flexor carpi radialis (FCR) muscle sEMG was measured using pediatric-sized surface
electrodes (3 mm electrode diameter, F-E9M 11 mm, GRASS Technologies, Astro-Med, Inc.,
Warwick, USA) with an inter-electrode distance of 10 mm. To ensure skin-electrode impedance
below 10 kO, the skin surface was shaved and abraded with electrolyte gel (NuPrep1, Weaver
and Company, Aurora, USA) and cleansed with isopropyl alcohol pads (Dukal Corporation,
Ronkonkoma, USA). The motor point of the FCR was located by applying a low-level electrical
stimulus to the muscle belly. The point that elicited the largest sEMG response at the lowest
stimulation intensity was identified as the motor point. Electrodes were placed in a bipolar
electrode configuration, as it is reliable for both evoked and voluntary measures [
], one on
the electrically identified motor point, and the other placed immediately distal (10 mm). The
electrodes were secured to the skin surface with two-sided tape and electrolyte gel (Signa
Gel1, Parker Laboratories, Fairfield, USA). The electrodes were then taped (Transpore™, 3M,
St. Paul, USA) to ensure contact with the skin throughout the experimental protocol. A 100
mm circular self-adhesive ground electrode (Dermatrode, Delsys, Boston, USA) was placed on
the dorsal side (back) of the hand for electrical safety and to minimize noise.
Placed medial to the elbow crease, the anode and cathode were connected in series with an
isolation unit (Grass Telefactor SIU8, Astro-Med Inc., Warwick, USA) and constant current
stimulator (Grass Telefactor S88, Astro-Med Inc., Warwick, USA) that delivered a square-wave
pulse 0.5 ms in duration. Median nerve stimulation was progressively increased until reaching
a plateau in the peak-to-peak M-wave amplitude (Mmax). The H-reflex was elicited using
lowlevel electrical stimulation until reaching peak-to-peak amplitude of 10% of Mmax.
6 / 17
Surface EMG signals were amplified (Grass P511, Astro-Med, Inc., Warwick, USA) and
bandpass filtered between 3 and 1000 Hz. Torque, sEMG, and displacement signals were acquired
at a sampling rate of 2500 Hz using an analog to digital converter (model BNC-2110, National
Instruments, Austin, USA), and simultaneously recorded on a personal computer using
custom-designed software (LabView 2011, National Instruments, Austin, USA). Torque was
lowpassed filtered (24 Hz, 3 dB) using a 2nd order Butterworth digital filter offline in MATLAB1
(The Mathworks Inc., Natick, USA). Temperature (Tre, Tsk, Tloc) data were collected at 1 Hz
and electrocardiogram at 1000 Hz (PowerLab, ADInstruments, Colorado Springs, USA), and
stored on a personal computer to be analyzed and processed offline using LabChart (Version
8, ADInstruments, Colorado Springs, USA).
For the brief, 3-s contractions, data were obtained from the average of two contractions for
both tasks. Eight participants reached their thermal tolerance at +1.5ÊC Tre and their data were
grouped in the `H-H' condition. Due to the unequal sample size beyond +1.5ÊC Tre and to
maintain consistency between contraction types (3-s and 1-min), the +1.5ÊC condition was
not used for analysis. As the 3-s contractions were not performed at BASE, H-H, H-C, and
POST timepoints, the first 3-s of the 1-min contractions were used for these thermal states.
Data for the 1-min contraction were obtained for the mean of 4, 1-s windows at 14, 29, 44, and
Normal distribution was assessed by skewness and kurtosis measures, and by visual inspection
of histograms. Normality was defined as skewness and kurtosis less than ± 3 and ± 9,
respectively. Results are presented in mean ± SD with sample size (n). Two participants were
excluded from the 1 min tasks as they were not able to complete the full min of contraction,
therefore a sample size of 18 was used. Paired samples t-tests were performed to compare
torque and sensorimotor responses before (PRE) and following (POST) the passive heating and
cooling protocol to determine whether overall fatigue from the thermal manipulation or
repeated contractions affected responses. The primary comparisons of interest were the
patterns of responses of the force and position tasks to the heating and cooling protocol. All
sEMG data collected were analyzed with a 2-way (Tre x contraction type) repeated measures
analysis of variance (ANOVA). Bonferonni post-hoc corrections were performed for multiple
comparisons where significant main effects were found. Complex interactions were explored
using orthogonal polynomials to evaluate trends within the data, and were performed using
SPSS 23 (SPSS Inc., Chicago, USA). Paired samples t-tests and repeated measures ANOVAs
were performed with GraphPad Prism (version 7.0, GraphPad Software Inc., La Jolla, USA).
The protocol was successful in eliciting the desired manipulations of Tre and Tsk (Fig 3, S1
File). During passive heating, Tre increased > 1.5ÊC for all participants (p < 0.001, n = 20),
from 37.1 ± 0.3ÊC at baseline to 38.8 ± 0.3ÊC; 8 of the 20 participants terminated heating at
this point, while the remaining 12 tolerated 1.9±2.1ÊC increase in Tre. The point at which
participants reached their highest Tre during the heating phase was defined as `H-H'. Tsk was
increased (p < 0.001, n = 20) from 31.2 ± 0.4ÊC to 38.5 ± 0.5ÊC during passive heating (H-H),
and was rapidly cooled to 34.1 ± 1.1ÊC while Tre remained at 38.7 ± 0.4ÊC (H-C).
7 / 17
Fig 3. Rectal temperature (A) and skin temperature responses (B) of mean skin (open squares) and local forearm
temperature (closed squares) responses to the passive heating and cooling protocol. Significantly different versus initial Tre
and Tsk (BASE; p < 0.001). ²Significantly different versus hot Tre, hot Tsk (H-H; p < 0.001).
Maximal voluntary torque and evoked potential responses at baseline (PRE) and following
(POST) the passive heating and cooling protocol are outlined in Table 1. All values were
similar PRE and POST. V-wave peak-to-peak amplitude was also similar for both contraction
types with four participants (p > 0.050, n = 4).
Brief 3-s force and position task
RMS amplitude, mean power frequency (MPF), and median frequency (MDF) for the 3-s
force and position task at each 0.5ÊC Tre interval are presented in Fig 4. Trend analysis of the
RMS amplitude (Fig 4A) revealed a significant quadratic (p < 0.001, n = 20) and cubic
(p = 0.025, n = 20) trend for the isometric force task, whereas no significant trend was
identified for the position task. A significant Tre x contraction interaction was found for RMS
amplitude (p > 0.010, n = 20). Trend analysis of MPF (Fig 4B) and MDF (Fig 4C) for both the
isometric force and position tasks revealed a quadratic trend (all p 0.043, n = 20) with no
other significant trend identified. No Tre x contraction interaction effects were observed for
MPF or MDF (p 0.356, n = 20). For the isometric force task, both MPF and MDF were
higher at H-H compared to BASE (both p 0.002, n = 20).
Sustained 1-min force and position task
RMS amplitude, MPF, and MDF activity for the 1-min sustained tasks are presented in Fig 5.
Trend analysis for RMS amplitude (Fig 5A) revealed no significant trends for the isometric
force task, whereas a significant cubic trend was identified for the position task (p = 0.009,
n = 18). RMS amplitude of the position task was significantly higher at BASE compared to the
isometric force task (p = 0.038, n = 18), but was not significant for all other time points. RMS
amplitude for the position task was lower at H-H compared to BASE (p = 0.038, n = 18). Both
contraction types demonstrated a significant quadratic trend (p 0.004, n = 18) for MPF (Fig
5B); whereas only the isometric force task demonstrated a significant quadratic trend
(p < 0.001, n = 18) for MDF (Fig 5C). For the 1-min task, MPF was significantly higher with
hyperthermia (H-H, H-C) for both the force and position tasks (p 0.003, n = 18) compared
to BASE. MDF was higher at H-C compared to BASE and POST (both p 0.022, n = 18) for
the isometric force task.
This study compared the relative contributions of core and skin thermal afferents on sEMG
responses to isometric force and position contractions during passive hyperthermia. A passive
28.5 ± 11.2
4.6 ± 1.7
0.39 ± 0.17
27.8 ± 10.6
4.5 ± 1.8
0.40 ± 0.16
PLOS ONE | https://doi.org/10.1371/journal.pone.0195219
9 / 17
Fig 4. Electromyographic responses of root-mean-square amplitude (RMS; A), mean power frequency (MPF; B)
and median power frequency (MDF; C) to passive heating and cooling for the 3-second isometric force (closed
bars) and position (open bars) task in 20 participants. Temperature states: initial Tre and Tsk (BASE); hot Tre, hot Tsk
(H-H); hot Tre, cool Tsk (H-C); and end of the protocol where Tre returned to normal and Tsk was cool (POST). Force
task significantly different from position task (p < 0.05). aSignificantly different from baseline (BASE). bSignificantly
different from hot core-hot skin (H-H). cSignificantly different from hot core-cool skin(H-C). dSignificantly different
from end of protocol (POST).
heating and cooling protocol permitted the investigation of sEMG activity from normothermia
(~37.1ÊC) to a hyperthermia of 1.5ÊC at ~0.5ÊC intervals, and then again at 0.5ÊC intervals
during passive cooling back to 37.8ÊC; thus, comparisons could be made at similar core
temperatures with both warm and cool skin temperatures. Similar maximal torque production and
sensorimotor responses before and following the passive heating and cooling protocol suggest
that any changes to sEMG responses would not be a result of central or peripheral fatigue
accumulating over the protocol, but more so directly influenced by the thermal manipulations of Tre
and Tsk. This is further supported by the observed increase in MPF with elevated core
temperatures during the sustained 1-min contractions. During fatigue, it has been shown that MPF
], yet this study revealed the opposite results. Also, similar V-wave
peak-topeak amplitudes during the preliminary analysis on a subset of participants revealed that the
central outflow to the muscle was similar between contraction types, supporting the hypothesis that
any central changes between contractions types were a result of thermal manipulations. A novel
aspect of this study was the use of a submaximal isometric position task as a more representative
model of dynamic muscular contractions compared to maximal isometric force tasks. The
primary finding was that the sEMG RMS amplitude responded differently to progressive
hyperthermia between isometric force and position tasks. Secondly, it appears that Tre and Tsk thermal
afferents were the main contributors to changes in the isometric force task and position task
RMS amplitude, respectively. These data support central nervous system activation being largely
driven by elevations in core temperature during isometric contractions [4±7], whereas decreases
in skin temperature drive decreases in torque output during isokinetic contractions [
Isometric force task
RMS amplitude was used as an indirect measure of motor unit recruitment, firing rate, and
conduction velocity [
]. In this study, trend analysis of the 3 s task revealed a dominant
quadratic pattern of increasing amplitude with passive hyperthermia, supporting previous work
where changes in voluntary activation using interpolated twitches decreased in a quadratic
pattern with increases in core temperature during a 3±5 s MVC [
]. Todd et al. [
] reported that
higher muscle temperature was associated with a ~20% increase in peak relaxation rate and
consequently, higher motor unit firing rates would be required to achieve fusion force. Thus,
the progressive increase in RMS amplitude with increasing Tre during our 3-s submaximal
isometric force task may reflect temperature-induced changes in muscle properties eliciting a
compensatory increase in the recruitment and firing rates of motor units in order to maintain
the target level of force [
MPF and MDF for the isometric force task was elevated with hyperthermia. This suggests a
shift towards higher frequencies driven by a combination of both high core and high skin
temperatures, possibly from an increase in muscle and nerve conduction velocity [
increase could be attributed to sodium and potassium channels opening and closing at faster
rates, causing a decreased amplitude and duration of motor unit action potentials .
However, the combination of increases in power spectrum parameters along with RMS amplitude
may also reflect an increase in relative force output at higher core temperatures [
]. In this
11 / 17
Fig 5. Electromyographic responses of root-mean-square amplitude (RMS; A), mean power frequency (MPF; B)
and median power frequency (MDF; C) to passive heating and cooling for the 1-minute isometric force (black
bars) and position (open bars) task in 18 participants. Temperature states: initial Tre and Tsk (BASE); hot Tre, hot Tsk
(H-H); hot Tre, cool Tsk (H-C); and end of the protocol where Tre returned to normal and Tsk was cool (POST).
aSignificantly different from baseline (BASE). bSignificantly different from hot core-hot skin (H-H). cSignificantly
different from hot core-cool skin(H-C). dSignificantly different from end of protocol (POST).
study, workload was held constant throughout at 60% of MVC at baseline. If MVC output
decreases at high Tre, the relative workload would have progressively increased during each
stage of the heating phase. This would cause an increase in the recruitment of larger diameter
motor units as force level is increased, resulting in increases in both motor unit amplitude and
firing frequency [
Interestingly, during brief 3-s contractions, RMS amplitude did not change during progressive
passive hyperthermia, yet decreases in Tsk upon the initiation of skin cooling±despite elevated
core temperature±elicited increases in RMS amplitude. This concurs with Cheung and Sleivert
], who showed decreases in isokinetic force output when the skin was rapidly cooled while
core temperature remained elevated, and is fundamentally distinct from the isometric force
task. The increase in amplitude was also supported by Winkel and Jorgensen [
reported a two-fold increase in sEMG amplitude during dynamic contractions with skin
cooling compared to warm skin when core temperature remained at a thermoneutral baseline. The
differences in the pattern of the RMS amplitudes during isometric force and position tasks in
response to core and skin temperature changes highlight the distinction between thermal
afferents that drive different modalities of muscular contraction.
MPF and MDF did not significantly change throughout the heating protocol for the 3-s
position task. The only observable difference was an increase in MPF and MDF partway
through the cooling phase (-0.5ÊC) compared to baseline, suggesting that the contribution of
frequencies to the sEMG signal were relatively constant until this point, whereupon there was
a transient shift to higher frequencies. A potential explanation for the trend in spectral
parameters observed for the position task is conduction velocity. Although not measured in this study,
Rutkove et al. [
] reported that with cooler skin temperatures, conduction velocity is slightly
decreased and ion-gated channels remains open for longer periods of time. This allows the
increase in ion flux through the channel, producing a larger depolarization and action
potential. This change in conduction velocity with cooling of the skin may explain the transient shift
towards higher frequencies.
Comparing force and position tasks responses to heating
Differences in the sEMG signal between contraction modalities may lie in central control of
different types of contractions, and their inherent motor unit characteristics [
et al. [
] sought to examine motor unit characteristics between isometric and dynamic
movements with the same torque-time characteristics. This provided a critical methodological
control to isolate central control mechanisms between the two contractions. The authors
demonstrated that the motor unit recruitment threshold was lower in dynamic movements
than in isometric contractions. Therefore, a potential mechanism underlying the stable sEMG
responses of a position task to progressive hyperthermiaÐreflected by RMS amplitude, MPF,
and MDF±is that the lower motor unit recruitment threshold requires less additional motor
unit firing to compensate for changes in muscle properties. While core temperature had no
13 / 17
impact on neural drive during a position task, the rapid decrease in skin temperature during
cooling transiently increased the neural drive required to maintain the submaximal load. This
suggests a significant contribution of skin temperature on motor unit recruitment and firing
rate required for a position task.
The intramuscular pressure generated during an isometric contraction prevents blood flow
to allow for metabolites, interfere with the contractile mechanism, to be cleared [
contrast, blood flow during a dynamic contraction has been shown to be maintained by enhancing
the venous return by the contracting muscle and removing metabolic by-products [
]. It has
been suggested that blood flow occlusion with isometric contractions could result in motor
unit rotation, where the sEMG signal shifts towards higher frequencies as other motor units
become impaired by ischemia [
]. This assumption could explain the progressive increase in
spectral parameters during the force task and the lack of change during the position task.
Two significant methodological controls in this study enabled the comparison of
neuromuscular function between tasks. There were no changes in maximal torque or Mmax amplitude
before and following the protocol, demonstrating that changes to the sEMG signal resulted
from thermal manipulations of Tre and Tsk rather than a result of central or peripheral fatigue
accumulating over the protocol. This project was done in a single session always in the order
of baseline, progressive hyperthermia, and returning to baseline Tre, thus we cannot exclude
the possibility of this order influencing results. However, this progressive heating and cooling
protocol has successfully been employed previous by our lab [
]. These studies have
all demonstrated a significant impairment of maximal voluntary torque with similar levels of
passive hyperthermia as in the present study. To avoid over-stressing participants and overall
neuromuscular fatigue, we did not perform maximal testing at peak hyperthermia; thus, it
remains possible that we did not cause significant hyperthermic impairment. Secondly,
matching the load between the isometric force and position task allowed the comparison of central
control mechanisms of each contraction independent of any changes to the difference in
force/length curves between contraction types [
One reason for selecting the FCR as the tested muscle was because of its thin and superficial
nature, such that there can be a reasonable assumption of a correlation between skin and
muscle temperature. However, as we did not directly measure muscle temperature, and also did
not directly quantify muscle or nerve conduction velocity, the potential effects of
thermallyinduced velocity changes on our frequency data remain speculative.
A young population (22±29 years old) was tested due to changes in motor unit
characteristics with aging, including: lower motor unit discharge rate, decreases in nerve conduction
velocity, and a decrease in the number of motor units present [
]. Therefore, results from this
study may not be generalizable to an older population. It could be assumed that the increase in
neural drive to compensate for changes in muscle properties during heat stress may be
dampened in older individuals due to these changes in motor unit properties.
Peripheral vasodilation and electrode location during passive heat stress increases the
distance between the muscle and the sEMG electrode [
], and could potentially act as additional
tissue filtering, dampening the signal to lower frequencies recorded at the surface of the skin.
Spectral analysis during hyperthermia may need to be interpreted with caution as the
frequency spectrum may be underestimated with additional tissue filtering. However, the
differences between contraction type as core remained elevated with both hot and cool skin is a
strong indication that changes to the sEMG amplitude and spectral parameters were a result of
the experimental manipulations.
14 / 17
This study demonstrated that core temperature was the primary thermal afferent influencing
sEMG amplitude during a brief isometric force task, independent of changes to skin
temperature. In contrast, sEMG amplitude during a position task were not affected by core
temperature increases, but rapidly lowering skin temperature increased sEMG amplitude. These
findings suggest that the neural drive to the muscle during hyperthermia is dependent on both
the task being performed and the relative role of core and skin thermal afferents. Importantly,
this suggests that, for future studies investigating the role of hyperthermia on neuromuscular
function during whole-body dynamic exercise, it is critical to move beyond reliance on
maximal isometric force contractions models and develop and utilize neuromuscular testing
protocols that more closely reflect the unique nature of dynamic contraction.
S1 File. Supplementary individual thermal and neuromuscular data.
We express our gratitude to the participants for their efforts throughout the study, and to D.G.
Stewart for assistance during data collection.
Conceptualization: Nico A. Coletta, Matthew M. Mallette, David A. Gabriel, Christopher J.
Tyler, Stephen S. Cheung.
Data curation: Nico A. Coletta, Matthew M. Mallette, Stephen S. Cheung.
Formal analysis: Nico A. Coletta, Matthew M. Mallette, David A. Gabriel, Christopher J.
Tyler, Stephen S. Cheung.
Funding acquisition: Stephen S. Cheung.
Investigation: Nico A. Coletta, Matthew M. Mallette, Stephen S. Cheung.
Methodology: Nico A. Coletta, Stephen S. Cheung.
Project administration: Nico A. Coletta, Stephen S. Cheung.
Resources: Stephen S. Cheung.
Supervision: David A. Gabriel, Christopher J. Tyler, Stephen S. Cheung.
Validation: Matthew M. Mallette.
Visualization: Matthew M. Mallette.
Writing ± original draft: Nico A. Coletta, Stephen S. Cheung.
Writing ± review & editing: Nico A. Coletta, Matthew M. Mallette, David A. Gabriel,
Christopher J. Tyler, Stephen S. Cheung.
15 / 17
16 / 17
1. Galloway SD , Maughan RJ . Effects of ambient temperature on the capacity to perform prolonged cycle exercise in man . Med Sci Sports Exerc . 1997 ; 29 : 1240 ± 1249 . PMID: 9309637
2. Ely MR , Cheuvront SN , Roberts WO , Montain SJ . Impact of weather on marathon-running performance: Med Sci Sports Exerc . 2007 ; 39 : 487 ± 493 . https://doi.org/10.1249/mss.0b013e31802d3aba PMID: 17473775
3. Cheung SS , Sleivert GG . Multiple triggers for hyperthermic fatigue and exhaustion . Exerc Sport Sci Rev . 2004 ; 32 : 100 ± 106 . PMID: 15243205
4. Morrison S , Sleivert GG , Cheung SS . Passive hyperthermia reduces voluntary activation and isometric force production . Eur J Appl Physiol . 2004 ; 91 : 729 ± 736 . https://doi.org/10.1007/s00421-004-1063-z PMID: 15015001
5. Nybo L , Nielsen B . Hyperthermia and central fatigue during prolonged exercise in humans . J Appl Physiol . 2001 ; 91 : 1055 ± 1060 . https://doi.org/10.1152/jappl. 2001 . 91 .3.1055 PMID: 11509498
6. Thomas MM , Cheung SS , Sleivert GG , Elder GE . Voluntary muscle activation is impaired by core temperature rather than local muscle temperature . J Appl Physiol . 2006 ; 100 : 1361 ± 1369 . https://doi.org/ 10.1152/japplphysiol.00945. 2005 PMID: 16339343
7. Todd G , Butler JE , Taylor JL , Gandevia SC . Hyperthermia: a failure of the motor cortex and the muscle . J Physiol . 2005 ; 563 : 621 ± 631 . https://doi.org/10.1113/jphysiol. 2004 .077115 PMID: 15613373
8. Cheung SS . Hyperthermia and voluntary exhaustion: integrating models and future challenges . Appl Physiol Nutr Metab . 2007 ; 32 : 808 ± 817 . https://doi.org/10.1139/H07-043 PMID: 17622299
9. HaÈkkinen K. Neuromuscular fatigue in males and females during strenuous heavy resistance loading . Electromyogr Clin Neurophysiol . 1994 ; 34 : 205 ± 214 . PMID: 8082606
10. Mero A , Luhtanen P , Viitasalo JT , Komi PV . Relationships between the maximal running velocity, muscle fiber characteristics, force production and force relaxation of sprinters . Scand J Sports Sci . 1981 ; 3 : 16 ± 22 .
11. Hunter SK , Ryan DL , Ortega JD , Enoka RM . Task differences with the same load torque alter the endurance time of submaximal fatiguing contractions in humans . J Neurophysiol . 2002 ; 88 : 3087 ± 3096 . https://doi.org/10.1152/jn.00232. 2002 PMID: 12466432
12. Hunter SK , Lepers R , MacGillis CJ , Enoka RM . Activation among the elbow flexor muscles differs when maintaining arm position during a fatiguing contraction . J Appl Physiol . 2003 ; 94 : 2439 ± 2447 . https:// doi.org/10.1152/japplphysiol.01038. 2002 PMID: 12547844
13. Nakazawa K , Kawakami Y , Fukunaga T , Yano H , Miyashita M. Differences in activation patterns in elbow flexor muscles during isometric, concentric and eccentric contractions . Eur J Appl Physiol . 1993 ; 66 : 214 ± 220 .
14. Nardone A , Romanò C , Schieppati M. Selective recruitment of high-threshold human motor units during voluntary isotonic lengthening of active muscles . J Physiol . 1989 ; 409 : 451 ± 471 . PMID: 2585297
15. Murphy AJ , Wilson GJ . Poor correlations between isometric tests and dynamic performance: relationship to muscle activation . Eur J Appl Physiol . 1996 ; 73 : 353 ± 357 .
16. Murphy AJ , Wilson GJ , Pryor JF . Use of the iso-inertial force mass relationship in the prediction of dynamic human performance . Eur J Appl Physiol . 1994 ; 69 : 250 ± 257 .
17. Morrison SA , Sleivert GG , Cheung S. Aerobic influence on neuromuscular function and tolerance during passive hyperthermia: Med Sci Sports Exerc . 2006 ; 38 : 1754 ± 1761 . https://doi.org/10.1249/01.mss. 0000230120 .83641.98 PMID: 17019297
18. Morrison SA , Sleivert GG , Neary JP , Cheung SS . Prefrontal cortex oxygenation is preserved and does not contribute to impaired neuromuscular activation during passive hyperthermia . Appl Physiol Nutr Metab . 2009 ; 34 : 66 ± 74 . https://doi.org/10.1139/H08-139 PMID: 19234587
19. Racinais S , Oksa J . Temperature and neuromuscular function . Scand J Med Sci Sports . 2010 ; 20 : 1± 18 . https://doi.org/10.1111/j.1600- 0838 . 2010 . 01204 . x PMID : 21029186
20. Rutkove SB , Kothari MJ , Shefner JM . Nerve, muscle, and neuromuscular junction electrophysiology at high temperature . Muscle Nerve . 1997 ; 20 : 431 ± 436 . PMID: 9121500
21. Rutkove SB . Effects of temperature on neuromuscular electrophysiology . Muscle Nerve . 2001 ; 24 : 867 ± 882 . PMID: 11410914
22. Cheung SS , Sleivert GG . Lowering of skin temperature decreases isokinetic maximal force production independent of core temperature . Eur J Appl Physiol . 2004 ; 91 : 723 ± 728 . https://doi.org/10.1007/ s00421-004 -1062-0 PMID: 15015000
23. Green LA , McGuire J , Gabriel DA . Flexor carpi radialis surface electromyography electrode placement for evoked and voluntary measures: novel FCR electrode placement . Muscle Nerve . 2015 ; 52 : 818 ± 825 . https://doi.org/10.1002/mus.24631 PMID: 25736453
24. Roman-Liu D. The influence of confounding factors on the relationship between muscle contraction level and MF and MPF values of EMG signal: a review . Int J Occup Saf Ergon . 2016 ; 22 : 77 ± 91 . https:// doi.org/10.1080/10803548. 2015 .1116817 PMID: 26654476
25. Bigland-Ritchie B , Woods JJ . Changes in muscle contractile properties and neural control during human muscular fatigue . Muscle Nerve . 1984 ; 7 : 691 ± 699 . https://doi.org/10.1002/mus.880070902 PMID: 6100456
26. Bigland-Ritchie B. EMG /force relations and fatigue of human voluntary contractions . Exerc Sport Sci Rev . 1981 ; 9 : 75 ± 117 . PMID: 6749525
27. Kent-Braun JA , Fitts RH , Christie A . Skeletal muscle fatigue . Compr Physiol . 2012 ; 2 : 997 ± 1044 . https://doi.org/10.1002/cphy.c110029 PMID: 23798294
28. Winkel J , Jørgensen K. Significance of skin temperature changes in surface electromyography . Eur J Appl Physiol . 1991 ; 63 : 345 ± 348 .
29. Bell DG . The influence of air temperature on the EMG/force relationship of the quadriceps . Eur J Appl Physiol . 1993 ; 67 : 256 ± 260 .
30. Jackson AS , Pollock ML . Generalized equations for predicting body density of men . Br J Nutr . 1978 ; 40 : 497 ± 504 . PMID: 718832
31. Jackson AS , Pollock ML , Ward A . Generalized equations for predicting body density of women . Med Sci Sports Exerc . 1980 ; 12 : 175 ± 181 . PMID: 7402053
32. Casa DJ , Armstrong LE , Hillman SK , Montain SJ , Reiff RV , Rich BS , et al. National athletic trainers' association position statement: fluid replacement for athletes . J Athl Train . 2000 ; 35 : 212 ± 224 . PMID: 16558633
33. Ramanathan NL. A new weighting system for mean surface temperature of the human body . J Appl Physiol . 1964 ; 19 : 531 ± 533 . https://doi.org/10.1152/jappl. 1964 . 19 .3.531 PMID: 14173555
34. Green LA , Parro JJ , Gabriel DA . Quantifying the familiarization period for maximal resistive exercise . Appl Physiol Nutr Metab . 2014 ; 39 : 275 ± 281 . https://doi.org/10.1139/apnm-2013 -0253 PMID: 24552367
35. Watanabe K , Akima H . Neuromuscular activation of vastus intermedius muscle during fatiguing exercise . J Electromyogr Kinesiol . 2010 ; 20 : 661 ± 666 . https://doi.org/10.1016/j.jelekin. 2010 . 01 .003 PMID: 20133154
36. Enoka RM . Activation order of motor axons in electrically evoked contractions . Muscle Nerve . 2002 ; 25 : 763 ± 764 . https://doi.org/10.1002/mus.10117 PMID: 12115963
37. Dimitrov GV , Arabadzhiev TI , Hogrel J-Y , Dimitrova NA. Simulation analysis of interference EMG during fatiguing voluntary contractions. Part IIÐchanges in amplitude and spectral characteristics . J Electromyogr Kinesiol . 2008 ; 18 : 35 ± 43 . https://doi.org/10.1016/j.jelekin. 2006 . 07 .002 PMID: 16963280
38. Moritani T , Muro M. Motor unit activity and surface electromyogram power spectrum during increasing force of contraction . Eur J Appl Physiol . 1987 ; 56 : 260 ± 265 .
39. Buchanan TS , Lloyd DG . Muscle activity is different for humans performing static tasks which require force control and position control . Neurosci Lett . 1995 ; 194 : 61 ± 64 . PMID: 7478214
40. Ivanova T , Garland SJ , Miller KJ . Motor unit recruitment and discharge behavior in movements and isometric contractions . Muscle Nerve . 1997 ; 20 : 867 ± 874 . PMID: 9179159
41. Masuda K , Masuda T , Sadoyama T , Inaki M , Katsuta S. Changes in surface EMG parameters during static and dynamic fatiguing contractions . J Electromyogr Kinesiol . 1999 ; 9 : 39 ± 46 . PMID: 10022560
42. Fallentin N , Jørgensen K , Simonsen EB . Motor unit recruitment during prolonged isometric contractions . Eur J Appl Physiol . 1993 ; 67 : 335 ± 341 .
43. Hunter SK , Pereira HM , Keenan KG . The aging neuromuscular system and motor performance . J Appl Physiol . 2016 ; 121 : 982 ± 995 . https://doi.org/10.1152/japplphysiol.00475. 2016 PMID: 27516536