Content uploaded by Alison R Buick
Author content
All content in this area was uploaded by Alison R Buick on Nov 26, 2019
Content may be subject to copyright.
J Physiol 00.0 (2019) pp 1–25 1
The Journal of Physiology
TOPICAL REVIEW
Neuromuscular electrical stimulation-promoted plasticity
of the human brain
Richard G. Carson1,2,3 and Alison R. Buick2
1Trinity College Institute of Neuroscience and School of Psychology, Trinity College Dublin, Dublin 2, Ireland
2School of Psychology, Queen’s University Belfast, Belfast BT7 1NN, UK
3School of Human Movement and Nutrition Sciences, University of Queensland, Brisbane, QLD 4072, Australia
Edited by: Ole Petersen & Diego Contreras
Richard G. Carson is Chair in Cognitive Neuroscience of Ageing in the School of Psychology and
the Institute of Neuroscience at Trinity College Dublin. He grew up near Belfast, graduated from the
University of Bristol, and was subsequently awarded his Ph.D. by Simon Fraser University in 1993. He
then held a series of research fellowships at the University of Queensland, before moving to Queen’s
University Belfast in 2006, and to Trinity College Dublin in 2011. His research focuses upon human
brain plasticity, with a particular emphasis upon changes that occur across the lifespan. Alison R. Buick
received her Ph.D. degree from the Queen’s University Belfast, Northern Ireland. Her doctoral work was
focused on novel interventions using electrical stimulation and physical therapy for stroke survivors.
Her current research interests are in the integration of technology with healthcare, particularly the use
of in-home, mobile EEG.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society DOI: 10.1113/JP278298
2 R. G. Carson and A. R. Buick J Physiol 00.0
Abstract The application of neuromuscular electrical stimulation (NMES) to paretic limbs has
demonstrated utility for motor rehabilitation following brain injury. When NMES is delivered
to a mixed peripheral nerve, typically both efferent and afferent fibres are recruited. Muscle
contractions brought about by the excitation of motor neurons are often used to compensate
for disability by assisting actions such as the formation of hand aperture, or by preventing
others including foot drop. In this context, exogenous stimulation provides a direct substitute for
endogenous neural drive. The goal of the present narrative review is to describe the means through
which NMES may also promote sustained adaptations within central motor pathways, leading
ultimately to increases in (intrinsic) functional capacity. There is an obvious practical motivation,
in that detailed knowledge concerning the mechanisms of adaptation has the potential to inform
neurorehabilitation practice. In addition, responses to NMES provide a means of studying CNS
plasticity at a systems level in humans. We summarize the fundamental aspects of NMES, focusing
on the forms that are employed most commonly in clinical and experimental practice. Specific
attention is devoted to adjuvant techniques that further promote adaptive responses to NMES
thereby offering the prospect of increased therapeutic potential. The emergent theme is that
an association with centrally initiated neural activity, whether this is generated in the context
of NMES triggered by efferent drive or via indirect methods such as mental imagery, may in
some circumstances promote the physiological changes that can be induced through peripheral
electrical stimulation.
(Received 12 May 2019; accepted after revision 16 August 2019; first published online 0 xxxx 2019)
Corresponding author R. G. Carson: Trinity College Institute of Neuroscience and School of Psychology, Trinity College
Dublin, Dublin 2, Ireland. Email: richard.carson@tcd.ie
Abstract figure legend The delivery of electrical current via a peripheral nerve (or across a muscle belly) activates
contractile muscle fibres indirectly by depolarizing motor axons (1b). As the sensory axons in the same mixed nerve
bundle have lower activation thresholds, ascending afferent volleys are also generated at intensities of electrical
stimulation that exceed themotor threshold (1a). These volleys are followed by (secondary) reafference arising from the
invoked muscle contraction (2). The goal of this review is to address the means through which the sensory-mediated
consequences of the stimulation alter the state of ‘sensory’ networks, and induce sustained ‘neuroplastic’ modifications
within central ‘motor’ networks. Figure redrawn and adapted from the author’s original artwork, which is available
at: https://commons.wikimedia.org/wiki/File:Neuromuscular_electrical_stimulation_promoted_brain_plasticity.jpg
(original figure published under a Creative Commons Attribution-Share Alike 4.0 International license).
Background
Although historical antecedents are often ascribed to
Galvani’s Commentarius, published in the late 18th
century, the practice of employing electricity to stimulate
human nerves can be traced to ancient times (Finger
& Piccolino, 2011). Murals depicting the Nile catfish,
Malopterurus electricus,havebeendiscoveredinEgyptian
tombs dating from the Fifth Dynasty (around 2400 BCE).
In extant records, however, it is not until 46 BCE that
the utilization of the (saltwater) torpedo ray’s electric
discharge for electrotherapy is noted by the Roman
physician Scribonius Largus (Cambiaghi & Sconocchia,
2018). Writing some 30 years later, Dioscorides (see
Gunther, 1934) provided perhaps the first explicit
reference to the use of the torpedo’s electric discharge
for artificial muscle stimulation in relating a remedy
for propalsus ani (Kellaway, 1946). The introduction of
the Leyden jar in 1746 provided a platform for the
modern progression of electrotherapy, with Benjamin
Franklin observing in 1774 that muscle contractions
could be brought about by exposure to static electricity
(Isaacson, 2003). Subsequently, Faraday’s application of
the principle of magnetic induction provided a means
of delivering electric current to the human body in a
controlled fashion – for which the term ‘Faradization’
was coined. Prominent among 19th century practitioners
investigating the physiology of ‘localized electrization’,
Duchenne de Boulogne employed a Faradic stimulating
machine to stimulate a wide range of muscles trans-
cutaneously via a pair of ‘humid rheophore’ electrodes
(e.g. Clarac et al. 2009). Performing his studies in cat
and monkey, Sherrington (1894) observed that a third
to one-half of the myelinated fibres of peripheral nerves
failed to degenerate following section of their ventral
(motor) spinal roots. As the application of maximal
Faradic currents to these remaining fibres failed to elicit
‘motor reactions’, he concluded that they must provide
sensory innervation. The presence of both sensory and
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 3
motor axons in the same (‘mixed’) nerve bundle, as
revealed by Sherrington, is a key factor determining the
physiological effects of contemporary forms of neuro-
muscular electrical stimulation (NMES).
The Greek name for the torpedo ray, nark`
e, meaning
numbness, suggests the nature of the initial therapeutic
applications of electrotherapy (Debru, 2006). Scribonius
Largus, for example, records use of the torpedo’s electric
discharge as a treatment for the pain associated with intra-
ctable headache and gout (Kellaway, 1946). In the guise
of ‘transcutaneous electrical nerve stimulation’ (TENS),
modern devices designed to achieve the same analgesic
goals are now widely available. These typically generate
high frequency (>50 Hz) trains of electrical stimulation,
at current intensities that are insufficient to evoke overt
motor responses. A contemporaneous historical lineage
for the therapeutic application of electrotherapy in motor
rehabilitation can also be traced – from Dioscorides
through Duchenne de Boulogne to the present day. Modes
of electrical nerve stimulation used for this purpose (which
tend to differ from those employed typically for pain
relief – by using lower frequencies and higher intensities
of stimulation) constitute the subject matter of the present
review (Table 1).
In a contemporary therapeutic context, applications of
NMES in motor rehabilitation can be conceived of as being
adaptive or restorative (Pomeroy et al. 2011). The term
functional electrical stimulation (FES) refers typically to
instances in which tetanic muscle contractions are induced
to assist or reinstate movement, thereby enabling an
otherwise quiescent limb to be engaged in goal-directed
actions. This form of stimulation is deemed to be adaptive,
as it provides direct compensation for the motor disability.
In the period since Liberson and colleagues (1961)
demonstrated that stimulation delivered to the common
peroneal nerve reduced the degree of foot-drop during
the swing phase of gait, numerous applications of FES
have been developed successfully to assist movement of the
upper and lower extremities (Prochazka, 2018). Yet NMES
may also be used restoratively, with a view to promoting
neural changes that lead ultimately to increased (intrinsic)
functional capacity. This is the primary focus of the current
review.
The delivery of electrical current to neuromuscular
tissue (i.e. via a peripheral nerve or across a muscle
belly) activates contractile muscle fibres indirectly by
first depolarizing motor axons. As the sensory axons
in the same mixed nerve bundle have lower activation
thresholds, ascending afferent volleys are also generated at
intensities of electrical stimulation that exceed the motor
threshold (MT) (Dawson, 1956). These are followed by
(secondary) reafference arising from the invoked muscle
contraction. While the capacity of NMES to provide
a direct substitute for (descending) endogenous neural
drive to muscles in circumstances of CNS injury or
disease can be readily appreciated, our goal is to address
means through which the sensory-mediated consequences
of NMES induce sustained ‘neuroplastic’ modifications
within central motor pathways.
Given an empirical literature that is characterized by
extraordinary diversity with respect to the stimulation
protocols that are employed (varying in relation to such
features as stimulation frequency, intensity, duration and
temporal pattern), there is little consensus with respect
to the cellular mechanisms engaged by NMES. Beyond
providing insights in relation to the expression of CNS
plasticity at a systems level in humans, there is an
obvious practical motivation for seeking the elucidation
of these processes. Detailed knowledge concerning the
mechanisms of adaptation clearly has the potential to
inform the development of neurorehabilitation practice.
Scope of the review
While the intent of this narrative review is to examine
general principles, the scope of the analysis is necessarily
restricted – for the most part to the effects of trans-
cutaneous (surface) electrical stimulation delivered using
intensities at or above the threshold for a motor response.
The emphasis is largely upon the upper limb, and upon
supraspinal adaptations (cf. Bergquist et al. 2011). To the
extent that specific clinical applications are considered,
these will generally be drawn from the domain of stroke
rehabilitation.
Evidently NMES exhibits the capacity to generate
changes in the excitability of descending (e.g. cortico-
spinal) projections from the cortex to the spinal cord
(Chipchase et al. 2011a). It has generally been assumed
that such changes in excitability reflect, at least in
part, modifications in the organization of the same
brain networks that serve ultimately as a basis for the
improvements in functional capacity that may be brought
about by neuromuscular stimulation (Traversa et al. 1997;
Van g et al. 1999; Barker et al. 2012). Although, as we
shall see, there are grounds to be cautious about such
assumptions (Carson et al. 2016), we include a survey
of studies that have characterized the neurophysiological
effects of NMES in terms of corticospinal excitability.
Most often these have been assessed through muscle
responses evoked by transcranial magnetic stimulation
(TMS) delivered over primary motor cortex (M1). We also
consider instances in which the effects of NMES have been
registered using various brain imaging methodologies. In
the closing sections, we return to the issue of whether
theneuralpathwaysuponwhichNMEShasthemost
readilydetectableeffectsarenecessarilyalsothosethatplay
an instrumental role in mediating changes in functional
capacity.
In the course of the review, specific attention is devoted
to adjuvant techniques that further promote restorative
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
4 R. G. Carson and A. R. Buick J Physiol 00.0
Table 1. Common variants of peripheral electrical stimulation
Type of
stimulation Typical intent
Typical
frequency
range Typical intensity
NMES Activation of sensory and motor axons for diverse purposes 1–100 Hz At or above motor threshold
FES Activation of both sensory and motor axons with the specific goal
of assisting motor function
20–60 Hz Above motor threshold
EST Activation of both sensory and motor axons with the specific goal
of preventing muscle weakness
35–100 Hz Above motor threshold
TENS Activation of sensory axons for the goal of pain relief. >50 Hz Below motor threshold
EST, electrostimulation strength training; FES, functional electrical stimulation; NMES, neuromuscular electrical stimulation; TENS,
transcutaneous electrical nerve.
responses to NMES. The emergent theme is that an
association with centrally initiated neural activity, whether
this is generated in the context of NMES triggered by
efferent drive or via indirect methods such as mental
imagery, can in some circumstances be efficacious in
promoting neural adaptations upon which changes in
functional capacity may be based.
Exemplars
We do not seek to be comprehensive with respect to the
characteristics of NMES that can be altered in either an
experimental or a clinical context. Rather, the empirical
literature is circumscribed with a view to emphasizing a
limited number of key concepts. It being evident that the
‘dose’ of NMES has a significant bearing on the changes in
brain activity thus invoked, we consider both protocols in
which the level of stimulation is just above motor threshold
and those in which it is of sufficient magnitude to elicit
overt movement.
Stimulation at motor threshold intensity. Sensory axons
in a mixed nerve bundle innervating skeletal muscle are
typically depolarized at levels of electrical stimulation
below those which are necessary to recruit motor axons
(Panizza et al. 1989, 1992; Veale et al. 1973). At intensities
of NMES at or above MT, therefore, ascending afferent
volleys will be generated directly by the depolarization
of sensory axons (e.g. Collins, 2007). Some degree of
secondary reafference arising (indirectly) from the invoked
muscle contraction will follow. While the nature and the
extent of the reafference will in turn be determined by
the characteristics of the joint movement thus induced
(which will itself be influenced by the posture of the limb,
degree of restraint and so on), a more general point is
that the relationships between the intensity of stimulation
and the level (and distribution) of brain activity arising
from (1) the direct sensory afference and (2) the indirect
secondary reafference are unlikely to be the same. Indeed,
both are context dependent and must be determined
empirically. Their relative contributions notwithstanding,
it is the sensory corollaries of NMES that provide the
principal means by which sustained (central) neuroplastic
adaptations are induced (Bergquist et al. 2011).
If the magnitude of a single electrical stimulus
delivered transcutaneously to a peripheral nerve is set
to approximately three times perceptual threshold, direct
motor responses in the innervated muscles are typically
observed (e.g. Ridding et al. 2001; McKay et al. 2002;
Litvak et al. 2007). At such intensities, extended (up
to 2 h) sequences of stimulation are necessary to bring
about sustained increases in the excitability of cortico-
spinal projections to the muscles in which the responses
are evoked (see also Luft et al. 2002). For example, Ridding
et al. (2000) delivered trains of pulses (10 Hz, 1 ms pulse
width)totheulnarnerveatthewrist,atarateofonetrain
per second, using a 50% duty cycle (i.e. 1 s on, 1 s off), for a
period of 2 h. The area of the scalp over which TMS elicited
MEPs in the ulnar nerve-innervated first dorsal inter-
osseus (FDI) and abductor digiti minimi (ADM) muscles
increased as a consequence of the intervention. Using
precisely the same protocol, Kaelin-Lang et al. (2002)
obtained increases in the amplitude of MEPs elicited in
ADM (but not in FDI). As these were not accompanied by
corresponding changes in the size of potentials evoked
by stimulation by corticospinal axons at the level of
the cervicomedullary junction, a cortical locus for the
adaptation was inferred (see also Ridding et al. 2000).
The capricious nature of the changes in corticospinal
excitability induced using these stimulation durations and
intensities is emphasized by the wide variation in response
across individuals reported by Charlton et al. (2003), when
FDI afferents were stimulated via the skin overlying the
muscle, rather than via the nerve trunk at the wrist (using
a protocol that was otherwise equivalent). Furthermore, if
the frequency at which the trains are delivered and the total
duration of the intervention is reduced, reliable elevations
of MEP amplitude are not obtained (Uy & Ridding, 2003).
If, however, the effective dose (if not the specificity) of
NMES is increased by delivering pulses simultaneously to
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 5
both the radial and ulnar nerves, a progressive increase
in the amplitude of potentials evoked in FDI occurs over
the time course of the intervention (McKay et al. 2002).
Furthermore, this dual stimulation technique increases
reliably both the area of the scalp over which TMS-elicited
MEPs can be obtained in FDI (and other hand muscles)
and the amplitude of the MEPs recorded following the
cessation of NMES (Ridding et al. 2001). Indeed, when
motor point stimulation is delivered simultaneously to
FDI and ADM via the skin overlying the muscles, an
intervention of 1 h duration is sufficient to induce
reliable increases in corticospinal excitability (Schabrun
& Ridding, 2007; cf. Charlton et al. 2003). As there
are no accompanying changes in the size of responses
elicited by cervicomedullary stimulation, a spinal locus
for the adaptation appears to be precluded (Ridding et al.
2001).
Stimulation at supra-motor threshold intensities. FES
typically comprises short bursts of electrical pulses
delivered at a frequency above that necessary to yield
a fused contraction (12 Hz) (Peckham & Knutson,
2005; Sheffler & Chae, 2007). The assumption that given
an adequate dose of NMES persistent elevations in the
excitability of corticospinal projections can be induced
is supported by studies that have employed stimulation
at an intensity and frequency sufficient to induce tetanic
motor responses (see Chipchase et al. 2011afor a review).
While it is not possible to exclude the possibility that
such supra-threshold intensity stimulation generates anti-
dromic impulses that modify synapses in the ventral horn
(Rushton, 2003), the consensus view is that the observed
changes in corticospinal excitability are driven primarily
by cortical reorganization (e.g. Luft et al. 2005).
For example, Schabrun et al. (2012) applied 30 min
of NMES to the skin overlying the abductor pollicis
brevis (APB) muscle at 30 Hz (4 s on, 6 s off) with
six periods of stimulation being applied every minute.
The intensity of stimulation was that which produced a
mid-range abduction of the thumb. The amplitudes of
MEPsevokedinAPBfollowingtheinterventionwere
substantially greater than those obtained prior to the
stimulation. Corresponding effects have been reported
when biceps brachii is the target of stimulation (Chipchase
et al. 2011b). When NMES is applied to APB in this
mannerforperiodsof20or40min,theinducedchangesin
corticospinal excitability are maintained for at least 20 min
following the cessation of the intervention (Andrews et al.
2013).
While it is clear that increases in the dose of stimulation
that is administered may be achieved by increases in the
current/voltage of individual shocks, and/or by a higher
frequency of delivery, it has been proposed (Chipchase
et al. 2011b) that increases in corticospinal reactivity are
generated reliably only by those forms of NMES giving
rise to a motor response that mimics a voluntary muscle
contraction. As noted previously, in addition to the initial
ascending afferent volley induced directly by electrical
stimulation of the nerve, such protocols encapsulate
secondary reafference arising from the muscle contra-
ctions (Schabrun et al. 2012). The extent of the neural
activity induced in M1 by such reafference can be sub-
stantially greater than that brought about directly by the
ES-mediated depolarization of the sensory axons (Shitara
et al. 2013). De Kroon and colleagues (2005) in their
review of the relationships between electrical stimulation
characteristics and clinical outcomes hypothesized that
supra-motor stimulation is more likely than sub-motor
stimulation to lead to improvements in motor control, as
a consequence of muscle and joint afferent feedback, i.e. in
addition to that derived from cutaneous afferents, which
are also engaged at lower intensities of stimulation.
Indeed, repeated changes in muscle length brought
about passively by mechanical joint rotation also induce
both acute (Lewis et al. 2001) and chronic (Mac´
eet al.
2008) increases in corticospinal excitability. Collectively,
these observations suggest that the secondary mediation
of Ia (muscle spindle) afferent projections to higher
brain centres is instrumental in augmenting the direct
depolarizing effects of NMES. Although it has been
proposed that cutaneous afferents make a greater
contribution than muscle spindle afferents to cortical
potentials produced by electrical stimulation of mixed
nerves in the upper limb (e.g. Halonen et al. 1988; Allison
et al. 1991), it is the precise brain circuits that exhibit a
change in state as a result of peripheral stimulation which
is likely to assume particular functional significance. It is
believed that Ia afferent input has its most direct effects
upon both area 4 (primary motor cortex) (Jones & Porter,
1980) and area 3a (in primary somatosensory cortex)
(Heath et al. 1976; Hore et al. 1976), whereas, input from
cutaneous receptors and low threshold mechanoreceptors
first alters the excitability of neurons in areas 3b and 1
(Kaas & Pons, 1988). We thus turn our attention to the
brain circuitry that is engaged by NMES, and to the impact
of its parametric variation.
Brain circuitry engaged by NMES
Somatosensory cortex. On the basis of findings derived
using a variety of neuroimaging techniques, it has been
surmised that electrical stimulation of peripheral afferents
engages circuits in the primary somatosensory cortex
(S1 – including Brodmann areas 3, 1 and 2) within the
postcentral gyrus, the second somatosensory area (S2 –
including parts of Brodmann areas 40 and 43) within the
parietal operculum on the ceiling of the lateral sulcus,
and the posterior parietal cortex (Korvenoja et al. 1999;
Boakye et al. 2000; Nihashi et al. 2005). In relation to the
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
6 R. G. Carson and A. R. Buick J Physiol 00.0
complex cortical responses that are extracted from electro-
encephalographic (EEG) and magnetoencephalographic
(MEG) recordings, there is consensus that short-latency
potentials occurring within the first 40 ms following
stimulation of the median nerve (e.g. at the wrist)
at intensities sufficient to elicit a muscle twitch arise
principally from contralateral S1 (Allison et al. 1991). The
presence of synchronized neuronal population activity
in S2 (registered by MEG) during this period, while
consistent with an influence of cortical afferents from S1,
does not, however, preclude the possibility of mediation
via additional parallel thalamocortical projections to S2
(Karhu & Tesche, 1999). With respect to the medium
latency (>40 ms) components, there is a distributed
pattern of activation that includes not only S1, but also
S2 bilaterally and contralateral posterior parietal cortex
(Hari et al. 1984; Allison et al. 1989a, 1989b, 1992; Forss
et al. 1994). It is currently believed that cortico-cortical
connections mediated by transcallosal projections play a
major role in shaping the bilateral character of the S2
response profile (Del Vecchio et al. 2019). These sources
continue to be active simultaneously during a period
70–140 ms following the onset of stimulation (Maugui`
ere
et al. 1997). When a sequence of stimuli is administered,
the offset of the sequence gives rise to a (P100 and N140)
stimulus evoked potential (SEP) signature distinct from
that associated with the individual stimuli (Yamashiro
et al. 2008, 2009).
The functional magnetic resonance imaging
(fMRI)-derived blood oxygenation level-dependent
(BOLD) response measured in contralateral S1 scales with
the intensity of ES (at least up to MT) (Krause et al. 2001,
see also Nelson et al. 2004). In contrast, bilateral activity
evident in S2 and posterior parietal cortex does not
appear to vary in this manner. A BOLD signal is, however,
registered in S2 at lower levels of stimulation than in S1.
This is augmented when attention is directed explicitly
to the stimulation (Backes et al. 2000). In circumstances
in which ES is applied in a range between the sensory
threshold (ST) and 1.2 ×MT, the amplitude of the
N9, N20 and N20-P25 SEP components derived from
EEG recordings increases in proportion to stimulation
intensity (Gatica Tossi et al. 2013; cf. Lakhani et al. 2012).
This effect remains present at 2.5 ×MT (Urasaki et al.
1998). Components of the SEP recorded in S1 saturate
at a level below the pain threshold (Parain & Delapierre,
1991), while the asymptote of the S2 response occurs at
lower stimulation intensities than for the S1 response (Lin
et al. 2003).
It is now broadly accepted that the initial (i.e. N20)
EEG responses to NMES are dominated by cutaneous
afferent input (Gandevia & Burke, 1990; Kunesch et al.
1995). The origin of the N20 response to cutaneous
inputs is considered to be a deep tangential generator
in area 3b (e.g. Desmedt & Ozaki, 1991; McLaughlin
& Kelly, 1993), whereas, it is probable that the source
generator for cortical potentials invoked by muscle spindle
afference is principally area 3a, although additional
contributions from area 2 cannot be excluded (Mima et al.
1996; MacKinnon et al. 2000). This is consonant with
evidence drawn from comparative studies that that the
most significant input to area 3a is from muscle spindle
afferents (Kaas, 1983). Thus surface electrical stimulation
at intensities above motor threshold will give rise to
cutaneous afferent-mediated activity in area 3b of primary
somatosensory cortex (S1), and also to activity in area
3a and area 2 (Wiesendanger & Miles, 1982), including
that arising by virtue of muscle contraction-induced
reafference.
Cortico-cortical connections from somatosensory cortex
to M1. Studies in cat indicate that stimulation of sensory
cortex can induce long-lasting potentiation of synaptic
potentials evoked in the motor cortex (Sakamoto et al.
1987). Detailed investigations in non-human primates
(e.g. Jones et al. 1978; Pons and Kaas, 1986; Ghosh et al.
1987; Huerta and Pons, 1990) and in cat (Grant et al.
1975; Zarzecki et al. 1978; Waters et al. 1982; Burton
and Kopf, 1984; Yumiya and Ghez, 1984; Porter and
Sakamoto, 1988; Avenda˜
no et al. 1992; Schwark et al.
1992) have revealed extensive networks of cortico-cortical
connections between SI and primary motor cortex (M1)
(Burton & Fabri, 1995). Neurons that exhibit short-latency
excitatory postsynaptic potentials (EPSPs), indicative of
direct input, in response to microstimulation of area 3a,
are found in all laminae of the motor cortex, with the
exception of layer I (Herman et al. 1985; Huerta & Pons,
1990; Porter et al. 1990). By comparison, only cells in
the superficial layers of M1 (II and III) respond in this
fashion to stimulation of area 2 (Kosar et al. 1985; Porter
et al. 1990). It has thus been proposed that area 3a should
be viewed as a relay to motor cortex (Jones & Porter,
1980), or even as a part of area 4 (Jones et al. 1978, cf.
Kuehn et al. 2017). This intimacy of association provides a
means through which muscle spindle input that is relayed
through area 3a can exert a direct influence on pyramidal
and multipolar neurons in deep (V and VI) layers of M1
(Porter et al. 1990). In contrast, while there are reciprocal
connections between area 3b and area 1 in particular, and
further projections to area 2 (which are ostensibly not
reciprocated), projections from area 3b to M1 are sparse
(Darian-Smith et al. 1993; Burton & Fabri, 1995), if indeed
detectable (Jones et al. 1978).
Cerebello-thalamo-cortical and thalmo-cortical connec-
tions. Although the possibility of direct activation of
the primary motor cortex via sensory afferents from the
periphery (Padel & Relova, 1991) cannot be excluded,
studies in non-human primates indicate that the ventral
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 7
posterior complex of the thalamus, the major sensory
thalamic relay, has relatively few direct projections to
M1 (Darian-Smith & Darian- Smith, 1993; Huffman &
Krubitzer, 2001a). In this regard, it is worth noting that
while S1 areas 1, 2 and 3b are represented acrossthe ventro-
basal complex of the thalamus, area 3a has connectional
relationships similar to those for area 4 (Jones et al. 1979).
For example, area 3a receives projections from nuclei of
the thalamus classically associated with the motor system,
including indirect input from the cerebellum and basal
ganglia via the ventral lateral (VL) nucleus (Huffman &
Krubitzer, 2001b). Thalamic processing of somatosensory
input extends beyond the relaying of primary afferent
signals to the cortex. For example, at levels of ES above
perceptual threshold, thalamic SEPs can be elicited over
intervals greater than 75 ms following the peripheral
shock, with the duration extending to 150 ms when the
intensity is set to MT (Klostermann et al. 2009).
Through receipt of convergent inputs from both the
sensorimotor cortex and the spinal cord, the interpositus
nucleus of the cerebellum also exerts a modulating
influence upon motor network responses to sensory
stimulation via thalamic projections to premotor and
primary motor cortices (Luft et al. 2005). Hemi-
cerebellectomy blocks the modulation of cortical motor
output associated with repetitive ES of the sciatic nerve in
the rat (Ben Taib et al. 2005). It has also been proposed
that the state of the motor cortex itself, acting via the inter-
mediate cerebellum, may further serve to tune the gain of
polysynaptic responses to peripheral stimulation (Manto
et al. 2006). This is a possibility to which will return in the
sections that follow.
Motor network. In view of the patterns of connectivity
outlined above, one might surmise that the electrical
stimulation of peripheral afferents has clear potential to
alter the state of circuits not only within somatosensory
cortex, but also within the (classically defined) motor
network. Although it does not provide a basis upon which
to resolve the specific mediating pathways that are engaged,
empirical support can now be drawn from human neuro-
imaging data. For the present purposes it will suffice to
provide a brief, and necessarily partial, representation of
the relevant findings. The picture that emerges is of a
multi-stage hierarchical process in which various elements
of the cortical motor network are consistently engaged
(Avanzini et al. 2018).
When median nerve stimulation at motor threshold
intensity (0.5–2.7 Hz; 0.2–0.3 ms pulse duration) is
employed, elevated activity registered concurrently by
fMRI (Spiegel et al. 1999) and by MEG (Kawamura
et al. 1996) is evident in both contralateral S1
and M1. Similar protocols also yield an elevated
BOLD response in supplementary motor area (SMA)
(Manganotti et al. 2012). Notwithstanding the likelihood
of prior disease- and drug treatment-related adaptations
in brain organization, recent reports of intracerebral
recordings from epilepsy patients have provided hitherto
unanticipated opportunities to resolve the spatiotemporal
characteristics of motor network responses to peripheral
nerve stimulation. These recordings indicate that in
addition to enhanced gamma band power in areas 3a and
3b (exceeding that of areas 1 and 2), 1 Hz median nerve
stimulation (0.2 ms pulse duration) at MT (and 20% below
MT) gives rise to elevated activity in M1, and in large
sectors of dorsal and ventral premotor cortex, and SMA
(Avanzini et al. 2016). Further detailed analysis of the time
course of these responses (Avanzini et al. 2018) indicates
that M1 (BA4) exhibits an initial (peaks 30–40 ms)
phasic response to median (and tibial) nerve stimulation
that closely resembles those registered for areas 3a and
3b, whereas the responses recorded from premotor areas
occur somewhat later. It is also notable that while median
nerve stimulation just above MT gives rise to elevated
gamma band activity (50–150 Hz) in ipsilateral dorsal pre-
motor cortex (PMd), no such response has been detected
in ipsilateral M1 (Del Vecchio et al. 2019; see also Klingner
et al. 2011).
There is an apparent dose-dependent character to
the BOLD response to NMES observed for M1. For
example, it appears to increase monotonically as the
level of stimulation applied over the motor point of the
quadriceps muscle is increased from sensory threshold
to that eliciting a maximum motor response (Smith
et al. 2003). Using functional levels of stimulation
sufficient to bring about alternating flexion and extension
of the wrist, Blickenstorfer et al. (2009) reported
simultaneously registered BOLD activation peaks in
regions defined as contralateral primary motor cortex,
primary somatosensory cortex and premotor cortex, the
ipsilateral cerebellum, bilateral secondary somatosensory
cortex, supplementary motor area and anterior cingulate
cortex (see also Del Gratta et al. 2000; Arienzo et al. 2006;
Joa et al. 2012). Patterned NMES (50 Hz with 200 [s
pulses) sufficient to invoke finger flexion elevates the
BOLD response in contralateral M1 and S1 and bilaterally
in S2 (Iftime-Nielsen et al. 2012). A recent report suggests
that 100 s of 30 Hz stimulation at intensities sufficient to
generate wrist flexion (against gravity), gives rise to sub-
sequent changes in EEG/EMG-registered corticomuscular
coherence (Xu et al. 2018).
It has also been shown that in some instances the
physiological changes reflected in the BOLD response may
be sustained. Two hours of median nerve stimulation
(10 Hz trains, 50% duty cycle at 1 Hz, intensity just
above MT) applied at the wrist was observed (in the
context of a thumb movement task) to bring about
an increase in signal intensity and number of voxels
activated in M1, S1 and PMd, which persisted for
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
8 R. G. Carson and A. R. Buick J Physiol 00.0
up to 60 min after the stimulation had ended (Wu
et al. 2005). Employing a protocol in which mesh-glove
stimulation was applied at a level below sensory threshold
for 30 min, Golaszewski et al. (2004) observed that the
magnitude of the BOLD response registered in primary
motor and primary somatosensory regions of both hemi-
spheres during a finger-to-thumb tapping task was greater
than when the task was performed in the absence of
prior stimulation. The elevated activity registered for the
contralateral primary motor region remained present 2 h
following the cessation of stimulation.
In general the spatial extent of the BOLD registered
response (i.e. number of voxels) and the magnitude of the
signal change (i.e. relative to rest) are larger for voluntary
movement than those brought about by FES (Francis
et al. 2009; Joa et al. 2012; Wegrzyk et al. 2017), although
the particular regions of interest for which the greatest
differences are obtained tend to vary somewhat across
studies. In addition, S2 activation that is greater during
FES than during voluntary contractions has been reported
(Iftime-Nielsen et al. 2012; Christensen & Grey, 2013). At
least with respect to ankle dorsiflexion, the spatial extent of
the BOLD-registered activity in M1, S1, S2, SMA, cingulate
motor area (CMA), bilateral dorsal and ventral premotor
areas, and cerebellum VI is greater during FES-generated
movements than during passive movements (Francis et al.
2009; see also Gandolla et al. 2014). The nature of the
brain activation that characterizes combined NMES and
voluntary or imagined movement is a matter to which we
will return in the sections that follow.
Corticospinal projections. In circumstances in which the
expressed intent has been to bring about changes in the
state of the CNS (rather than produce overt movements)
(see Bergquist et al. 2011), the effects of parametric
variations in NMES upon the state of corticospinal
projections have been investigated. When delivered in
a4sonand6soffcyclefor20minat30Hz,
median nerve stimulation applied at the wrist gave rise
toincreasesintheamplitudeofMEPsrecordedinAPB
when the intensity was 110% of MT, but not when it
was 90% of MT (Sasaki et al. 2017). Applying 30 min
of mesh-glove whole-hand stimulation, Golaszewski et al.
(2012) noted that 50 Hz stimulation at sensory threshold,
and 2 Hz stimulation at motor threshold, gave rise to
increases in corticospinal excitability extending to 1 h
following. Such changes were not obtained when 50 Hz
stimulation at a level below the sensory threshold or 2 Hz
stimulation at sensory threshold was used. The outcomes
of this specific form of intervention (i.e. using mesh
glove stimulation), in which afferent fibres of multiple
types, with widespread innervation zones, are likely to
be involved, are not necessarily emblematic of those
obtained when a single nerve is stimulated. Specifically,
the magnitude of the change in corticospinal excitability
depends on the stimulation frequency (for intensity
MT). When applied at 100 Hz and in the range of
20–50 Hz, increases in corticospinal excitability (CSE)
in excess of 50% are routinely observed. This is not
generally the case for stimulation applied at 10 Hz or less
(Jaberzadeh et al. 2017).
If the intensity of peripheral nerve stimulation applied
in humans is between 30% and 50% of that required to
produce a maximum compound muscle action potential
(M-max), MEPs evoked subsequently by TMS over M1
are facilitated at inter-stimulus intervals (ISIs) from 25 to
60 ms in abductor pollicis brevis (APB) following median
nerve stimulation at the wrist (Deletis et al. 1992). A
similar outcome was noted (Komori et al. 1992) for the
thenar muscle at ISIs between 50 and 80 ms when the
peripheral shock was set to 10% of M-max. Devanne et al.
(2009) reported than even when stimulation intensity is
set just above motor threshold, median nerve stimulation
(at the wrist) gives rise to marked facilitation of MEPs
recorded in the APB, FDI and extensor carpi radialis
(ECR) muscles when ISIs ranging from 40 to 80 ms are
employed. At ISIs extending beyond 200 ms (and below
25 ms – around the latency of the N20 component of the
somatosensory evoked potential), a diminution of MEP
amplitude is generally obtained (e.g. Turco et al. 2018).
It is of particular interest in the present context that after
NMES is delivered over the ulnar nerve (100 Hz in a 20s
on, 20s off duty cycle; intensity 15% of that to elicit
a maximum m-wave) for 40 min, short-latency afferent
inhibition (SAI: ISI 18–25 ms) is markedly diminished,
whereas for those ISIs (28–35 ms) at which there occurred
potentiation of MEP amplitudes following a (single)
conditioning peripheral nerve stimulus, the NMES inter-
vention served to further increase the amplitude of the
TMS-evoked response (Mang et al. 2012). These findings
are consistent with the possibility highlighted above, that
the state of M1 (potentially acting via the intermediate
cerebellum) may influence the gain of polysynaptic circuits
that modulate the effects of peripheral stimulation (Manto
et al. 2006).
It remains unclear at present whether sustained changes
in corticospinal excitability brought about by prolonged
NMES interventions are instrumentally related to changes
in behaviour. Veldman et al. (2016) applied trains to
the radial and median nerves (proximal to the elbow)
consisting of five square wave pulses at 10 Hz (pulse width,
1 ms) 50% duty cycle, at intensities just below MT. In three
separate interventions the stimulation was applied for 20,
40 or 60 min. Changes in the performance of a visuomotor
tracking task (post-intervention relative to baseline) were
compared to a fourth group of participants who did not
receive stimulation. Although some improvements in task
completion and in measures of CSE were observed over
the course of the following week, there was no evidence
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 9
that these outcomes were related. A more general issue (to
which we will return) is thereby illustrated. Variations in
CSE, as revealed by TMS, are not necessarily indicative
of the functional adaptations (in this case brought about
by NMES) that mediate improvements in performance
(Carson et al. 2016).
In light of the assumption that contractions of an
intensity sufficient to mimic some features of those
brought about by voluntary activation are necessary to
cause reliable changes in CSE (Chipchase et al. 2011a),
it may appear paradoxical that FES (primarily lower
limb) protocols bring about immediate effects that are
of lesser magnitude than those associated with 20–50 or
100 Hz stimulation delivered closer to MT (Jaberzadeh
et al. 2017). Nonetheless, it is also the case (i.e. as with
intensity MT) that supra-motor threshold stimulation
is more effective at increasing CSE when delivered at
30 Hz than at 10 Hz (Chipchase et al. 2011b). It has been
reported that while 20 and 40 min of stimulation (30 Hz)
at intensities sufficient to generate a ‘voluntary-like’
contraction in APB increased CSE, this was not the case
for 60 min of stimulation (Andrews et al. 2013). Although
perhaps counterintuitive, a similar but less pronounced
non-monotonic effect of duration is, however, also pre-
sent for MT level stimulation (Jaberzadeh et al. 2017).
In other words, there comes a point at which increasing
the intensity or duration of stimulation brings about no
further gains, at least in terms of the excitability of cortico-
spinal projections to the target muscles.
There exist forms of NMES (typically delivered over the
muscle belly) that have been developed with the express
aim of preventing skeletal-muscle weakness, for example
during acute critical illness. They are sufficient to generate
high levels of force (and thus sometimes designated
electrostimulation strength training). Usually utilizing
frequencies between 35 and 100 Hz, the stimulation
can be applied for up to an hour daily, over periods
ranging between 1 and 6 weeks (Maffiuletti et al. 2011,
2013). There are comprehensive reviews dealing with
the nature of the central and peripheral adaptations
that may mediate the observed increases in functional
capacity that can be accrued by these methods (e.g.
Hortob´
agyi & Maffiuletti, 2011). The present aim is not
to recapitulate these analyses. It is, however, pertinent to
highlight one of the key observations to emerge in the
course of this research. As noted in preceding sections,
bilateral alterations in the state of brain circuits that
constitute the classical motor network in both hemispheres
are frequently observed following unilateral NMES. It is
therefore particularly salient that these NMES variants
can increase the force-generating capacity of homologous
muscles in the limb opposite to the one in receipt of
stimulation (Cabric and Appell, 1987; Hortob´
agyi et al.
1999; Zhou et al. 2002; Huang et al. 2007; Kadri et al. 2017).
In recent studies conducted with the aim of determining
the mechanistic basis of such effects, there has been an
understandable initial focus upon the degree to which
less ‘intense’ forms of unilateral NMES might bring about
bilateral changes in CSE. Veldman et al. (2015) applied
trains consisting of five square wave pulses delivered to
the radial and median nerves of the right arm (above
the elbow) at 10 Hz (pulse width, 1 ms) 50% duty
cycle, using an intensity equal to twice the perceptual
threshold (i.e. presumed to be below MT) in five blocks of
5 min duration. They noted increases in the amplitude of
MEPs recorded in both right and left ECR following the
intervention, which were accompanied by improvements
in the performance of a visuomotor tracking task (i.e.
for both limbs). There was, however, no evidence of
a statistical association between these measures (see
also Summers et al. 2017). Using a largely equivalent
stimulation protocol, Veldman et al. (2018) also observed
improvements in the performance of the opposite limb,
in this case during a retention test conducted 2 days
following the intervention. And as in the preceding study,
electrophysiological measures (in this case EEG derived)
of directional oscillatory coupling (representing ‘cortico-
cortical connectivity’), between posterior parietal and
primary somatosensory cortex to the primary motor
cortex, did not vary in accordance with the changes in
behaviour.
A reflection on the brain circuitry engaged by NMES. It
isevidentthatthereexistvariantsofNMESthatprovidea
means of altering the state of elements within an extended
brain network (encompassing not only classically defined
somatosensory and motor areas), and the excitability of
circuits with projections to the spinal cord (e.g. Schabrun
et al. 2012). What remains to be determined are the
causal relations between the changes in brain state that
canberegisteredbymodernneuroimagingandelectro-
physiological techniques, and alterations in functional
capacity that can in some circumstances be brought about
by NMES. In recent years there has perhaps been an
undue haste to infer that intervention-induced changes
in corticospinal excitability are indicative of the neural
adaptations that mediate sustained changes in behaviour
(Carson et al. 2016). Indeed, well-powered individual
studies (e.g. Ruddy et al. 2016) and several meta-analyses
(e.g. Veldman et al. 2014; Berghuis et al. 2017; Manca et al.
2018) have failed to demonstrate an association between
changes in CSE and improvements in motor performance.
There is consequently a growing recognition that in our
empirical investigations we must devote greater attention
to paths and structures other than the ones that can be
assayed easily by such techniques as TMS (Veldman et al.
2016) or conventional brain imaging analysis approaches.
For example, a case can be made for considering the
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
10 R. G. Carson and A. R. Buick J Physiol 00.0
individual differences in functional or structural brain
connectivity associated with variations in the expression
of performance changes (e.g. Ruddy et al. 2017) that follow
the administration of NMES. This would be in contrast to
simply registering brain regions or pathways that exhibit
a change in state following stimulation.
Adjuvant techniques
The production of voluntary movement has two essential
components: central efferent drive that is initiated at the
level of the cortex and consequentially muscle contra-
ctions that displace joints and thus give rise to afference.
Electrical stimulation of peripheral nerves provides a
means of producing muscular contractions without the
initial central drive by direct depolarization of motor
axons located below the stimulating electrodes. It has
been noted previously that the effectiveness (both adaptive
and restorative) of NMES may be enhanced through
the use of specific protocols (e.g. pulse width/frequency
combinations) that promote synaptic recruitment of
spinal motoneurons by the electrically evoked sensory
volley (e.g. Collins, 2007). The adaptive benefits are
readily appreciated. For example, afference-mediated (i.e.
synaptic) recruitment of spinal motoneurons is likely
to occur in normal physiological order, and thus pre-
ferentially include fatigue-resistant motor units. The
restorative benefits, while perhaps less obvious, are,
however, also potentially significant. In this regard,
emphasishasbeenplacedonacapacityfortherepeated
evocation of sensory volleys by NMES to induce increased
activity in spinal and supraspinal circuits, and in turn bring
about acute and chronic neuroplastic adaptations that are
sufficient to enhance function (e.g. Bergquist et al. 2011).
While in this scheme the accent is on the cumulative effects
of stimulus repetition per se, there are further possibilities.
In recent years, there has been particular interest
in associative forms of neural plasticity, such as those
in which the repeated coincidence of experimentally
induced activity in both sensory circuits (by peripheral
nerve stimulation) and motor circuits (by TMS applied
over M1) gives rise to sustained changes in cortico-
spinal excitability (e.g. Stefan et al. 2000). In terms
of the phenomenology of the induced effects, there is
notionally a resemblance to Hebbian plasticity (Hebb,
1949), whereby a presynaptic input onto a postsynaptic
neuron is strengthened as a consequence of both the
pre- and postsynaptic neurons being active simultane-
ously. In seeking to provide a more mechanistic account
of this paired associative stimulation (PAS), it has
been proposed that it shares key features with spike
timing-dependent plasticity (STDP) (M¨
uller-Dahlhaus
et al. 2010) – as this has been elaborated in animal
models and reduced (e.g. slice) preparations. In STDP,
the polarity of the induced change in synaptic efficacy
is determined by the sequence of pre- and postsynaptic
neuronal activity (for reviews see Dan & Poo, 2006;
Markram et al. 2011). In prototypical representations of
STDP (e.g. Song et al. 2000), potentiation occurs if a pre-
synaptic neuron fires no more than 50 ms in advance of the
postsynaptic neuron (Feldman, 2000). Depression arises if
postsynaptic spikes precede presynaptic action potentials
(or transpire without activity in the presynaptic neuron)
(Levy & Steward, 1983; Bi and Poo 1998; Cooke & Bliss,
2006). There is also held to be a sharp transition from a
weakening of synaptic efficacy (long term depression) to
strengthening of synaptic efficacy (long term potentiation)
at time differences in the vicinity (within 5 ms) of zero
(Feldman, 2012).
Inthesectionsthatfollow,weusetheconceptual
framework of associative plasticity to consider the impact
of adjuvant techniques upon responses to NMES. The
argument is made that an association of NMES-generated
afference with centrally initiated neural activity, such as
that which occurs if the stimulation is triggered by efferent
drive, or is delivered following instructions to engage in
mental imagery, may promote neural adaptations upon
which changes in functional capacity may be based. In
doing so, we first make the critical point that the induction
of associative effects that can be observed at a systems level
in humans does not require adherence to the defining
characteristics of STDP. In particular, associative effects
are expressed when the relative timing of the activity
induced in sensory and motor circuits is not precisely
circumscribed.
Extending the concept of associative stimulation. There
are a number of recent and comprehensive reviews of
paired associative stimulation (e.g. Carson & Kennedy,
2013; Suppa et al. 2017). It is not our intent to reprise their
contents. There are nonetheless important points that can
be gleaned from these reviews, and from empirical findings
that have appeared subsequently. Foremost among these is
the observation variants of PAS in which the timing of the
contributory elements is not strictly confined, for example
when extended trains of peripheral nerve stimuli are used
(e.g. Ridding & Taylor, 2001; Carson et al. 2013; McNickle
& Carson, 2015; Shulga et al. 2016; Carson & Rankin,
2018; Tolmacheva et al. 2019), produce elevations in CSE
that are comparable to, if not greater than, those obtained
when the ISI separating the peripheral and cortical events
is precisely circumscribed. The associative nature of the
effects are, however, emphasized by the fact that in these
studies the NMES alone (typically at an intensity MT)
does not bring about changes in CSE. The conclusion
that the relative timing need not be either precise or
restricted is further emphasized by reports that the
nerve stimulation component of PAS can be replaced by
movement-generated afference, without loss of generality
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 11
(Edwards et al. 2014; see also McNickle & Carson, 2015).
In this vein, cortical microstimulation experiments in
freely behaving non-human primates reveal that changes
in synaptic strength between stimulated sites in precentral
and/or postcentral cortex can be brought about without
adherence to STDP rules (Seeman et al. 2017). These recent
findings also serve to emphasize what should perhaps be
apparent on apriorigrounds alone, that when applied in
vivo, there are multiple pathways via which the corollaries
of (i.e. peripheral) stimulation may reach and influence the
cortex (Carson & Kennedy, 2013), and as a consequence
relative timing is likely to be only one of many factors
that govern the induction of neuroplastic adaptations
(Feldman, 2012).
It is in this light that the outcomes yielded by associative
stimulation protocols can be more easily reconciled with
the results of studies demonstrating that the combined
effects of NMES and forms of exogenous cortical
stimulation other than TMS are greater than those of each
stimulation modality alone. Rizzo et al. (2014) described
a protocol in which NMES (500 [s pulse duration, at
5 Hz for 5 min, 1500 stimuli, intensity 2×ST)
was delivered to the median nerve simultaneously with
transcranial direct current stimulation (tDCS). When
the cortical electrode montage was such that the anode
was positioned on the scalp over M1 contralateral to
the site of peripheral nerve stimulation, the elevation
in CSE recorded following the cessation of the inter-
vention was markedly greater than that induced by tDCS
alone (NMES +sham tDCS did not alter CSE). In
addition, the duration of the elevation in CSE brought
about by the combined stimulation persisted for at least
1 h (considerably longer than following anodal tDCS
alone). Employing an NMES variant in which 1 ms pulses
(intensity MT) were applied simultaneously to the FDI
and ABP motor points (at frequencies between 0.35 and
6.7 Hz 6345 pairs) for a period of 30 min, and anodal
tDCS delivered for the final 25 min, Hoseini et al. (2016)
observed subsequent improvements in performance of
the Purdue pegboard test (used to assess dexterity) that
were not seen following either NMES +sham tDCS or
anodal tDCS +sham NMES. For cases in which tDCS is
applied (i.e. continuously) over an extended period during
which NMES is also delivered at various intervals, there
exists no discrete timing relationship between peripheral
and cortical stimulation events. Yet associative effects are
nonetheless obtained. Although not in accordance with
STDP-based models of associative plasticity, this general
pattern of findings is, however, consistent with recent
analyses showing that not only the phase, but also the
power of the cortical oscillatory beta cycle (e.g 16–17 Hz)
at the moment stimulation is delivered influences the
increase in CSE caused by TMS (Khademi et al. 2019).
There is a more general point. Since a single relative
timing relationship between the corollaries of cortical
and peripheral stimulation is not a prerequisite for the
induction of associative effects, when NMES is paired
with endogenously generated elevations in motor network
excitability, similar neuroplastic adaptations are likely to
occur. There is now a considerable body of evidence to
support this conjecture, and to suggest that the adaptations
may be functionally significant.
Augmenting NMES at motor threshold intensity. For the
present purposes, we consider two endogenous means of
altering the state of the motor network: voluntary contra-
ctions and mental imagery. With a view to confining the
limits of the discussion, ‘cognitive’ factors such as the focus
of attention, which are believed to have an influence on
the efficacy of associative stimulation protocols (e.g. Stefan
et al. 2004), will not be treated in any detail.
It has for some time been appreciated that when NMES
is applied in the context of voluntary contractions, the
consequential changes in the state of efferent projections
from the brain to the spinal cord are greater than those
achieved through NMES alone (de Kroon et al. 2005).
Although the majority of empirical studies conducted in
this domain have employed levels of stimulation sufficient
to evoke overt motor responses, it can also be shown
that these features emerge when much lower intensities
of NMES are used. For example, Taylor and colleagues
(2012) delivered biphasic pulses (50 Hz; 200 [s pulse
duration; intensity MT; 50% duty cycle for 6 s) over
the wrist extensors (ECR and extensor carpi ulnaris) at the
onset of 60 isometric wrist extension contractions (to 15%
MVC) – triggered when the surface EMG recorded from
the target muscles exceeded 25 [V. In a control condition
NMES was delivered in isolation. An elevation of CSE was
observed following EMG-triggered delivery of NMES, but
not following NMES alone. Similar findings have been
obtained for the lower limb, when NMES is delivered
either over the tibialis anterior (TA) muscle or to the
(common peroneal) nerve during ballistic dorsiflexions
of the ankle (Jochumsen et al. 2016). In this regard, it
is notable that the acute augmentation of CSE appears
to be greater when NMES is combined with shortening
contractions than with isometric contractions (Saito et al.
2014).
Of greater practical relevance are the changes in
functional capacity that arise from the combination of
NMES and voluntary contractions. Carvalho et al. (2018)
conducted a double-blind, sham-controlled, randomized
trial engaging healthy adults, in which median nerve
stimulation (random frequency ranges (1–4, 8–12 and
60–90 Hz) and intensity levels (2–6 mA)) at the wrist
was applied during 20 min practice of a serial reaction
time task (SRTT) requiring keypress responses. This was
followed by a similar ‘consolidation’ session of 30 min
duration.Itwasnotedthatexplicitrecallofthelearned
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
12 R. G. Carson and A. R. Buick J Physiol 00.0
sequence improved following both initial training and
consolidation. No such improvements were obtained for
either a group that received ‘off line’ NMES, or a group
that was given sham stimulation.
That the origin of the neuroplastic effects of combined
voluntary contraction and NMES is likely to be pre-
dominantly supraspinal rather than spinal, at least when
relatively low levels of electrical stimulation are employed,
is indicated by a series of studies in which the delivery
of NMES has been in the context of motor imagery tasks
performed by the recipient. Employing a task in which
the participants were asked to imagine that they were
squeezing and relaxing a ball (motor imagery), while
watching a video of the action (observation) (during which
time the ball was held ‘passively’), Yasui et al. (2019)
applied NMES (trains of 20 pulses at 10 Hz; 1 ms pulse
duration; intensity 90% MT; 50% duty cycle of 2 s on,
3 s off) during four blocks of 5 min duration. A cumulative
increase in the amplitude of MEPs recorded from FDI was
obtained in this condition, but not for NMES alone (or
imagery/observation alone). Corresponding effects that
are sustained for at least 30 min following cessation of
combined NMES/motor imagery have also been reported
for the lower limb (Takahashi et al. 2019). In a small-scale
study (without a control group), Okuyama et al. (2018)
observed increases in upper extremity function in
10 chronic stroke survivors, following an intervention
(10 trials per day for 10 days) in which stimulation (MT
of the extensor digitorum communis) of the radial nerve,
innervating wrist and finger extensors, was combined with
motor imagery/observation.
A compelling case that these effects are associative
in nature can be made on the basis of reports that
they can be obtained when the delivery of NMES is
triggered by EEG-registered movement-related cortical
potentials (MRCPs) – generated when individuals follow
an instruction to imagine the ‘kinaesthetics’ of ballistic
movements. Deploying an intervention of this type,
Niazi et al. (2012) triggered stimulation (1 ms pulse
duration; intensity MT) of the common peroneal
nerve upon detection of the initial negative phase of
the MRCP, as 50 self-paced imagined movements were
performed. The intervention gave rise to increases in
the excitability of corticospinal projections to TA. No
such changes were induced by NMES alone or by motor
imagery alone (see also Mrachacz-Kersting et al. 2017).
Comparable results are obtained if the timing of the
NMES is yoked (using an estimate of the contingent
negative variation) to the onset of a cued imagined
movement (Mrachacz-Kersting et al. 2012). In a recent
investigation using MRCP-triggered NMES (equivalent
to the protocol of Niazi et al. 2012), increases in CSE
persisting for one hour were registered (Olsen et al. 2018;
see also Jochumsen et al. 2018). In addition to giving rise
to increases in CSE in both chronic (Mrachacz-Kersting
et al. 2016) and sub-acute (Mrachacz-Kersting et al. 2019)
stroke survivors, imagery-related MRCP triggered NMES
appears capable of promoting positive changes in motor
function. As far as we are aware, however, it has not been
established that any changes in CSE brought about by these
techniques are instrumentally related to improvements in
performance.
Given the very large number of brain imaging studies
that have been conducted, there are several meta-analyses
(e.g. Grezes and Decety, 2001; Caspers et al. 2010;
Molenberghs et al. 2012; H´
etu et al. 2013; Hardwick et al.
2018) that provide a basis upon which to survey the
brain regions engaged during voluntary movement, action
observation and motor imagery. As has been highlighted
recently, however (Savaki & Raos, 2019), by and large
these meta-analyses are based upon studies in which the
three task contexts have been investigated independently
of one another. On the basis of these analyses it
appears reasonable to draw the conclusion that voluntary
movement, action observation and motor imagery all
give rise to consistent activation of a brain network
encompassing premotor, parietal and somatosensory areas
(e.g. Hardwick et al. 2018). In the present context, we
follow the lead of Savaki and Raos (2019) in suggesting
that there is additional information to be gained by giving
particular weight to the small number of studies in which
fMRI has been used to assay the whole brain when all three
variants of the same ‘motor’ task have been performed by
the same group of participants. In a recent study in which
there were no aprioriconstraints upon regions of inter-
est (ROIs) deemed to be of interest, Simos et al. (2017)
determined that during both motor imagery and execution
of a geometric tracing task performed by the right index
finger, BOLD activity in the following regions surpassed
the assigned threshold: bilateral dorsal and ventral pre-
motor cortex, left supplementary motor cortex (SMA
proper), bilateral BA 7 in the superior and BA 40 in
the inferior parietal cortex, bilateral BA 8 in the middle
frontal gyrus, bilateral BA 22 in the posterior part of
superior temporal gyrus including the temporo-parietal
junction, bilateral BA 37 in the posterior part of the middle
temporal gyrus including the extrastriate body area, the
left extrastriate visual BA 19 in the cuneus, the right lingual
gyrus (LG) and the left middle occipital gyrus (MOG),
left BA 7 in the posterior precuneus and right BA 37
in the fusiform gyrus. The left secondary somatosensory
cortex (SII) was also deemed engaged in both tasks. As
might be anticipated, while the upper limb representations
of the primary motor and somatosensory cortical areas
(2/3) exhibited bilateral activity, the magnitude of the
BOLD response was larger during execution than during
imagery. In contrast, during motor imagery there was
relatively greater BOLD response magnitude bilaterally in
prefrontal, premotor and parieto-temporal cortices. In a
related investigation in which the technique of multi-voxel
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 13
pattern analysis was used in conjunction with apriori
selection of ROIs (excluding MI and SI), Filimon et al.
(2015) reported that during both execution and motor
imagery of reaching to visual targets, the BOLD response
is registered across both ventral and dorsal premotor, and
parietal areas.
It is readily apparent therefore that during both the
execution of (upper limb) movements and motor imagery
there is a high degree of overlap with those brain regions
that are believed to exhibit increased activity in response
to NMES (see preceding sections). As such, and the
consequential changes in CSE that have been observed
in some cases notwithstanding, it cannot be assumed that
the M1 or S1 is the principal locus of the associative inter-
actions that occur when NMES is delivered during either
motor imagery tasks or voluntary contractions. Indeed, it
is clear that there are many potential loci. At present there
is no empirical basis upon which to resolve the various
possibilities. It is important to emphasize that during
all motor tasks, the notionally ‘active’ (i.e. in a BOLD
registration context) brain regions constitute a network
of functional connections (e.g. Simos et al. 2017), such
that the task-relevant contribution of any specific region
of interest cannot sensibly be considered in isolation (e.g.
Anderson, 2008). In closing this section, it should also be
noted that there have been very few randomized clinical
trials (with appropriate blinding) in which the combined
effects on function of either voluntary contractions or
motor imagery and NMES at motor threshold intensity
have been evaluated.
Augmenting NMES at supra-motor threshold intensities.
Empirical studies, in which the focus has been upon the
combined effects of voluntary contractions and NMES
delivered at intensities sufficient to generate functional
levels of muscle tension (i.e. FES), have typically been
undertaken in a clinical context. In many such instances
thefocushasbeenuponthepromotionofmovement
capacity in stroke survivors. In light of the relatively large
number of investigations of this kind that have been under-
taken, several systematic reviews have been compiled.
Although initial summaries of this nature (e.g. de Kroon
et al. 2005) tended to suggest that clinical outcomes
obtained for FES triggered by voluntary contraction (e.g.
via EMG registration) were superior to those following
FES alone, it was not generally the case that cumulative
effect size estimates were obtained. In a more recent
analysis that was restricted to the outcomes of randomized
controlled trials (RCTs) engaging chronic stroke survivors,
Yan g et al. (2019) reported that the changes in function
(as assessed by the Fugl–Meyer test) and activity (e.g. as
assessed by the Action Research Arm Test) arising from
‘cyclic’ FES (not triggered by voluntary contraction) and
EMG-triggered FES could not be distinguished in terms
of their quantified effects (although both were superior
to control). In their systematic reviews, both Monte-Silva
et al. (2019) and Nascimento et al. (2014) arrived at a
same conclusion. In the single RCT of which we are aware
(Wilson et al. 2016) that compared their relative efficacy
inacutestrokesurvivors(<6 months post-stroke), the
improvements in Fugl–Meyer scores and the Arm Motor
Ability Test, registered following an 8-week intervention
period, did not differ between administrations of ‘cyclic’
FES and EMG-triggered FES.
It is particularly notable, therefore, that when the
delivery of stimulation at levels sufficient to produce
joint displacement is triggered by contractions of the
opposite (i.e. non-impaired) limb, improvements in
clinical outcomes greater than those induced by NMES
alone have been obtained in several trials. Knutson et al.
(2016) employed with chronic stroke survivors a method
whereby opening of the ipsilesional hand (monitored
using an instrumented glove) modulated the intensity of
stimulation applied to the finger (and wrist) extensors
of the paretic hand, such that both hands opened
synchronously. Fugl–Meyer scores and performance of the
Arm Motor Ability Test exhibited by following a 12-week
intervention (10 h of stimulation per week) were greater
than those exhibited by patients who received cyclic FES
(see also Knutson et al. 2012). In the context of a trial of
3 weeks duration (5 sessions per week; 20 min per session)
engaging acute (3 post) stroke survivors, Shen et al.
(2015) implemented a protocol whereby a wrist extension
movement executed by the non-impaired limb triggered
the delivery of stimulation (50 Hz; 200 [s pulse duration;
intensity up to that sufficient to produce full range wrist
extension) to the impaired limb. In the NMES group,
matched levels of stimulation were applied. Although
both groups exhibited clinically relevant improvements in
capacity (Fugl–Meyer assessment, the Hong Kong version
of functional test for the hemiplegic upper extremity
(FTHUE-HK) and active range of motion), the magnitude
of these changes was substantially greater in the group
for whom NMES was triggered by movement of the
opposite limb. In a more recent trial using the same
methodology that engaged individuals within 15 days of
stroke, the combination of routine rehabilitation with
NMES triggered by movement of the ipsilesional limb
gave rise to better outcomes than routine rehabilitation
combined with matched levels of electrical stimulation
(Zheng et al. 2019).
The contrasting effects (relative to FES alone) of FES
triggered by voluntary engagement of the same limb and
of FES triggered by movement of the opposite limb might
also be considered in light of the following. Systematic
reviews of randomized or quasi-randomized controlled
trials examining the effects of electrical stimulation
delivered at intensities close to sensory threshold (e.g.
TENS) on motor recovery following a stroke suggest that
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
14 R. G. Carson and A. R. Buick J Physiol 00.0
clinical outcomes are superior when it is combined with
voluntary movement (e.g. Ikuno et al. 2012; Laufer &
Elboim-Gabyzon, 2011). Taken together, these findings
suggest that the functional impact of combining NMES
with voluntary contraction depends on the intensity
of the electrical stimulation. When it is insufficient
to generate muscle contractions, additive effects are
obtained. In contrast, when FES intensities are employed,
the combined effects are comparable to those induced
by FES alone. There are at least two possible accounts
of this phenomenon. The first is that there is a ceiling
effect. That is, if the effects of FES alone on the state of
the motor network approach asymptotic levels, there may
be little scope for endogenous activity generated in the
context of voluntary contractions to promote additional
restorative changes. The additive effects of contractions
performed by the opposite limb, however, suggest that
this explanation is insufficient. As described above, in
both acute and chronic stroke survivors, when the delivery
of FES is triggered by contractions of the ipsilesional
limb, the benefits in term of clinical outcomes are greater
than those brought about by FES alone. Similarly, when
NMES at functional intensities is delivered during mental
imagery (triggered by very low ‘incidental’ levels of EMG),
improvements in function achieved by chronic stroke
survivors are greater than those achieved using FES alone
(Hong et al. 2012; You & Lee, 2013; cf. Park, 2019).
There is no ceiling effect. An alternative possibility is
that when voluntary contractions are combined with,
or initiate (e.g. EMG-triggered FES), NMES delivered at
intensities sufficient to produce joint displacement, there
is a mismatch between the anticipated consequences of
the efferent drive and the afferent feedback that arises
from the combined effects of the voluntary contraction
and stimulation-driven recruitment of motoneurons (e.g.
Iftime-Nielsen et al. 2012). As a corollary, the degree
of any such ‘mismatch’ is likely to depend not only
on the intensity of the stimulation, but also on the
degree to which the pattern of its application mimics
natural muscle synergies. For example, it is known that
in the context of tasks in which a large number of
degrees of freedom (muscular and biomechnical) must be
coordinated such as the formation of a grasp, if electrical
stimulation (of the intrinsic and extrinsic flexor muscles)
is imposed upon a voluntary contraction, maximal grip
force diminishes (Boisgontier et al. 2010). In other tasks
in which a relatively small number of muscles actuate a
single joint (e.g. Barker et al. 2008, 2017), the discrepancy
maybesmaller.Theassumptionisthat,tothedegree
to which a mismatch is present, further augmentation of
the effects of NMES through associative mechanisms is
precluded.
There have been relatively few studies in which imaging
techniques have been used to compare patterns of brain
activity arising when FES is delivered both in isolation and
in combination with voluntary contractions. Employing
the method of near-infrared spectroscopy with healthy
adults, Lin et al. (2016) reported that when NMES was
delivered at a level sufficient to augment force output
during isometric knee extension contractions, the O2
demand in the contralateral premotor cortices and SMA
was greater than the sum of that observed during NMES
alone and during voluntary movement alone. Oxy-Hb
increases in ‘sensory-motor cortex’ (relative to rest) of
greater magnitude during EMG-triggered FES than for
voluntary contractions alone (and FES alone) have also
been reported for chronic stroke survivors (Hara et al.
2013). There are two further studies (of which we are
aware) in which fMRI has been employed during upper
limb movements (for the lower limb see Gandolla et al.
2014). Joa et al. (2012) reported that FES combined with
voluntary wrist extension gave rise to a greater BOLD
signal in ipsilateral cerebellum, contralateral MI (‘primary
central gyrus’) and SI (‘post central gyrus’) than during
FES alone. Christensen and Grey (2013) noted that a larger
BOLD response was registered during combined FES
and voluntary (finger flexion–extension) movements than
during voluntary movements alone in the following brain
regions: superior temporal gyrus, supramarginal gyrus,
insula, rolandic operculum and angular gyrus. There were
no regions for which a larger BOLD response was obtained
during voluntary movements alone. Of particular interest
in the present context is the observation that following
administration of an ischaemic nerve block that removed
sensory feedback (but preserved the capacity for voluntary
movement), there were no differences in BOLD response
between the two conditions (i.e. voluntary movement with
and without FES). This pattern of outcomes supports
the conjecture that the additional brain activity otherwise
evident during combined voluntary movement and FES
(i.e. compared to voluntary movement) is related to
the integration of afferent feedback (i.e. relative to that
anticipated on the basis of the efferent command).
Gandolla et al. (2014) present a somewhat similar line of
argument.
For completeness, we highlight briefly the finding
that for both healthy adults and survivors of stroke,
the excitability of corticospinal projections to muscles
in receipt of FES is greater when it is combined with,
or triggered by, voluntary contraction than when it is
delivered in isolation (Khaslavskaia & Sinkjaer, 2005; Barsi
et al. 2008; Stein et al. 2013; McGie et al. 2015). Although
these data were not obtained in the context of the clinical
trials described above, they do serve to emphasize an
important point. Two variants of an intervention that
can be distinguished clearly in terms of the changes in
corticospinal excitability to which they give rise do not
necessarily lead to different treatment outcomes when
they are deployed over multiple sessions in a rehabilitation
setting.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 15
A reflection on the augmentation of NMES at supra-motor
threshold intensities. There was a period during which it
was widely assumed that NMES at supra-motor threshold
intensities in combination with voluntary contractions,
particularly when there was a contingent relation (as
in EMG-triggered FES), gave rise to outcomes super-
ior to those that could be achieved by NMES alone. In
such circumstances it was natural to seek explanatory
constructs. For example, De Kroon et al. (2005), in what
was then a comprehensive review of the available data,
hypothesized that there may be an additional cognitive
element present in EMG-triggered NMES that is not a
feature of NMES alone. In was suggested that an additional
investment of mental resources and attention improves
performance. Any explanation of this type should apply
in equal measure to instances in which the effects of
NMES delivered at lower (e.g. MT) intensities are
accentuated by simultaneous voluntary contractions. The
inconvenient truth is, however, that there is currently little
by way of systematic evidence to indicate that EMG (or
movement)-triggered FES is more efficacious than cyclic
FES(whichisnotyokedtovoluntarymovement).
Clarac et al. (2009, page 367) remark that Wundt
(1863) was among the first to note explicitly that passive
movements and active movements differ in respect of
their perceptual consequences. More particularly, they
are distinguished by the relationship between efferent
impulses and the referent response. Duchenne de
Boulogne also promoted the concept of an ‘efferent sense’
of central origin, which precedes a muscle contraction
and is necessarily distinguishable from the sensation that
arises as a result of the contraction (Clarac et al. 2009).
On the basis of electrophysiological recordings obtained
using modern methodologies, Lebedev et al. (1994)
established that during self-initiated movement, activity in
the primary somatosensory cortex becomes evident before
theinitiationofmotoroutput.Thiswasinterpretedaspre-
paration for receipt of the imminent changes in afferent
inflow that will result from the movement (see also Nelson
et al. 1991; Nelson, 1996). fMRI-based investigations in
healthy volunteers further reveal that during active but not
passive movement, a BOLD response in the Brodmann
area 2 subregion of S1 is closely associated with that
registered in premotor and supplementary motor areas,
the parietal cortex and the cerebellum, in the absence of
common mediation by area 3b (Cui et al. 2014). These
and similar observations have been taken as evidence in
support of the construct of efference copy – conceived of
by von Holst and Mittelstaedt (1950) as the internal copy
of an outgoing, action-producing ‘command’ generated
by the motor system. In an extension of the concept, it
is proposed that the CNS instantiates forward internal
models that utilize efference copy in order to anticipate the
sensory consequences of an action (e.g. Miall & Wolpert,
1996).
The conventional contemporary line of thinking is
that brain computer interfaces (BCI) that instantiate
closed-loop control (i.e. brain–efference–change in
muscle length/joint displacement–afference–brain) offer
concordance between the efference copy and sensory
consequences of an action. It is furthermore assumed
that (repeated) concomitance of voluntarily generated
brain activity, and movement-related afference (even if
generated by artificial means) can promote neuroplastic
adaptation and in some cases restoration of function (e.g.
Jackson & Zimmermann, 2012). A key requirement in
this regard is that there is a persistent causal relationship
between the initiating endogenous neural activity (e.g.
descending drive leading to recruitment of motoneurons
– as registered by EMG) and the consequential endogenous
neural activity (e.g. afference generated by EMG-triggered
FES). A further necessity is temporal congruency. That
is, the delay between the initiating and consequential
neural activity must be consistent with the natural latency
between the efference copy of a motor command and the
reafferent sensory feedback (e.g. Leube et al. 2003). It has
been highlighted recently that, even in circumstances in
which the afference generated by EMG-triggered NMES
is dominated by that which arises from direct activation
of sensory axons (i.e. for intensities MT), conduction
delays within the central and peripheral nervous systems
dictate that stimulus-evoked activity is unlikely to be able
alter the state of circuits in M1 sooner than 60 ms following
the voluntary activity that generated the triggering EMG
(Brown et al. 2016). In the event that the afference
generated by NMES is dominated by reafference produced
by the resulting contraction (i.e. such as with FES), and
given electromechanical delays in the order of 40 ms
(Cavanagh & Komi, 1979), the latency will be very much
greater. If the FES is triggered by joint displacement (rather
thanbyEMG),itwillbelongerstill.Wehaveemphasizedin
preceding sections that the induction of associative effects
that can be observed at a systems level in humans does not
require adherence to the defining characteristics of STDP
(i.e. precisely circumscribed relative timing, with pre-
synaptic firing occurring no more than 50 ms in advance
of postsynaptic firing; Feldman, 2000). Nonetheless, if the
interval over which the contingent relationship is defined
exceeds certain bounds, the effects of the association are
likely to be diminished (Carson & Rankin, 2018). Indeed,
even in STDP schemes, it is predicted that the magnitude
of potentiation is inversely related to the delay between
pre- and postsynaptic activity (Markram et al. 1997).
Such considerations raise the possibility that the
failure of EMG triggered FES to bring about functional
adaptations that are greater than those achieved by cyclic
FES is attributable to the extended delay between initiation
of the voluntary command (that generates the EMG), and
the reafference produced by the resulting contraction. In
the case of EMG-triggered NMES delivered at intensities
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
16 R. G. Carson and A. R. Buick J Physiol 00.0
sufficient to activate only a relatively small proportion of
motor axons (i.e. around motor threshold), in which the
resulting afference is dominated by that which arises from
direct activation of sensory axons, the delay following the
voluntary command will be shorter. It is notable therefore
that such protocols appear (at least based on evidence
currently available) to more consistently yield positive
changes in functional capacity that exceed those brought
about by NMES alone.
General conclusions
It is widely held that the application of NMES in a
rehabilitation setting can bring about effects that are
both adaptive and restorative. Direct compensation for
motor disability (i.e. the ‘adaptive’ response) aside,
assessment of the evidence gathered in contemporary
systematic reviews and meta-analyses suggests that NMES
delivered at levels sufficient to generate fused contractions
(Nascimento et al. 2014; Howlett et al. 2015; Monte-Silva
et al. 2019; Yang et al. 2019) is capable of promoting
restorative changes in a number of neurological disorders
that are at least equivalent to those brought about by
conventional therapy. It also appears to have a positive
effect on the functional status of older adults who do
not have neurological conditions (Langeard et al. 2017).
There is preliminary evidence that it may elevate serum
levels of brain-derived neurotrophic factor (BDNF) – a
neurotrophin that plays a well-documented role in the
expression of neural plasticity (Kimura et al. 2019).
With respect to lower levels of stimulation (e.g. using
intensities in the vicinity of motor threshold), the picture
is less clear. This is partly due to the fact that the
widely heterogeneous (in terms of stimulation parameters
and target muscles) nature of the studies that have
been conducted, generally precludes their combination
in meta-analyses (Chipchase et al. 2011a;Wattchow
et al. 2018). Given conflicting evidence concerning the
efficacy of stimulation delivered at intensities that evoke
paresthesia but generally no motor response (Grant et al.
2018), a restorative effect of low-level NMES cannot
necessarily be assumed. Nonetheless, on the basis of a
small scale meta-analysis of studies restricted to those that
adopted a variant of the stimulation protocol described
by Ridding et al. (2000) (i.e. ulnar, median or radial nerve
stimulation (10 Hz, 1 ms pulse width, duty cycle 1 s, 500 ms
on–500 ms off) for period of 2 h), it can be inferred
that NMES at an intensity close to MT may improve
upper limb motor function in (chronic) stroke survivors
(Conforto et al. 2018). Although not yet supported by
sufficient evidence derived from RCTs, there are some
indications that adjuvant techniques, such as voluntary
contractions, and mental imagery may further promote
restorative responses to NMES delivered at around motor
threshold.
What is common to all forms of NMES is the absence
of a clear understanding of the mechanisms that mediate
its influence on motor function. Evidence derived using a
range of methodologies both in humans and non-human
primates indicates clearly that NMES alters the state of
circuits in many parts of the brain, often extending beyond
the classical sensory and motor networks. An increase in
the excitability of corticospinal projections from primary
motor cortex (generally assayed using TMS) is a pervasive
feature of the immediate physiological response to NMES.
Nonetheless, there is presently no indication of which we
are aware that the increases in CSE brought about by
NMES are instrumentally related to any improvements
infunction.Webasethisconclusiononthefactthatthere
have been no reports of statistical associations between
alterations in CSE and motor function following the
administration of NMES. In addition, there are inter-
ventions that can be distinguished in terms of the changes
in CSE to which they give rise that do not differ with
respect to the changes in movement function that they
bring about. This analysis highlights the more general
concern (e.g. Carson et al. 2016) that TMS is perhaps
not the best tool for the purpose of discriminating neural
mechanisms that mediate the restorative effects of NMES
(Veldman et al. 2016).
The augmentation of the effects of NMES that occurs
when it is combined with adjuvant techniques such as
voluntary contractions and mental imagery bears the
hallmarks of associative plasticity. As we have noted
elsewhere (Carson & Kennedy, 2013) the induction of
associative effects that can be observed at a systems
level in humans does not necessarily require protocols
that adhere to the defining characteristics of STDP. The
appeal to constructs that have been elaborated in the
context of reduced sliceor animal preparations is, however,
seductive. It can also be reinforced (perhaps inadvertently)
by the identification at a systems level of features that
bear a resemblance to those that have been studied and
manipulated in vitro. It appears that in closed loop
control, such as EMG-triggered FES, temporal congruency
of the initiating (i.e. efferent) and consequential (i.e.
afferent) endogenous neural activity is critical for the
induction of restorative effects. This should not, however,
be taken as reflecting adherence to STDP rules as they
apply to individual presynaptic and postsynaptic neurons.
In seeking to provide a deeper understanding of the
mechanisms that mediate the effects of NMES, it might
also be useful to consider the influence of the integrative
properties of the brain (e.g. cortical ‘rhythms’) that
only emerge as a consequence of its topological network
properties, and the coupling of individual neural (and
non-neural) elements to which this architecture gives
rise (e.g. Guggenberger et al. 2018; Kraus et al. 2018).
There is certainly also scope for greater consideration
of the potential role of subcortical structures such as the
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 17
thalamus in mediating the changes in functional capacity
that can be induced by NMES (e.g. Kimura et al. 1999; see
also Veldman et al. 2018).
References
Allison T, McCarthy G & Wood CC (1992). The relationship
between human long-latency somatosensory evoked
potentials recorded from the cortical surface and from the
scalp. Electroencephalogr Clin Neurophysiol 84, 301–314.
Allison T, McCarthy G, Wood CC, Darcey TM, Spencer DD &
Williamson PD (1989a). Human cortical potentials evoked
by stimulation of the median nerve. I. Cytoarchitectonic
areas generating short-latency activity. J Neurophysiol 62,
694–710.
Allison T, McCarthy G, Wood CC & Jones SJ (1991). Potentials
evoked in human and monkey cerebral cortex by stimulation
of the median nerve. A review of scalp and intracranial
recordings. Brain 114, 2465–2503.
Allison T, McCarthy G, Wood CC, Williamson PD & Spencer
DD (1989b). Human cortical potentials evoked by
stimulation of the median nerve. II. Cytoarchitectonic areas
generating long-latency activity. J Neurophysiol 62, 711–722.
Anderson ML (2008). Circuit sharing and the implementation
of intelligent systems. Connection Science 20, 239–251.
Andrews RK, Schabrun SM, Ridding MC, Galea MP, Hodges
PW & Chipchase LS (2013). The effect of electrical
stimulation on corticospinal excitability is dependent on
application duration: a same subject pre-post test design. J
Neuroeng Rehabil 10, 51.
Arienzo D, Babiloni C, Ferretti A, Caulo M, Del Gratta C,
Tartaro A, Rossini PM, Romani GL (2006). Somatotopy of
anterior cingulate cortex (ACC) and supplementary motor
area (SMA) for electric stimulation of the median and tibial
nerves: An fMRI study. Neuroimage 33, 700–705.
Avanzini P, Abdollahi RO, Sartori I, Caruana F, Pelliccia V,
Casaceli G, Mai R, Lo Russo G, Rizzolatti G & Orban GA
(2016). Four-dimensional maps of the human somatosensory
system. Proc Natl Acad Sci U S A 113, E1936–E1943.
Avanzini P, Pelliccia V, Russo GL, Orban GA & Rizzolatti G
(2018). Multiple time courses of somatosensory responses in
human cortex. Neuroimage 169, 212–226.
Avenda˜
no C, Isla AJ & Rausell E (1992). Area 3a in the cat II.
Projections to the motor cortex and their relations to other
corticocortical connections. JCompNeurol321, 373–386.
Backes WH, Mess WH, van Kranen-Mastenbroek V & Reulen
JPH (2000). Somatosensory cortex responses to median
nerve stimulation: fMRI effects of current amplitude and
selective attention. Clin Neurophysiol 111, 1738–1744.
Barker RN, Brauer SG, Barry BK, Gill TJ & Carson RG (2012).
Training-induced modifications of corticospinal reactivity in
severely affected stroke survivors. Exp Brain Res 221,
211–221.
Barker RN, Brauer SG & Carson RG (2008). Training of
reaching in stroke survivors with severe and chronic upper
limb paresis using a novel nonrobotic device: a randomized
clinical trial. Stroke 39, 1800–1807.
BarkerRN,HaywardKS,CarsonRG,LloydD&BrauerSG
(2017). SMART arm training with outcome-triggered
electrical stimulation in subacute stroke survivors with
severe arm disability: a randomized controlled trial.
Neurorehabil Neural Repair 31, 1005–1016.
Barsi GI, Popovic DB, Tarkka IM, Sinkjær T & Grey MJ (2008).
Cortical excitability changes following grasping exercise
augmented with electrical stimulation. Exp Brain Res 191,
57–66.
Berghuis KM, Semmler JG, Opie GM, Post AK & Hortob´
agyi T
(2017). Age-related changes in corticospinal excitability and
intracortical inhibition after upper extremity motor
learning: a systematic review and meta-analysis. Neurobiol
Aging 55, 61–71.
Bergquist AJ, Clair JM, Lagerquist O, Mang CS, Okuma Y &
Collins DF (2011). Neuromuscular electrical stimulation:
implications of the electrically evoked sensory volley. Eur J
Appl Physiol 111, 2409.
Ben Taib NO, Nordeyn OBT, Manto M, Mario M, Pandolfo M,
Massimo P, Brotchi J & Jacques B (2005).
Hemicerebellectomy blocks the enhancement of cortical
motor output associated with repetitive somatosensory
stimulation in the rat. JPhysiol567, 293–300.
Bi GQ & Poo MM (1998). Synaptic modifications in cultured
hippocampal neurons: dependence on spike timing, synaptic
strength, and postsynaptic cell type. JNeurosci18,
10464–10472.
Blickenstorfer A, Kleiser R, Keller T, Keisker B, Meyer M,
Riener R & Kollias S (2009). Cortical and subcortical
correlates of functional electrical stimulation of wrist
extensor and flexor muscles revealed by fMRI. Hum Brain
Mapp 30, 963–975.
Boakye M, Huckins SC, Szeverenyi NM, Taskey BI & Hodge CJ
(2000). Functional magnetic resonance imaging of
somatosensory cortex activity produced by electrical
stimulation of the median nerve or tactile stimulation of the
index finger. JNeurosurg93, 774–783.
Boisgontier M, Vuillerme N & Iversen M (2010). Superimposed
electrical stimulation decreases maximal grip force. JSports
Med Phys Fitness 50, 152–158.
Brown KI, Williams ER, de Carvalho F & Baker SN (2016).
Plastic changes in human motor cortical output induced by
random but not closed-loop peripheral stimulation: the
curse of causality. Front Hum Neurosci 10, 590.
Burton H & Fabri M (1995). Ipsilateral intracortical
connections of physiologically defined cutaneous
representations in areas 3b and 1 of macaque monkeys:
projections in the vicinity of the central sulcus. JComp
Neurol 355, 508–538.
Burton H & Kopf EM (1984). Ipsilateral cortical connections
from the second and fourth somatic sensory areas in the cat.
JCompNeurol225, 527–553.
Cabric M & Appell HJ (1987). Effect of electrical stimulation of
high and low frequency on maximum isometric force and
some morphological characteristics in men. Int J Sports Med
8, 256–260.
Cambiaghi M & Sconocchia S (2018). Scribonius Largus
(probably before 1CE-after 48CE). JNeurol265, 2466–2468.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
18 R. G. Carson and A. R. Buick J Physiol 00.0
Carson RG & Kennedy NC (2013). Modulation of human
corticospinal excitability by paired associative stimulation.
Front Hum Neurosci 7, 823.
Carson RG, Nelson BD, Buick AR, Carroll TJ, Kennedy NC &
Mac Cann R (2013). Characterizing changes in the
excitability of corticospinal projections to proximal muscles
of the upper limb. Brain Stimul 6, 760–768.
Carson RG & Rankin ML (2018). Shaping the effects of
associative brain stimulation by contractions of the opposite
limb. Front Psychol 9, 2249.
Carson RG, Ruddy KL & McNickle E (2016). What do TMS-
evoked motor potentials tell us about motor learning? Adv
Exp Med Biol 957, 143–157.
Carvalho S, French M, Thibaut A, Lima W, Simis M, Leite J &
Fregni F (2018). Median nerve stimulation induced motor
learning in healthy adults: A study of timing of stimulation
and type of learning. Eur J Neurosci 48, 1667–1679.
Caspers S, Zilles K, Laird AR & Eickhoff SB (2010). ALE
meta-analysis of action observation and imitation in the
human brain. Neuroimage 50, 1148–1167.
Cavanagh PR & Komi PV (1979). Electromechanical delay in
human skeletal muscle under concentric and eccentric
contractions. Eur J Appl Physiol Occup Physiol 42, 159–163.
Charlton CS, Ridding MC, Thompson PD & Miles TS (2003).
Prolonged peripheral nerve stimulation induces persistent
changes in excitability of human motor cortex. JNeurolSci
208, 79–85.
Chipchase LS, Schabrun SM & Hodges PW (2011a). Peripheral
electrical stimulation to induce cortical plasticity: a
systematic review of stimulus parameters. Clin Neurophysiol
122, 456–463.
Chipchase LS, Schabrun SM & Hodges PW (2011b).
Corticospinal excitability is dependent on the parameters of
peripheral electric stimulation: a preliminary study. Arch
Phys Med Rehabil 92, 1423–1430.
Christensen MS & Grey MJ (2013). Modulation of
proprioceptive feedback during functional electrical
stimulation: an fMRI study. Eur J Neurosci 37, 1766–1778.
Clarac F, Massion J & Smith AM (2009). Duchenne, Charcot
and Babinski, three neurologists of La Salpetri`
ere Hospital,
and their contribution to concepts of the central
organization of motor synergy. JPhysiolParis103, 361–376.
Collins DF (2007). Central contributions to contractions
evoked by tetanic neuromuscular electrical stimulation.
Exerc Sport Sci Rev 35, 102–109.
Conforto AB, dos Anjos SM, Bernardo WM, Silva AAD, Conti
J, Machado AG & Cohen LG (2018). Repetitive peripheral
sensory stimulation and upper limb performance in stroke:
A systematic review and meta-analysis. Neurorehabil Neural
Repair 32, 863–871.
Cooke SF & Bliss TVP (2006). Plasticity in the human central
nervous system. Brain 129, 1659–1673.
CuiF,ArnsteinD,ThomasRM,MauritsNM,KeysersC&
Gazzola V (2014). Functional magnetic resonance imaging
connectivity analyses reveal efference-copy to primary
somatosensory area, BA2. PLoS One 9, e84367.
Dan Y & Poo MM (2006). Spike timing-dependent plasticity:
from synapse to perception. Physiol Rev 86, 1033–1048.
Darian-Smith C & Darian-Smith I (1993). Thalamic
projections to areas 3a, 3b, and 4 in the sensorimotor cortex
of the mature and infant macaque monkey. JCompNeurol
335, 173–199.
Darian-Smith C, Darian-Smith I, Burman K & Ratcliffe N
(1993). Ipsilateral cortical projections to areas 3a, 3b, and 4
in the macaque monkey. JCompNeurol335, 200–213.
Dawson GD (1956). The relative excitability and conduction
velocity of sensory and motor nerve fibres in man. JPhysiol
131, 436–451.
Debru A (2006). The power of torpedo fish as a pathological
model to the understanding of nervous transmission in
Antiquity. C R Biol 329, 298–302.
de Kroon JR, IJzerman MJ, Chae J, Lankhorst GJ & Zilvold G
(2005). Relation between stimulation characteristics and
clinical outcome in studies using electrical stimulation to
improve motor control of the upper extremity in stroke.
JRehabilMed37, 65–74.
Deletis V, Schild JH, Beri´
c A & Dimitrijevi ´
c MR (1992).
Facilitation of motor evoked potentials by somatosensory
afferent stimulation. Electroencephalogr Clin Neurophysiol
85, 302–310.
Del Gratta C, Della Penna S, Tartaro A, Ferretti A, Torquati K,
Bonomo L, Romani GL & Rossini PM (2000). Topographic
organization of the human primary and secondary
somatosensory areas: an fMRI study. Neuroreport 11,
2035–2043.
Del Vecchio M, Caruana F, Sartori I, Pelliccia V, Russo GL,
Rizzolatti G & Avanzini P (2019). Ipsilateral somatosensory
responses in humans: the tonic activity of SII and posterior
insular cortex. Brain Struct Funct 224, 9–18.
Desmedt JE & Ozaki I (1991). SEPs to finger joint input lack
the N20-P20 response that is evoked by tactile inputs:
contrast between cortical generators in areas 3b and 2 in
humans. Electroencephalogr Clin Neurophysiol 80,
513–521.
Devanne H, Degardin A, Tyvaert L, Bocquillon P, Houdayer E,
Manceaux A, Derambure P & Cassim F (2009). Afferent-
induced facilitation of primary motor cortex excitability in
the region controlling hand muscles in humans. Eur J
Neurosci 30, 439–448.
Edwards DJ, Dipietro L, Demirtas-Tatlidede A, Medeiros AH,
Thickbroom GW, Mastaglia FL, Krebs HI & Pascual-Leone
A (2014). Movement-generated afference paired with
transcranial magnetic stimulation: an associative stimulation
paradigm. JNeuroengRehabil11, 31.
Feldman DE (2000). Timing-based LTP and LTD at vertical
inputs to layer II/III pyramidal cells in rat barrel cortex.
Neuron 27, 45–56.
Feldman DE (2012). The spike-timing dependence of plasticity.
Neuron 75, 556–571.
Filimon F, Rieth CA, Sereno MI & Cottrell GW (2015).
Observed, executed, and imagined action representations
can be decoded from ventral and dorsal areas. Cereb Cortex
25, 3144–3158.
Finger S & Piccolino M (2011). The Shocking History of
Electric Fishes: From Ancient Epochs to the Birth of Modern
Neurophysiology. Oxford University Press, New York.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 19
Forss N, Hari R, Salmelin R, Ahonen A, H¨
am¨
al¨
ainen M, Kajola
M, Knuutila J & Simola J (1994). Activation of the human
posterior parietal cortex by median nerve stimulation. Exp
Brain Res 99, 309–315.
Francis S, Lin X, Aboushoushah S, White TP, Phillips M,
Bowtell R & Constantinescu CS (2009). fMRI analysis of
active, passive and electrically stimulated ankle dorsiflexion.
Neuroimage 44, 469–479.
Gandevia SC & Burke D (1990). Projection of thenar muscle
afferents to frontal and parietal cortex of human subjects.
Electroencephalogr Clin Neurophysiol 77, 353–361.
Gandolla M, Ferrante S, Molteni F, Guanziroli E, Frattini T,
Martegani A, Ferrigno G, Friston K, Pedrocchi A & Ward NS
(2014). Re-thinking the role of motor cortex: context-
sensitive motor outputs? Neuroimage 91, 366–374.
Gatica Tossi MA, Lillemeier A-S & Dinse HR (2013). Influence
of stimulation intensity on paired-pulse suppression of
human median nerve somatosensory evoked potentials. 24,
451–456.
Ghosh S, Brinkman C & Porter R (1987). A quantitative study
of the distribution of neurons projecting to the precentral
motor cortex in the monkey (M. fascicularis). JCompNeurol
259, 424–444.
Golaszewski SM, Bergmann J, Christova M, Kunz AB,
Kronbichler M, Rafolt D, Gallasch E, Staffen W, Trinka E &
Nardone R (2012). Modulation of motor cortex excitability
by different levels of whole-hand afferent electrical
stimulation. Clin Neurophysiol 123, 193–199.
Golaszewski SM, Siedentopf CM, Koppelstaetter F, Rhomberg
P, Guendisch GM, Schlager A, Gallasch E, Eisner W, Felber
SR & Mottaghy FM (2004). Modulatory effects on human
sensorimotor cortex by whole-hand afferent electrical
stimulation. Neurology 62, 2262–2269.
Grant G, Landgren S & Silfvenius H (1975). Columnar
distribution of U-fibres from the postcruciate cerebral
projection area of the cat’s group I muscle afferents. Exp
Brain Res 24, 57–74.
Grant VM, Gibson A & Shields N (2018). Somatosensory
stimulation to improve hand and upper limb function after
stroke—a systematic review with meta-analyses. Top St r o k e
Rehabil 25, 150–160.
Grezes J & Decety J (2001). Functional anatomy of execution,
mental simulation, observation, and verb generation of
actions: a meta-analysis. Hum Brain Mapp 12, 1–19.
Guggenberger R, Kraus D, Naros G, Le˜
ao MT, Ziemann U &
Gharabaghi A (2018). Extended enhancement of cortico-
spinal connectivity with concurrent cortical and peripheral
stimulation controlled by sensorimotor desynchronization.
Brain stimulation 11, 1331–1335.
Gunther RT (1934). The Greek Herbal of Dioscorides.Illustrated
by a Byzantine A.D. 512; Englished by John Goodyer A.D.
1655, Edited and first printed A.D. 1933, by Dr. Robert T.
Gunther. Pp. x +701 (Oxford: Dr. Robert T. Gunther, 5
Folly Bridge, 1934.) n.p.
Halonen JP, Jones S & Shawkat F (1988). Contribution of
cutaneous and muscle afferent fibres to cortical SEPs
following median and radial nerve stimulation in man.
Electroencephalogr Clin Neurophysiol 71, 331–335.
Hara Y, Obayashi S, Tsujiuchi K & Muraoka Y (2013). The
effects of electromyography-controlled functional electrical
stimulation on upper extremity function and cortical
perfusion in stroke patients. Clin Neurophysiol 124,
2008–2015.
Hardwick RM, Caspers S, Eickhoff SB & Swinnen SP (2018).
Neural correlates of action: Comparing meta-analyses of
imagery, observation, and execution. Neurosci Biobehav Rev
94, 31–44.
Hari R, Reinikainen K, Kaukoranta E, H¨
am¨
al¨
ainen M,
Ilmoniemi R, Penttinen A, Salminen J & Teszner D (1984).
Somatosensory evoked cerebral magnetic fields from SI and
SII in man. Electroencephalogr Clin Neurophysiol 57,
254–263.
Heath CJ, Hore J & Phillips CG (1976). Inputs from low
threshold muscle and cutaneous afferents of hand and
forearm to areas 3a and 3b of baboon’s cerebral cortex. J
Physiol 257, 199–227.
Hebb DO (1949). The Organization of Behavior: A
Neuropsychological Theory. John Wiley & Sons, New York.
Herman D, Kang R, MacGillis M & Zarzecki P (1985).
Responses of cat motor cortex neurons to cortico-cortical
and somatosensory inputs. Exp Brain Res 57, 598–604.
H´
etu S, Gr´
egoire M, Saimpont A, Coll MP, Eug`
ene F, Michon
PE & Jackson PL (2013). The neural network of motor
imagery: an ALE meta-analysis. Neurosci Biobehav Rev 37,
930–949.
von Holst E & Mittelstaedt H (1950). The reafference principle.
Interaction between the central nervous system and the
periphery. In The Behavioural Physiology of Animals and
Man: Selected Papers of Erich von Holst, Vol. 1, pp. 39–73.
Methuen & Co., London.
Hong IK, Choi JB & Lee JH (2012). Cortical changes after
mental imagery training combined with electromyography-
triggered electrical stimulation in patients with chronic
stroke. Stroke 43, 2506–2509.
Hore J, Preston JB & Cheney PD (1976). Responses of cortical
neurons (areas 3a and 4) to ramp stretch of hindlimb
muscles in the baboon. J Neurophysiol 39, 484–500.
Hortob´
agyi T & Maffiuletti NA (2011). Neural adaptations to
electrical stimulation strength training. Eur J Appl Physiol
111, 2439–2449.
Hortob´
agyi T, Scott K, Lambert J, Hamilton G & Tracy J
(1999). Cross-education of muscle strength is greater with
stimulated than voluntary contractions. Motor Control 3,
205–219.
Hoseini N, Munoz-Rubke F, Wan HY & Block HJ (2016).
Combined motor point associative stimulation (MPAS) and
transcranial direct current stimulation (tDCS) improves
plateaued manual dexterity performance. Neurosc i Lett 633,
134–140.
Howlett OA, Lannin NA, Ada L & McKinstry C (2015).
Functional electrical stimulation improves activity after
stroke: a systematic review with meta-analysis. Arch Phys
Med Rehabil 96, 934–943.
Huang LP, Zhou S, Lu Z, Tian Q, Li X, Cao LJ, Yu JH & Wang
H (2007). Bilateral effect of unilateral electroacupuncture
on muscle strength. J Altern Complement Med 13, 539–546.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
20 R. G. Carson and A. R. Buick J Physiol 00.0
Huerta MF & Pons TP (1990). Primary motor cortex receives
inputfromarea3ainmacaques.Brain Res 537, 367–371.
Huffman KJ & Krubitzer L (2001a). Thalamo-cortical
connections of areas 3a and M1 in marmoset monkeys. J
Comp Neurol 435, 291–310.
Huffman KJ & Krubitzer L (2001b). Area 3a: topographic
organization and cortical connections in marmoset
monkeys. Cereb Cortex 11, 849–867.
Iftime-Nielsen SD, Christensen MS, Vingborg RJ, Sinkjaer T,
Roepstorff A & Grey MJ (2012). Interaction of electrical
stimulation and voluntary hand movement in SII and the
cerebellum during simulated therapeutic functional electrical
stimulation in healthy adults. Hum Brain Mapp 33, 40–49.
Ikuno K, Matsuo A & Shomoto K (2012). Sensory electrical
stimulation for recovery of hand and arm function in stroke
patients: a review of the literature. JNovPhysiotherS1, 007.
Isaacson W (2003). Benjamin Franklin: An American Life.
Simon and Schuster, New York.
Jaberzadeh S, Mezhov D & Zoghi M (2017). What is the effect
of motor level peripheral electrical stimulation on cortico-
spinal excitability and functional outcome measures in both
healthy participants and those with neurological disorders?
A systematic review and meta-analysis. JClinPhysiotherRes
2, 1–14.
Jackson A & Zimmermann JB (2012). Neural interfaces for the
brain and spinal cord—restoring motor function. Nat Rev
Neurol 8, 690.
Joa K-L, Han Y-H, Mun C-W, Son B-K, Lee C-H, Shin Y-B, Ko
H-Y & Shin Y-I (2012). Evaluation of the brain activation
induced by functional electrical stimulation and voluntary
contraction using functional magnetic resonance imaging. J
Neuroeng Rehabil 9, 48.
Jochumsen M, Cremoux S, Robinault L, Lauber J, Arceo J,
Navid M, Nedergaard RW, Rashid U, Haavik H & Niazi I
(2018). Investigation of optimal afferent feedback modality
for inducing neural plasticity with a self-paced brain-
computer interface. Sensors 18, 3761.
Jochumsen M, Niazi IK, Signal N, Nedergaard RW, Holt K,
Haavik H & Taylor D (2016). Pairing voluntary movement
and muscle-located electrical stimulation increases cortical
excitability. Front Hum Neurosci 10, 482.
Jones EG, Coulter JD & Hendry SHC (1978). Intracortical
connectivity of architectonic fields in the somatic sensory,
motor and parietal cortex of monkeys. JCompNeurol181,
291–347.
Jones EG & Porter R (1980). What is area 3a? Brain Res Rev 2,
1–43.
Jones EG, Wise SP & Coulter JD (1979). Differential thalamic
relationships of sensory-motor and parietal cortical fields in
monkeys. JCompNeurol183, 833–881.
Kaas JH (1983). What, if anything, is SI? Organization of first
somatosensory area of cortex. Physiol Rev 63, 206–231.
Kaas JH & Pons TP (1988). The somatosensory system of
primates. In Neurosciences: Comparative Primate Biology,ed.
Steklis HD & Erwin JM, pp. 421–468. Wiley-Liss, New York.
Kadri MA, No ´
e, F., Nouar MB & Paillard T (2017). Effects of
training programs based on ipsilateral voluntary and
stimulated contractions on muscle strength and monopedal
postural control of the contralateral limb. Eur J Appl Physiol
117, 1799–1806.
Kaelin-Lang A, Luft AR, Sawaki L, Burstein AH, Sohn YH &
Cohen LG (2002). Modulation of human corticomotor
excitability by somatosensory input. JPhysiol540, 623–
633.
Karhu J & Tesche CD (1999). Simultaneous early processing of
sensory input in human primary (SI) and secondary (SII)
somatosensory cortices. J Neurophysiol 81, 2017–2025.
Kawamura T, Nakasato N, Seki K, Kanno A, Fujita S, Fujiwara
S & Yoshimoto T (1996). Neuromagnetic evidence of pre-
and post-central cortical sources of somatosensory evoked
responses. Electroencephalogr Clin Neurophysiol 100, 44–50.
Kellaway P (1946). The part played by electric fish in the early
history of bioelectricity and electrotherapy. Bull Hist Med
20, 112–137.
Khademi F, Royter V & Gharabaghi A (2019). State-dependent
brain stimulation: Power or phase? Brain Stimul 12, 296–299.
Khaslavskaia S & Sinkjaer T (2005). Motor cortex excitability
following repetitive electrical stimulation of the common
peroneal nerve depends on the voluntary drive. Exp Brain
Res 162, 497–502.
Kimura A, Grigor’yan R & Asanuma H (1999). Long-term
increases in neuronal activity in the motor cortex evoked by
simultaneous stimulation of the thalamus and somato-
sensory cortex in cats. Neurosci Behav Physiol 29, 475–481.
Kimura T, Kaneko F, Iwamoto E, Saitoh S & Yamada T (2019).
Neuromuscular electrical stimulation increases serum
brain-derived neurotrophic factor in humans. Exp Brain Res
237, 47–56.
Klingner CM, Huonker R, Flemming S, Hasler C, Brodoehl S,
Preul C, Burmeister H, Kastrup A & Witte OW (2011).
Functional deactivations: multiple ipsilateral brain areas
engaged in the processing of somatosensory information.
Hum Brain Mapp 32, 127–140.
Klostermann F, Wahl M, Schomann J, Kupsch A, Curio G &
Marzinzik F (2009). Thalamo-cortical processing of
near-threshold somatosensory stimuli in humans. Eur J
Neurosci 30, 1815–1822.
Knutson JS, Gunzler DD, Wilson RD & Chae J (2016).
Contralaterally controlled functional electrical stimulation
improves hand dexterity in chronic hemiparesis: a
randomized trial. Stroke 47, 2596–2602.
Knutson JS, Harley MY, Hisel TZ, Hogan SD, Maloney MM &
Chae J (2012). Contralaterally controlled functional
electrical stimulation for upper extremity hemiplegia: an
early-phase randomized clinical trial in subacute stroke
patients. Neurorehabil Neural Repair 26, 239–246.
Komori T, Watson BV & Brown WF (1992). Influence of
peripheral afferents on cortical and spinal motoneuron
excitability. Muscle Nerve 15, 48–51.
Korvenoja A, Huttunen J, Salli E, Pohjonen H, Martinkauppi S,
Palva JM, Lauronen L, Virtanen J, Ilmoniemi RJ & Aronen
HJ (1999). Activation of multiple cortical areas in response
to somatosensory stimulation: combined magneto-
encephalographic and functional magnetic resonance
imaging. Hum Brain Mapp 8, 13–27.
Kosar E, Waters RS, Tsukahara N & Asanuma H (1985).
Anatomical and physiological properties of the projection
from the sensory cortex to the motor cortex in normal cats:
the difference between corticocortical and thalamocortical
projections. Brain Res 345, 68–78.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 21
KrausD,NarosG,GuggenbergerR,Le
˜
ao MT, Ziemann U &
Gharabaghi A (2018). Recruitment of additional cortico-
spinal pathways in the human brain with state-dependent
paired associative stimulation. JNeurosci38, 1396–1407.
Krause T, Kurth R, Ruben J, Schwiemann J, Villringer K,
Deuchert M, Moosmann M, Brandt S, Wolf K, Curio G &
Villringer A (2001). Representational overlap of adjacent
fingers in multiple areas of human primary somatosensory
cortex depends on electrical stimulus intensity: an fMRI
study. Brain Res 899, 36–46.
Kuehn E, Dinse J, Jakobsen E, Long X, Sch¨
afer A, Bazin PL,
Villringer A, Sereno MI & Margulies DS (2017). Body
topography parcellates human sensory and motor cortex.
Cereb Cortex 27, 3790–3805.
Kunesch E, Knecht S, Schnitzler A, Tyercha C, Schmitz F &
Freund HJ (1995). Somatosensory evoked potentials elicited
by intraneural microstimulation of afferent nerve fibers. J
Clin Neurophysiol 12, 476–487.
Lakhani B, Vette AH, Mansfield A, Miyasike-daSilva V &
McIlroy WE (2012). Electrophysiological correlates of
changes in reaction time based on stimulus intensity. PLoS
One 7, e36407.
Langeard A, Bigot L, Chastan N & Gauthier A (2017). Does
neuromuscular electrical stimulation training of the lower
limb have functional effects on the elderly?: A systematic
review. Exp Gerontol 91, 88–98.
Laufer Y & Elboim-Gabyzon M (2011). Does sensory
transcutaneous electrical stimulation enhance motor
recovery following a stroke? A systematic review.
Neurorehabil Neural Repair 25, 799–809.
Lebedev MA, Denton JM & Nelson RJ (1994). Vibration-
entrained and premovement activity in monkey primary
somatosensory cortex. J Neurophysiol 72, 1654–1673.
Leube DT, Knoblich G, Erb M, Grodd W, Bartels M & Kircher
TT (2003). The neural correlates of perceiving one’s own
movements. Neuroimage 20, 2084–2090.
Levy WB & Steward O (1983). Temporal contiguity
requirements for long-term associative potentiation/
depression in the hippocampus. Neurosc ience 8, 791–797.
Lewis GN, Byblow WD & Carson RG (2001). Phasic
modulation of corticomotor excitability during passive
movement of the upper limb: effects of movement frequency
and muscle specificity. Brain Res 900, 282–294.
Liberson WT, Holmquest HJ, Scot D & Dow M (1961).
Functional electrotherapy: stimulation of the peroneal nerve
synchronized with the swing phase of the gait of hemiplegic
patients. Arch Phys Med 42, 101–105.
Lin TY, Wu JS, Lin LL, Ho TC, Lin PY & Chen JJJ (2016).
Assessments of muscle oxygenation and cortical activity
using functional near-infrared spectroscopy in healthy
adults during hybrid activation. IEEE Trans Neural Syst
Rehabil Eng 24, 1–9.
Lin YY, Shih YH, Chen JT, Hsieh JC, Yeh TC, Liao KK, Kao
CD, Lin KP, Wu ZA & Ho LT (2003). Differential effects of
stimulus intensity on peripheral and neuromagnetic cortical
responses to median nerve stimulation. Neuroimage 20,
909–917.
Litvak V, Zeller D, Oostenveld R, Maris E, Cohen A, Schramm
A, Gentner R, Zaaroor M, Pratt H & Classen J (2007).
LTP-like changes induced by paired associative stimulation
of the primary somatosensory cortex in humans: source
analysis and associated changes in behaviour. Eur J Neurosci
25, 2862–2874.
Luft AR, Kaelin-Lang A, Hauser T-K, Buitrago MM, Thakor
NV, Hanley DF & Cohen LG (2002). Modulation of rodent
cortical motor excitability by somatosensory input. Exp
Brain Res 142, 562–569.
Luft AR, Manto M-U & Oulad ben Taib N (2005). Modulation
of motor cortex excitability by sustained peripheral
stimulation: the interaction between the motor cortex and
the cerebellum. Cerebellum 4, 90–96.
Mac´
e MJ-M, Levin O, Alaerts K, Rothwell JC & Swinnen SP
(2008). Corticospinal facilitation following prolonged
proprioceptive stimulation by means of passive wrist
movement. J Clin Neurophysiol 25, 202–209.
MacKinnon CD, Verrier MC & Tatton WG (2000). Motor
cortical potentials precede long-latency EMG activity evoked
by imposed displacements of the human wrist. Exp Brain Res
131, 477–490.
Maffiuletti NA, Minetto MA, Farina D & Bottinelli R (2011).
Electrical stimulation for neuromuscular testing and
training: state-of-the art and unresolved issues. Eur J Appl
Physiol 111, 2391–2397.
Maffiuletti NA, Roig M, Karatzanos E & Nanas S (2013).
Neuromuscular electrical stimulation for preventing
skeletal-muscle weakness and wasting in critically ill
patients: a systematic review. BMC Med 11, 137.
Manca A, Hortob ´
agyi T, Rothwell J & Deriu F (2018).
Neurophysiological adaptations in the untrained side in
conjunction with cross-education of muscle strength: a
systematic review and meta-analysis. JApplPhysiol124,
1502–1518.
Mang CS, Bergquist AJ, Roshko SM & Collins DF (2012). Loss
of short-latency afferent inhibition and emergence of
afferent facilitation following neuromuscular electrical
stimulation. Neurosci Lett 529, 80–85.
Manganotti P, Storti SF, Formaggio E, Acler M, Zoccatelli G,
Pizzini FB, Alessandrini F, Bertoldo A, Toffolo GM, Bovi P,
Beltramello A, Moretto G & Fiaschi A (2012). Effect of
median-nerve electrical stimulation on BOLD activity in
acute ischemic stroke patients. Clin Neurophysiol 123,
142–153.
Manto M, Oulad Ben Taib N & Luft AR (2006). Modulation of
excitability as an early change leading to structural
adaptation in the motor cortex. JNeurosciRes83, 177–180.
Markram H, Gerstner W & Sj ¨
ostr¨
om PJ (2011). A history of
spike-timing-dependent plasticity. Front Synaptic Neurosci
3,4.
Markram H, L ¨
ubke J, Frotscher M & Sakmann B (1997).
Regulation of synaptic efficacy by coincidence of
postsynaptic APs and EPSPs. Science 275, 213–215.
Maugui`
ere F, Merlet I, Forss N, Vanni S, Jousm¨
aki V, Adeleine
P & Hari R (1997). Activation of a distributed somatosensory
cortical network in the human brain. A dipole modelling
study of magnetic fields evoked by median nerve
stimulation. Part I: Location and activation timing of SEF
sources. Electroencephalogr Clin Neurophysiol 104, 281–289.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
22 R. G. Carson and A. R. Buick J Physiol 00.0
McGie SC, Zariffa J, Popovic MR & Nagai MK (2015).
Short-term neuroplastic effects of brain-controlled and
muscle-controlled electrical stimulation. Neuromodulation
18, 233–240.
McKay D, Brooker R, Giacomin P, Ridding M & Miles T (2002).
Time course of induction of increased human motor cortex
excitability by nerve stimulation. Neuroreport 13, 1271–1273.
McLaughlin DF & Kelly EF (1993). Evoked potentials as indices
of adaptation in the somatosensory system in humans: a
review and prospectus. Brain Res Rev 18, 151–206.
McNickle E & Carson RG (2015). Paired associative
transcranial alternating current stimulation increases the
excitability of corticospinal projections in humans. JPhysiol
593, 1649–1666.
Miall RC & Wolpert DM (1996). Forward models for
physiological motor control. Neural Netw 9, 1265–1279.
Mima T, Terada K, Maekawa M, Nagamine T, Ikeda A &
Shibasaki H (1996). Somatosensory evoked potentials
following proprioceptive stimulation of finger in man. Exp
Brain Res 111, 233–245.
Molenberghs P, Cunnington R & Mattingley JB (2012). Brain
regions with mirror properties: a meta-analysis of 125
human fMRI studies. Neurosci Biobehav Rev 36, 341–349.
Monte-Silva K, Piscitelli D, Norouzi-Gheidari N, Batalla MAP,
Archambault P & Levin MF (2019). Electromyogram-related
neuromuscular electrical stimulation for restoring wrist and
hand movement in poststroke hemiplegia: A systematic
review and meta-analysis. Neurorehabil Neural Repair 33,
96–111.
Mrachacz-Kersting N, Jiang N, Stevenson AJT, Niazi IK, Kostic
V, Pavlovic A, Radovanovic S, Djuric-Jovicic M, Agosta F,
Dremstrup K & Farina D (2016). Efficient neuroplasticity
induction in chronic stroke patients by an associative
brain-computer interface. JNeurophysiology115, 1410–
1421.
Mrachacz-Kersting N, Kristensen SR, Niazi IK & Farina D
(2012). Precise temporal association between cortical
potentials evoked by motor imagination and afference
induces cortical plasticity. JPhysiol590, 1669–1682.
Mrachacz-Kersting N, Stevenson AJ, Jørgensen HR, Severinsen
KE, Aliakbaryhosseinabadi S, Jiang N & Farina D (2019).
Brain state–dependent stimulation boosts functional
recovery following stroke. Ann Neurol 85, 84–95.
Mrachacz-Kersting N, Voigt M, Stevenson AJT,
Aliakbaryhosseinabadi S, Jiang N, Dremstrup K & Farina D
(2017). The effect of type of afferent feedback timed with
motor imagery on the induction of cortical plasticity. Brain
Res 1674, 91–100.
M¨
uller-Dahlhaus F, Ziemann U & Classen J (2010). Plasticity
resembling spike-timing dependent synaptic plasticity: the
evidence in human cortex. Front Synaptic Neurosci 2, 34.
Nascimento LR, Michaelsen SM, Ada L, Polese JC &
Teixeira-Salmela LF (2014). Cyclical electrical stimulation
increases strength and improves activity after stroke: a
systematic review. JPhysiother60, 22–30.
Nelson AJ, Staines WR, Graham SJ & McIlroy WE (2004).
Activation in SI and SII; the influence of vibrotactile
amplitude during passive and task-relevant stimulation.
Brain Res Cogn Brain Res 19, 174–184.
Nelson RJ (1996). Interactions between motor commands and
somatic perception in sensorimotor cortex. Curr Opin
Neurobiol 6, 801–810.
Nelson RJ, Smith BN & Douglas VD (1991). Relationships
between sensory responsiveness and premovement activity
of quickly adapting neurons in areas 3b and 1 of monkey
primary somatosensory cortex. Exp Brain Res 84, 75–90.
Niazi IK, Mrachacz-Kersting N, Jiang N, Dremstrup K & Farina
D (2012). Peripheral electrical stimulation triggered by
self-paced detection of motor intention enhances motor
evoked potentials. IEEE Trans Neural Syst Rehabil Eng 20,
595–604.
Nihashi T, Naganawa S, Sato C, Kawai H, Nakamura T, Fukatsu
H, Ishigaki T & Aoki I (2005). Contralateral and ipsilateral
responses in primary somatosensory cortex following
electrical median nerve stimulation—an fMRI study. Clin
Neurophysiol 116, 842–848.
Okuyama K, Ogura M, Kawakami M, Tsujimoto K, Okada K,
Miwa K, Takahashi Y, Abe K, Tanabe S, Yamaguchi T & Liu
M (2018). Effect of the combination of motor imagery and
electrical stimulation on upper extremity motor function in
patients with chronic stroke: preliminary results. Ther Adv
Neurol Disord 11, 1756286418804785.
Olsen S, Signal N, Niazi IK, Christensen T, Jochumsen M &
Taylor D (2018). Paired associative stimulation delivered by
pairing movement-related cortical potentials with peripheral
electrical stimulation: An investigation of the duration of
neuromodulatory effects. Neuromodulation 21, 362–367.
Padel Y & Relova JL (1991). Somatosensory responses in the cat
motor cortex. I. Identification and course of an afferent
pathway. J Neurophysiol 66, 2041–2058.
Panizza M, Nilsson J & Hallett M (1989). Optimal stimulus
duration for the H reflex. Muscle Nerve 12, 576–579.
Panizza M, Nilsson J, Roth BJ, Basser PJ & Hallett M (1992).
Relevance of stimulus duration for activation of motor and
sensory fibers: implications for the study of H-reflexes and
magnetic stimulation. Electroencephalogr Clin Neurophysiol
85, 22–29.
Parain D & Delapierre G (1991). Effects of stimulus intensity
increase on short-latency somatosensory evoked potentials:
application of polynomial curvature coefficients. Brain
Topog r 4, 31–35.
Park JH (2019). Effects of mental imagery training combined
electromyogram-triggered neuromuscular electrical
stimulation on upper limb function and activities of daily
living in patients with chronic stroke: a randomized
controlled trial. Disabil Rehabil 4, 1–6.
Peckham PH & Knutson JS (2005). Functional electrical
stimulation for neuromuscular applications. Annu Rev
Biomed Eng 7, 327–360.
Pomeroy V, Aglioti SM, Mark VW, McFarland D, Stinear C,
Wolf SL, Corbetta M & Fitzpatrick SM (2011). Neurological
principles and rehabilitation of action disorders:
rehabilitation interventions. Neurorehabil Neural Repair
25(Suppl), 33S–43S.
Pons TP & Kaas JH (1986). Corticocortical connections of area
2 of somatosensory cortex in macaque monkeys: a
correlative anatomical and electrophysiological study.
JCompNeurol248, 313–335.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 23
Porter LL & Sakamoto K (1988). Organization and synaptic
relationships of the projection from the primary sensory to
the primary motor cortex in the cat. JCompNeurol271,
387–396.
Porter LL, Sakamoto T & Asanuma H (1990). Morphological
and physiological identification of neurons in the cat motor
cortex which receive direct input from the somatic sensory
cortex. Exp Brain Res 80, 209–212.
Prochazka A (2018). Motor Neuroprostheses. Compr Physiol 9,
127–148.
Ridding MC, Brouwer B, Miles TS, Pitcher JB & Thompson PD
(2000). Changes in muscle responses to stimulation of the
motor cortex induced by peripheral nerve stimulation in
human subjects. Exp Brain Res 131, 135–143.
Ridding MC, McKay DR, Thompson PD & Miles TS (2001).
Changes in corticomotor representations induced by
prolonged peripheral nerve stimulation in humans. Clin
Neurophysiol 112, 1461–1469.
Ridding MC & Taylor JL (2001). Mechanisms of motor-evoked
potential facilitation following prolonged dual peripheral
and central stimulation in humans. JPhysiol537, 623–631.
Rizzo V, Terranova C, Crupi D, Sant’Angelo A, Girlanda P &
Quartarone A (2014). Increased transcranial direct current
stimulation after effects during concurrent peripheral
electrical nerve stimulation. Brain Stimul 7, 113–121.
Ruddy KL, Leemans A, Woolley DG, Wenderoth N & Carson
RG (2017). Structural and functional cortical connectivity
mediating cross education of motor function. JNeurosci37,
2555–2564.
Ruddy KL, Rudolf AK, Kalkman B, King M, Daffertshofer A,
Carroll TJ & Carson RG (2016). Neural adaptations
associated with interlimb transfer in a ballistic wrist flexion
task. Front Hum Neurosci 10, 204.
Rushton DN (2003). Functional electrical stimulation and
rehabilitation—an hypothesis. Med Eng Phys 25, 75–78.
Saito K, Sugawara K, Miyaguchi S, Matsumoto T, Kirimoto H,
Tamaki H & Onishi H (2014). The modulatory effect of
electrical stimulation on the excitability of the corticospinal
tract varies according to the type of muscle contraction
being performed. Front Hum Neurosci 8, 835.
Sakamoto T, Porter LL & Asanuma H (1987). Long-lasting
potentiation of synaptic potentials in the motor cortex
produced by stimulation of the sensory cortex in the cat: a
basis of motor learning. Brain Res 413, 360–364.
Sasaki R, Kotan S, Nakagawa M, Miyaguchi S, Kojima S, Saito
K, Inukai Y & Onishi H (2017). Presence and absence of
muscle contraction elicited by peripheral nerve electrical
stimulation differentially modulate primary motor cortex
excitability. Front Hum Neurosci 11, 146.
Savaki HE & Raos V (2019). Action perception and motor
imagery: Mental practice of action. Prog Neurobiol 175,
107–125.
Schabrun SM & Ridding MC (2007). The influence of
correlated afferent input on motor cortical representations
in humans. Exp Brain Res 183, 41–49.
Schabrun SM, Ridding MC, Galea MP, Hodges PW &
Chipchase LS (2012). Primary sensory and motor cortex
excitability are co-modulated in response to peripheral
electrical nerve stimulation. PLoS One 7, e51298.
Schwark HD, Esteky H & Jones EG (1992). Corticocortical
connections of cat primary somatosensory cortex. Exp Brain
Res 91, 425–434.
Seeman SC, Mogen BJ, Fetz EE & Perlmutter SI (2017). Paired
stimulation for spike-timing-dependent plasticity in primate
sensorimotor cortex. JNeurosci37, 1935–1949.
Sheffler LR & Chae J (2007). Neuromuscular electrical
stimulation in neurorehabilitation. Muscle Nerve 35,
562–590.
Shen Y, Yin Z, Fan Y, Chen CF, Dai W, Yi W, Li Y, Zhang W,
Zhang Y, Bian R, Huang Y, Machado S, Yuan TF & Shan C
(2015). Comparison of the effects of contralaterally
controlled functional electrical stimulation and
neuromuscular electrical stimulation on upper extremity
functions in patients with stroke. CNSNeurolDisordDrug
Ta r g e t s 14, 1260–1266.
Sherrington CS (1894). On the anatomical constitution of
nerves of skeletal muscles; with remarks on recurrent fibres
in the ventral spinal nerve-root. JPhysiol17, 210–258.
Shitara H, Shinozaki T, Takagishi K, Honda M & Hanakawa T
(2013). Movement and afferent representations in human
motor areas: a simultaneous neuroimaging and transcranial
magnetic/peripheral nerve-stimulation study. Front Hum
Neurosci,7, 554.
Shulga A, Zubareva A, Lioumis P & M¨
akel¨
a JP (2016). Paired
associative stimulation with high-frequency peripheral
component leads to enhancement of corticospinal
transmission at wide range of interstimulus intervals. Front
Hum Neurosci 10, 470.
Simos PG, Kavroulakis E, Maris T, Papadaki E, Boursianis T,
Kalaitzakis G & Savaki HE (2017). Neural foundations of
overt and covert actions. Neuroimage 152, 482–496.
Smith GV, Alon G, Roys SR & Gullapalli RP (2003). Functional
MRI determination of a dose-response relationship to lower
extremity neuromuscular electrical stimulation in healthy
subjects. Exp Brain Res 150, 33–39.
Song S, Miller KD & Abbott LF (2000). Competitive Hebbian
learning through spike-timing-dependent synaptic
plasticity. Nat Neurosci 3, 919.
Spiegel J, Tintera J, Gawehn J, Stoeter P & Treede RD (1999).
Functional MRI of human primary somatosensory and
motor cortex during median nerve stimulation. Clin
Neurophysiol 110, 47–52.
Stefan K, Kunesch E, Cohen LG, Benecke R & Classen J (2000).
Induction of plasticity in the human motor cortex by paired
associative stimulation. Brain 123, 572–584.
Stefan K, Wycislo M & Classen J (2004). Modulation of
associative human motor cortical plasticity by attention. J
Neurophysiol 92, 66–72.
Stein RB, Everaert DG, Roy FD, Chong S & Soleimani M
(2013). Facilitation of corticospinal connections in
able-bodied people and people with central nervous system
disorders using eight interventions. J Clin Neurophysiol 30,
66–78.
Summers SJ, Schabrun SM, Marinovic W & Chipchase LS
(2017). Peripheral electrical stimulation increases
corticomotor excitability and enhances the rate of
visuomotor adaptation. Behav Brain Res 322, 42–50.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
24 R. G. Carson and A. R. Buick J Physiol 00.0
Suppa A, Quartarone A, Siebner H, Chen R, Di Lazzaro V, Del
Giudice P, Paulus W, Rothwell JC, Ziemann U & Classen J
(2017). The associative brain at work: evidence from paired
associative stimulation studies in humans. Clin Neurophysiol
128, 2140–2164.
Takahashi Y, Kawakami M, Yamaguchi T, Idogawa Y, Tanabe S,
Kondo K & Liu M (2019). Effects of leg motor imagery
combined with electrical stimulation on plasticity of
corticospinal excitability and spinal reciprocal inhibition.
Front Neurosci 13, 149.
Taylor L, Lewis GN & Taylor D (2012). Short-term effects of
electrical stimulation and voluntary activity on corticomotor
excitability in healthy individuals and people with stroke. J
Clin Neurophysiol 29, 237–243.
Tolmacheva A, M¨
akel¨
a JP & Shulga A (2019). Increasing the
frequency of peripheral component in paired associative
stimulation strengthens its efficacy. Sci Rep 9, 3849.
Traversa R, Cicinelli P, Bassi A, Rossini PM & Bernardi G
(1997). Mapping of motor cortical reorganization after
stroke: a brain stimulation study with focal magnetic pulses.
Stroke 28, 110–117.
Turco CV, El-Sayes J, Savoie MJ, Fassett HJ, Locke MB & Nelson
AJ (2018). Short-and long-latency afferent inhibition; uses,
mechanisms and influencing factors. Brain Stimul 11, 59–74.
Urasaki E, Wada SI, Yasukouchi H & Yokota A (1998). Effect of
transcutaneous electrical nerve stimulation (TENS) on
central nervous system amplification of somatosensory
input. JNeurol245, 143–148.
Uy J & Ridding MC (2003). Increased cortical excitability
induced by transcranial DC and peripheral nerve
stimulation. JNeurosciMethods127, 193–197.
Vang C, Dunbabin D & Kilpatrick D (1999). Correlation
between functional and electrophysiological recovery in
acute ischemic stroke. Stroke 30, 2126–2130.
Veale JL, Mark RF & Rees S (1973). Differential sensitivity of
motor and sensory fibres in human ulnar nerve. JNeurol
Neurosurg Psychiatry 36, 75–86.
Veldman MP, Maffiuletti NA, Hallett M, Zijdewind I &
Hortob´
agyi T (2014). Direct and crossed effects of somato-
sensory stimulation on neuronal excitability and motor
performance in humans. Neurosci Biobehav Rev 47, 22–35.
Veldman MP, Maurits NM, Zijdewind I, Maffiuletti NA, van
Middelkoop S, Mizelle JC & Hortob´
agyi T (2018). Somato-
sensory electrical stimulation improves skill acquisition,
consolidation, and transfer by increasing sensorimotor
activity and connectivity. J Neurophysiol 120, 281–290.
Veldman MP, Zijdewind I, Solnik S, Maffiuletti NA, Berghuis
KMM, Javet M, N´
egyesi J & Hortob´
agyi T (2015). Direct and
crossed effects of somatosensory electrical stimulation on
motor learning and neuronal plasticity in humans. Eur J
Appl Physiol 115, 2505–2519.
Veldman MP, Zijdewind I, Maffiuletti NA & Hortob´
agyi T
(2016). Motor skill acquisition and retention after
somatosensory electrical stimulation in healthy humans.
Front Hum Neurosci 10, 115.
Waters RS, Favorov O, Mori A & Asanuma H (1982). Pattern of
projection and physiological properties of cortico-cortical
connections from the posterior bank of the ansate sulcus to
the motor cortex, area 4γ,inthecat.Exp Brain Res 48,
335–344.
Wattchow KA, McDonnell MN & Hillier SL (2018).
Rehabilitation interventions for upper limb function in the
first four weeks following stroke: a systematic review and
meta-analysis of the evidence. Arch Phys Med Rehabil 99,
367–382.
Wegrzyk J, Ranjeva JP, Four´
e A, Kavounoudias A, Vilmen C,
Mattei JP, Guye M, Maffiuletti NA, Place N, Bendahan D &
Gondin J (2017). Specific brain activation patterns
associated with two neuromuscular electrical stimulation
protocols. Sci Rep 7, 2742.
Wiesendanger, MARIO & Miles TS (1982). Ascending pathway
of low-threshold muscle afferents to the cerebral cortex and
its possible role in motor control. Physiol Rev 62, 1234–
1270.
Wilson RD, Page SJ, Delahanty M, Knutson JS, Gunzler DD,
Sheffler LR & Chae J (2016). Upper-limb recovery after
stroke: a randomized controlled trial comparing EMG-
triggered, cyclic, and sensory electrical stimulation.
Neurorehabil Neural Repair 30, 978–987.
Wu CW-H, Van Gelderen P, Hanakawa T, Yaseen Z & Cohen
LG (2005). Enduring representational plasticity after
somatosensory stimulation. Neuroimage 27, 872–884.
Wundt WM (1863). Vorlesungen ueber Menschen-und
Thierseele, vol 2. Voss, Leipzig.
Xu R, Wang Y, Wang K, Zhang S, He C & Ming D (2018).
Increased corticomuscular coherence and brain activation
immediately after short-term neuromuscular electrical
stimulation. Front Neurol 9, 886.
YamashiroK,InuiK,OtsuruN,KidaT,AkatsukaK&KakigiR
(2008). Somatosensory off-response in humans: an ERP
study. Exp Brain Res 190, 207–213.
Yamashiro K, Inui K, Otsuru N, Kida T & Kakigi R (2009).
Somatosensory off-response in humans: an MEG study.
Neuroimage 44, 1363–1368.
Yang JD, Liao CD, Huang SW, Tam KW, Liou TH, Lee YH, Lin
CY & Chen HC (2019). Effectiveness of electrical
stimulation therapy in improving arm function after stroke:
a systematic review and a meta-analysis of randomised
controlled trials. Clin Rehabil 33, 1286–1297.
Yasui T, Yamaguchi T, Tanabe S, Tatemoto T, Takahashi Y,
Kondo K & Kawakami M (2019). Time course of changes in
corticospinal excitability induced by motor imagery during
action observation combined with peripheral nerve
electrical stimulation. Exp Brain Res 237, 637–645.
You SJ & Lee JH (2013). Effects of mental activity training
linked with electromyogram-triggered electrical stimulation
on paretic upper extremity motor function in chronic stroke
patients: a pilot trial. Tu r k J P hy s Med Re h a b 59, 133–139.
Yumiya H & Ghez C (1984). Specialized subregions in the cat
motor cortex: anatomical demonstration of differential
projections to rostral and caudal sectors. Exp Brain Res 53,
259–276.
Zarzecki P, Shinoda Y & Asanuma H (1978). Projection from
area 3a to the motor cortex by neurons activated from group
I muscle afferents. Exp Brain Res 33, 269–282.
Zheng Y, Mao M, Cao Y & Lu X (2019). Contralaterally
controlled functional electrical stimulation improves wrist
dorsiflexion and upper limb function in patients with
early-phase stroke: a randomized controlled trial. JRehabil
Med 51, 103–108.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
J Physiol 00.0 Neuromuscular electrical stimulation 25
Zhou S, Oakman A & Davie AJ (2002). Effects of unilateral
voluntary and electromyostimulation training on muscular
strength on the contralateral limb. Hong Kong J Sports Med
Sports Sci 14, 1–11.
Additional information
Competing interests
There are no competing interests of which the authors are aware.
Author contributions
Both authors contributed to the conception and writing of this
review. Both authors have read and approved the final version
of this manuscript and agree to be accountable for all aspects of
the work in ensuring that questions related to the accuracy or
integrity of any part of the work are appropriately investigated
and resolved. All persons designated as authors qualify for
authorship, and all those who qualify for authorship are listed.
Funding
The support of the Northern Ireland Department for
Employment and Learning is acknowledged. Richard Carson
thanks Atlantic Philanthropies for their generous support of his
research through their funding of the NEIL (Neuroenhancement
for Independent Living) programme at Trinity College Institute
of Neuroscience.
C
2019 The Authors. The Journal of Physiology C
2019 The Physiological Society
A preview of this full-text is provided by Wiley.
Content available from The Journal of Physiology
This content is subject to copyright. Terms and conditions apply.
