Source: Comprehensive Guide to Neuromuscular Physiology and EMG Analysis (University of Florida)
Difficulty: Intermediate | Prerequisites: Neuromuscular physiology notes (motor units, recruitment, NMJ). Basic understanding of electrical signals is helpful but not essential.
Electromyography (EMG) is the tool that lets researchers and clinicians observe what motor units are doing inside a living muscle, in real time. Where the neuromuscular physiology notes explain the biology, these notes cover how that biology is measured, processed, and interpreted. EMG bridges the gap between textbook physiology and clinical or laboratory practice. If you are comfortable with what a motor unit is and how recruitment works, everything here will build naturally on top of that.
EMG records the electrical activity of skeletal muscles using electrodes, amplifies the tiny signals with a bioamplifier, and presents them as either raw waveforms or processed (integrated) curves. Raw EMG is noisy and complex; integrated EMG is smoothed and easier to compare across conditions. EMG amplitude generally increases with force because more motor units are recruited and fire faster, but signals cannot be compared between individuals without normalisation.
Electromyography (EMG)
A technique for recording the electrical activity produced by skeletal muscles during contraction. In simple terms, it is a way of "listening" to your muscles' electrical chatter through electrodes on the skin (surface EMG) or inserted into the muscle (intramuscular EMG).
Bioamplifier
An electronic device that takes the tiny electrical signals picked up by EMG electrodes (on the order of microvolts) and amplifies them to a level that can be displayed and analysed (millivolts). It also filters out unwanted noise.
Raw EMG
The unprocessed electrical signal recorded from the muscle. It appears as a complex, irregular waveform with both positive and negative deflections, and includes noise alongside the real muscle signal.
Integrated EMG (iEMG)
A processed version of the raw signal. The signal is rectified (all values made positive), then smoothed over time to produce a cleaner curve that tracks overall muscle activation. Think of it as a running average of how "active" the muscle is.
Maximal voluntary contraction (MVC)
The greatest force a subject can produce voluntarily. Used as the 100% reference point so that submaximal efforts can be expressed as a percentage (e.g., 50% MVC).
Motor unit recruitment (in EMG context)
The progressive activation of additional motor units as force demand increases, visible in the EMG as a rise in signal amplitude.
Firing rate / rate coding
The frequency at which a motor neuron sends action potentials to its muscle fibres. Higher firing rates produce greater force and alter the frequency content of the EMG signal.
Muscle fatigue (EMG perspective)
A decline in force-producing capacity during sustained or repeated contraction. In the EMG, fatigue often shows as increasing amplitude (more motor units recruited to compensate) and decreasing frequency (slower conduction velocity in fatigued fibres).
Normalisation
The process of expressing an EMG value as a percentage of a reference value (typically MVC). Required for meaningful comparisons between muscles, sessions, or individuals.
Synergistic muscles
Muscles that assist the primary mover (agonist) during a given action. For example, during a biceps curl, the brachialis and brachioradialis are synergists to the biceps brachii.
What EMG measures
EMG records the summed electrical potentials from all active motor units within range of the electrodes
The signal reflects three things: how many motor units are active (recruitment), how frequently each is firing (rate coding), and the conduction properties of the muscle fibres themselves
Surface vs intramuscular EMG
Surface EMG uses electrodes placed on the skin over the muscle. It captures the aggregate signal from many motor units and is non-invasive. This is the type used in most undergraduate lab settings
Intramuscular EMG uses fine-wire or needle electrodes inserted directly into the muscle. It can isolate individual motor unit action potentials but is more invasive
Role of the bioamplifier
Muscle electrical signals are very small (typically 0.01 to 5 mV at the skin surface), so they must be amplified before analysis
The bioamplifier also applies filters to reduce artefacts from sources such as mains electricity (50/60 Hz), electrode movement, and nearby electronic equipment
The output can be displayed in real time on a screen or stored digitally for later analysis
Raw EMG
The unprocessed signal straight from the electrodes and amplifier
Appears as a complex, oscillating waveform with rapid positive and negative spikes
Contains real motor unit potentials mixed with noise and artefacts
Difficult to interpret by eye because of its complexity, but useful for identifying individual motor unit action potentials (with intramuscular electrodes)
Signal processing steps
Filtering: high-pass and low-pass filters remove frequencies outside the range of interest (typically 20 to 500 Hz for surface EMG). This strips out slow movement artefacts and high-frequency electronic noise
Rectification: all negative values are flipped to positive, turning the oscillating signal into one that only goes upward. This is necessary because the raw signal's average would otherwise be close to zero
Smoothing / integration: the rectified signal is averaged over a sliding time window (e.g., 50 to 200 ms), producing a smooth envelope that tracks the overall level of muscle activation over time
Why integrated EMG is useful
Produces a clean curve that rises and falls with muscle activity, making it straightforward to compare activation across different force levels, different exercises, or different time points
Allows quantification of total muscle effort during a task
Enables meaningful comparison of activation between conditions (e.g., 25% vs 75% MVC), though comparison between individuals still requires normalisation
Establishing a baseline: maximal voluntary contraction (MVC)
The subject grips a dynamometer (or similar device) as hard as possible for approximately 5 seconds
Both the peak force and the corresponding EMG amplitude are recorded
This value becomes the 100% MVC reference for all subsequent measurements
Submaximal contraction trials
The subject performs contractions at set percentages of their MVC: typically 25%, 50%, 75%, and 100%
As the target force increases, EMG amplitude rises because more motor units are recruited and existing units fire more rapidly
The relationship between force and EMG amplitude is roughly linear at low to moderate forces but may become non-linear at high forces
Fatigue protocol
The subject maintains a steady contraction at 50% MVC for 2 minutes
Over time, force output tends to decline even though the subject tries to maintain the target
EMG amplitude may increase during fatigue (the nervous system recruits additional motor units to compensate for those losing force) or may decrease (if conduction along fatigued fibres slows)
Frequency content of the EMG signal shifts downward during fatigue, a hallmark sign used in research and clinical assessment
Exercise comparison
Different types of biceps curls (e.g., standard curl, hammer curl, concentration curl) produce different EMG patterns
This occurs because each exercise changes the joint angle, the involvement of synergistic muscles (brachialis, brachioradialis), and the mechanical advantage of the biceps brachii
Comparing EMG across exercises reveals which movements maximise activation of a target muscle
Why EMG cannot be compared directly between individuals
Skin impedance varies from person to person (and even day to day on the same person)
Electrode placement is never perfectly identical between sessions or subjects
Subcutaneous fat thickness affects how much of the signal reaches the electrode
Muscle size and fibre composition differ between individuals
Normalisation to MVC (or another reference contraction) is the standard solution
Clinical applications
Diagnosing neuromuscular diseases: muscular dystrophy, peripheral neuropathies, and nerve injuries all produce characteristic abnormal EMG patterns
Abnormal findings include spontaneous activity at rest (fibrillation potentials, fasciculations), reduced recruitment (fewer motor units firing than expected), and altered motor unit morphology
Nerve conduction studies paired with EMG help localise the site of a lesion (nerve root, plexus, peripheral nerve, or muscle)
Interpreting fatigue in EMG
During sustained contraction, EMG amplitude often rises as the nervous system compensates for force loss by recruiting additional motor units
Alternatively, if muscle fibre conduction velocity drops significantly, the EMG amplitude may decrease
The median frequency of the EMG power spectrum shifts downward during fatigue, a more reliable fatigue marker than amplitude alone
Exercise-specific EMG patterns
Different exercises recruit synergistic muscles to varying degrees
A biceps curl performed with a supinated grip preferentially activates the biceps brachii; a neutral grip shifts load toward the brachioradialis
EMG data from multiple muscles can map the coordination pattern of a movement
Physiotherapists use EMG biofeedback to help patients relearn muscle activation after stroke or surgery: the patient watches their own EMG signal on a screen and tries to increase it, turning an invisible process into a visible one. In sports science, EMG is used to compare exercises and determine which movements produce the highest activation of a target muscle, informing training programme design. Ergonomists use EMG to assess which workplace postures cause excessive muscle loading, guiding the design of tools and workstations that reduce fatigue and injury risk.
Students often think higher EMG amplitude always means more force. This is roughly true within one person and one session, but raw EMG values cannot be compared between people or even between sessions without normalisation.
A common mistake is confusing raw EMG with integrated EMG on an exam. Raw EMG is the noisy, oscillating signal. Integrated EMG is the processed, smoothed version. Know which is which and when each is useful.
Students sometimes assume that if EMG amplitude increases during a fatigue trial, the muscle must be getting stronger. The opposite is true: amplitude rises because the nervous system is recruiting additional motor units to compensate for failing ones.
Some students believe EMG directly measures force. It does not. EMG measures electrical activity, which correlates with force but is influenced by many other factors (electrode placement, skin impedance, subcutaneous tissue, fibre type composition).
⚠️ Be able to explain why EMG amplitude increases with force (more motor units recruited, higher firing rates)
⚠️ Know the difference between raw and integrated EMG, including the processing steps (filtering, rectification, smoothing)
⚠️ Explain why EMG signals cannot be compared between individuals without normalisation, and list the factors that differ (skin impedance, electrode placement, muscle size, subcutaneous tissue)
⚠️ Describe what happens to EMG during a fatigue protocol: amplitude may rise (compensation) while median frequency shifts downward (slowed conduction)
⚠️ Understand the MVC protocol: what it establishes, why it is performed first, and how submaximal efforts are expressed relative to it
⚠️ Be able to name at least two clinical uses of EMG (diagnosing muscular dystrophy, peripheral neuropathy, nerve injury)
True or false: Raw EMG produces a smooth curve that is easy to interpret. (False. Raw EMG is noisy and complex. The smooth curve comes from integrated EMG.)
Fill in the blank: The device that amplifies the tiny muscle signals and filters noise is called a ______. (bioamplifier)
True or false: You can directly compare raw EMG values between two different people. (False. Differences in skin impedance, electrode placement, and subcutaneous tissue make direct comparison invalid without normalisation.)
Fill in the blank: During a fatigue protocol, the median frequency of the EMG signal tends to shift ______. (downward / decrease)
True or false: EMG directly measures the force produced by a muscle. (False. EMG measures electrical activity, which correlates with force but is not a direct measure of it.)
Q: What are the three main things EMG provides insight into?
A: Motor unit recruitment (how many motor units are active), firing rates (how frequently motor neurons are activating their muscle fibres), and muscle fatigue (changes in amplitude and frequency over time).
Q: Describe the signal processing steps that convert raw EMG into integrated EMG.
A: First, the raw signal is filtered (band-pass, typically 20 to 500 Hz) to remove artefacts and noise outside the frequency range of muscle activity. Next, the filtered signal is rectified (all negative values converted to positive). Finally, the rectified signal is smoothed by averaging over a sliding time window, producing a clean envelope that tracks overall activation.
Q: Why is normalisation to MVC necessary when comparing EMG data?
A: Raw EMG amplitude depends on factors unrelated to muscle effort, including skin impedance, electrode placement, subcutaneous fat thickness, and muscle geometry. These vary between individuals and between recording sessions. Expressing EMG as a percentage of MVC removes these confounding variables and allows meaningful comparison.
Q: During a sustained 50% MVC contraction, a student notices that EMG amplitude gradually increases while force stays roughly constant. Explain what is happening physiologically.
A: As individual motor units fatigue and produce less force, the nervous system recruits additional motor units and increases the firing rate of existing ones to maintain the target force. This compensatory recruitment generates more total electrical activity, so EMG amplitude rises even though force output remains the same.
Q: Name two clinical conditions diagnosed with the help of EMG and describe one abnormal finding you might see.
A: Muscular dystrophy and peripheral neuropathy are two examples. In peripheral neuropathy, EMG may show reduced recruitment (fewer motor units firing than expected for the effort) because damaged nerves fail to activate their motor units. Spontaneous electrical activity at rest (fibrillations) may also appear, indicating denervated muscle fibres firing on their own.
Everything in this document rests on the neuromuscular physiology covered in the companion notes. Motor unit recruitment and the Size Principle explain why EMG amplitude increases with force, and excitation-contraction coupling explains the electrical events that EMG is detecting.
If your course covers biomechanics or kinesiology, EMG data is central to understanding which muscles are active during specific movements and how coordination patterns change with training, fatigue, or injury.
The clinical EMG material connects to neuroanatomy and pathology. Understanding which nerve supplies which muscle (peripheral nerve distributions, dermatomes, myotomes) is necessary for interpreting clinical EMG findings and localising lesions.
Electromyography, EMG, surface EMG, intramuscular EMG, bioamplifier, raw EMG, integrated EMG, iEMG, rectification, signal processing, filtering, smoothing, motor unit recruitment, firing rate, rate coding, muscle fatigue, median frequency, power spectrum, maximal voluntary contraction, MVC, submaximal contraction, normalisation, skin impedance, electrode placement, subcutaneous tissue, neuromuscular disease, muscular dystrophy, peripheral neuropathy, nerve injury, fibrillation potentials, fasciculations, reduced recruitment, synergistic muscles, brachialis, brachioradialis, biceps brachii, biceps curl, EMG biofeedback, ergonomics, clinical EMG, nerve conduction study