What Electromyography Can And Cannot Tell You

Comparisons claiming one exercise activates a muscle more than another usually rest on surface electromyography. The technique measures something real and considerably narrower than those claims require.

What the signal actually is

Electrodes on the skin detect the summed electrical activity of muscle fibers beneath them as motor units fire.

That signal is processed into an amplitude, which is then interpreted as how much the muscle is working during the task.

It is a measure of electrical events rather than of force, and the two are related in a way that changes with contraction type, speed and fatigue.

Why the amplitude is unstable

Electrode placement changes the reading substantially, and a few millimeters of difference alters which fibers dominate the signal.

The amount of tissue between muscle and electrode filters the signal, so the same muscle produces different amplitudes in different people.

Skin preparation, sweat during a session, and movement of the skin relative to the muscle all shift the recording within a single trial.

Normalization and its limits

Because raw amplitudes cannot be compared, values are usually expressed relative to a maximal voluntary contraction performed by the same person.

That reference is itself variable, depending on effort and position, and an underestimate of the maximum inflates every subsequent value.

Comparisons across sessions are weaker still, since electrodes cannot be replaced in exactly the same position on a different day.

Deep muscles are effectively unavailable to a surface electrode, and signal from neighboring muscles bleeds into the recording, which makes attributing activity to one structure uncertain.

Why activation does not predict growth

Higher recorded activity during an exercise is often presented as evidence that it produces more growth, and that inference does not hold.

Amplitude rises with fatigue, so a lighter exercise taken further toward failure can register higher activity while imposing less mechanical demand.

Longer-term studies comparing exercises on measured size have repeatedly failed to line up with the ranking that activation studies produced.

What it is legitimately used for

Timing is where the technique is strongest, since onset and offset of activity are detected reliably even when amplitude is not.

It is used clinically for that reason, in gait analysis and in assessing whether a muscle is activating at the appropriate moment.

Research settings also use it to study fatigue, since the frequency content of the signal shifts in characteristic ways as a contraction is sustained.

Reading an activation study as a ranking of exercises asks the method for something it was not built to provide.