JOVANA
Explore Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
All guides

The Biophysics of Electrical Stimulation: What a Pulse Actually Does

The activating function, charge-balanced pulses and safety limits, strength-duration recruitment, the volume of tissue activated, and the honest, still-unsettled question of how DBS works.

The activating function: why axons fire first

When an electrode injects current, it sets up an extracellular potential V_e across the tissue. Along a nerve fiber, what depolarizes the membrane is not V_e itself but its second spatial derivative — the activating function. Where this curvature is positive, current is driven into the fiber and the membrane depolarizes.

f_n \;\propto\; \left.\frac{\partial^2 V_e}{\partial x^2}\right|_{n} \;\approx\; \frac{V_{e,n-1} - 2\,V_{e,n} + V_{e,n+1}}{\Delta x^2}

Rattay's activating function at node n of a myelinated axon: the drive to depolarize is the discrete second difference of the extracellular potential along the fiber. Sharp spatial gradients — near the electrode, at fiber bends and terminations — dominate.

An axon fires not where the outside voltage is highest, but where that voltage changes shape most sharply along the fiber. This formula measures exactly that curvature — the second difference — so bends, endings, and the spot right under the electrode light up first.

f_n
The drive to fire at node n — how strongly this spot is pushed toward an action potential.
V_e
The voltage the electrode creates outside the fiber, in the surrounding tissue.
\frac{\partial^2 V_e}{\partial x^2}
The curvature of that outside voltage along the fiber; big wherever it bends sharply.
\Delta x
The spacing between neighboring nodes used to approximate the curvature.

Along a straight stretch of fiber far from the electrode the voltage changes smoothly, the curvature is near zero, and nothing fires; right at a bend the curvature spikes and that node crosses threshold first.

Charge, safety, and the biphasic pulse

Stimulators deliver a charge-balanced biphasic pulse: a cathodic phase that excites, followed by an equal-and-opposite anodic phase that reverses the electrochemistry at the electrode. The point of balancing is to drive the net injected charge to zero, so that no irreversible faradaic products accumulate to corrode the electrode or damage tissue.

Q = I \cdot t_{\mathrm{pw}}, \qquad \int I(t)\,dt = 0

Charge per phase Q is current times pulse width; charge balance requires the time integral of current over the biphasic pulse to vanish. Q (and charge density Q/area) are the quantities that matter for both excitation and safety.

Two rules of a safe stimulation pulse. First, the "dose" that actually excites tissue is charge — current multiplied by how long the pulse lasts, not current alone. Second, a biphasic pulse must push exactly as much charge one way as the other, so nothing accumulates and corrodes the electrode.

Q
Charge per phase — the true excitation "dose" of one pulse.
I
The current amplitude during the pulse.
t_{\mathrm{pw}}
Pulse width — how long the current flows.
\int I(t)\,dt = 0
The charge-balance condition: the net charge over the whole biphasic pulse is zero.

A 100 µA current held for 200 µs delivers a charge Q of 20 nanocoulombs; a matching reverse phase then returns that same amount so the net charge is zero.

How much is safe? The empirical Shannon criterion links charge per phase Q and charge density D = Q/A to the onset of tissue damage. Both matter: a small electrode can reach damaging charge density at a modest total charge.

\log_{10} D = k - \log_{10} Q

The Shannon damage boundary in the charge-density / charge-per-phase plane. The dimensionless k (roughly 1.5 for a conservative safe zone, up to ~2 near injury) sets how aggressive a waveform is. Staying below the line is a necessary — not sufficient — safety condition.

A safety line drawn in log-log space: the higher the charge density D you use, the smaller the charge-per-phase Q you're allowed, and vice versa. The constant k marks how close to the danger line you sit — staying underneath is necessary for safety, but not a full guarantee.

D
Charge density — charge spread over the electrode's surface area (charge per unit area).
Q
Charge per phase — the same "dose" as in the previous formula.
k
The Shannon constant setting how aggressive the waveform is (about 1.5 conservative, up to ~2 near injury).

Small electrodes squeeze charge into a tiny area, driving D up, so the line forces Q (and thus each pulse) to stay small — one reason microelectrodes must use gentle pulses.

Strength-duration and selectivity

Threshold current trades off against pulse width along a strength-duration curve. Longer pulses fire at lower current; very short pulses need much more. Two parameters summarize it: the rheobase I_{rh} (threshold for an infinitely long pulse) and the chronaxie t_{chr} (the pulse width at twice rheobase).

I_{\mathrm{th}}(t_{\mathrm{pw}}) = I_{\mathrm{rh}}\left(1 + \frac{t_{\mathrm{chr}}}{t_{\mathrm{pw}}}\right)

The Weiss-Lapicque strength-duration relation. Because chronaxie is shorter for large myelinated fibers than for small ones and for cell bodies, narrow pulses (short t_pw) bias recruitment toward large axons — a knob for selectivity as well as energy efficiency.

To make an axon fire, a briefer pulse must be stronger. This curve says the threshold current equals a "floor" current I_{\mathrm{rh}} (for a very long pulse), scaled up by a factor that blows up as the pulse gets short. Because large fibers have a shorter chronaxie, narrow pulses recruit them first — a selectivity knob.

I_{\mathrm{th}}
The threshold current needed to fire, for a given pulse width.
I_{\mathrm{rh}}
Rheobase — the minimum current that still works when the pulse is very long.
t_{\mathrm{chr}}
Chronaxie — the pulse width at which threshold is exactly twice rheobase; a fiber's characteristic time.
t_{\mathrm{pw}}
The pulse width you choose.

When t_{\mathrm{pw}} equals the chronaxie, the term in parentheses becomes two, so you need twice the rheobase current; make the pulse ten times shorter and the required current climbs about elevenfold.

The volume of tissue activated, and steering it

Integrate the activating-function threshold over space and you get the volume of tissue activated (VTA) — the region where fibers reach threshold. The VTA grows with amplitude and its shape depends on electrode geometry and on anisotropic tissue conductivity (current runs along white-matter tracts). Modern directional leads split a ring into segments so that current steering can sculpt the VTA toward the therapeutic target and away from structures that cause side effects, such as the internal capsule (which causes dysarthria and muscle pulling).

So how does DBS actually work?

Here honesty matters: there is no single agreed mechanism. The original story was functional inhibition — high-frequency stimulation silences the target, mimicking a lesion (which is why lesioning worked before DBS). But the activating function says we mostly excite axons, so a cleaner account is an informational lesion: rather than switching the nucleus off, ~130 Hz stimulation overwrites the pathological low-frequency bursting with a regular high-frequency pattern the downstream circuit reads as noise, decoupling the pathological rhythm. Network, synaptic-depression and astrocytic effects all contribute.