ERP pipeline · PSYCH 390 step 1 / 11

Introduction to Electricity

How does electricity work, and how is electrophysiology similar and different?

Figure 1

Ohm's Law in a Wire and in the Head

World
Wire circuitThe head
Charge carriersElectrons in metalIons (Na+, K+, Cl−) in salty fluid
ConductorA thin wire; current has one pathThe whole head, a volume conductor; current spreads everywhere at once
VoltagesVolts (a 9 V battery, 120 V at the wall)Microvolts at the scalp: EEG 20–100 µV, ERP effects 1–10 µV
AC or DCBatteries are DC; wall current is 60 Hz ACSkin potentials drift like DC; the brain signal itself changes constantly, so EEG is AC-like
What is measuredVoltage across two points of the wireActive minus reference (A − R); the ground electrode drops out of the subtraction
What limits the signalResistance of the loadSkull resistance and the impedance of the electrode, gel and skin junction

All values are simulated; the head is reduced to a single loop with one skull resistor on each side and no capacitance in the tissue.

What You Are Looking At

Start with the wire. The battery supplies voltage, a push that makes charge move, and the moving dots are electrons carrying current around the loop. The resistor opposes that flow and turns the energy into heat, so it glows as the power rises. The hose underneath shows the same three quantities as water. Voltage is pressure, current is how much water comes out per second, and resistance is how narrow the hose is. Ohm's law ties the three together, I = V / R, and power is P = V × I.

Switch to the head and the same parts are all present, made of different materials. The battery is a small patch of cortex whose synchronized neurons behave like a tiny dipole. Current is carried by ions drifting through salt water rather than electrons in copper, the skull is a large resistor in the path, and the scalp is a thin conductive layer that spreads the current sideways. Each electrode sits on gel and skin, a junction that acts as a resistor and a capacitor in parallel, and the amplifier does not read the active electrode by itself. It reads the difference between the active electrode and a reference electrode, A minus R. Nearby room wiring carries 60 Hz alternating current, and a changing current makes a changing magnetic field, which induces a current in any conductor it crosses. That hum is common to both electrodes and should cancel in A minus R, but the cancellation only works if the two electrodes have similar, low impedance. As impedance rises the mismatch grows, less of the hum cancels, and skin potentials get larger too (Luck, 2014, Chapter 5). The blue wire above the circuit carries the hum, and at high impedance it swamps the brain signal in the teal amplifier trace at the bottom.

Try This

  1. In the wire circuit, raise the voltage with the resistance fixed. The electrons speed up, the hose pressure gauge climbs, and the current readout follows I = V / R.
  2. Now raise the resistance with the voltage fixed. The hose narrows, less water flows, and the dots slow down. The pressure is still there; it just cannot push as much through.
  3. Think of the same hose with the flow held fixed and the constriction tightened. The pressure would have to rise. A constant neural current pushed through a higher resistance produces a larger voltage in the same way.
  4. Switch to the head. Slide the source strength from about 200 neurons to 100,000 and watch the scalp voltage climb from a tenth of a microvolt to tens of microvolts.
  5. Raise the electrode impedance to 50 kΩ. The scalp voltage from the brain barely changes, but the 60 Hz hum grows until the amplifier trace is mostly noise. Bring it back below 5 kΩ and the trace clears.

Why It Matters for the Pipeline

Three facts from this page come back on almost every later one. Electricity follows the path of least resistance, so current from a cortical source spreads through the whole head instead of staying under one electrode, and a single channel never sees a single source. The skull is a poor conductor, so current spreads sideways through the lower-resistance scalp and the pattern reaching the electrodes is blurred. And the amplifier reports A minus R, so the choice of reference changes every waveform. Before a recording starts, the experimenter checks each electrode's impedance and works the gel until the impedance is low and roughly matched across sites, usually below 5 to 10 kΩ. EEGLAB does not store impedances; they are checked on the acquisition computer before recording and written into the lab notes, so keep a record of any electrode that was hard to bring down. A channel that was noisy from the start is the one you will later repair with pop_interp, and a noisy channel that is left in will cause artifact detection to reject far too many trials. Low, matched impedance is the cheapest noise reduction available (Luck, 2014, Chapter 5).

References

Luck, S. J. (2014). An introduction to the event-related potential technique (2nd ed.). MIT Press.