Event-Related Potential Analysis
Also known as: ERP, evoked potential, averaged EEG
Event-Related Potential (ERP) analysis is a method for extracting stereotyped brain electrical responses time-locked to stimulus presentation or behavioral events from EEG recordings. Formalized in the cognitive neuroscience literature by researchers including Sutherland and Picton, ERP analysis enables millisecond-level temporal resolution of neural processing and has become foundational for studying perception, attention, memory, and decision-making.
Key highlights
- Exceptional temporal resolution (milliseconds); reveals the time course of neural processing with precision unmatched by other neuroimaging
- Direct measure of neural activity; ERPs reflect postsynaptic potentials, not hemodynamic proxy signals
- High signal-to-noise ratio through averaging; single-trial noise easily suppressed by averaging ~30–50 trials
- Cost-effective and portable; EEG equipment is affordable and can be used in diverse settings
- Rich theoretical foundation; ERP components linked to well-characterized cognitive processes (N1=attention, P300=classification, N400=semantic integration)
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
ERP is ideal for studying time-sensitive cognitive processes with discrete onset events, when millisecond-level temporal precision is required, and when cognitive stages unfold sequentially. Use ERP for studying perception, attention, memory encoding, language processing, and decision-making. Avoid ERP when studying continuous processes without clear stimulus boundaries or when sustained tonic activity is the focus (use spectral power instead).
Strengths & limitations
- Exceptional temporal resolution (milliseconds); reveals the time course of neural processing with precision unmatched by other neuroimaging
- Direct measure of neural activity; ERPs reflect postsynaptic potentials, not hemodynamic proxy signals
- High signal-to-noise ratio through averaging; single-trial noise easily suppressed by averaging ~30–50 trials
- Cost-effective and portable; EEG equipment is affordable and can be used in diverse settings
- Rich theoretical foundation; ERP components linked to well-characterized cognitive processes (N1=attention, P300=classification, N400=semantic integration)
- Poor spatial resolution; EEG averages activity across large cortical areas; pinpointing source localization requires additional assumptions
- Volume conduction: electrical activity from a source spreads to many electrodes, making local topography interpretation ambiguous
- Small signal-to-noise ratio requires many trials; some processes (rare events, decisions) yield few epochs, reducing statistical power
- Baseline definition arbitrary; ERP amplitude is relative to a chosen baseline period, which can influence results
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
What is the P300 and what does it mean?
The P300 is a positive-polarity component peaking around 300 ms after stimulus presentation, largest when stimuli are task-relevant and unexpected. It reflects stimulus classification and decision processes. Amplitude varies with task difficulty and attentional allocation; latency varies with stimulus evaluation time. High P300 suggests intact attention; low/absent P300 may indicate cognitive impairment.
How many trials do I need to average for a reliable ERP?
Minimum 20–30 trials per condition for most components. High-noise data or rare events may require 50–100+ trials. More averaging improves signal-to-noise, but more trials increase experiment duration. Balance signal quality with practical constraints. Report number of trials retained after artifact rejection.
What baseline should I use for ERP measurement?
Common practice: -200 to 0 ms (pre-stimulus baseline). Baseline reduces slow drifts and DC offsets. Choice is somewhat arbitrary; sensitivity to baseline selection indicates unreliable effects. Report baseline chosen; if results depend critically on baseline, effects are weak. Some advocate correcting entire epoch before baselining.
How do I interpret ERP topography?
Topography shows which scalp locations have largest amplitudes but does NOT directly reveal neural source due to volume conduction. Multiple sources can produce similar topographies; similar sources can produce different topographies. Use topography for hypothesis generation, then verify with source localization (sLORETA, LORETA) or independent evidence.
Sources
- 1.Luck, S. J. (2005). An Introduction to the Event-Related Potential Technique. MIT Press.
- 2.Picton, T. W., Bentin, S., Berg, P., et al. (2000). Guidelines for using human event-related potentials to study cognition: recording standards and publication criteria. Psychophysiology, 37(2), 127–152.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 3). Event-Related Potential Analysis. ScholarGate. https://scholargate.app/neuroimaging/event-related-potential-analysis