INSTRUCTIONS

ERP AVERAGING

CONCLUSION

BACKGROUND

In an ERP experiment, the goal is to measure the brain activity related the time-locked event – we call this the “signal”. However, ERP data also contain “noise”, including brain activity unrelated to the time-locked event and external sources of electrical interference. The purpose of this simulation is to demonstrate how averaging affects the ERP signal-to-noise ratio.

SIMULATION

CREDITS

UNDERSTANDING EEG WAVEFORMSSIMULATION 4: ELECTRODE POSITION AND DIPOLE DISTANCE AND ORIENTATION

Noise refers to electrical signal picked up in an EEG that is related to other brain activity, as well as external sources of electrical interference. 

The Problem of NoiseUsing only single trials, the difference between the target condition and the standard condition is hidden by noise. This demonstration will show how averaging ERP data across multiple trials within each condition reduces noise and makes the condition difference more visible.

ERP AVERAGING

ConclusionAveraging ERP waveforms across trials within a condition increases the signal-to-noise ratio, better revealing the event-related signal. Because the event signal is time-locked to the ERP waveform, averaging does not affect the signal related to the event. That is, to the extent that the event-related signal is consistent across trials, it will maintain its strength and polarity when averaged.   However, neural sources of noise are not time-locked to the ERP waveform, and to the extent that the timing of these sources is inconsistent across trials, these noise sources will decrease with averaging. Thus, averaging increases the signal-to-noise ratio by maintaining the signal and reducing the noise. 

Created by: Cindy Bukach Nadia Bukach David Leland Jane Couperus Cathy Reed

The Oddball ParadigmThis demonstration uses an oddball paradigm that produces a positive waveform at 300 ms, called the P3. The oddball paradigm presents rare targets among common standards. Participants typically press one button for targets, and another button for standards.

In the oddball paradigm, standards are the frequently occurring stimulus that serves as a baseline. Targets are rare stimuli that the participant is instructed to detect. In this example, standards are black circles and targets are red circles.

P3 or P300 is a large positive ERP component that occurs around 300-600 ms after a stimulus is presented. It is associated with categorization, allocation of attention, and memory updating operations.

The bottom right panel displays the building average ERP of all trials. The black line represents the averaged ERP for standard trials, and the red line represents the averaged ERP for target trials.

The bottom left panel diplays the single trial ERP - a segment of data (epoch) from 200 ms before the stimulus appears to 800 ms after it appears.

Play/Pause

Slow/Fast (coming soon)

Next Trial

The top panel of the simulation contains two seconds of EEG data for each trial, one second before and after the onset of the stimulus event. Directly above the EEG data the trial stimulus (red or black circle) will appear for 200 ms. Each new trial overwrites the previous trial. This panel also includes action buttons as labeled above.

Last Trial

Previous Trial

First Trial

Instructions This demonstration has three panels that display EEG and ERP data in real time as the participant responds to trials in the oddball paradigm.