Electroencephalography

EEG lab setup

The technology of electrophysiology is used in the Jolicoeur research laboratory to collect digital electroencephalograms that are used in various types of studies in cognitive neuroscience, including those using Event-Related Potentials (ERP), Time-Frequency analyses (wavelets), Principal and Independent Components Analysis (PCA, ICA), etc. Each active neuron in the brain produces a tiny electrical signal. When many neurons are active at the same time, their combined activity can be measured on the scalp using an array of extremely sensitive electrodes placed in contact with the scalp (the science of electroencephalography (EEG)). These weak signals are amplified and saved in computer files along with "marker" signals that indicate when different stimuli and responses occurred. Analyses can recover the time-course of neuronal events that generate the electric signals.

Brisson et al, 2007

Each experiment is designed to isolate specific aspects of brain activity during controlled tasks. The stimuli may be visual, auditory, or tactile, etc. To tease out the details of a phenomenon, there are various types of trials whose results are compared. This involves carefully choosing two or more conditions and possibly a control. Only this comparison of the results from the extracted time-locked segments of many trials in each condition gives the researcher the statistical power to infer a significant difference between the various conditions.

squiggles

A typical example of a simple ERP experiment is to display a stream letters on a computer screen, one after the other. The majority of the letters are one particular letter. Occasionally, a different letter is inserted into the stream. If one examines each of the traces registered after each letter, the raw data shows no discernible pattern. The signals seem to be random noise, just squiggles, as they are called (above left). This is because many parts of the brain are working simultaneously. For instance, just the action of maintaining ones balance while sitting in a chair will have an associated electrical signal.

However, if the signals in the segments for the more and the less frequent letters in the stream of the experiment above are each averaged, two smoother traces are obtained. Only what is common to each trial is retained (unsynchronized activity drops out) and the two averaged curves are clearly different. The less frequent letter contributes a "pop-out" effect, which can affect ERP components (i.e. Positive- or Negative-going waveforms) such as the N2 (at approximately 200 ms post-stimulus) or the P3 (~300 ms). This phenomenon can be combined with other conditions to study other phenomena such as attention, memory, or other cognitive processes. Another way to separate out extraneous brain activity is to have stimuli on both sides of the screen, draw attention to one side or the other (without moving the eyes), and then subtract the signals obtained from the ignored side from the attended side. The ignored side provides a control for the attended side.

Predovan, 2009

The diagram at the left is an example of the above concepts but for a memory experiment. The form of the traces and the scalp distributions (of one side subtracted from the other and also averaged over the 600 ms after the stimulus) vary with condition.

Because electroencephalography is based upon neuronal electrical signals, EEG has the advantage of very high temporal resolution (on the order of milliseconds). This compares favourably against fMRI technology, which tracks changes in blood flow, a much slower measure of brain activity. Unfortunately, the spatial resolution of EEG is much poorer. The electrical signal is smeared by the skull. Also, we can only pick up the signal as it exits the scalp. It is not a 3-dimensional picture (as is fMRI) and signals from deeper in the brain are attenuated. Other advantages of EEG over fMRI are that it is silent, tolerant of subject movement and much lower in cost (MRI systems are very loud, a problem in auditory research).

Electrodes

Electrophysiology is a very safe procedure. In the Jolicoeur lab, the electrical contact with the scalp is made by using a small amount of low-impedance conducting gel in each hole in the head cap that holds electrodes in place at predetermined well-established locations. Because the signal on the scalp is so weak, the experiments are conducted in a shielded booth (to minimize interference from external electromagnetic noise due to machines, computers, etc.).

Faraday cage

Even the equipment inside the booth can be a source of unwanted noise. The monitor is placed inside a Faraday cage in which the surrounding metal screen is grounded. Also, the signal produced by the subject blinking or moving his/her eyes can add an artifact that is much larger than the usual 10-100 µV brain signal. This means that the subject can only move her/his eyes between trials. Trials that include these artifacts must be dropped from any averages and/or corrected. The signal from each channel or electrode is sent to a differential amplifier prior to filtering and digitization.

Careful experimental design is critical in properly isolating cognitive phenomena. Dr. Steven J. Luck has written an excellent book, The Event-Related Potential Technique in Cognitive Neuroscience for the field of ERP. It details the various principles for EEG lab and experimental design (for eg., noise reduction, reference localization, choice of number of electrodes, minimization of impedence, avoiding confounds) and ERP data analysis (including the pitfalls of confusing overlapping components, comparison difficulties, time and amplitude shifts of components due to filtering, averaging and other issues).


The large majority of the papers published in the Jolicoeur lab at Université de Montréal are in the field of electrophysiology. Below are links to a few.

The following paper involved separating out the various components (N2pc and SPCN):
Jolicoeur, P., Brisson, B., & Robitaille, N. (2008). Dissociation of the N2pc and Sustained posterior Contralateral Negativity in a Choice Response Task. Brain Research, 1215, 160-172.

The same two components as above exhibited (PRP) interference with visual attention and memory using a dual-task paradigm:
Brisson, B., & Jolicoeur, P. (2007). A psychological refractory period in access to visual short-term memory and the deployment of visual-spatial attention: Multitasking processing deficits revealed by event-related potentials. Psychophysiology, 44, 323-333.

Here, the SPCN was used to test whether a representation in VSTM was being mentally rotated:
Prime, D. J., & Jolicoeur, P. (2010). Mental rotation requires visual short-term memory: Evidence from human electric cortical activity. Journal of Cognitive Neuroscience, 22, 2437-2446.

The paper by Robitaille et al, 2010 showed the correspondence between EEG and data collected with other imaging technologies:
Robitaille, N., Marois, R., Todd, J. J., Grimault, S., Cheyne, D., & Jolicoeur, P. (2010). Distinguishing between lateralized and nonlateralized brain activity associated with visual short-term memory: fMRI, MEG, and EEG evidence from the same observers. NeuroImage, in press.

Finally, a different paradigm is used to study the capture of visual spatial attention:
Leblanc, É., Prime, D., & Jolicoeur, P. (2008). Tracking the location of visuospatial attention in a contingent capture paradigm. Journal of Cognitive Neuroscience, 20, 657–671.

Last update: Feb. 14, 2011.