*Please note: the content of this writeup is derived solely from my experience in the two EEG laboratories where I have worked (both of which were labs in universities which relied on scalp rather than intracranial electrodes). Other labs probably have different standards for subject preparation. Also, be aware that this is really long. That said...
How to set up and record an EEG
There is one underlying principle to keep in mind each time you record an EEG to ensure that you are properly prepared: Nervous, uncomfortable people make terrible EEG subjects. At the scalp, the electrical signal from the brain is very weak. Any sort of muscular tension will create an electrical signal much larger than the signal you are trying to measure. Anything you can do to prepare ahead of time, to ease your subject's mind, and to make the lab seem like a well-oiled machine will only increase the likelihood that you will get clean, usable data.
Arrive at least 1/2 hour prior to the subject's arrival. Turn on all the computers. Reboot the ancient machine that displays the stimuli. Reboot it again. Set up all the appropriate programs from stimuli presentation and signal recording so that there is as little left to do when the subject arrives as possible. If possible, get things set up so that all you have to do is enter a single subject code before starting the stimuli presentation program. Make sure all of your supplies are easily accessible and in working order. You will need:
- electrode cap(s): Electrode caps are generally made of nylon mesh. A number of electrodes are sewn into the cap (the most common electrode counts are 16, 32, 64, and 128). These electrodes consist of donut-shaped metal contacts surrounded by plastic mounts. The metal contacts generally sit about 1-2 millimeters up from the surface of the plastic mount, providing space for conductive substances that improve the signal. The caps also feature straps which can be snapped onto a strap fitted around the chest, preventing the cap from moving around on the head too much. Check the electrodes in the cap. The metal bits should be clean and shiny. If they aren't, scrape the gunk off (gently) with one of those orange sticks that people use to push in their cuticles. Check the wires, looking for obvious breaks.
- free electrodes: free electrodes are placed in areas of little electrical activity to serve as a ground on which to base the scalp recordings (caps with more than 16 electrodes often feature an electrode that is designed to be used as a ground, but it's always a good idea to have a redundant ground electrode). Free electrodes can also be placed on various parts of the face to record eye and facial movement, allowing data that contains movement artifacts to be discarded. It is also a good idea to have some free electrodes that are designed to replace faulty cap electrodes, in case there is a problem in the wiring that cannot be detected from a surface check. Make sure that you have enough electrodes for all the grounds and movement detectors you plan to place on the subject, and that these electrodes are not all tangled up.
- electrogel: This is a conductive substance used to improve the connectivity between the scalp and the electrode. It is essentially a colloid made with salt water. It has a funny hospital smell, and is a sort of translucent tan color. If the electrogel is too thick, you can add a little distilled water to thin it a bit, but be careful -- if the gel is too thin, it will be unusable.
- plastic syringes: it's usually a good idea to have out at least two syringes per subject. Have more available if there will be more than one experimenter preparing the subject for the recording. Load up the syringes with electrogel, attempting to avoid getting air bubbles in the gel in the syringe. Air bubbles will make it difficult to fill the electrode wells because they make it impossible to produce a controlled flow of gel. Make sure the syringes are in good shape. The plungers in old syringes don't move smoothly through the barrel of the syringe, again impeding the production of a controlled flow of gel from the syringe.
- Blunt-ended needles: blunt ended needles are used on the syringes to aid in the directed squirting of electrogel into the electrode wells, without much threat of stabbing someone in the head and hurting them. You should have as many of these as you do syringes. Make sure not to open the packaging until the subject is present, because it's one thing to be told that fresh, sterile needles are used on each subject, and another to see the experimenter opening the packaging on the fresh, sterile needles.
- long-stick q-tips: the kind doctors use to swab your throat when they are checking you for strep throat. The back ends (the wooden bit) of these are used to gently abrade the dead skin off of the area of scalp just beneath the electrode. *Never tell the subject that you are abrading dead skin off their head! As an alternative, say something about "working the gel closer to the skin surface" or just "reducing impedence".
- cloth tape: This is just your standard medical tape, the kind you can buy in any drug store. It is used to tape on the ground and movement-detection electrodes.
- measuring tape: This is used to measure the circumference of the head in order to determine what size cap should be used, as well as to aid in proper electrode positioning.
- chest strap: this is used to anchor the cap in place.
- foam donuts: There is probably a technical term for this, but I don't know that term. In the labs where I have worked, they were always referred to as foam donuts. These are placed around the plastic mountings of the cap electrodes located on the forehead, in order to make the subject more comfortable.
- latex gloves: these should not be put on until the subject arrives for the same reasons that you should wait to open the needle packaging. Make sure your gloves are the right size, or else it will be difficult to manipulate the tiny objects that you'll be working with.
- rubbing alcohol and cotton balls: use these to clean the areas where ground and movement-detection electrodes will be placed. This will increase the connectivity between the electrode and the skin, and decrease the amount of abrasion necessary to get a good connection.
- paper towels: useful for cleaning up excess gel from the electrodes, and for cleaning off the subject's head after you are done.
- consent forms: The forms that describe what you will be doing to the subject (generally without giving away the big secret point of the experiment) and leave a space for the would-be subject to sign their name, voiding your legal responsibility to pay them off if they start to feel anxious and claustrophobic. These are probably only necessary in a research setting. *For confidentiality purposes, the consent form should be the only place on which the subject's full name appears. All other forms containing information about the subject should be labeled with some kind of subject code.
- subject information forms: These consist of question regarding the subject's handedness, age, how well they slept the night before (alpha waves, the bane of EEG recordings. Well, unless you are studying alpha waves, I suppose), how they are feeling, whether they have done any drugs or drank any alcohol within the past 24 hours or so.
- debriefing forms: these forms should provide a fairly detailed explanation of what is being studied, and also have an e-mail address and phone number the subject can use to contact the primary investigator in case s/he has any questions or concerns about the experiment and the procedure.
Now that all your materials are prepared, you can sit around waiting for the subject to show up. When s/he arrives, do the following:
- introduce yourself and the experiment. Look the subject in the eye. Be confident. Smile. Explain the experiment and the procedure as much as possible. Ask the subject if s/he has any questions, and if any come up, answer them thoroughly and to the best of your ability.
- give the subject the consent form. Ask them to read it completely, and ask any questions they think of before they sign it. Point out the clause in the consent form that says they can decide to stop at any time should they become uncomfortable. Remind them that they can ask more questions at any time throughout the experiment. If they do not consent, thank them for their time, ask them if they have any more questions, and let them leave. If they consent...
- ask the subject to fill out the subject information form. Reassure the subject that their answers are confidential, and will not be associated with their name in the records. If all goes well, you can put on your gloves and get down to the dirty work.
- measure the subject's head. Measure the head circumference just above the ears, and crossing the forehead. If the measurement is on the borderline between two cap sizes, choose the larger cap, as this will be more comfortable for the subject. In my experience, female subjects were upset if it turned out they needed the large cap, and male subjects were upset if they needed the small cap, so it is best to refer to the cap by color (they are generally color coded for size) rather than tell the subject what size cap you use.
- place the ground electrodes. In the labs where I worked, the ground electrodes were placed on the mastoids, the bony area behind the ears. Clean these areas using cotton balls and rubbing alcohol. Do not be grossed out by all the yucky stuff you clean off, you are probably dirty behind your ears, too. Tape the electrode in place with a good long strip of tape, as the electrogel will eventually prevent the adhesive from sticking to the skin in the area immediately around the electrode. Stick the other end of the wire into the appropriate slot on the amplifier. Check the readout monitor--do you see a regular, repeating increase in activity? If so, you have placed the electrode too close to the blood vessel back there, and you will need to reposition the electrode. When the electrode is positioned reasonably well, put a sterile needle onto one of the syringes and squirt a small amount of gel into the electrode well (just enough to fill it up). Check the impedence. The goal is to get the impedence below 2 (I'm not sure if that's ohms or milliohms, sorry). If it is above 2, insert the back end of a q-tip into the hole in the top of the electrode and gently scrape at the skin underneath (this loosens dead skin cells, and works any air bubbles out of the gel). You should make every effort to do this without hurting the subject. Put in more gel as needed to keep the electrode well full. Keep doing this until the impedence reads below 2. Repeat this for the other ground.
- place the eye movement detection electrodes. Eye movement detection electrodes should be placed on one cheekbone, just below the eye, and on the other side of the face, on the bony space between the eye socket and the temple. (If you put your fingers on your face in these places and blink and look around and stuff, you can feel the movement pretty easily. The electrical signal associated with this movement can be detected by the EEG at many places on the scalp. By recording at these locations, where the signal would be exceptionally strong should the subject blink or look around, you'll be able to detect the data that may not accurately reflect brain activity.) Clean with rubbing alcohol and apply electrode and gel as described in the previous step. If you must abrade the skin with the q-tip, be extra gentle. After all, that is someone's face!
- put on the chest strap. Generally the rule in our labs was to just ask the subject to raise his/her arms, and then place the strap around the chest, ensuring the snaps were in the front. It might be a little weird if the gender of the experimenter doesn't correspond to the gender of the subject. It probably helps to wear a lab coat or something so that the subject doesn't get weirded out that you are almost touching their chest. You can also ask the subject to put the chest strap on themselves, explaining that the snaps should be centered on the front of their chest.
- put on the cap. Foam donuts should be placed around the two frontmost electrodes so that the cap will be more comfortable. The two previously mentioned electrodes in the front should be placed equidistant from the midline of the forehead about 2.5 centimeters above the line of the eyebrows. There is a line of electrodes behind these front electrodes that should rest along the midline of the head. If these two things are done, the cap will be properly positioned. Be careful not to knock the motion-detection and ground electrodes off as you position the cap. Snap the cap straps to the chest strap (or ask the subject to do so). It is easiest to keep track of which electrodes you have already worked on if you do it in some standardized manner. I usually worked the electrodes from front to back, one row at a time. Scalp electrodes generally take longer to get good impedences on, so I suggest you squirt in gel, work with a q-tip, and re-squirt gel before the first time you check the impedences. It is important not to use too much gel, especially if you are using a cap with more than 16 electrodes, as excess gel can cause the signal to bleed from one electrode to the next. As with the grounds, work the scalp electrode sites until the impedence is less than 2.
- check on the subject. When you have finished the preparations, ask the subject if s/he is okay, and if s/he has any further questions before the experiment starts. Explain what they should expect to happen. If the EEG is being employed as an experimental tool, the stimuli programs will generally be designed with fixation points, and breaks. Explain that the subject should try to minimize movement between breaks, and that, when possible, the subject should try to limit blinking to the time period when the fixation point is displayed. Tell the subject that if they have any questions during the experiment, they should let you know.
- start the programs. Always make sure that you have both the recording program and the stimuli presentation program running at the same time, or else you will be very sad.
- clean up. When the program has finished running, offer to show the subject their own brain activity, and after doing this (if the subject wants to take a peek), start cleaning up the subject. Ask them to unsnap the cap straps and undo the chest strap. Remove the cap. It's a good idea to ask the subject to remove the motion-detection and ground electrodes, since hair invariably gets caught beneath the tape, and they can come off painfully if they are taken off by someone else. Clean the gel out as well as you can. If you are using a cap with more than 16 electrodes, you should consider having shampoo and towels available so that subjects can wash their hair in a nearby sink.
- debrief the subject. (and no, I don't mean that you should take away their underpants.) Give the subject the debriefing form, and any payment or other reward that is due to them for participation. Answer any additional questions they have.
- save your work. After your subject has left the area, save the data you have collected. If you forget to save your data, you will be very sad.
It is important to keep the subject relaxed and awake. The easiest way to do this is to have a conversation. Practice your patter. Ask about where the subject is from, what they are doing for the holidays, where they work, anything.
If you notice a problem with the recording (e.g. an electrode has fallen off or become weird), wait until one of the breaks to deal with it. Being interrupted unexpectedly while concentrating on the experiment will be very disruptive for the subject, and decrease the quality of the rest of your recording.
If the subject becomes uncomfortable in the middle of the experiment and asks to leave, let them go without making any sorts of unhappy noises. Unless the EEG is being recorded for medical purposes, the subject has volunteered his/her time, and it is important to be respectful of him/her. It is sad and frustrating to lose data, but that is not sufficient cause to be mean to someone.
If you notice a lot of alpha activity (it will show up as wide-wavelength, repetitive activity on the monitor of the recording computer), your subject is sleepy. Wait until a break, tell the subject the problem, and offer them a soda or some coffee, or suggest that they take a big stretch and shift around a bit during this break.
Happy brain recording!