Origin and technical aspects of the EEG 1 

neuron it leads to a local reduction of the transmembrane potential

ORIGIN OF THE EEG
(depolarization) and is called an excitatory post-synaptic potential
The EEG records electrical activity from the cerebral cortex. Inasmuch (EPSP), typically located in the dendrites. Note that during an EPSP the
as electrocortical activity is measured in microvolts (µV), it must be inside of the neuronal membrane becomes more positive while the extra-
amplified by a factor of 1,000,000 in order to be displayed on a com- cellular matrix becomes more negative. Inhibitory post-synaptic poten-
puter screen. Most of what we record is felt to originate from neurons, tials (IPSPs) result in local hyperpolarization typically located on the cell
and there are a number of possible sources including action potentials, body of the neuron. The combination of EPSPs and IPSPs induces cur-
post-synaptic potentials (PSPs), and chronic neuronal depolarization. rents that flow within and around the neuron with a potential field
Action potentials induce a brief (10 ms or less) local current in the axon sufficient to be recorded on the scalp. The EEG is essentially measuring
with a very limited potential field. This makes them unlikely candidates. these voltage changes in the extracellular matrix. It turns out that the
PSPs are considerably longer (50–200 ms), have a much greater field, typical duration of a PSP, 100 ms, is similar to the duration of the
and thus are more likely to be the primary generators of the EEG. Long- average alpha wave. The posterior dominant rhythm (PDR), consisting
term depolarization of neurons or even glia could also play a role and of sinusoidal or rhythmic alpha waves, is the basic rhythmic frequency
produce EEG changes. of the normal awake adult brain.
In the normal brain an action potential travels down the axon to the It is easy to understand how complex neuronal electrical activity
nerve terminal, where a neurotransmitter is released. However, it is the generates irregular EEG signals that translate into seemingly random
synaptic potentials that are the most important source for the electro- and ever-changing EEG waves. Less obvious is the physiological expla-
encephalogram. The resting membrane potential (electrochemical equi- nation of the rhythmic character of certain EEG patterns seen both in
librium) is typically –70 mV on the inside. At the post-synaptic membrane sleep and wakefulness. The mechanisms underlying EEG rhythmicity,
the neurotransmitter produces a change in membrane conductance and although not completely understood, are mediated through two main
transmembrane potential. If the signal has an excitatory effect on the processes. The first is the interaction between cortex and thalamus. The
1
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

activity of thalamic pacemaker cells leads to rhythmic cortical activation. For long-term monitoring, especially if the patient is mobile, cup elec-
For example, the cells in the nucleus reticularis of the thalamus have the trodes are affixed with collodion (a sort of glue), and a conductive gel
pacing properties responsible for the generation of sleep spindles. The is inserted between electrode and scalp through a small hole in the elec-
second is based on the functional properties of large neuronal networks trode itself. This procedure maintains recording integrity over prolonged
in the cortex that have an intrinsic capacity for rhythmicity. The result periods.
of both mechanisms is the creation of recognizable EEG patterns, varying Other types of electrodes are available including plastic, as well as
in different areas of neocortex that allow us to make sense of the needle electrodes. In fact, new plastic electrodes are MRI compatible.
complex world of brain waves. Needle electrodes, which in the past were often used in ICUs, have been
redeveloped and consist of a painless (really!) subdermal electrode.
TECHNICAL CONSIDERATIONS
ELECTRODE PLACEMENT
The essence of electroencephalography is the amplification of tiny cur- Electrode placement is standardized in the United States and indeed in
rents into a graphic representation that can be interpreted. Of course, most other nations. This allows EEGs performed in one laboratory to
extracerebral potentials are likewise amplified (movements and the like), be interpreted in another. The general problem is to record activity from
and these are many times the amplitude of electrocortical potentials. various parts of the cerebral cortex in a logical, interpretable manner.
Thus, unless understood and corrected for, such interference or artifacts Thanks to Dr. Herbert Jasper, a renowned electroencephalographer at
obscure the underlying EEG. Like the archeologist, the epileptologist the Montreal Neurological Institute, we have a logical, generally accepted
seeks to fully understand artifacts in order to discern the truth. Later, system of electrode placement: the 10-20 International System of Elec-
we will discuss artifacts in detail and illustrate clearly their many guises. trode Placement (Figure 1-1). The numbering has been slightly modified
At this point we will consider the technical factors that are indispensable since the last edition to a 10-10 system (Figure 1-2). The system was
in obtaining an interpretable record. modified so that if additional electrodes are to be placed on the scalp,
there is a logical numbering system with which to do so.
ELECTRODES Both the 10-10 and the 10-20 system depend on accurate measure-
Electrodes are simply the means by which the electrocortical poten- ments of the skull, utilizing several distinctive landmarks. Essentially, a
tials are conducted to the amplification apparatus. Essentially, standard measurement of the skull is taken in three planes – sagittal, coronal, and
EEG electrodes are small, non-reactive metal discs or cups applied to horizontal. The summation of all the electrodes in any given plane will
the scalp with a conductive paste. Several types of metals are used includ- equal 100%. Electrodes designated with odd numbers are on the left;
ing gold, silver/silver chloride, tin, and platinum. Electrode contact those with even numbers are on the right. Standard electrode designa-
must be firm in order to ensure low impedance (resistance to current tions and placement should be memorized during the student’s first day
flow), thus minimizing both electrode and environmental artifacts. of his or her elective (Table 1-1).

2
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
NASION
NZ

FPZ
FP1 FP2
Fpi Fp2
AF7 AF8
AF3 AFZ AF4
F9 F10
F7 F8
F7 F8 F5 F3 F1 F2 F4 F6
Fz
F3 Fz F4
FT9 FT7 FT8 FT10
FC5 FC3 FC1 FCz FC2 FC4 FC6

A1 A2
T3 C3 Cz C4 T4 A1 T9 T7 C5 C3 C1 CZ C2 C4 C6 T8 T10 A2
C5 C6

CP5 CP3 CP1 CPz CP2 CP4 CP6

TP9 TP7 TP8 TP10

P3 P1 Pz P2 P4
P5 P6
P3 Pz P4 P7 P8
T5 T6 P9 P10
PO3 POZ PO4
PO7 PO8

O1 O2
OZ
O1 O2

IZ
INION

Figure 1-2 10-10 system. The 10-20 system has been modified to standardize a
 

Figure 1-1 10-20 system. A single-plane projection of the head showing all
method for adding more electrodes.
 

standard positions and the locations of the Rolandic and Sylvian fissures. The outer
circle was drawn at the level of the nasion and inion. The inner circle represents
the temporal line of electrodes.

3
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Table 1-1 Standard electrode designations

 

Left Right Electrode

Parasagittal/supra-sylvian electrodes
Fp1 Fp2 Frontopolar, located on the forehead – postscripted numbers are different than other electrodes in this sagittal line (3,4)
F3 F4 Mid-frontal
C3 C4 Central – roughly over the central sulcus
P3 P4 Parietal
O1 O2 Occipital-postscripted numbers are different from other electrodes in this sagittal line (3,4)
Lateral/temporal electrodes
F7 F8 Inferior frontal/anterior temporal
T7 T8 Mid-temporal – formerly T3, T4
P7 P8 Posterior temporal/parietal – formerly T5,T6
Other electrodes
Fz, Cz, Pz Midline electrodes: Frontal, central and parietal.
A1 A2 Earlobe electrodes. Often used as reference electrodes from contralateral side. Of note, they record ipsilateral mid-temporal activity.
LLC RUC Left lower canthus/right upper canthus (placed on the lower and upper outer corners of the eyes). These electrodes are used to detect eye
movements and can help distinguish eye movements from brain activity. Sometimes designated LOC, ROC.

How to measure for electrode placement measurement (Fpz) and the point above the inion that also is 10%
Sagittal plane: The sagittal measurement starts at the nasion (the of the total (Oz). These locations are used as coordinates to help iden-
depression at the top of the nose) over the top the head to the inion (the tify the other designated electrode destinations. Divide and mark
prominence in the midline at the base of the occiput). With a red the remaining 80% into four segments, each 20% of the total measure-
wax pencil, mark the point above the nasion that is 10% of the total ment. The first 20% point is Fz, the second Cz and the third Pz – the
4
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
midline electrodes (z = zero). The final 20% is the distance between Pz electrodes. The remaining 20% segments represent the distance from C3
and your point 10% above the inion (Oz). Thus, the total is 100% to Cz and Cz to C4 (Figure 1-3B).
(Figure 1-3A). Horizontal plane: The trickiest measurements are in the horizontal
Coronal plane: The coronal plane extends from the point anterior to plane. The horizontal plane is generated with a measurement from Fpz
the tragus (the cartilaginous protrusion at the front of the external ear) to T7 to Oz on the left and from Fpz to T8 to Oz on the right. Fp1 and
to the same point on the opposite side, making sure that the tape Fp2 are placed on either side of Fpz, both a distance of 5% of the total
measure traverses the Cz point on the sagittal measurement. The inter- horizontal circumference from Fpz. Similarly, O1 and O2 are placed at
section of the halfway (50%) points of the sagittal and coronal measure- a 5% distance of the total horizontal circumference from Oz. The dis-
ments is the location of the vertex and thus the Cz electrode. The first tances from Fp1 to F7 to T7 to P7 to O1 on the left and from Fp2 to
10% points up from the tragus define T7 and T8, the mid-temporal F8 to T8 to P8 to O2 on the right are all 10% of the total horizontal
electrodes. The next 20% points then define C3 and C4, the central circumference (Figure 1-3C).

Cz Figure 1-3 Measurements in the

 

C 10-10 system in the (A) sagittal, (B)

20% % 20%
20 coronal, and (C) horizontal plane. (A)
F 20 C
% C4 3
Lateral view of the skull to show the
method of measurement from the
nasion to inion at the mid-line. Fp is
%
20

the frontal pole position, F is the

20%

20%
frontal line of electrodes, C is the
Fp

central line, P is the parietal line, and

20%

O is the occipital line. Percentages

10%

indicate proportions of the total

T 7 10%
10% T8

measurement from the nasion to the

inion. The central line is 50% of this
O

distance. (B) Frontal view of the skull

10

showing the coronal measurements.

%

A B
5
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Finally, F3 and F4 are defined by the halfway points between F7 and Fpz
Fp
Fz on the left and F8 and Fz on the right. Similarly, P3 and P4 are defined F p1 5% 5% 2
by the halfway points between P7 and Pz on the left and P8 and Pz on
the right. 10
%

%
10
An observation: The F7 and F8 electrodes are probably placed too
high for optimal definition of anterior temporal activity. Likewise, the
P7 and P8 electrodes are probably too high for good definition of pos-

F7

F8
terior temporal activity. Thus, it is possible to logically place additional
electrodes (F9/F10, T9/T10, and P9/P10), which are placed 10% inferior

10%
10%
to the standard (F7/8, T7/8, P7/8, respectively) electrodes. In some labo-
ratories, these additional electrodes are routinely used.
In the 10-10 system, there are remaining electrode positions in the

T7
10% intermediate lines between the existing standard coronal and sagit-

T8
tal lines. Best to look at Figure 1-2 while reading the next several sen-
tences. Coronally, these electrode positions are named by combining the

10%
10%
designation of the coronal lines anterior and posterior. For example, the
coronal line between the parietal (P) and occipital (O) chain is designated
PO. The only exception is in the first intermediate coronal line, which

P7
is named AF (anterior frontal) rather than FpF or FF. In the sagittal line,

P8
the same postscript numbers are used; for example, AF3, F3, FC3, C3,

10
CP3, P3, and PO3. From the midline moving laterally the postscript
%

%
10
begins at z followed by the numbers 1, 3, 5, 7, 9 on the left and 2, 4,
6, 8, 10 on the right. We now have the 10-10 system where each letter
5% 5% O2
O1 Oz
appears on only one coronal line and each postscripted number on a C
sagittal line (except for Fp1/Fp2 and O1/O2). The 10-10 system locates
Figure 1-3, cont’d (C) Superior view with cross-section of the skull through the
each electrode at the intersection of a specific coronal (identified by the  

temporal line of electrodes.

letter) and sagittal (identified by the number) line.
While the 10-10 system may sound ever so slightly complicated, in
practice it is quite easily carried out. Nonetheless, there is nothing like
actually measuring and placing the electrodes yourself under the guid-
ance of an experienced EEG technologist. We recommend that all
6
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
residents perform at least two to three supervised EEGs during their EEG electrical currents that flow in and around the cells. The flow of current
rotations. Fellows should do more until they are confident in their ability creates a field that spreads out from the origin of an electrical event
to measure accurately and apply electrodes properly. (such as a spike or slow wave), much the same as the concentric rings
created on a glassy pond when one tosses a pebble onto its surface.
Potential fields are usually oval in shape and may be quite restricted or
POTENTIAL FIELDS very widespread. The field’s effect diminishes as the distance from the
Before discussing how we display the electrical information recorded by source increases. This means that events producing maximal volt­age
the electrodes, the reader should understand the concept of the potential on a particular electrode will affect adjacent electrodes as well, but to a
field. The summation of IPSPs and EPSPs in a neuronal net creates lesser extent as the potential wanes from the point of origin (Figure 1-4).

120 Figure 1-4 A potential field. (A) The

 

figure illustrates a maximum negative

potential of -100 µV at F8. The field
Fp2 100 spreads to involve T8 at a lower potential
of –70 µV and then to Fp2 and P8 at
–30 µV. The background averages –20 µV.
80 (B) Another way to depict the same data.
F8
Potential in V

Note the steep rise from Fp2 to F8,

100 V declining successively to T8 and P8.
60
70 V

Background
20 V T8 40

20
30 V

P8 0
Fp2 F8 T8 P8 O2
Electrodes
A B
7
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

AMPLIFICATION
Easiest to understand is the simple amplifier. Input from a single active
electrode is conducted to the amplifier and compared with ground
(earth). Thus, the output consists of the potential difference between the
active electrode and ground. Electrocortical potentials, as well as other
environmental potentials affecting the electrode (e.g., 60 Hz interfer-
ence), are displayed in the output. In differential amplification, signals
from two active leads are conducted to the amplifier, thus measuring the
potential difference between the two (Figure 1-5). In this case, any signal
that affects both inputs identically (say 60 Hz) will result in no potential
difference and thus will not be displayed or be much reduced. This
phenomenon is termed in-phase cancellation. A Ground
We are now in a position to consider methods of recording electro-
cortical potentials so that we can make sense of them. Recalling that
amplifiers record potential difference between two incoming signals, we
can record the potential difference between two electrodes on the scalp
(bipolar recording). On the other hand, we can record the potential
difference between a scalp electrode and another point (the reference)
that, ideally, is unaffected by cerebral potentials or other interference
(referential recording). Unfortunately, it is virtually impossible to achieve
this ideal, but certain references (e.g., the ears) are quite serviceable.
These two types of recording, along with their advantages and disad-
vantages, are discussed below.

BIPOLAR RECORDING B Ground

Bipolar recordings electronically link successive electrodes (known as a
chain or line). The voltage at one electrode is compared with the voltage Figure 1-5 (A) Simple amplifier. Input from each active electrode is compared with
 

affecting adjacent electrodes (potential difference). Each amplifier has ground. (B) Differential amplifier. Here, potential difference is measured between
two active electrodes.
two inputs, I and II. By convention, the rules for understanding the
display are:
8
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
If input I becomes negative with respect to input II, there is an
upward deflection.
If input II becomes negative with respect to input I, there is a
downward deflection.

Contrary to our well-used Cartesian coordinate system, the conven-

tion in neurophysiology is that an upward deflection is negative and a –100
downward deflection is positive. –20 T8
In the simplest example, consider a spike with a very limited potential
field involving only T8 (Figure 1-6). The electrode pairs (or derivations)
in this case are F8–T8 and T8–P8. F8–T8 is Channel 1, and T8–P8 is
Channel 2. In Channel 1 T8 is in input II, and in Channel 2 T8 is in P8
input I. The voltage at T8 (–100 µV) is compared with the background
activity at F8 and P8 (–20 µV). Therefore, in this example:
Figure 1-6 Principle of bipolar localization. The figure depicts a spike discharge of
Channel 1: F8–T8 = –20 µV– (–100 µV) = 80 µV (downward
 

–100 µV at T8. The potential is conducted to input II in the first amplifier and to
deflection) input I in the second amplifier. Other electrodes are not affected by the event. The
Channel 2: T8–P8 = (–100 µV) – (–20 µV) = –80 µV (upward result is known as a phase reversal.
deflection)

In this case, the adjacent channels containing the T8 electrode record to the voltage P8, which is unaffected by the spike at F8 but is recording
the same potential but in opposite directions. This creates the phase the background activity (–20 µV). In Channel 4, the voltage at P8 is
reversal. Most spike discharges at the surface are negative in sign, and compared with O2, both unaffected by the F8 spike. Thus, there is no
negative phase reversals resemble two sharp points touching or nearly potential difference and no deflection.
touching. Channels 1 and 2 are displaying the same potential but with
opposite deflections. Again, this is phase reversal – the localization Channel 1: Fp2–F8 = (–50 µV) – (–100 µV) = 50 µV (downward
principle of bipolar recording. deflection)
Let us now analyze the display when a spike at F8 has a wider poten- Channel 2: F8–T8 = (–100 µV) – (–50 µV) = –50 µV (upward
tial field that also affects Fp2 and T8 (Figure 1-7A). In Channel 1, the deflection)
voltage at Fp2 (–50 µV) is compared with the voltage at F8 (–100 µV). Channel 3: T8–P8 = (–50 µV) – (–20 µV) = –30 µV (a smaller
In Channel 2, the voltage at F8 (–100 µV) is compared with the voltage upward deflection)
at T8 (–50 µV). In Channel 3, the voltage at T8 (–50 µV) is compared Channel 4: P8–O2 = (–20 µV) – (–20 µV) = 0 µV (no deflection)
9
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Channel 1 I
Channel 1
II
I
–100
II Channel 2
Channel 2
–100
–80
–80
–50
–20 –50 I
T8 –20
Channel 3
Channel 3 T8 II
I
P8 P8 Channel 4
II
Channel 4

A B
Figure 1-7 (A) Phase reversal in longitudinal bipolar montage. Here, a spike of –100 µV at F8 spreads to involve Fp2 and T8, each at –50 µV. The potential difference
 

between F8 and the other two electrodes is 50 µV. The display demonstrates a phase reversal at F8 (Channels 1 and 2) with representation of the spike in Channel 3 (the
potential difference between T8 and P8 is –30 µV). (B) Referential montage. The same spike displayed in a referential montage. In a referential montage, each electrode is
compared to a reference electrode. The potential at the active electrode is conducted to input I of each amplifier. The reference electrode is conducted to input II. The
amplitude of the displayed spike is proportional to the voltage at each active electrode.

Other channels (e.g., F4–C4 and C4–P4) may be affected by the conducted to input II. Thus, in referential recording, we record the
declining potential field generated at F8. Thus, phase reversals at lower potential difference between a particular scalp electrode and a referential
amplitude would be recorded at these sites. Note that these considera- electrode. Reference montages produce a higher amplitude EEG record-
tions apply to any potential at any point on the scalp. ing because of the longer interelectrode distances. Theoretically, the
reference can be located anywhere, but there are practical considera-
REFERENTIAL RECORDING tions. A reference placed at any distant point will be contaminated with
In referential recording the amplifiers are not linked as in bipolar record- ambient electrical noise, 60 Hz artifact (50 Hz in Europe). A reference
ing. Signals from each of the scalp electrodes are conducted to input placed on, say, the shoulder or chest would also pick up high-voltage
I of the associated amplifier, while signals from the reference are EKG artifact. Interference from an EKG would render the EEG
10
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
unreadable. The ears are relatively free from both these artifacts, A note on ear and vertex referential recording: A recorded event
although it must be said that EKG is sometimes a contaminant at the (spike, slow wave) is best represented when the reference is distant from
ear electrodes. Moreover, due to the proximity of the ears to the mid- the exploring electrode. Considering the ipsilateral ear reference (A1 or
temporal lobes, the ears do pick up cerebral activity. A2), the ear is close to the midtemporal electrodes T7 or T8. When
Now, utilizing the ears as a contralateral reference, let us compare examining a spike at T7, the ipsilateral ear reference (A1) is not an
the voltage of an event occurring at F8 with that at a contralateral ear appropriate choice, as the potentials at T7 and A1 are very similar. A
reference, A1 (Figure 1-7B). In this example we will assume that A1 is vertex reference or a contralateral ear reference (A2) is more appropriate
recording the same as the background at –20 µV. Here we have a spike for the examination of that T7 spike. Similarly, a spike that is maximal
discharge with an amplitude of –100 µV at F8. The potential field of at C3 will be ill served by placing it in a reference montage using the
the spike spreads to Fp2 and T8 with an amplitude of –50 µV. Beyond Cz electrode, as the reference and the active electrode are too close
these points there is no representation of the field associated with the together. For a C3 spike, either ear electrode would be an appropriate
spike. reference. The reference chosen for a particular spike should be as
distant as possible from that spike.
Channel 1: Fp2–A1 = (–50 µV) – (–20 µV) = –30 µV (small A widely used reference is the common average reference. In this
upward deflection) scheme, the voltage of an event occurring under a particular elec-
Channel 2: F8–A1 = (–100 µV) – (–20 µV) = –80 µV (big trode (input I) is compared with the average voltage recorded by all the
upward deflection) electrodes on the scalp (input II). This creates a situation in which a
Channel 3: T8–A1 = (–50 µV) – (–20 µV) = –30 µV (small focal spike discharge, maximal at T8, will result in an upward deflec-
upward deflection) tion at T8 as T8 will be more electronegative than the average re­
Channel 4: P8–A1 = (–20 µV) – (–20 µV) = 0 µV (no deflection) ference. Neighboring electrodes involved in the field, for example at
F8, will have upward deflections as well, but these will be lower in am­
In referential recording, the localization principle is amplitude. That plitude. Note that the upward deflections thus recorded define the po-
is, the electrode recording the greatest amplitude of the wave in question, tential field of the event. Electrodes not involved in the negative spike
in this case a spike at F8, defines the focus. discharge at T8 will be relatively electropositive compared with
References other than the ears are also in common use. One is the the average reference and thus will have a downward deflection
vertex (Cz), often used in a referential montage to complement the ear (Figure 1-8).
reference. The astute reader will recognize that the vertex resides in a We now present the paradox of bipolar recording and stress how
sea of cerebral activity. Thus, the background of the EEG recorded by important it is to use the various montages in a complementarily fashion.
the vertex electrode will be input II of all channels. As long as this is The paradox is a result of the previously mentioned in-phase cancella-
recognized, one is able to determine the location of a waveform that tion – that is, potentials that are equal in the two inputs of an amplifier
stands out from the background (e.g., a spike or delta wave). are isoelectric in the display. In other words, there is no potential
11
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Potential Figure 1-8 The common average reference. Recording of a spike discharge at T8 of
 

difference –100 µV. The reference (going into input II at each channel) is the common average
(PD) voltage, which in this case is –20 µV. In Channels 2, 3, and 4 the amplitude is
Fp2 I
proportional to the recorded voltage at each electrode. The downward deflection
–10 II 10
in channels 1 and 5 is due to the fact that Fp2 and O2 are relatively electropositive
F8 I (–10 µV) compared with the average voltage (–20 µV).
–30
–50
II
Average
voltage –100 I
–80
20 T8 II

I
–30
P8 II
I
O2 II 10

difference! The unwary, when examining Channels 2 and 3 of Figure montage) (Figure 1-10A). Note: arrows are often used in North America
1-9A, might conclude that little if anything is occurring at F8, T8, and for convenience: the tail of the arrow indicates input I; the point of the
P8. On the other hand, when one looks at the same situation with a arrow input II.
referential recording, it becomes clear that the maximum abnormality Adjacent electrodes are connected from front to back, including the
underlies those very electrodes (Figure 1-9B). temporal (lateral) chain and the parasagittal (supra-sylvian) chain. The
EEG is displayed in various ways. In this example, the four channels of
the temporal chain on one side are followed by the temporal channels
MONTAGE SELECTION on the opposite side. Similarly, the four channels of the parasagittal
Montage refers to the pattern of systematic linkage of the scalp elec- chain also alternate. In North America, the left side is written out first
trodes designed to obtain a logical display of the electrical activity. followed by the right. In Europe the opposite is the case. Some labora-
Unlike the 10-10 system of electrode placement described earlier, there tories write out the eight channels of left-sided electrodes followed
is no international standard of montages to be used in EEG laboratories. by the right-sided electrodes. Still others prefer alternating homologous
Certain montages, however, are in widespread use. In bipolar recording channels, for example, Fp1 → F7; Fp2 → F8, and so on. Overall, the
the longitudinal arrangement is perhaps the most popular (known in the latter tends to be a bit more confusing – but electroencephalographers
trade as the “double banana,” and by some as the Queen Square experienced with a particular electrode arrangement have no difficulty.
12
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Potential Potential
difference difference
I (PD) (PD)
I
II II

I I
II –80

–100 II
–100
I
–20 T8 I –20 –80
T8 II
II
I
P8 P8 II –80
I

II –80 I
II
A B
Figure 1-9 The paradox of bipolar recording. (A) Representation of a –100 µV spike that affects F8, T8, and P8 equally. Inasmuch as there is no potential difference
 

between F8–T8 and T8–P8, the spike is not recorded in Channels 2 and 3 and gives the impression that there is no abnormality at T8. (B) Same discharge in referential
recording. Note equal deflection in Channels 2, 3, and 4. The true picture is thus displayed.

A second popular arrangement is the transverse bipolar montage. occur at the end of the longitudinal bipolar chain: Fp1, Fp2, O1 or O2
This links adjacent electrodes in transverse chains, starting anteriorly (Figures 1-10C and 1-11).
and progressing posteriorly. Each chain starts with the left side and With respect to referential recording, the recording is usually dis-
progresses to the right (i.e., F7 → F3 → Fz → F4 → F8). The transverse played in both A-P and transverse arrangements, reprising commonly
montage is particularly well suited to record abnormalities occurring at used bipolar montages. A variety of other montages are employed at the
or near the vertex (e.g., midline spikes) (Figure 1-10B). One additional discretion of the individual electroencephalographer. The idea, in short,
bipolar montage comes to mind: the circumferential montage. As the is to highlight certain areas of interest in the best possible way. If the
name implies, the circumferential montage encircles the head and student is familiar with the 10-10 system and is apprised of the montage,
is particularly useful for examining spikes and sharp waves, which he or she should have no difficulty in interpreting the record.

13
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Figure 1-10 (A) A typical longitudinal bipolar

 

montage. The numbers refer to channels, reflecting

voltage difference between two electrodes. Both
2 temporal and supra-sylvian chains alternate from
left to right. The arrows represent the inputs to
1 5 1 3 each amplifier. The tail is in input I and the arrow is
9 13
in input II. (B) A transverse bipolar montage. The
chains run from left to right, beginning anteriorly
and proceeding posteriorly. Note that the midline
4 5 6 7
2 10 17 14 6 electrodes are incorporated into the second, third,
and fourth chains, thus allowing good
8 9 10 11 12 13 representation of midline events.

3 11 18 15 7
14 15 16 17

12 16
4 8 18 20

19

A B

In the era of digital EEG, specific montage selection by the technolo- In summary, the technologist may record an EEG in a set sequence
gist is not as critical as it was in the analog days. All recording is actually of montages but the reviewing electroencephalographer can review the
done referentially. The software allows display of recorded potentials in EEG in any montage desired. Furthermore, a given page or discharge
any desired montage. Thus, the technician and reader can now easily can be examined in a variety of montages to help understand its meaning.
switch from one montage to another to examine the characteristics of a Much as we would circle a complex sculpture in a museum, we circle
particular phenomenon. A low-amplitude temporal spike during bipolar an EEG wave by using different montages. Remember, the central idea
recording can rapidly be inspected on a referential montage with the is to maximize the opportunity to display an abnormality for optimal
flick of the computer mouse. recognition.

14
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Clearly, some order was required so that EEGs obtained in one labora-
tory are easily interpretable at another. For many years nearly all labo-
8
ratories in North America, and indeed in many laboratories throughout
FP1 FP2
the world, have used similar electronic settings for routine work.
9 7 Following is a brief discussion of the most important recording
parameters.
F7 F8
F3 Fz F4 Calibration
10
(11) (13) (15)
6 Calibration is a way to accurately measure EEG potentials by adminis-
trating a standard signal through each amplifier. Once this is performed,
A1 T7 C3 CZ C4 T8 A2 the voltage of an EEG potential is compared against this known voltage.
Calibration is currently built into the software of most digital EEG
(12) (14) (16) systems and is performed automatically. Additionally, an impedance
1 5
Left Right check should appear at the start of every recording. The impedance
P3 Pz P4 check is a way of establishing the integrity of each electrode. Impedances
P7 P8
should not exceed 5 kohms.
2 4
01 02
Display
3
In most North American and many European laboratories the standard
display timebase is 30 mm/sec with 10 seconds of EEG per display. There
C is nothing magic about the number – in fact, some laboratories (par­
Figure 1-10, cont’d (C) Circumferential bipolar montage.
 
ticularly in Europe) prefer a timebase of 15 mm/sec. The appearance of
the EEG is considerably altered in the latter case (i.e., the alpha rhythm
at 30 mm/sec looks like rhythmic beta activity at 15 mm/sec). The
important point is that the reader knows what timebase is selected. It
should be said that there are instances when use of a shorter timebase
OVERVIEW OF ELECTRONICS
is quite useful (e.g., in the identification of periodicity, or even rhythmic-
We often say that the EEG display can be manipulated at will and ity of a particular phenomenon [e.g., in ICUs or for neonatal EEGs]).
made to demonstrate a severe abnormality or to show a normal pattern. Likewise, increasing the timebase to, say, 60 mm/sec may allow one to
This manipulation refers to changing the electronic circuitry with the analyze more accurately wave configuration, particularly when a phe-
press of a button in order to alter sensitivity, filtration, and timebase. nomenon is “crowded” as in grouped spikes.

15
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Figure 1-11 Occipital spike. Spike discharges at the “end of chain (Fp1, Fp2, O1,
 

O2)” can be easy to miss in the standard longitudinal bipolar montage. Here, a
Fp2 right occipital (O2) spike discharge is displayed at –100 µV. (A) In a standard
longitudinal bipolar montage, the deflection is always downward. (B) The
Background
discharge can be confirmed by placing it in a circumferential bipolar montage.
–20 V F8 Phase reversal at O2 confirms the spike maximum at this location.

30 V

T7 T8

Sensitivity
P7 50 V P8
The sensitivity of each channel refers to the amplitude of the display
100 V produced by the received signal. The measurement is expressed in
O1 O2
voltage per deflection. Standard sensitivity is 7 µV/mm.
Sensitivity may be altered for any particular channel depending on
the specific need. For example, the sensitivity of a channel recording the
Longitudinal Potential Circumferential Potential
bipolar difference posterior difference EKG would have to be decreased due to the much higher voltage of this
montage (PD) halo (PD) signal (measured in millivolts). In general, the sensitivity of all channels
Fp2-F8 No PD T8-P8 20 V recording the EEG may be changed simultaneously by a stepped gain
control. For example, one might wish to increase sensitivity in situations
where the general voltage of the EEG is low. Similarly, some EEG phe-
F8-T8 10 V P8-O2 50 V
nomena reach very high voltages (e.g., generalized spike-wave dis-
charges), requiring a decrease in sensitivity (15 µV/mm) in order to
T8-P8 20 V O2-O1 50 V properly analyze the waveforms. Please note, raising the gain from, for
example, 7 µV to 15 µV is the same thing as lowering the sensitivity,
and the EEG will appear lower in amplitude.
P8-O2 50 V O1-P7 20 V

A B High-frequency filters (HFFs) or low pass filters

This circuit attenuates undesirable high frequencies (e.g., muscle action
potentials) and passes low frequencies (Figure 1-12A). In an HFF circuit,
the input signal is placed across the combination of a resistor and a
16
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
High-frequency filter Low-frequency filter capacitor in series and the output signal is measured across the capacitor
alone (Figure 1-12A). At high frequencies, the impedance of any capaci-
High Low tor is low. Measuring output across the capacitor with a high frequency
filter filter will be essentially zero, as the voltage does not change, so the potential
C difference is zero. The standard HF setting is 70 Hz. Other standard
R settings are 35 Hz and 15 Hz, the latter severely attenuating a broad
range of high frequencies. As a practical matter, recording at a HFF
vin C vout vin R vout setting of 15 Hz should not be employed save in rare and unusual cir-
cumstances. An unwanted consequence would be a marked attenuation
of spike potentials. Unfortunately, the authors have inspected EEGs from
1.0 outside sources in which an HFF setting of 15 Hz was used throughout.
Passband Stopband Stopband Passband
Such records look “clean” but fail to convey needed information. Don’t
0.1 Nominal filter Nominal filter do it.
frequency frequency
Gain

(gain  0.71) (gain  0.71)

0.01 Low-frequency filters (LFFs) or high-pass filters
In an LFF, there is marked attenuation of slow potentials below the
0 cutoff frequency (such as those caused by sweat artifact, respirations,
0.1 1.0 10 100 1000 10,000 .01 0.1 1.0 10 100 and tongue movement), with little effect on rapid potentials such as
Frequency, Hertz Frequency, Hertz spikes or muscle artifacts. In an LFF circuit, the input signal is placed
across the combination of a capacitor and a resistor in series and the
A B
output signal is measured across the resistor alone (Figure 1-12B). The
Figure 1-12 (A) High-frequency filter. In a high-frequency filter, the output voltage
 
impedance of any capacitor is very high at low frequencies. In this circuit
for high frequencies is lower than the input voltage for high frequencies. In the arrangement, low-frequency input signals are essentially blocked. At
log–log graph, frequencies below the cutoff are unchanged while frequencies higher frequencies, the impedance at the capacitor is low and the signal
above the cutoff are attenuated. (B) Low-frequency filter. In a low-frequency filter, is measured across the resistor essentially unchanged from the input. The
the output voltage for low frequencies is lower than the input voltage for low LFF is typically set at 1 Hz.
frequencies. In the log–log graph frequencies above the cutoffs are unchanged
while frequencies below the cutoff are attenuated. (Adapted with permission from
Notch filter
Lippincott Williams & Wilkins/Wolters Kluwer Health: Schomer, Lopes da Silva,
Niedermeyer’s Electroencephalography, 2010.) In addition, a notch filter setting of 60 Hz (US) or 50 Hz (Europe) is
usually employed, selectively reducing environmental interference.
17
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

NOTES ON RECORDING THE EEG that are only evident when the patient is awake – a circumstance some-
times encountered in patients with dementia.
Many special problems confront the technologist in his or her efforts to
obtain an EEG that can be interpreted successfully by the electroen- ARTIFACTS
cephalographer. We emphasize that the electroencephalographer is Recognition of artifacts is one of the vexing and strangely satisfying
totally dependent on the quality of the recording – that is, regardless of aspects of EEG interpretation, as well as one of the most important. As
the expertise of the reader, he or she is unable to use that expertise in a beginner, you may find the differentiation of artifacts from physiologi-
the face of a technically inadequate tracing. The ability to properly place cal phenomena quite difficult. A distinguishing characteristic of the
electrodes in conformity with the 10-10 International System (including, experienced electroencephalographer is the ability reliably to recognize
importantly, accurate measurements of electrode location) is critical if artifacts. For the most part the reader will soon master artifact recogni-
one is to compare electrical activity between the two hemispheres with tion, particularly after understanding their characteristics and referring
accuracy. If epilepsy is suspected, the technologist should attempt to to the mini-atlas, and should not be too daunted by the seeming impos-
record drowsiness and sleep if possible. Moreover, because focal epilep- sibility of this task!
tiform activity is often activated by the interface between wake and Artifacts come in many different forms and have diverse causes. The
drowsiness, the technologist should gently alert the drowsy patient on major underlying problem is the enormous amplification required to
several occasions in an attempt to provoke spikes. Similarly, if a patient record brain waves. As a result, amplified non-cerebral potentials – for
is sleeping at the onset of the test, he or she should be aroused after example vigorous movements by the patient producing random excur-
some minutes of recording. This ensures that a relative waking record sions of the electrode leads – may render the EEG uninterpretable.
is obtained. Unfortunately, sleep may obscure background abnormalities Specific artifacts are detailed in Figures 1-13–1-30.

18
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
300 uV
Cz-Pz
1 sec
ECGL-ECGR

Figure 1-13 Chewing artifact. Generalized muscle action potentials (arrows) with repetitive chewing motions.
 

19
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz

LOC-A1 100 uV
ROC-A2
02/27/2014 15:39:42
1 sec
02/27/2014 15:39:48
ECGL-ECGR Asleep FREQUENT SNORES WITH AROUSALS

Figure 1-14 EKG artifact. Diffuse sharp potentials (arrows) coincident with the EKG. The artifact is particularly prominent in channels connected to the ears. It also may be
 

diffuse. If there is no EKG monitor, and if the patient has atrial fibrillation or frequent premature contractions, the artifact may be confounding, be inconsistent, and
masquerade as spike discharges. Look for phase relationships that do not comport with those of true spikes. EKG artifact is particularly prominent in the obese and those
with hypertension.

20
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
140 uV
Cz-Pz
1 sec

Figure 1-15 Eye blink artifact. High-voltage potentials, maximal in the frontal derivations. The deflection results from the cornea-retinal potential (the cornea is
 

electropositive with respect to the retina, measured in millivolts), along with a minor contribution of the electroretinogram (ERG). During an eyeblink the globes turn
slightly upward (Bell’s phenomenon). Thus, the frontopolar electrodes become momentarily positive (to understand deflections, recall the rule for bipolar recording.)
Figure shows eye opening (thin arrow), eye closure (thick arrow) and disappearance and reappearance of PDR (arrowheads).

21
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz

LOC-A1 140 uV
ROC-A2
1 sec
05/13/2014 12:23:08 05/13/2014 12:23:14
ECGL-ECGR Eyes Open Eyes Closed

Figure 1-16 Prosthetic eye. In a patient with a right prosthetic eye, the blink artifact is expressed on only one side. Arrows point to missing right-sided eye blink artifact.
 

One will also see limited eye blink potentials in those with a third nerve palsy.

22
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Figure 1-17 Eyelid flutter.
 

Fp1-F7 In (A) eyelid-flutter

F7-T7 produces a rhythmic
T7-P7 bifrontal frequency, here
at 3–4 Hz (thin arrow). Eye
P7-O1
leads are out of phase as
LOC is positioned on the
Fp2-F8 left lower canthus and
ROC is positioned on the
F8-T8 right upper canthus. (B)
T8-P8 Shows frontally
P8-O2 predominant generalized
rhythmic delta activity
(GRDA) (thick arrow). Eye
Fp1-F3
leads show synchronous
F3-C3 (in phase) delta as both
C3-P3 eye electrodes are anterior
P3-O1 to the frontal lobe and
recording very similar
activity.
Fp2-F4
F4-C4
C4-P4
P4-O2
140 uV
Fz-Cz
1 sec
Cz-Pz

LOC-A1

ROC-A2
A B
23
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Figure 1-18 Lateral eye

 

Fp1-F7 movement artifact (reading).

Recognizable in the
F7-T7 frontotemporal derivations as
sharply contoured potentials
T7-P7 that are out of phase. This
figure shows three left saccades
P7-O1
(thick arrows). When the eyes
saccade to the left, the globe
Fp2-F8 on the left approaches the left
anterior temporal electrode (F7)
while the right globe turns
F8-T8
away from the right anterior
T8-P8 temporal electrode (F8). A
positive potential is therefore
P8-O2 recorded at F7 and a negative
potential at F8. (Remember, the
cornea is positive with respect
Fp1-F3 to the retina.) Thus, in bipolar
recording, the resultant
F3-C3 waveforms deviate away from
each other in the two channels
C3-P3 connected to F7, while the
P3-O1 opposite is the case with the
channels connected to F8. Note
also that very rapid spike
potentials may occur during
Fp2-F4
lateral eye movements with
potential maxima at the F7/F8
F4-C4
electrodes. These result from
C4-P4 movements of the lateral rectus
140 uV muscles and are known as
P4-O2 lateral rectus spikes (thin
1 sec arrow).

24
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz

LOC-A1 100 uV
ROC-A2
1 sec

Figure 1-19 Nystagmus. In this patient with nystagmus, again there are sharply contoured potentials (arrows) in the frontotemporal derivations, which are out of phase.
 

There is a rapid rise on the right side followed by a gradual fall, which is the corrective movement. The steeper positive phase reversal, seen here on the right, indicates
the direction of the fast component of the nystagmus.

25
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
140 uV
Cz-Pz
1 sec
ECGL-ECGR

Figure 1-20 Roving eye movements. Slow, lateral eye movements during drowsiness that produce slow waves with alternating phase relationships in the frontotemporal
 

derivations.

26
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz
140 uV
09/24/2014 11:03:35
[*] RELAXED JAW
1 sec

Figure 1-21 Muscle artifact. Muscle artifact (arrows) maximal in the frontal and temporal regions due to electrode placement over the frontalis and temporalis muscles.
 

When the technician asks the patient to relax his jaw, the artifact dissipates. Muscle potentials are less than 20 ms, whereas cerebral spike potentials are longer, lasting
20–70 ms.

27
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
140 uV
Cz-Pz
1 sec

Figure 1-22 Tooth grinding artifact. Alternating tooth grinding produces this checkerboard muscle artifact pattern.
 

28
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz 140 uV

1 sec
ECGL-ECGR

Figure 1-23 Patting artifact. Rhythmic potentials resembling an ictal discharge seen here in the right occipital electrodes (arrows), usually produced by a mother who
 

holds her baby on her lap during the EEG. Notice the lack of a field anterior to the artifact.

29
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz

LOC-A1 100 uV
ROC-A2
1 sec
ECGL-ECGR

Figure 1-24 Ventilator artifact. Wide excursions (arrows) that may resemble delta waves. A check on the rhythmicity (usually in the range of 12 per min), along with a
 

stereotyped waveform, makes the diagnosis. Note that the artifact, in cases where the patient overrides the respirator, may demonstrate irregularity. Amplitude can vary.

30
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz 60 uV

1 sec
ECGL-ECGR

Figure 1-25 Respiratory artifact. Note the periodic bursts of sharply contoured theta/alpha frequency activity prominently seen over the anterior regions. In this patient
 

(same patient as in Figure 1-23), this activity correlated with ventilator rate (chest rising movement) and disappeared with suction. This artifact is caused by the
movement of fluids within the upper respiratory tract and/or the tube and can also occur irregularly in a patient overriding the respirator. Concomitant use of video and/
or audio (sometimes you can hear gurgling sounds) can help to prevent misinterpreting these artifacts as cerebral rhythm.

31
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7

P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz 140 uV

1 sec

Figure 1-26 Shiver artifact. Bursts of rhythmic widespread spikes at 10–14 Hz, which are too brief to be cerebral in origin.
 

32
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz
100 uV
LOC-A1 06/24/2014 16:17:55
TALKED
ROC-A2 1 sec

ECGL-ECGR

Figure 1-27 Glossokinetic artifact. The tip of the tongue is negatively charged, and movement of the tongue can cause synchronous delta activity (arrow) in the frontal
 

derivations.

33
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz 140 uV

1 sec
A B
Figure 1-28 60 Hz artifact. Rhythmic frequency at 60 Hz (or 50 Hz in Europe) secondary to nearby electrical apparatus or poor grounding, usually expressed because of high
 

electrode impedance but sometimes (particularly in the ICU) difficult to eliminate. (A) Shows EEG with a great deal of 60 Hz artifact. In (B) the notch filter has been applied.

34
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
Cz-Pz

LOC-A1
140 uV
ROC-A2
1 sec
ECGL-ECGR

Figure 1-29 Tremor artifact. 4–6 Hz tremor artifact (arrows) posteriorly in this 66-year-old woman with Parkinson’s disease. Note how there is little field anteriorly, which
 

would be very unusual for a cerebrally generated wave.

35
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
ROWAN’S PRIMER OF EEG

Fp1-F7
F7-T7
T7-P7
P7-O1

Fp2-F8
F8-T8
T8-P8
P8-O2

Fp1-F3
F3-C3
C3-P3
P3-O1

Fp2-F4
F4-C4
C4-P4
P4-O2

Fz-Cz
140 uV
Cz-Pz
1 sec
ECGL-ECGR

Figure 1-30 Electrode artifact. A faulty electrode contact (arrow) results in a recording with an exact mirror image referable to the common electrode (in this case, F4). In
 

referential recording, only one channel reflects the discharge. In both cases there is no potential field. A faulty electrode can also “pop” resulting in a mirror image for only
a moment.

36
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.
 Origin and technical aspects of the EEG 1 
Further reading Ebner, A., Sciarretta, G., Epstein, C.M., et al., 1999. EEG instrumentation. The
International Federation of Clinical Neurophysiology. (Practice Guideline).
Adrian, E.D., Matthews, B.H.C., 1934. The Berger rhythm: potential changes from the
Electroencephalogr. Clin. Neurophysiol. Suppl. 52, 7–10.
occipital lobes in man. Brain 57, 355–385.
Goldensohn, E.S., 1979. Neurophysiological substrates of EEG activity. In: Klass, D., Daly, D.
American Clinical Neurophysiology Society Guidelines. www.acns.org.
(Eds.), Current Practice of Clinical Neurophysiology. Raven, New York, pp. 421–440.
American, E.E.G., 1986. Society Guidelines in EEG, 1–7 (Revised 1985). J. Clin.
Goldman, D., 1950. The clinical use of the “average” reference electrode in monopolar
Neurophysiol. 3, 131–168.
recording. Electroenceph Clin Neurophysiol 2, 211–214.
Andesen, P., Andersson, S.A., 1968. Physiological Basis of the Alpha Rhythm. Appleton, Halliday, A.M., Butler, S.R., Paul, R. (Eds.), 1987. A Textbook of Clinical Neurophysiology.
New York. Wiley, Chichester, pp. 3–22.
Beaussart, M., Guiev, J.D., Section, I.I.I., 1977. Artefacts. In: Remond, A. (Ed.), Handbook Homan, R.W., Herman, J., Purdy, P., 1987. Cerebral localization of international 10-20
of Electroencephalography and Clinical Neurophysiology, vol. 11A. Elsevier, system electrode placement. Electroenceph Clin Neurophysiol 55, 376–382.
Amsterdam, pp. 80–96.
Jasper, H.H., 1958. Report of the committee on methods of clinical examination in
Berger, H., 1929. Ueber das elektroenkephalogramm des menschen. Arch Psychiatr electroencephalography. Electroenceph Clin Neurophysiol 10, 370–375.
87, 527–570.
Jasper, H.H., 1958. The ten-twenty electrode system of the International Federation.
Binnie, C.D., 1987. Recording techniques: montages, electrodes, amplifiers and filters. Electroenceph Clin Neurophysiol 10, 371–375.
In: Halliday, A.M., Butler, S.R., Paul, R. (Eds.), A Textbook of Clinical Neurophysiology. Klass, D.W., 1977. Symposium on EEG montages: which, when, why and whither.
John Wiley, New York, pp. 3–22. Introduction. Am. J. EEG Technol. 17, 1–3.
Binnie, C.D., Rowan, A.J., Gutter, T., 1982. A Manual of Electroencephalographic Lesser, R.P., Lueders, H., Dinner, D.S., et al., 1985. An introduction to the basic
Technology. University Press, Cambridge. concepts of polarity and localization. J. Clin. Neurophysiol. 2, 45–61.
Brittenham, D., 1974. Recognition and reduction of physiological artifacts. Am. J. EEG Litt, B., Cranstoun, S.D., 2003. Engineering Principles. In: Ebersole, J., Pedley, T.A. (Eds.),
Technol. 14, 158–165. Current Practice of Clinical Electroencephalography. Lippincott Williams & Wilkins,
Buzsáki, G., Anastassiou, C., Koch, C., 2012. The origin of extracellular fields and Philadelphia, pp. 32–71.
currents – EEG, ECoG, LFP and spikes. Nature Rev Neurosci 13, 407–420. Moruzzi, G., Magoun, H.W., 1949. Brain stem reticular formation and activation of the
Buzsáki, G., Traub, R., Pedley, T., 2003. The Cellular Basis of EEG activity. In: Ebersole, J., EEG. Electroenceph Clin Neurophysiol 1, 455–473.
Pedley, T.A. (Eds.), Current Practice of Clinical Electroencephalography. Lippincott Saunders, M.F., 1979. Artifacts: activity of noncerebral origin in the EEG. In: Klass, D.W.,
Williams & Wilkins, Philadelphia, pp. 1–11. Daly, D.D. (Eds.), Current Practice of Clinical Electroencephalography. Raven Press,
Creutzfeldt, O., Houchin, J., Section, I., 1974. Neuronal basis of EEG-waves. In: Remond, New York, pp. 37–68.
A. (Ed.), Handbook of Electroencephalography and Clinical Neurophysiology, vol. 2C. Silverman, D., 1960. The anterior temporal electrode and the ten-twenty system.
Elsevier, Amsterdam, pp. 5–55. Electroenceph Clin Neurophysiol 12, 735–737.
Dempsey, E.W., Morison, R.S., 1942. The production of rhythmically recurrent cortical Stones, E.A., Whitehead, M.K., MacGillivray, B.B., 1967. The nature of the eye blink
potentials after localized thalamic stimulation. Am. J. Physiol. 135, 293–300. artefact. Proc Electrophysiol Technol Assoc 14, 208–214.
Ebersole, J.S., 2003. Cortical Generators and EEG Voltage Fields. In: Ebersole, J., Pedley, Westmoreland, B.F., Espinosa, R.E., Klass, D.W., 1973. Significant prosopo-
T.A. (Eds.), Current Practice of Clinical Electroencephalography. Lippincott Williams & glossopharyngeal movements affecting the electroencephalogram. Am. J. EEG
Wilkins, Philadelphia, pp. 12–31. Technol. 13, 59–70.

37
Downloaded from ClinicalKey.com at University of Cincinnati - OhioLINK February 13, 2017.
For personal use only. No other uses without permission. Copyright ©2017. Elsevier Inc. All rights reserved.