JPRS ID: 9214 TRANSLATION EVOKED POTENTIALS IN PSYCHOLOGY AND PSYCHOPHYSIOLOGY BY E.M. RUTMAN

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000200100045-9
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 231
• Document Creation Date: November 1, 2016
• Sequence Number: 45
• Content Type: REPORTS

APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9
i T E ~ ~ _ r
~ ~
,~~L~' E. ~UT~~~~ ~

APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
~
FOR OFFICIAL USE ONLY
JPRS L/9214
24 July 1980
Tr~nslation
EVC~KED POTENTIALS IN PSYCHOLOGY
AND PSYCHOPHYSIOL~GY
gy -
~ E,M. Rutman
~BI~ FGREIGN BROADCAST INFORMATiON SERVICE
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
NOTE
JPRS publications contain information primarily from foreign ,
newspapers, periodicals and books, but also from news agency _
transmissions and broadcasts. Materials irom foreign-language
sources are translated; those from English-language so~irces
are transcribed or reprinted, with the original phrasing and
other characteristics retained.
Headlines, editorial reports, and mat~rial enclosed in brackets
are supplied by JPRS. Proce~sing indicators such as [Text]
or [Excerpt] in the first line of each item, or following the
last l~nc of a brief, indicate how the original information was
- processed. Where no processing indicator is given, the infor-
ma.tion was summarized or extracted.
- Unfamiliar names rendered phonetically or transliterated are
enclosed in parentheses. Words or names preceded by a ques-
tion mark and enclosed in parentheses were not clear in the
original but have baen supplied as appropriate in context.
Other unattributed parenthetical notes within the body of an "
item originate with the source. Times within items are as
given by source.
The conCents of this publication in no way represent the poli- -
cies, views or attitudes of the U.S. Government.
For farther information on report content -
call (703) 351-2938 (economic); 3~+6~3
(political, sociological, military); 2726
(life sciences); 2725 (physical sciences).
COPYRIGHT LAWS AND REGULATIONS GOVERNING OWNERSHIP OF
MATERIALS REPRODUCED HEREIN REQUIRE THAT DISSEMINATION
OF THIS PUBLICATION BE RESTRICTED FOR OFFICIAL USE ONLY.
.
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
_ FOR OFFICIAL USE ONLY
JPRS L/9214
24 July 1980
EVOKED POTENTIALS IN PSYCHOLOGY AND PSYCHOPHYSIOLOGY
Moscow VYZVANNYE POTENTSIALY V P5IKHOLOGII I PSIKHOFIZIOLOGII -
in Russian 1979 s~gned to press 24 Mar 79 pp 1-216 ~
[Book by E.M. Rutman, USSR Academy of Sciences, Physiclogy
Division, Izdatel'stvo "Nauka", 1,850 copies]
- CONTENT~
Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
General Problems of the Method
Chapter I. The Method for Recording Averaged EP's From the Scalp 4
Chapter II. Monopolar and Bipolar EP Recordi.ng Techniques 13
Chapter III. Descriptior~ of an EP . . . . . . . . . . . . . . . . . . 18
Chapter IV. EP Processing and Analysis Techniques . . . . . . . . . . 23
Chapter V. The Neurophysiological Nature of EP's Record~d From
the Scalp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Chapter VI. Variability of EP's . . . . . . . . . ~ . . . . . . . . . 55
Morphology and Neurogenesis of EP's to Stimuli of Differe~nt Modalities
Chapter VII. Auditory EP's . . . . . . . . . . . . . . . . . . . . . . 62
- Chapter VIII. Somatosensory EP's . . . . . . . . . . . . . . . . . . . 76
Chapter IX. Visual EP' s . . . . . . . . . . . . . . . . . . . . . . . 87 -
_ Chapter X. Some Modally Nonspecific Changes in EP's Associsted
With Stimulus Characteristics . . . . . . . . . . . . . . . . . . . 103
Use of EP's in Psychology and Psychophysiology
Chapter XI. Use of EP's to Assess Sensory Functions 110 -
Chapter XII. EP's and Intelligence Research . . . . . . . . . . . . . 114
Chapter XIII. Use of EP's in Research on Functional Interhem~spheric
- Mutual Relationships (Function Lateralization, Hemispheric
Dominance, and So On) , , , , , , , , , , , , , , , , , , , , , , , 124 -
Chapter XIV. EP'S And Attention Research . . . . . . . . . . . . ..132
- Chapter XV. Psychophysiological "Functi~nal Analysis" of EP ~
Components. The "Orientation" 7~P0 and the "Attention" LPO 152 -
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
- a- LI - USSR - C FOUO]
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
F OR OFF iC IAL US E ONLY
PUBLICATION DATA
_ English title : EVOKED POTENTIALS IN PSYCHOLOGY AND
PSYCHOPHYSIOLOGY
Russian title . VYZVAI~NYE POTE:ITSIALY V PSIKHOLOGII I
PSIKHOF~ZIOLOGII
Author (s) , E. M. Rutman
Editor (s) . D. A. Farber
Publishing House , Nauka
- Place of Publication . Moscow
Date of Publication . 1~'~
~
Signed to press . 24 Mar 79
Copies , 1,850
COPYRIGHT . Izdatel'stvo "Nauka", 1979
- b -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
ANNOTATION
_ This book generalizes basic knowledge on the method of reporting evoked
potentials (EP) from the human scalp, accumulated in the 30 years of its
exis~ence. The greatest emphasis is placed on the basic results of research
employing EP in psychology and psychophysiology in the last 5-10 years. The
difficulties, limitations, and passible future uses of the method are
examined. Much attention is devoted to current methods for analyzing recorded
EP, and to the neurogenesis and functional meaning of individual components.
Basic information is given on the morphology of EP to widely employed sti.muli.
This publication is intended for psychologists and psychophysiologists, and -
it will also be of interest to neurophysiologists, psychologists, and neuro-
logists.
Six tai~les, 21 figures, 642 bibliographic references. -
1
FOR OFFICIr,; USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
~ r~~ii ~rt~ rc;Tnr, ~rtir; (1Nl,Y -
FOREWG~D
_ The recording of evoked potentials (EP) from the human scalp ha:: opened a
new� irreplaceable "window into the brain." The method, which was fi~st
used in 1947, came into broad use in the 1960's, and it is still developing
swiftly today. On one hand it is being used to solve an ever- increasing
range of problems in the most diverse areas of psychology and psychophysiology,
while on the o~her hand research on the nature of EP's and on their relation-
ship to cerebral structures and functions is continuing. Concurrently the _
methods for recording, processing, and analyzing EP's are being improved,
and new methods are arising. The literature illuminating all of these
~ directions has become unfathomable. And yet a researr.her who wishes to
employ this method or determine the suitability of its application to
concrete problems must have the possibility for quickly and, at the same
time, sufficiently fully acquaint himself with today's ideas about the
nature of EP's, with the existing methods for processing and analyzinq them,
and with tne basic results. The objective of this inonograph i~ to afford
such a possibility.
The book consists of three parts. The first examines the method's general -
problems,With considerable attention being devoted to complex and not entirely
cZear ones (neurogenesis, variability, assessment methods, ~nd so on). It
briefly presents the history of the development and supersession of ideas -
about the neurogenesis of EP components with the objective of emphasizing
_ the importance of maintaining a critical attitude toward published data and -
ca�t;on when interpreting one's own results, and the necessity for
psychologists and psychophysiologists to constantly be within the main-
stream of current neurophysiological research on the nature of EP's, and
to maintain an awareness of present ideas concerning the relationsh~p be- -
tween ascillations recorded from the scalp and the activity of brain
structures. A significant fraction of this part is de~?oted to ctrxrent
- methods of analyzing and mathematically pro:.essing EP's, multidimensional
analysis in particular, which is recently enjoying increasingly greater use.
- The second part contains basic inforniation on the m~rphology and neuroqenesis
of modally specific components of EP's in response to auditory, visual, and
somatosensory stimuli. Tt provides normative data on the morphology of EP's
- arising in response to the basic fornis of stimuli used in experimental
2
- FOR OFFICIr~L iiSE ONLY _
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR ~FFICIAL USE ONLY
research, and it examines changes occurring in EP's in co:uiection with
modally nonspecific characteristics of stimulation, such as stimulus proba-
bility and the duration of interstimulus intervals.
The third part provides a brief review of a number of research directions
in psychology and psychophysiology making use of EP, and it examines the
_ basic achievements, the possible causes of failures, and the prospects of
- subsequent research.
Because several monographs fully or partially devoted to the EP method and
its use in a number of areas have been published (5,23,30,76), a number of
problems that are illuminated sufficiently well in these books are not
treated here, thus making fuller examination of others possible. As a
consequence the depth to which the material is covered is to a certain
extent nonuniform, but as a whole the presentation appeared suitable and
justified. ~
The author_ expresses deep gratefulness to I. V. Rabich-Shcherbo i~r his help
and his friendly participation in the work on this book.
3
~
FOR OFFICIti;. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
~ FOR OFFICIAL USE ONLY
GENERAL PRUBLEMS OF TFiE METHOD
~ CHAPTER I
THE METHOD FOR RECORDING AVERAGED E1?'S FROM THE SCALP
Determination of EP's. Sensory EP's, Motor J.~oten~ial, the E-Wave
The term "evoked potential" is applied in neurophysiology to potential -
oscillations in any division of the nervous system, arising in response
to an external influence and existing in a relatively strict temporal
relationship with it. EP's are recorded from the human scalp by ordinary
electrodes and amplifiers used to obtain electroencephalograms (EEG)--
- recordings of brain bioelectric activity taken from the scalp (36,75).
Potential oscillations--manifestations of so-called spontaneous, or back-
ground, brain bioelectric activity--are known to be constantly present on
the human scalp apart from any sort ~f external stimuli. Thus EP's aris2
on the background of spontaneous oscillations. Therefore an EEG reflects
concurrently both potentia~. oscillatioas elici.ted by an external stimulus
. and spontaneous (or background) manifestations of bioelectric activity. -
Inasmuch as the amplitude of responses to external stimuli is lower as a
rule than the amplitude of the oscillations of background bi.;,electric
activity, single responses to a stimulus are usually indistinguishable
on the background of the latter.
_ Registration of EP's became possible owing to the use of various tecn.:~iques
for isolating a si9nal from noise. Evoked activity is interpreted in the
first approximation as oscillations "suspended" in spontaneous activity (368),
though in fact a rather complex re3.ationship daes exist between the EEG
and EP's. The technique for isolating EP's from the general pattern of bio-
electric activity is based on the premise that repetition of a stimulus
elicits repetition of a similar response, while spontaneous background
activity is random. Therefore summation of a sufficient number of EEG
segments covering the moment of stimulus presentation should result in
isolation of the response to the stimulus, no matter how small it is in
comparison with the background (random) activity (Figure 1). In light of
the above, EP's recorded from the human scalp can be defined as the sum of
_ individual responses evoked by repeated stimuli, where each response taken
separately is not necessarily distinguishable (368).
4
FOR OFFICIA;, USF ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
I~UK ~.~E'F1Ci~1, USI? ONI,Y -
~ B -
~
~ D
E -
Figure 1. EEG Averaging, With an Artificial Signal Mixed With
Artificial I~oise as the Example: The background
EEG (A) is synthesized out of a sequence of 90
random numbers from 0 to 10, which were used to
create nine 1-second EEG segments. Segments repre-
senting a background EEG coupled with a response to -
a light flash (B) are created by the addition of
an artificial EP (E) with an average of 0 and a
standard deviation of 10 to each EEG segment.
Averaging (C,D) is achieved by summing the values
of all curves at particular points in time, followed
_ by division of the siun by the number of curves (116).
The term "evoked potentials" is also used to describe ,potentials that are
not responses to external stimuli. Chronologically, techniques for isolating
a signal from noise were first used to record responses of a nerve to
- electric stimulation, and later to record responses of the brain to external
stimuli; in both cases potentials evoked by external stimuli were involved.
Meanwhile the possibility for employing the technique for isolating a signal -
froan noise to extract information from the human EEG was found to be much
broader.
In fact, it became possible to use these techniques to isolate, from EEG's,
changes in potential that are rather rigidly associated in time with some
fixed event. The latter could include a person's response, or some sort of
fluctuations in his autonomic functions, and not necessarily the action of
a sensory stimulus. In particular, scientists were able to record oscilla-
tions associated with the activity of the motor zone of the cortex.(a motor. �
potentiai) and with a particular state which can be described as intention,
5
FOR OFFICI~,:. USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICiAL USL ONLY
anti.cipation, or readiness (the E-wave). These oscillations are recorded on
th~ basis of the same principle of isolating a signal from noise that is
applied to responses to sensory stimuli. The only difference in the case of
a motor EP is that we sum the EEG segments not after the stimulus, but
rather before the motor response, the moment of which can be determined from
a myogram or from a resulting movement. The moment a pushbutton :is pressed
(or an EMG response) serves as the point of reckoning in the same way that
_ the moment of stimulus delivery is used in relation to sensory EP's (reverse
averaging). Thus a number of authors (242,306; Vaughan, et al., 1968;
Deeke, 1969) nave recorded a certain sequence of oscillations before initia-
- . tion of movement, some of v,~hich are most highly pronounced above the motor
zone of the cortex corresponci~.ng to the project~on of the group of muscles
responsible for the movement.
An E-wave was revealed with the recording (by means of a constant current
amplifier) of slow oscillations arising in the period between a preparatory
stimulus and a triggering stimulus in a situation calling for motor reactions
occurring over constant intervals. A negative potential deviation that
arises regularly in response to a preparatory signal and grows at a varying
rate depending on the length of the interval between the preparatory and
triggering signals was recorded (518). In this case the external signal
- is used as a time "marker" associat.ing development of the process at the
basis of the E-wave (or reflected within it) with a certain moment in time.
(which is necessary for sumanation). The accumulated data on the E-wave
implied an association between it and states that can be described as in-
_ tention, readiness, anticipation, and attention (492).
Po'tential oscill~tions arising at the moment an expected stimulus is missed
have also been recorded. Weinberg et al. (525) suggested the term
"emitted events" or "emitted (generated) potentials" to designate such
oscillations, since they arise not in response to an external stimulus but
rather in connection with anticipation of a stimulus or a reaction. Vaughan
(1969) suggested the term "event-related potentials" to designate any bio-
electric responses recorded with the help of techniques for isolating a
signal from noise. These terms have not as ye~ established themselves in
the literature, and motor potentials and other bioelectric oscillations not
associated direc~ly with an external stimulus are often interpreted as
variants of the EP.
This book will examine only sensory and associative EP's--that is, bioelectric
oscillations arising in the human brai.n in response to stimulation of af~erent
sensory systems and associated with reception and processing of sensory in-
formation contained within the stimulus.* As a rule when'we refer to the
* This book does not examine so-called steady state potentials, described
in detail in Regan's book (394), and stable auditory potentials (Picton
et al., 1978).
6
FOR OFFICI6L USE ONLY ~
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200100045-9
FOR UrrICLAI. USE ONLY
recording of EP's from the scalp, we imply responses obtained by one of the -
techniques for isolating a signal from noise, and:not singular responses. _
For the sake of brevity we will use the term "evoked potential"--EP--
irrespective of precisely what technique for isolating a signal fro~n noise
was used to obtain it, if the situation does not call for any special remarks.
In cases where we will be dealing with "raw" EEG recordings made following a -
stimulus, we will refer to singular EP's. (The situation is generally some-
what paradoxical, since a single EEG recording made following a stim~ilus
does not in a sense contain a"pure" response to the stimulus--th~t is, an
EP; instead, it represents the sum of the response and background activity.)
As a rule, recent works deal with EP's obtained as a result of su~nation, -
or of su~ation and averaging. The differences will become clear later on,
and as a rule they are inconsequential in relation to analyzing and inter-
. preting the results.
The Method of Summation and Averaging
The su~nation and averaging method is used to isolate a signal--an event
strictly associated .in time with a stimulus, from noise--events not ha.ving
a str~ct temporal relationship to the stimulus (this division of EEG "event~"
into signal and noise does not necessarily correspond to their "si~~naling" _
significance to the work of the brain). The averaging principle was em-
ployed for signal isolc.tion back in the last century by Laplace, while `
Dawson (179) was the first to use it to register zesponses in the nervous _
system. Application of the averaging method to EP's is based on a number
of assumptions that are not in fact entirely valid. A certain degree of
inadequacy of the averaging assumptions manifests itself in the discrepancy
between the real averaging results and the theoretically expected �results,
- but it is not so great as i:o make the averaging method unacceptable.
According to the axioms of averaging, a response to a stimulus, S, is inter-
preted as a constant with zero variability. Noise (that is, a spon-
- taneous EEG) is interpreted as a random stationary uncorr?lated process -
described by the value of its average, n, which, given a sufficient number
of tria]_s, tends toward zero: E'{n) = 0, and by its variance, ct2.
Let us examine how a signal is isolated from noise by means of summation
and averaging under the conditions described above. Assume we have a set
of N EEG segments after a stimulus f(i). What would be the arithmetic
mean of the sum of these segments f(2)? In keeping with the premises pre-
sented above,
1 N ~ N
- ff = N ~ l fk = N ~ ~SIk nik~
k~i p~l
7
FOR OFFICIA;. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200100045-9
FOR OFrICIAL USE ONLY ~
_ where S is signal intensity and n is noise intensity. Hence
N N -
fi = N~ Sik T N~ nik = Si nh
k~l _ p~l
~ and inasmuch as SZ = SZ ~ then fi = SZ + ytti (in view of an axiom of averaging-- ~
the constancy of the response to the stimulus and the additive nature of
siqnal and noise). The mathematical expectation of the average o~ the sum
of EEG segments may be expressed as -
~'(f;)=S,+E(n;) -S,+E(n),
Inasmuch as the mean of the noise is zero, E(f2) = Si. This means that when ~
we approach the limit, given a sufficiently large N, summation should pro-
duce a signal without noise artefacts in the averaged EP for any real n.
Inasmuch as the mean of the noise is presumed to be zero, the sk~are of noise
artefacts depends on its variance:
1 N N
QZ ~I1~ = Nz ~i ~ ~Cik! iP~ - EZ i~
k=1 p=1
- N N
. - ~ ~ {E (S~S;p) E (n;~nip)} - Si.
Since we assume that noise segments n(t) do not correlate with one another,
when k~ p, E(nLknZp) = 0, and so
a s 1 s 1 2 1 Z
~ (fr) = S~ -I- N E (ni) - Si = N a (nr) = N Q (n)�
This means that when we average N EEG segments fol.lowing a stimulus, the
expected mean of the amplitude at each moment in time would be equal to
the amplitude of the reaction to the stimulus (signal), and that the
- variance would be 1/NQ2(n). In other words while the mathematical expecta-
tion of the mean of the signal, SZ, is the same both for a singular EEG
recording following the stimulus and for the sum of N EEG segments, averaging
causes the mathematical expectation of the variance to decline in proportion
to the number of singular reactions averaged. Inasmuch as the mean of the
noise is zero, the ratio of signal to noise, S/n, can be defined as the ratio
of the signal's mathematical expectation to the mathematical expectation of -
the variance of noise. Then if for a singular rec~rding S/n = SZ/6(n),
for an averaged EP derived from N EEG segments, S/n = SZ/~a(n). The latter _
formula shows that the signal/noise ratio increases as ~ in response to -
averaging,
_ 8
FOR OFFICIE,L USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
rOR UFFICIAL OSE ONI.Y
~ However, this entire line of reasoning would be fully valid only on the
condition that all axioms upon which the averaging method is based are valid.
But this is not the case. In particular, the premise that the re is absolutely
no correlation exhibited in noise is wrong, for which reason averaging would
- lead to a lesser increase in the signal/noise ratio than shown above. Were
- we to interpret the EEG as noise with a Gaussian distribution filtered within
a particular frequency band (the extreme variant of this would be a narrow
frequency band with in the algha-rhythm zone), then an increase in the signal/
~ noise ratio proportionate to f would be possible only at the limit, and _
- for practical purposes this increase will always be smaller. Possible in-
fluence of noise correlation upon the EP, the influence of the alpha-rhythm
in particular, was specially studied in a number of works. Ruchkin (416)
used a mathematical model to examine interaction between a signal and noise
_ having the characteristics of an alpha-rhythm, and he came tc the conclu- ~
sion that aperiodic stimulation could reduce noise correlation in EEG seg-
ments being averaged. A mathematical model has also been used to analyze -
the variance of an EP as a function of N in cases where steady noise is
superimposed over the reaction (265). It has been demonstrated that intro-
duction of aperiodicity into stimulation changes the function describing
the relationship between variance and N in the same way that replacement of
noise in a narrow frequency band by noise with a very broad frequency
spectrum does; in tnis case the variance becomes proportional to 1/N.
However, on examining a real EP recording situation, the authors concluded
that the natural irregularity of the alpha-rhythm is sufficient to make the
difficulties associated with delivering aperiodic stimulation outweigh.its
merits. Only in some cases, when special stimulation conditions promote
an increase in the correlation of noise segments--for example when the
signal is delivered at a particular moment.v~zithin the alpha-rhythm, can
introduction of zperiodicity into the signal be found to be suitable. _
- Nor is the notion of the additive nature of signal and noise entirely valid
either: A stimulus alters a spontaneous EEG (148). Another idea at the =
basis of the averaging technique--that a stable relationship exists between
_ siqnal and noise--is also contrary to fact (121,394). This is especially
significant in cases where singular trials to be subjected to the averaging
technique are recorded over such a long period of time tihat tiring, sleepiness,
sleep, and other similar phenomena develop--that is, when the subject's state
changes so much that both the spontaneous EEG and the responses to the
stimulus change.
Incic'.entally the idea of reaction constancy, also at the basis of the
ave-raginq technique, is apparently not entirely correct in principle.(see _
Chapt~r VI). For practical purposes averaged EP's always contain significant
noise artefacts, and change in the signal/noise ratio does not always change
strictly in proportion to E~ipirical research on this problem has pro-
- duced highly variable results. For example Perry and Childe rs (368) present
a figure in their book which shows linear increase in the amplitude of a
visual EF' with growth in N from 120 to 960 in the presence of rhythmical
stimulation at a frequency of four flashes per second. Recording auditory
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICI~'1L IISr ~1NI,Y
EP's, Milner (344) found the am~litude of the EP's to increase in proportion
to N only at frequencies not greater than 1 stimulus in every 7.5 seconds;
beginning at a frequency of 1 stimulus every 3 seconds, further growth in
istimulation frequency caused an increasing change in this dependence, and
at a frequency of 2 stimuli per second the EP amplitude increased in propor-
- tion to -
The question arises as to what is the least number of summed EP's that would
be needed for the averaging technique. A researcher trying to find an
answer to this question perpetually finds himself between Scyla and Charybdis:
An increase in the number of signal deliveries necessary to reduce the pro-
portion of noise in the recorded EP is inevitably associated with an un-
desirable increase in experiment duration. EP characteristics change de-
pending on the state of the subject, and under otherwise equal conditions
_ the probability of this change grows as the research time lengthens. In
this connection when we attempt to correct for variability of EP's associated
with too small a number of summed responses (that is, a large proportion of
, noise), from a certain moment on we witness a concurrent increase in EP
variance clue to fluctuations in the subject's state. Perry and Childers
(368) write in their visual EP analysis handbook that in most cases su~nation
of 50-100 reactions would be enough to obtain stable averaged EP's.
Ellingson et al. (209), who studied variability of visual EP's in a~3ults
and children, note that the variability and form of EP's derived from
summation of 32 responses differ little fram the same EP characteristics
- when 50, 60, and 100 responses are summed. At the same time there are indi-
cations that 500 responses must be sum.med to register early components of
EP's to light flashes (359).
In principle, the number of stimulus repetitions necessary to acquire a
sufficiently distinct EP depends on the ratio of the oscillation amplitude
- of the potential evoked by a single stimulus delivery to the oscillation
amplitude of background activity (in other words on the signal/noise ratio).
Inasmuch as an EP consists as a rule of several~ oscillations of different
amplitudes, the minimum number of stimulus deliveries required for isolation
of an EP depends on the particular component in which the researcher is
interested: The greater the amplitude of the component under analysis, the
fewer stimulus deliveries he could get by with to register an EP. Here are
some simple computations. For example, when 100 responses are summed,
according to the formula S/n = SZ/f6(n) the signal/noise ratio should in-
crease by a factor of 10. In reality this number would be smaller. On
this basis, knowing the approximate amplitude of the components that have
to be registered and the amplitude (variability) of background activity;
we can approximate the needed number of summations.
Inasmuch as the mean of the noise amplitude does nc.t reach zero for practical
purposes, and since it depends on a number o� conditions (for example _
- whether the eyes are open or closed, the difficulty of the assignment, and
so on), Perry and Childers (358) recommend always recording, in parallel
with the EP and as a control, the same average number of EEG segments in
10
FOR OFFICI~. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR (ir'rICIAL USE ONLY
the absence of the stimulus. Th~~s makes it possible to obtain what could be
called the real mean of the noi~~e in the given conditions (technically
speaking, such registration i.s extremely simple--summation and averaging _
are performed in precisely the same way as with stimulus delivery, except that
the stimulus itself is not presented.) Figure 2 shows an example of such
a recording.
~ i z
.~I ~~1~ ~
~ d 9
h
rESO ~BBC
Figure 2. Isolation of a Visual EP (Signal) From an EEG (Noise):
~ 1--EEG together with an EP to a singular stimulus;
, 2--background EEG (stimulus is not presented);
3--summed visual EP to 150 stimulus presentations;
4--control (summation of 100 EEG segments in the -
absence of a stimulus) (368)
~ ~ ~ ~
�~-o~
, ~ �~,a.,~'�,~
~ ~o.no ~'O~"�O~O-dOO -
d ~
Figure 3. Demonstration of the Possibility for Obtaining
~dentical Averaged EP's By Summing Different
Singular Responses: A--Constant reactions;
B--variable latent times; C--variable amplitudes;
I--singular EEG reaction, II--averaged EEG reactions (116)
What can be treated as the criterion for presence of a signal (a response to
- a stimulus) in a comparison of such recordings? Perry and Childers (368)
believe that the main criterion of signal presence is a signal-to-noise ratio
not less than 2. It can be easily determined by visual comparison of EP's
and a corresponding control recording. It would not for practical purposes
be difficult to obtain a ratio of 4:1 and more by increasing the number of
summations or the intensity of the stimulus, or by changing background
_ activity, as well as with the help of frequency filters.
11
- FOR OFFICIti,�_. USE UNLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
To answer the question as to the minimum necessary number of stimulus deli-
veries, the researcher usually relies upon visual analysis of recorded EP's,
attempting to obtain EP's of distinctive shape that are sufficiently stable
with repeated registration. Understandably the same number of sturanations
may be fully sufficient to isolate some components and entirely ir~adequate
in relation to others of lower amplitude (see F'igure 9 below).
The S/n ratio is a good sign of signal isolation itself, but it says nothing
about the degree of correctness, precision. As Regan :~otes (394), it is
enticing but dangerous to interpret the shape of a wave, obtained by
averaging, necessarily as the result of summation of N identical evoked -
potentials in response to singular stimuli (Figure 3).
It is clear from the above that given the availability of a set of EEG
amplitude values following stimulus deliveries (for example, having a tape-
recorded EEG and an analog-digital converter), we could perform the averaging
operation with any sort of computer. EP registration based on the cross-
correlation method may be more convenient to some tasks.(227). Easier -
methods of EP registration can also be employed: algebraic stunmation of
responses on magnetic tape, accumulation of responses on photographic film,
and the method developed in the 1960's by Ertl et al. (see 23,394; Shagass
i974) .
Though not broadly accepted yet, selective averaging is very promising.(372).
It essentially entails the averaging of not all singular responses to
stimuli, but only those which satisfy certain requirements (a previously
established "template"). Selective averaging opens some fundamentally new
perspectives in the use of EP's. As an example after selectively averaging
EP's having a certain component with particular characteristics, we could
see what sort of correlations are revealed with the characteristics of
other components. These correlations mi~ht not be revealed with nonselective
averaging.
- 12
FOR OFFICIi,I, USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OrFICIAL USE ONLY
- CHAPTER II
MONOPOLAR AND BIPOLAR EP RECORDING TECHNIQUES
Division of recording techniques into monopolar and'~bipolar is highly _
conditional, inasmuch as the so-called indifferent contact points are in
fact active, and not neutral, as is theoretically presumed in monopolar
registration. Nevertheless it can be assumed as an approximation that mono-
polar recording provides an impression of absolute changes in potentials in -
the vicinity of the active electrode, while bipolar recording provides in-
formation on changes occurring in the difference between the potentials
at the two points being studied. Bipolar registration may reveal more-local
and smaller changes in potential than monopolar registration. At the same
time the absolute value of the.potential obtained with bipolar registration
is often lower than that obtained with monopolar registration, which means
that a larger number of responses must be summed. Another advantage of
monopolar registration is greater intra- and interindividual stability than
with bipolar recording (166). It is probably in part a product of the fact
that significant changes in EP's accompanying slight changes in electrode
position are more typical of bipolar recording (especially in relation to
early components). According to Benett et al. (98) movement of an electrode
3 cm upward, to the right, or to the left of the occipiLal protuberance did
not cause significant changes in an EP to light, recorded monopolarly.
On the whole, the two techniques should be thought of as supplementary, and
not competitive, since, judging fr~m experiments in which they were used
- together, bipolar recording is found to be more informative in some cases
(359,393), while monopolar recording is found to be more informative in
others (404).
It would be pertinent to cite two studies as a precaution against mistaken _
interpretations. In the first (512), it was discovered that the amplitude
of bipolarly recorded EP'w~ to visual structures presented in half of the
visual field is greater in the right temporal region than in the left,
irrespective of the side stimulated. The authors assessed their results
as evidence of a special role played by the right hemisphere in perception
of spatial information. An invEStigation of monopolarly recorded EP's to
13
FOR OFFICI~;. USE ONLY �
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFrICIAL USE ONLY
the same stimuli (Shagass et al., 1976) did not reveal interhemispheric
differences in amplitude. An analysis of the distribution of the amplitudes
of components recorded from different points on the scalp revealed a larger
gradient in the right hemisphere, which is what was responsible for the
. greater amplitude observed by Vella et al. with bypolar recording. -
The practical criterion used to select the point for the indifferent electrode
is absence of stable EP's at this point, when recorded "bipolarly" using a
series of other points, also tentatively indifferent. Usually the indifferent
electrode is placed on the earlobe, in which case it is recommended that
_ it be located on the side opposite that of the active electrode (245,321).
At the same time there are data showing that the ear.electrode is not in- -
different. When EP's are recorded with an ear electrode paired with another
electrode presumed to be indifferent (usually noncephalic some subjects
exhibit regular changes in potential in response to a light flash (260,314)
and stimuli of other modalities,(463j, In addition to the earlobe and the
mastoid process, the nose or the bridge of the nose is used as the location
_ of the indifferent electrode (Alli.son et al., 1962), which according to
- Stowell (1972) is inappropriate in the case of somatosensory EP's.
Noncephalic location of the indifferent electrode is also employed. A
so-called thoracic indifferent electrode is used to record EEG's.(470).
It was concluded from a special study that a thoracic electrode is preferable
to an ear electrode only for the recording of visual EP's; it does not have
advantages over an ear electrode in the two other modalities (314).
Comparing different locations of the indifferent electrode (earlobe, neck,
- nose, back, and so on), Broughton (112) concluded ~khat it is best to place
the indifferent electrode on the earlobe, though activity of intracerebral
origin and myoqenic potentials may doubtlessly be recorded in this area
following light and sound stimuli.
Considering the above, it would apparently be suitable to use the ear
electrode to record auditory and somatosensory EP's, and a thoracic electrode
for visual EP's. Incidentally we cannot fai] to note that a tremendous
quantity of doubtlessly valuable results have been obtained with the in-
different electrodes in the vicinity of the earlobe or the mastoid process,
and that the location of the indifferent electrode apparently becomes
significant only when it comes down to determining the sour�ce of the regis-
tered oscillation (see Chapter VII) (303).
A standard system for locating "active" electrodes to record EP's similar
to the international 10/20 system does not exist: All we can do is mention
the locations used most often in relation to each r~.odality. Standardization
will probably come along in the future (at least in relation to some parti-
cular set of objectives and situations). (Let us recall that prior to
creation of the international 10/20 system, electrodes used to record EEG's
_ were also arranged differently by different laboratories).
14
- FOR OFFICIti:. USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR uFrICIAL USE ONLY
_ The choice of electrode location depends fir~t of all on the components in
which the researcher is mostly intzrested and on the sort of stimuli employed.
Early EP components (in the first 60-100 msec) are expressed predominantly
in the cortical projection zone corresponding to the stimulus modality.
Auditory EP's, the early components of which are expressed no worse and,
accoxding to some data, even better in the parietal area than in the temporal
area, are an exception. Late EP components (occurring after 100 msec and
more) are registered sufficiently clearly over the entirz scalp. Thus when
early components are to be studied, placing the electrodes in the cortical
projection zones corresponding to the modality of the stimuli employed is
usually recommended (3Z1).
- The points on the scalp at which electrodes are located for optimum regis-
tration of different components differ slightly in concrete studies.
Shagass' recommendation of placing the electrode at the point where the
= component under analysis exhibits its greatest amplitude may be used as
the most general rule (this is the most widespread technique for locating
the electrode) (76). It should be recalled, however, that components of
- similar form and latency in different divisions of the brain may have
different origins, and therefore the registration point at which the
greatest amplitude is recorded within the given time period may not always
be the poi.nt at which the component under analysis is expressed best (if we
define this component as a potential oscillation having a hypothetically
known source). In addition to the YO/20 system (276), an atlas (184) and a
method developed by Remond (397) can be used to determine electrode locations. -
There are no unambiguous, universal recommendations for the choice of both
t~i- or monopolar registration and electrode location, nor can there be such
= recoimnendations. It is important to remember that in a number of cases the -
- electrode locations and the recording technique may turn out to be the
_ :~ecisive factor, and changes that are not revealed at all or are poorly re-
vealed by one recording technique may manifest themselves distinctly with
J the other. When a researcher is unable to conduct preliminary studies to
- resolve these questions in relation to the basic objectives of his own work,
Perry and Childers (368) advise using those points of contact that have been -
employed in other studies performed in the given area. This would at least
make is easier to compare the results. Of cour~~ this does carry the danger
of limiting oneself to an acceptable but in no way necessarily the best
technique.
Some data and recommendations, "retrieved" from concrete studies, on locating
active electrodes in correspondence with stimulation modality are presented
below.
Judging from the works of Benett et al. (98) �small differences in the
location of the active electrode are inconsequential in relation to mono-
polar recording of EP's to light flashes. They recorded EP's to light
flashes at four points: Over the occipital protuberar:ce, 3 cm above it,
15
FOR OFFICIA;, USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
huh c~~�r~ic ach LP oE a certain initial set may be expres~ed by a cer_tain number of
main comE~orients 3IiCl we.igY~~ fu:ld be impossible due i:o substantial independence of
individua]. components withiil any gr.oup. It is no accident that such descrip-
tions o~ L�'P's iri analyses of resu].ts have been c~ncc>untcred with incre~~ingly
e7real:~.r rar ity in rer:ent ye~~rs, and the indicaL-ion of separate components is
bec:oming common; correspondinnly, potential generai:ors are also being sought
iri relation t~ eacti comporzent taken individua].ly.
- 51
FOk OFFICIAL USE ON1,Y -
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
Attempts at determining the location of a generator producing oscillations
recorded from the scalp have boiled down in recent years to theoretical de-
- velopmen~ of a model of the brain as an isotropic three-dimensional conductor,
and to determination of a dipole or several possible dipc~~-.; that should pro-
duce the same potential distribution on the scalp as t':; ~ne actu~.lly observed
(or one close to it). After this the theo.r~tically calculated lacations of
the dipoles are compared with the neuroanatomical _nformation on the brain,
and then the locations that coincide with a physiologically adequate source
are assumed to be proven--that is, it is presumed that the model has "interpre-
tive value.�
~ An example of this approach can be found in research by Smith et al. (463).
- These authors recorded EP's to clicks, light flashes, and electrocutaneous
stimulation in 15 persons. The indifferer:i: electrode was extracranial. EP's
_ were recorded from the scalp at points C2, C;g, and Tg and, in additon, with ~
_ a nasopharyngeal electrode. EP's to stimuli of all modalities produced with.in
the first 80-100 msec following stimulus delivery rad similar shape in relation
to all electrodes; later components (occurring after 120 msec) were recerded
in the vicinity of the nasopharyngeal electrode with polarity opposite to
~ that of the rest of the electrodes. Using a mathematical model in which the
brain was *~iewed as a homogeneous conducting medium, the authors studied the
possible locations of a dipole serving as tne source of oscillations recorded
simultaneously by the employed electrodes in successive moments 20 msec apart.
According to this model the source for the early components of the visual EP's
(up to 120-140 msec) is located close to the surface of the cortex; 160 msec
_ after stimulus delivery the source shifts deep into the brain and closer to the
_ midline, moving after 240 msec somewhat upward, vertically along the midline.
Similar results were obtained with the other modalities. -
In the opinion of the authors, according to this model mesidl structures of the -
mesencephalon, the mesial temporal cortex, the hippocampus, and the lingual
gyrus can make an equal "claim" upon the role of the source of late oscillations.
The results of this work are not at all absolute proof that a dipole located
in the deeper structures of the brain is the source of the late components.
The very hypothesis that the source is a localized dipole rather than many
individual sources or dipole layers may be incorrect; viewing the brain as a
homogeneous medium also remains debatablP. However, if the initial assumptions
are valid, then the results indicate.the possibility that oscillations arising
in the brain's deeper divisions may be reflected on the scalp due to physical
conduction, a possibility which has been repudiated by some data (160).
The validity of such a possibility is demonstrated by Jewett et al. (285,286),
_ who registered,on the scalp, potential oscillations arising in the center of
the cochlea. Incidentally these authors suggested potentials of the "near"
and "far" fields (near and far relative to the recording points) as difference
criteria to be applied to scalp recordings. Near field potentials are typi-
fied by significant differences in shape (that is, amplitude and polarity)
associated with insignificant changes in the position of the recording elec-
trodes. Absence of significant differences in response to small changes in ~
recording points is typical of far field potentials. Thus b_y recording EP's
52
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR ,~FICIAL USE ONLY
with a set of electrodes located close together, we can localize the source
of a near field, while far field potentials would be the s~me beneath all
electrodes. At the same time a single electrode may record th~ activity of
different far field sources.
It is clear from this how complex and undefined the mutual relationships are
between potential oscillations on the scalp, in adjacent and remote divisions
of the cortex, and i~ profound brain structures. The situation becomes e~ten
more complex when we consider that construction of simple models of dipoles
in a homogeneous medium--an analytical approach that is widespread today ancl
presents the appearance of being successful--is, in Regan's opinion (394),
stupefying and dangerous. The initial success enjoyed with simple models of
single dipoles does not provide the grounds for presuming, Rsg~n emphasizes,
_ that any other model might not fit the data just as well. S:;ch another model, -
one ~hat is mathematically more complex, may contain a different equivalent
source, the neurophysiological interpret~ition of which would differ significant-
ly from the newaophysiological interpretaiion of a simple dipole model. If
an attempt is not made to test out more than ju5t one model, the mathematically
~ simplest one, we will never be sure that the obta~ined data do not also
correspond to so,.~e other hypotheses. Moreover, although components are usually
interpreted in the analysis and study of EP neurogenesis ~as certain structual
units relative to which generating structures and functional associations are
sought, it is not at all obvious a pzYior'i that a given component reflects de-
velopment of some particular bioelectric process in nerve elements adjacent
to the electrode, rather than a pause between two successive processes in the
same structures, or the algebraic sum of changes in potential occurring in a
number of structures different distances away from �~he recording points. This
possibility should be kept in mind, though for pz�ctical purposes the commonly
accepted interpretation has turned out to be ful~.~ r,z~ductive, and we are
able to associate some components with specific generators rather clearly and
unambiguously.
Specific current ideas about the generators of indivdual EP components are
presented in our examination of the neurogenesis of EP's to stimuli of differ-
ent modalities. It is impossible to categorically define the source of a _
number of components today. It would pay to emphasize here that even when the _
neurogen qsis of a component is known,caution should be exercised when inter-
~ preting changes in this component as reflections of changes in its hypothesized
generator. A change i.n a component may be a manifestation of changes in
neighboring components.
The Relationship of EP Oscillatyons to the Activity of 5ingle Nerve Elements
Animal research, which usually deals with EP's recorded directly from within
the cortex and not from the scalp, is the main source of our ideas on this
issue. As neurophysiology developed, the tendency to interpret slow oscilla-
- ticns as the integral result of spike (puls~ed) activity of neurons was super- -
ceded by notions of the dominant role played in EP generation by sufficiently
synchronous slow potential oscillations of the soma or dentrites, recorded with
macroelectrodes on the surface of the cortex or in subcortical formations
(2,211,212,388; Purpura, Cohen, 1962). (The idea that EP's reflect the nature
53
FOR OFFICIAL USE ONLY
I
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
F~R OFFICTAL USE ONLY
of impulses travelir~g along afferent ribers into projection zones is still
accepted today only in relation to the earliest EP components recorded from
the scalp.. ) However, although the role o.f slow, gradual potential changes
accompanying EP generation (and other slow activity in the cortex) has been
recognized, concrete experimental data on the nature of tnz relationship be-
tween the examined phenomena are far from ambiguous.
As we know~, gradual potential changes of two types may aris e in nerve ce11s;
sfimulatory postsynapt:ic potentials ( S.PSP) , manifested by depolarization or
negativeness, and iilhibitory postsynaptic potentials (IPSP)--hyperpolarization,
positiveness. Experimental studies describe different re lationships between
potential changes in neurons on one hand and the polarity and nature of slow
oscillations recorded with a macroelectrode from the structure under examina-
tion on the other. Among them, in addition to the frequently repeated and
replicable results, we can also find contradictions. Perhaps the reason for ~
this lies in differences in the experimental conditions, and a literature
analysis does not always clarify the differ~nces in these conditionsfully.
A hypothesis that had once been rather popular (167) was that alternation of
negative and positive EP phases on the scalp reflects alte rnation of SPSP's
and IPSP's in cortical neurons; it now requires additional study, since signi-
ficant diversity in simultaneous responses by different cortical neurons to
the same stimulus, the possibility of simultaneous existen ce of SPSP's in
some nEUrons and IPSP�s in others, and the corresponding incorrectness of re-
lating neuron events to EP components unambiguously have be en discovered (168).
In general, we have not as yet been able to reveal a relat ionship between slow
oscillations on the surface of the cortex on one hand and pulsed discharges or
slow changes in the potentials of single neurons on the oth er, though in some -
conditiona certain mutual relationships depending on what sup~rficial slow =
oscillations were considered (for example the EEG or EP's) , on the state of
the animal (anesthesia, sleep, alertness), and so on have been revealed. Data
aze appearing indicating possible significant participation of n~uroglia in
the genesis of superficial slow oscillations (135,158,247).
J Summarizing, we can say that for 4he moment,unfortunately, there is no "bridge"
- between EP's and the activity of single neurons. Existing points of view on
the possibility of using EP's to analyze cerebral mechanisms vary from almost
total skepticism to the opinion that EP's may serve as a"window" for re-
search on the central nervous system just as successfully a s recording the
activity of single nerve elements. At the present level o f knowledge, it is
impossible to establish the advantages of one method over the other (394),
54
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
~
FO.T OFFICIAL USE ONLY
CHAPTER VI
VARIABILITY OF EP'S
Every experimenter beginning his research on ~P�s encounters both inspi-ringly
comparable, stable EP�s and discouragingly different EP's in successive trials
i:nder what appear to be identical conditions experienced by the same subject,
_ and all the more so by different subjects.(figures 4 and 8). Interindividual
differences are especially great, but even interindividual differences can
also be highly gronounced, despite stability of the general pattern and latent
period of the responses.
~ A ~
y ~
z Ts
. ~ ~
d"~ V1 G~~
j~,~' -
~ V
sco~s~~~
Figure 8. Example of the Variability of an EP to a Light Flash ix~
Successive Recordings ~rom the Same Individual (318): A--
EP's summed from 25 responses: 1-6--successive averagings;
B--EP summed from 150 responses L
The beginning researcher would naturally want to have some sort of reference
points from which to conclude that the EP's he is recording, and their varia-
bility, do not differ too much from those encountered in most previous studies
- employing identical or similar recording conditions.
There has not been much systematic research on normative data in this aspect.
The existing data on variability (that is, the scatter of latent periods and
amplitudes) are presented in our examination of the typical znorphology of EP's
~ associated with each modality; in this section we will examine some general
issues.
55
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
Sources of Intraindividual Variability. Variability of Noise and "Intrinsic"
Variability of EP's. Variability Assessment Techniques
Inasmuch as ~the number of single reactions usually employed in averaging is
such that a certain fraction of noise always remains in the EP, variabili~y
caused by noise is unavoidable. The question lies in whether or not noise
artefacts are all that cause variability in EP's. The problem of dividing EP -
variability into reaction variability and noise variability (that is, back-
ground variability of an EEG recorded in the absence of reactions) is theoreti-
cally complex, and it is not fully resolvable in the same way that we cannot
answar the question as to whether an EP is a reflection of reorganization (or
sychronization) of neuron activity in response to a stimulus, or a reflection
of the "addition" of new "events" to constantly present spontaneous activity.
However, for practical purposes it is entirely possible to reveal the relation-
ship existing between the role of noise and the "intrinsic" variability of an
EP, for example by comparing the variability of the background EEG and the
variability of the EP. The more they cor�relate with one another, the better
are the grounds for assuming that the noise artefacts play a significant roie
in EP variability.
A large number of studies of this sort have been conducted (143,148,267,372,425;
Halliday, 1972; etc.). Some authors concluded that the variability of noise
has the domina~:t role, while others suggested variability of the responses
themselves to be the main source of variability. Analyzing the appropriate
data on sources of variability, Childers et al. (1972) noted that the role
of the variability of noise is emphasized predominantly in the studies making
- use of rhythmical stimulation, while when aperiodic stimulation is involved,
the variability of reactions is usually viewed as the main source of the varia- -
bility. This is to be expected on the basis of data indicating that aperiodic
sti.mulation eliminates some noise associated with the alpha-rhythm (265), and
that a significant proportion of the variability of visual EP's may stem from
insufficiently reduced alpha-activity at the time of averaging (500). On the
whole, the larger proportion of the data available today show that although
noise does make its "contribution" to the variability of EP's, the variability
of the reactions themselves is a significant source.
_ W'hat are the causes of intraindividual variability of EP's? First there are
the variations in the subject's state: changes in sensitivity, changes in
emotional state, fluctuations in attention, habituation, the action of the
orientation reaction, and so on. Changes in sensory input which are caused
by uncontrollable eye movements, changes in pupil size, and contractions of
ear muscles, and which are unrecognized by the subject and the experimenter
may be a source of variability (Tanaka et al., 1974). Stimulation of differ-~
ent divisions of the retina may be a source of variability associated with -
_ point stimulation and insufficient fixation of gaze (189,203,508; etc.).
The steps taken to "fight" variability become clear from the above discussion:
keeping vigilance at the same level, keeping the sensory input constant, dis-
carding responses to the first few stimuli (or at least to the first in a
series) before averaging, excluding EEG segments with distinct artefacts from
averaging, and so on. Mathematical assessment of the variability of signals
and noise (EP's and background) would be desirabl.a in all studies, but itis
56
FOR OFFICIAI~ USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOF OFFi~IAL USE ONLY
absolutely mandatory in cases where the causes of EP changes are unclear or '
contradicted by the bulk of existing data. Incidently there is a simple auto-
matic technique assessing EP stability (327). It is based on comparing two
halves of an EP, representing EP's averaged separately in relation to all odd
and all even stimulus deliveries.
A rather simple technique can be used to obtain visual and numerical character-
_ istics for variability of noise; this technique was suggested by Perry and
Childers (368) as a mandatory control over each EP: A number of EEG seg-
ments recorded in the absence of a stimulus equal to the number~nf EP's are
averaged in parallel. This method is being employed quite ~Luitfully,though
summation of EEG segments recorded in the absence of a stimulus may give wrong in-
formation on noise artefacts in an EP, inasmuch as a stimulus not only elicits
an electric reaction but also alters background activity. Noise variability
may be assessed with the help of a ratio method consisting of successive
_ addition and subtraction of EEG segments associated with the stimulus,
followed by their division by the total number of additions and subtractions
(422,433). The method of so--called cumulative averaging (121) may provide
information on the variability of single EP's in the course of averaging, and
on the validity of comparing averaged; responses. Incidentally an analysis
y performed w.ith the a~sistance of this method distinctly demonstrated signi-
ficant periodic oscillations of EP amplitude not corresponding to the premise
that the probability distribution of amplitudes at the basis of the averaging
process is constant. The research results demonstrated the danger of mistaking
EP differences revealed through different averagings and associated with
"spontaneous" amplitude oscillations for differences directly caused by
experimental manipulations.
A method has been developed for determinging the temporal intervals of an EP
= within which nonhomogeneity is observed. This method reveals only periodically
arising segments of nonhomogeneity: Were we to "mix" si*~gle responses from
two different EP's at random, the method would not reveal nonhomogeneity. A
technique has also been developed for separate averaging of different groups -
- of EP's, to be applied if the sequence of EP's to be averaged turns out to be
nonhomogeneous (417).
We are interested in assessing variability not only because high variability
must serve as a warning to the researcher (it may be the result of extra-
cerebral noise, abrupt variations in the subject's state, artefacts, and so on),
but also because the variability of the background governs the researcher's
treatment of potential oscillations in the EP as components: If their ampli-
tude is less than background variability (which is often the case in regard to
early components), the number of responses added together would have to be
_ increased. A comparative assessment of any two EP's would require a knowledge
of their variability; otherwise such an assessment would usually be meaningless
(394).
Finally, we must have an assessment of variability because, as numerous studies
have showt~,variability is a significant EP characteristic, and not just noise.
The importance of considering variahility as a special indicator of an EP -
was discovered for the first ti.me in connection with the fact that small _
doses of LSD-25 dramatically reduce the variability of visual EP's while having ~
- little influence upon their amplitude (302). -
FOR OFFICIAL USE ONLY -
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIt1.L L'SE ONZX
Significant changes in variability were later deinonstrated in connection with
_ different effects upon and features of the nervous system and behaviox. In
particular, inasmuch as the variability of EP's recorded from ch~.ldren is
obviously superior to that of adults, the associa~ion variabili~y has with
_ the degree of development or maturation of the nervous system has been studied
(128,209,210; Callaway, 1973). Research has also been conducted on the rela-
tionship of variability to intelligence (129), patholoyy (275), general nervous
system properties (54), and so on. The overwhelming majority of the s~udies
revealed correlation between variability charac~eristics and the indicators
under analy~is. Analyzing the results of these works (and when performing one's -
own research), it is important to recognize that different indicators are used
to measure variability. Calla~vay and Halliday (130) provide a detailed compara-
tive analysis of different variabil.ity indicators. They demonstrated in parti-
cular a relationship, varying in degree and sign, between five variability
indicators and EP amplitude. Moreover their analysis showed that different
variability factors (variations in latent periods and amplitudes of responses,
the proportion of noise present, and so on) should affect different indicators
- differently. A researcher desiring to use variabilii.y as an EP characteristi.c
should acquaint himself rather attentively with this work, since it is one of -
the most meticulous studies in this area.
Interindividual Variability
An interesting study by Childers et al. (143) sought the causes of intezindivi- -
dual variability. The authors based themselves on the idea that interindividu-
al.variability has two sources: one in common with intraindividual variability,
an$ the other governed specifically by interindividual differences in reactions
to a stimulus. The former source is the presence of numerous factors examined
_ above, identical for all subjects and manifesting themselves randomly in
the reco~ded EP at each given moment. It may be asswned for example that '
differences in the physiologicalstate of different subjects at the time of
EP registration are a significant cause of interindividual differences. How-
_ ever, if the same factor lies at the hasis of inter- and intraindividual differ-
ences, then averaging of an identical quantity {sufficiently large) of single
- EP's for the same subject and the same quantity of single EP's recorded from
different subjects should produce identical or very similar EP's. The results
showed that this does not happen: EP's obtained by summing 50 singular success-
ive responses by the same subject differed very significantly from EP's ob-
tained by averaging singular reactions of 50 different subjects. The ampli-
tude of individual EP's was as a rule 1.5 times greater than that of the group
EP. This means that interindividual variabiliLy is greater than intraindividu-
_ al variability, and that is has "its own" sources, apart from those generat-
ing intraindividual variability.
In order to eli.minate nonfunctional interindividuaJ. differences and amplitudes -
caused by thickness of bones or soft tissues and ot~:er like factoxs, Childers
et al. compared EP variability following normalization. Normalized intraindi-
vidual EP's were also found to be greater in magnitude than interindividual EP's. `
We are still unsure of what "specific" sources of inl:erindividual variability '
there are. One of them might be individual feature, associated with the
relationship between the surface of the brain and the surfac:e of the head: What ~
appears on the basis of external criteria to be the same point on the su-rface of
the heads of different people may be associated diff.ere nt]..y with the same region
of the brain. FOR OFFICIAL USE ONLY ~
58
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR ~~FICIAL USE ONLY
Evidence of this possibility can be found in the fact that interindividual
variability of visual EP's decreases withfi~l~ree-dimensional averaging (107),and ~
in simultaneous recording, by Childers et al., with numerous scalp elec-
tr�odes followed by three dimensional representati~yn of spatial-temporal EP's,
which reduced interindividual variability. Given the introduction of appro-
priate technical devices into practical research, this technique will appar-
ently aid research;in which interindividual variability is an interference. This
includes to some extent all research requiring group comparison of changes
in EP's occurring in response to some particular influences. Differences
that are distinctly evident in each subject may be smoothed ou~ in group
analysis, inasmuch as they fall within barely different time intervals (si.mi-
larly as with reduction of the amplitude of a component due to the variabili-
ty of its latent periods).
An example of such a situation can be found on comparing EP's in response
to monocular stimulation of healthy individuals and patients with one injured
eye. Comparison of EP's in response to stimulation of the stricken and
the healthy eye did not reveal a statistically significant difference in the
_ group as an average, while comparison of sets of EP's recorded in response
to stimulation o� the healthy and stricken eye of the same subject revealed,
in all cases, a lower amplitude for the stricken eye (368).
2t would obviously be sensible to select subjects with sufficiently similar ~
EP's for group studies (or, what would be the same thing, to exclude, from
the analysis, EP's that differed dramatically from those of the main group),
or to isolate basic "types"--subgroups of curves from the sample under analysis,
to perform intergroup statistical analysis on the basis of these subgroups,
and then to make an attempt to identify the components in the different sub-
groups, after which the results are analyzed for the group as a whole.
Inasmuch as some fraction of interindividual variability stems from differences
in the background EEG, it would possibly make sense to select subjects with
similar background activity, or to try to present stimuli on an identical -
background.
- The sources of interindividual differences continue to be an object of re-
search. Data showing the correZations between some EP parameters and intelli-
gence (see Chapter XII) and the genotppic dependence of EP shape (40,119) -
encourage a search for correlates, within individual features of EP's, of
functional-morphalogical features of brain systems important to behavior. -
When repeated recording is involved, significant interindividual variability
accompanied simultaneously by EP stability serves to some authors as the _
groundsfor suggesting individual specificity of EP's (97). Analogies are _
even made in this case with the individual specificity of fingerprints. _
Nevertheless it should be noted in this connection that individual specifici-
ty (in the meaning in which this concept is applied to fingerprints) had
never been demonstrated yet in the literature available to us. In order to ,
confirm the existence of individual specificity of EP's, we would have to
demonstrate the presence of a set of characteristics in a combination that
never repeats itself in any other members of a population sample of suffi-
ciently representative size.
59
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
~
FOR OFFICIAT~ USF. O~TI,Y
'~n ~he wh~le, EP varial~ility is concur-rently ~n~ "scou~~ge" o.~ research,
lzmiting the possibilities of the method's a~plicztion, and ~ significant
EP characteristic, associated in particular with the maturation level (or
state) of the nervous system (25,127,128},
60
FOR OFFICIAL USE ONLX
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR ~3rFtCtAL USE ONLY
MORPHOLOGY AND NEUROGENESIS OF EP'S TO STIMULI OF DIFFERENT MODALIT2ES
Because of the absence of standardization in EP recording procedures
(which is fully understandable and unavoidable in the given stage), EP
"standards" for what is "normal" do not exist. At the same time an in-
- creasingly greater amount of research is being conducted in this direction,
and typical EP characteristics may be isolated from the published data for
a certain set of conditions today. It would appear suitable to create a
set of standards for the types of stimuli most frequently encountered in
- laboratory practice as a beginning "guidebook" for work with EP's.
- Unfortunately the results of a literature analysis reflect the evolved
stereotypes of research (arbitrary on occasion) and fail to fall within a
well-ordered system: There are more data pertaining to some conditions or
modalities, and less pertaining to others, and the main sample of
"representative" results that are to play the role of "standard" would
have to be taken from the available set of results intuitively. In doing
so, we would of course consider the strictness ~nd "purity" of the selected
studies (registration of eye movements, control o� muscular artef.acts,
sample volume, number of summations, presence of averaged background
segments as a control, and so on), as well as the number of different works
offering similar results in relation to EP morphology. The results of
such an analysis are presented in the following chapters, which descrxbe
EP's in relation to different modalities of stimulation. We emphasize the
great complexity of selecting the "standards" so that the reader who uses
them would not abandon further analysis of EP morphology, a~zd believe the
presented data to be absolute. 5light discrepancies between EP's obtained
in a concrete st;ldy and thPir "standard" are fully possible, even in similar
recording conditions, but sign~ficant differences should signal the re-
searcher that something is wrong.
61
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
CHAPTER VII
AUDITORY EP' S
- Morphology of EP's to Clicks and the Beginning of a Tone
It was namely in response to sound that the first EP's were recorded and
described: the K-complex arising during sleep (322), and reactions arising -
in the EEG during wakefulness and in light drowsiness, in response to sound
(178)� The potential oscillations Davis described, which fall within a
period 100-120 msec following an acoustic stimulus, were named the "vertex
potential," since they exhibited their greatest amplitude when recorded in
the vicinity of the crown. They represent that part of the EP to sound in
which the amplitude is most pronounced. As the recording equipment
, developed, earlier auditory EP components were recorded as well: first
components arising in the 20-80 msec period following stimulus delivery
(237), and later the earliest component, arising in the first 10 msec
(286) .
A detailed work published in 1974 offers results that may be used as a
- "standard" for EP's to a short tone or click (375). The research was con-
ducted on 12 healthy adults. An EP from which 15 components could be
distinctly isolated were recorded from the vertex of each of them (the `
indifferent electrode was positioned on the mastoid process) in response to
_ a click with a loudness of 60 db and a stimulus delivery rate of 1 per
second; 1,024 responses were summated. To isolate the different components,
the EEG tape recording, which bore stimulus marks, was processes several
times with different periods of analysis. The recorded EP's are shown in
Figure 9A, B. The letter designations of auditory EP components used in
the subsequent discussion correspond to the designations of Picton et al.
According to the data of different authors, the similarity exhibited by EP
characteristics (Table 1) attests to the validity of using the data of
Picton et al. as a standard for an EP to tones or clicks recorded in the
vicinity of the crown. -
Goff et al. (1976) studied the characteristics of EP's to a click recorded
_ from 24 points on the scalp in the 10/20 system (Table 2, figures 11, 12).
62
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOF ~FFIC;IAL USE ONLY
A r:
~
Yll
I II YI
~ I~-z ~
8 P
~ ~
fl, Smse
06
Oa .
Oo
~o Io-~ uv _
~a a -
, , ~ ~o
�s � ~s
f, 0 OQ p~ .
J 4s
qz
~
- ~f q2 YI ~o
_ jZU~ Qs ~ m
1 ta p~ n, ~
~l ~o ~ ~r
O,S sec 5~~0 ZO SO !0 ZO SO f00200 ~f00 lOOG
Figur~ 9. Typical Auditory EP Recorded in Response to a 60 db
Click Delivered to One Ear With a Frequency of 1 per -
Second: A--R,eal EP of one subject, summated from 1,024
responses. Monopolar registration with electrodes in -
the vicinity of the crown and on the mastoid process;
B--schematic EP to the same stimulus delivered to a
group of eight persons, presented in logarithmic scale
(375) -
Figure 11 illustrates amplitude measurement and creation of isopotential
maps for individual EP components presented in figures 12, 14, and 17,
using an imaginary positive component with its amplitude maximum in the -
left parietal region. Examples of ineasuring amplitude from a base line to
the peaks of components recorded with different points of contact are
presented. At point C the component being measured is absent, it reveals ~
itself at point T6 butZits amplitude, measured frc�n the base line, is zero,
its amplitude is negative at point C4, and at point 02 this component mani-
fests itself as a small oscillation in another component. The amplitudes _
determined for all of these cases are shown. r^i:ne~.-stippling represents the
area in which the component's amplitude attains not less than 75 percent of
its maximum value, course stippling represents the region in which
amplitudes are 50-75 percent of maximum, and the region of maximum amplitude
is shaded. The same system of symbols is used in figures 12, 15, and 18.
Diagonal lines drawn through points of contact in these figures represent all
63
~ FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007/02148: CIA-RDP82-00850R040240100045-9
FOR OFFICIAL USE ONLY
- Table 1. Latent Periods of the Components of EP's to a Click or a Short
Tone, Frcm the Data of Different Authors, Msea (From Picton et al.,
1973)
\ ~ KOMROAPBfH
( 2 ~ Cax~ae pafwxe ~~~p, ro~
1 I II I III , IV I V I VI
!.5 2,~ 3,8 5,0 5,8 7,4 jPicton~et a1.,~1973
1,5 2,6 3,5 4,3 5,1 6,5 Jewett, Wilkinson, -
!97!
(Vb'1) 1,5 (~V2j ?,5 ,(~V3) 3,5 (W4) 5,0 (N5) 6,0 Lev, Sohmer, 1972
(~4 ) Cpe,�Hxe
Oo 17o Oa I78 On
8,9 12 !6 25 36 ~Picton et al., 1973
(1)13 (2) 23 (3) 28 (4) 39 Ruhm et al., 1967
13 22 32 45 ~iendel ef Golds- -
tein, i~J69�
i0 43 27 35 Goff et al., 1969
_ ~ ~ . . ~ ~ ~ ( 5 } ,I1~NxHaTarexr~te
6
l
~1 Ol I7s Os
_ 50 83 !61 290 Picton et al., i973
50 !00 175 300 Davis, Zerlin, 1966
_ 90 175 Rapin et al., 19~~
(II3a) (03s) (04a) (175a) Goff et ai., 1969
~ 8i 100 ~ 475 � -
Key :
1. Components 5. Long latent period
2. Earliest 6, p
3. Author, year N
4. Middle
points of contact at which the given component attains 90 percent of its
maximum amplitude in at least one subject of a group.
-
Thus the number of diagonal lines in each region indicates the number of
subjects for whom the amplitude of the given component attained 90 percent
of maximum in this region. The maps were created for e�ch subject and each
component. Group maps were created in similar fashion. The number of
subjects from whom EP's were zecorded to compile the particular isopotential
- map is shown in parentheses.
Figure 12 shows the topography of an EP tc a click. Components N75 and N95 -
are so variable aznong different individuals that the median ~litude did not
exceed 50 percent of maximum at any of the points of contact; thus it was
. found to be impossible to create an isopotential map for these components
64
FO1; OFFICIAL USE ONLY
~
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOF ~FFICIAL USE ONLY
Table 2. Auditory EP Components (=rom (246))
- ~1~ Jlare~(A ne-I ~3~ Maxc~myn~ e~ennirryAa, wKR ilpeAnanarar (6~
T(oYnoxeer I npoecxoHC-
paoR. wceK uetuiaxat4~l ~5~ Pxa6Doc Aee+ce
(7)i110 9,8�i,i -}-4,4 ~-0,5 -}-5,5 H -
~g)Oi5 15,8�2,2 -i,8 -0,8 -7,5 M _
I725 24,8�2,6 -}-3,0 -}-~,9 -{-5,7 H
- 030 29,0�2,5 -i,0 -{-0,1 -2,5 M
1135 32,9�2,7 -~'~,8 -}-2,0 -}-16,2 M
040 , 40,5�4,8 -3,5 -0,8 -8,5 M-}-H
Ii5(D, 51,2�4,7 -~-6,1 -}-1,0 -I-~4,2 H
060, 59,7�6,3 -3,2 -0,7 -7,7 1K-{'H
r770 69,r�5,3 -F-4+4 -}-~,2 ~-13,5 H
07~ 75,5�7,5 -2,2 -0,7 -8,4 M
II90 9i,2�7,8 -~-10,7 -{-i.4 -}-52,0 M
095 95,4�8,8 -i , 2 -{-0,5 -4,8 H
IIi10 i12�22 -}-13,0 -}-0,7 -~48,6 M
IIii5 114�7;7 -}-2,5 -}-0,5 -~-6,9 H `
0115 115+2! -7,1 -3,9 -9,$ H ~
~ Oi45 i46+16 -0,8 -O,i -2,5 H
IIi80 178�i6 -}-16,7 -{-4i,9 -}-24,4 H _
0230 230�18 -2,4 -{-1,5 -fi~8 H
I1270 266�23 , 8 -}-0, 8 -}-11 ~0 H _
0300 296+34 -5,9 -3,B -i9,2 H
I1340 338�30 -}-2 , 9 -1, 7 -{-5, 2 H
~4~ 394+26 -7 , 3 -3, 6 -9, 7 H _
Symbols: H--Neurogenic origin; M--myogenic origin.
Key:
1. Component 5. Scatter
2. Latent period, msec 6. Proposed origin
3. Maximum amplitude uv 7. P
4. Median 8. N
(Gn~f et al., 1976). The latent periods and variability of the components,
shown in Table 2, we~e determined f~n the values of the giv~n component at all
- recording points, while amglitude was determined only at the point where
it exhibited its greatest value.
Mention should be made of the work by Wolfpaw and Penry (53), who believe
- that the difference they revealed between EP's to a click recorded at the
vertex and in the temporal region is the result of addition of a group of
- "local" oscillations generated in tne temporal cortex to the oscillations
generated in broader regions of the cortex and reflected identically at the
7 vertex and in the temporal region. Subtracting the EP recorded at the
vertex from the EP recorded in the temporal region, they obtained a positive-
negative complex with latent periods of 105-110 and 150-160 msec
respectively, which they interpret as the temporalcomponent of the
auditory EP.
65
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
A ~ Z s B
BO ,11, ~ _
v~
70 mse
y000
60
sQ /000
y0 2S0
30
ZO b'J -
f0 ~ !6 ~
- � a-y a-y 2-o a-s ~-o j?-o,
u~ ~ 1 I.p-~.
~1 S JO S00 !2 S SO S00 "
_ ' msec ' m~:c
Figure 10. Effect of Some Stimulus Parameters on Audit~ry E? -
Components: A--Click Intensity: 1--Averaged from -
- 4,096 responses with an interstimulus interval of 50 -
msec; 2--averaged from 2,048 responses with a 200 msec
interval; 3--averaged from 256 responses with an inter-
stimulus int2rval of 2 seconds; B--Delivery rate: The
number of summated responses is (from the top down)
1,024, 1,024, 2,048, 4,096, and 8,192 (375).
Da.ta preseiited in (300) may be used to describe the approximate range of
changes in the EP in response to change in the intensity and duration of
stimuli, or the interstimulus intervals (Figure 10).
_ It should be noted that the analog filters used in most recording systems ~
cause a shift in the latent periods, which may attain, in relation to
early components for example, 13 msec and more (308).
Differences in the amplitude of EP's to tones of equal loudness may result,
according to data from different studies, from differences in the rise times
(484) .
Presentation of a tone of sufficient duration causes arisal of an EP to
stimulus cessation--the so-called off-reaction (157,296,358,434). EP's to
stimulus cessation have a shape identical to EP's to stimulus onset,
differing only by having a lower amplitude (by a factor of 2-3), and they are -
- not pronounced in a11 people. As an example EP's to stimulus cessation were
observed among 20 percent of subjects exposed to a 1,000 Hz, 50 db tone
- with a 1 second duration, and among 50 gercent of subjects exposed to a tone
of the same frequency with a loudness of 75 db and a duration of 1.5
seconds (157).
66 -
FOR C~'FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOF JFFICIAL USE ONLY
-J -I r2, 4, 3J
- C~
cy -------r - - -
-z, ~ v
~ + ~ ~s
- 'J
J * 9, /J, B,i
i + /0, /Z, /00
___L
P - * 6, 3 Ei
i t0~ Z,
~~+6~ 6 6T
n, ~
o *s, T se
Z
p -----�t +y, 6, SO
~ , -
~
~
'~_L_~
/ 1 /
1
?f
f, ~ Fj FY S .
1
- / 1 l
A
A~ ~ - lJ 0 - - �
- .
~ .
��.7.,.,_,
,"t':~'~~~.
, .
0 `
~ Figure 11.. Analysis of the Topography of EP Comp~nent Amplitudes:
The initial amplitudes (uv), amplitudes after trans-
formation (uv), and amplitudes expressed as percentages
in relation to maximum amplitude, adopted as 100 per-
cent, are shown in that order to the right of the
curves. See text for further explanation.
J~ somewhat shorter latent period for component N~1 in the EP to stimulus
cessation than that observed for the EP to stimulus onset was noted in a '
number of works. The amplitude of Nl-P2 in the EP to stimulus sensation
increases linearly with an increase in inten~ity, and the.latent period of
N1 decreases (434). It is interesting that the amplitude of the EP to
tone onset is not always areater than that to tone cessation. Thus when tones
of 500 msec dureation were presented with a 2,500 msec interval, the EP's to
tone onset were about three times greater than EP's to tone aessation but
when tones of a duration of 2,500 msec were presented with a 500 msec
interval, ~;~e responses to the start of the tone and its end were
identical (371).
67
FOR OFFICIAL USE ONLY -
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
PIO � N15 P25 N30
Ilzl ~.)Q"~'~. t12) t!z) ~--f~-.~ (12) -
~ l~( ;ll( � -
)'c. ~ P~ ~ ~ ~ � ~ ~ ~
_
~ ~o o _
0
P35 N40 , P50 ~ N60 ~
tl2l (12) (12) t7)
_ ~ ~ ~ o ~ ~ o.
~ ~ ~ ~ ~ ~ ~ ~ l" ~ � ~ ~ ~
,
~ . .
~
� ~ ~ a
P70 N75 P90 N95
(8) ~o-~ (10) (II) . t5) ~
_ ~ ~ � ~ ~ � , o � o
~
~
~ x ~ o ~ o o ~ ~ ~ ~ ~ / ~ ~ ~ .
0 0 0 0
- Figure 12. Topography of the Components of an EP to a Click:
Explanation in text ( see pages 62--64) _
An increase in the preparatory interval, a der,rease in the mean tone
delivery rate, or uncertainty of delivery would increase the EP to both
stimulus onset and cessation. This leads to the notion that amplitude is
associated not only with the type of response (onor off) but also with the
uncertainty of the moment of delivery.
Dependence of EP on Sound Frequency
- The amplitude of N1-P2 has been shown to decline in response to an increase
in tone frequency (86,221) . It ~,~as nypothesized in ~his connection that
one of the determinants of the amplitude of N~--P2 is the number of nerve
elements in ~he cochlea activated by the sta.muli ('1.21). Because tone
frequency is inversely proportional to the surface~~rea of the basilar
membrane activated, low tones elicit EP�s of greater. amplitude than do high
- tones. This explanation is in full agreement with tl;e fact that the thres-
holds for stimulation by high tones 000-5, 000 Hz) are higher than those
68
FrJR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOfi JFFICIAL U5E ONLY
~
PI10 P115 , N115 N145 I
(7) (11) ~~�"1~ (II) (3) ~~t~
o~ o a 1 0 -
. o � o
~ I ~
~ o � o~ ~ o o~ ;jy;: o 0
0 0 o a -
PI80 N230 I', P270 N300
- ( I I ) ( 9 ) ~(-1'!!. ( 9 ) ~o-~--o~ U 2 )
o�x
- 1'"~-~
~ ~ ~ ~ � � x ~ ~
� ~ ~ ~ _
. P340 N400~
( ~2) .~f-{-~. ( ~o)
_ ~~f~
~
0 3t ~ ~
r
. ~ ~ o
Figure 12. (Conclusion)
for low tones (250 Hz) (252), determined both from components N1-P2-N2 and
from subjective reports, and with the fact that sound pitch influences the
dependence of EP amplitude on intensity. It is interesting that the curve
describing th~ dependence of amplitude on intensity for frequencies typical
of speech lies somewhat higher than curves describing all other frequencies
(503). It was concluded ~rom an investigation of EP's to 375, 1,000, and
- 8,000 Hz tones, delivered either siunultaneously with a 250 Hz tone or
successively, that the amplitude of EP components N100-P180 reflects the
activity of the auditory cortex and its tonotopic organization with every
- frequency broadly represented (Butler, 1972). The same is implied by the
results obtained by Khechinashvili et al. (72).
- EP's in Reponse to Change in Frequency of Constant-Intensity Sound
A special comparative study was performed on EP's in response to amplitude
and frequency modulation of a tone in an effort to find the best type of
stimuli to be used in audiometry (484). A 1,000 Hz tone 30 db above the
subjective thresh~ld was presented. The rise ti.me was varied from 5 to
b9
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
250 msec; the stimulus had a subsequent plateau of 800 msec, identical in
the cases of amplitude and frequency modulation. The latent period of
oscillation N1 and the amplitude of N1-P2 were determined. The ~mplitud~ -
and latent period were found to be greater with frequency modul~tion th~:n
with changes in amplitude (14 uv as compared to 8 uv), given an identical
rise time. The greater the rate of growth in frequency cha.nges, the
greater was the amplitude and the shorter was the latent period. The latter
varied from 114 msec (at a rise time of 5 msec, and with a frequency change
of one octave) to 271 msec (at a rise time of 250 msec, with a frequency
change of a tenth of an octave). With amplitude madulation, an increase
_ in the rise time from 5 to 250 msec changed the amp].itude and latent
period from 8 uv and 112 msec to 4 uv and 198 msec. The amplitude and
latent period of EP`s to frequency modulation are mure a function of the
rate of frequency change than the magnitude of the change or the rise ti.me.
A rise time of 10 msec is, in the opinion of the authors, the most
appropriate to audiometry. Inasmuch as one of the main functions of the
human auditor~ system--discriminating speech sounds--is associated more
with discrimination in the frequency spectrum than in intensity, use of
frequency modulation as the stimulus in EP recording may turn out to be
more adequate than tones and clicks in regard to research on functions of
_ the auditory system and, in particular, speech perception.
Mention should also be made of research describing rhythmical oscillations
with a period corresponding to the frequency of the delivered tone (348).
Responses were recorded from the vertex, and 999 responses were summated.
Tones in the 250-2,000 Hz range were presented. These frequency-specific
reactions were distinguishable at an intensity of 10-20 db abo~e the -
subjective threshold, but an intensity of more ttian 40 db was required for
registration of distinct reactions. These reactions were also studied by
Sohmer and Pratt (465).
EP's to Verbal Stimuli ~
When we record EP's to verbal stimuli, we encounter the problem of "physical"
similarity of words in relation to the rise time of the leading edge of the -
sound, amplitude-frequency modulation, and so on. A longer rise time than -
with clicks and commonly employed tones is possibly the r.eason why EP's
to verbal stimuli are often of lower amplitude than k,P's to tone bursts o~
the same intensity (Figure 13).
Usually when we record EP's to verbal stimuli, the latter are recorded on
- tape, which is then played back to the subject in such a way that the
_ moment a word begins is synchronized with the beginning of the averaging
interval. For some reason studies involving the u::e of ~P's have still not
made use of artificial formants which could be subj~::cted to speech analysis,
and which could at the same time yield more readil.y L-o smoath physical
modulations than do pronounced words.
70
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR )rFICIAL USE ONLY
!~s
A B nL C
i7~
~ ~ fl " V / V` ~ nf ~ ~L -
al n ~o~
s a s~ Di s~ , L ~ -
- l000 ms.ec S00� msec SOO-msec
Figure 13. EP's in Response to Tones and Verbal Communication:
A--to 500 Hz, 90 db tones; B--to the beginning of a
verbal com�nunication containing an instruction; C--
to the beginning of the same communication played
backwards--that is, a meaningless speech fragment
(Sharrard, 1972)
Published data concerning the shape of EP's to verbal stimuli and the degree
of their siniilarity to EP's to nonverbal stimuli having comparable physical
characteristics are rather contradictory, which may be the product of =
different experi.mental conditions. Comparing EP's to verbal stimuli with
EP's to tones or clicks, some authors fail to note any significant
differences (222,356,414,446), while others describe entirely different
EP's to monosyllabic words and to noise of similar configuration and
intensity: N40, P80, N123, P213, N262, P322, N363, and P411 after words,
and P23, N47, P93, N185, P244, N374, P415, and N4 56 after noise (92). The
following EP's were recorded monopolarly in the temporal region in response _
= to spoken numbers and clicks presented at the same loudness (85 db): P42 _
(20-50), N101 (85-130), P205 (185-230), N266 (240-280), P317 (290-350) (356).
In a number of works (249,346,540) only two components are clearly iso-
lated in EP's to verbal stimuli: N69-153 and P160-306. Wood and Goff (1971)
- describe a triphasal complex: P100, N200, and P2 50-800. We can see from
. the available studies that it is fully possible to record EP's to verbal
signals, and that they can be used productively t~ solve the corresponding
problems .
It would be important to note the following. By evolved tradition, _
differences between EP's to verbal and nonverbal stimuli are usually
interpreted as changes in the amplitudes or the 1 atent periods of indivi-
dual ~omponents. Implicit to this tradition is the supposition that the _
component remains constant in the examined cases. We should expect,
however, that EP's to verbal stimuli would to have a somewhat different
origin from that of EP's to clicks or tones (in the same way that the
origins of early components of EP's to illumination changes and to changes
in the structure of the visual field are different), though experimental
proof of this does not exist as yet, if we discount data on interhemispheric
asymmetry (see Chapter XIII). This hypothesis i s based mainly on the ~fact
that the integrity of the auditory cortex is not siqnificant to discrimination
71
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL' USE ONLX
of the intensity and frequency of l.ong tones, ~ut i~t ~_s nPCessary -~o
discrimination of the temporal characteristics of a stimuJ.us, to perception
of short stimuli, and to analysis of verba:~. sti.muli~ D~tailed study of
EP's ta verbal stimuli, their desc.ription in relation to the most wiclesprezd
situations, and the study of their neurogenesis will ~roblably occur i_n the
near future. Without a doubt their. use will broadex~ in ~he neai fu~ure,
both in research on the mechanisms responsible for analysis of verbal
stimuli, development of speech, and dominance of the t~en9~spheres, and in
the study of inemory, attenkion, and so on.
_ Neurogenesis of the Components of EP's to Cliclcs o~, Short Tones
The hypothesis that oscillations arising on the scalp in the first 10 msec
following a stimulus reflect activation of the cochlea, transmission of
impulses along the auditory nerve, and activation at the brain stem's
auditory nuclei (286) was confirmed by an analysis af. the distribution oi
the amplitudes of these oscillations over the scalp (3'15), The neuro-
genesis of components falling within rhe 10-60 msec period following the
stinulus remains extremely unclear. The authors of the first description of -
auditory EP components falling within the 10-60 msec period following the
_ stimulus (238) interpreted them as a reflection of in~tracerebral potentials.
Following Bickford's work (101,102), which demonstra~L-ed the doubtless
presence of myogenic artefacts in this period, many stridies wer.e performed
to reveal the origin of the components within th~s periocl. These efforts
confirm the signific ant hazard of "contamination" by myogenic a~tefacts,
but they simultaneous ly present numerous data on Lhe intracerebral o-rigin
of components in thi s period (137,245,?.46,325,376,422,460).
Goff et al. (1976) conclude from their own data and trom an analysis of
Picton et al. (376) that components in the 1~-60 msec period are basically
_ neurogenic, but they may be "contaminated" by myoyenic artefacts,.thus in the -
end representing an a lloy of neurogenic and myogeiiic zctivity. One of the
sources of "contamina~ion" is muscular activity a.n trze ~oi:rse of the
startle reflex following loud sounds, in connectioi~ wii:h which Picton et al.
recommend avoiding loud sounds and rec~rding EP' s in ~the vezL-ex area (where
there are no muscles) or during sleep (when musc]_es z~re relaxed}, -
What is the origin of cerebral EP's fallir~g wi.thin thi.s period? Their ~
latent periods imply a relationsl~ip wit}~ the activ9.ty of the medial
qeniculate body, thal amic nuclei, ancl thc auditory and lssociative cortex.
Inasmuch as very low correlation was discovered betw~en the latent
periods of FP's recorded from the primary audii:ory cor~ex and EP's re-
corded from the scalp (137) , Picton et al. s,.iggest t!iat the activity of the
primary auditory cortex is reflecte~~ t~o a very sm~-~J 1 ext`nt in Ep's
recorded from the scalp. 'they believe that cornpoiie~~ts N, P, and N
(see Figure 9) reflect L-he activity of the medial re.zicu~ate�body ora
polysensory thalamic nuclea., and that osci.llation P~:�epresents activation
of neurons in the ass ociative cortex. 'I'he latter hypothesis is ground~d
72
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOI. ~FFICIAL USE ONLY .
in the fact that a pronounced positive oscillation with a latent period of
25-30 msec has been recorded intracranially from the associative cortex
of the frontal, occipital, and parietal regions (136,141,422), Thus in
their opinion nonspecific oscillations arise in the first 50 msec after the
auditory stimulus.
According to Goff et al. (1976) the 60-110 msec period is filled mainly with
myogenic oscillations; we were unable to find persuasive data in the
literature indicating the neurogenesis of components falling within this _
period. All components following component P115 have a"pure" neurogenic
origin. The largest number of neurogenesis studies is devoted to
components N1 and P2--the so-called vertex potential, the part of the
EP that has the most pronounced amplitude. As was noted earlier, for a
rather long period of time these oscillations were thought to be a
reflection of the activity of the brain's nonspecific activating system
(321). -
Investigation of tl~e dis~ribution of the amplitudes of all components of tonal
- EP's recorded from the scalp led to the conclusion that components N1-P2
arise in the auditory cortex (508,510). It was discovered that the
amplitude of N100-P200 had its maximum at the vertex, that it decreased as
- the e~.ectrode was moved away from the latter in a frontal direction,
attaining its zero value approximately at the level of the line passing
abca~~e the fissure of Sylvius, and then that it once again increased with
opposite sign. Earlier :?scillations also changed in similar fashion.
The authors computed the theoretical�distribution of potentials over the
scalp for all hypothetically possible sources of thESe oscillations, basing
themselves on data concerning the morphofunctional structure and temporal
characteristics of the auditory system, and current ideas about the
distribution of potentials over the surface of a three-dimensional
conductor. After this they selected, from the set of hypothetically
possible sources, that for which the theoretical amplitude distribution
agreed best of all with the empirically obtained distri.bution. A dipole _
layer perpendicular to the surfac~ of the cortex and parallel to the
orientation of the primary auditory cortex (Heschl's gyrus) was found to
be such a source. The authors concluded that the examined EP components
are generated in the primary auditory cortex. This was also in agreement
with the dominance they noted of EP amplitude on the side opposite the
stimulated ear for the case of monaural stimulation. Other authors came ~
to similar conclusions as well (245,534), discovering that integrity of the
primary auditory cortex is necessary if EP components falling within the
100-200 msec period following a stimulus are to be recorded in response to
sound from patients with cerebral afflii;tions.
It should be noted that Kooi et al. (303) were unable to detect a shift in
EP polarity above the fissure of Sylvius when EP's to sound were recorded
with the indifferent electrode located in the vicinity of the rib cage,
and they exFlained the shift in polarity observed by Vaughan and Ritter
by t.he fact that the indifferent ear electrode was in fact active, with its _
73
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAI. USE ONi.Y
activity exceeding that beneath electrodes below ti~e zissu.rc of Sylvius,
- which is what causes change in polarity. Polarity sh~..ft in the vicinity
of the fissure of Sylvius was lat2r shown to occur ~-~irii an e,.tracraniai
electrode (314,463).
Picton et al. believe that polarity shift is caused more li}cely by the
activity of the lower margin of the hemisphere than by a deeply located
dipole (one in the auditory cortex), and that comp~nents N1-P2 reflect
mainly the activity of the froz~tal and not the auditory associative cortex.
Their opinion is based on two groups of facts. On one hand they observed
these components to have their highest amplitude in the frontal and central
regions, and other authors (517,527) recorded oscil!ations with comparable
latent periods directly within the frontal cortex, ~~hile on the other
hand their research did not reveal a shift in the polarity of N1-P2 when
the recordings were made with a thoracic electrode, and that there was
no interhemispheric difference in EP's in response to monaural stimulation.
Nbreover, recording directly from within the auditory cortex, they noted
absence of oscillations with latent periods close to those of EP's
recorded from the scalp (137). The fact that these components are
absent or reduced among patients with damage to the primary auditory cortex
is interpreted by these authors as evidence of an influence played by the
primary auditory cortex upon the activity of the frontal associati_ve cortex
via cortical-subcortical-cortical or cortical-cortical pathways. Picton
et al. concurrently presumed that the auditory cortex also participates
to some extent in the generation of these compor~ents, in addition to the
. "main" source in the frontal associative cortex, and they agree ful].y with
the idea (510) that these components are mode-specific.
Significant evidence indicating participation of ~he prinu~ry auditory cortex
in the generation of N1-P2 is contained in the work of Peronnet et al.
(366), who concluded that components N1-P2 reflect the activity of the
auditory cortex and the parietotemporal associative cortnx, the dec~ree of
participation of these regions depending on the exper~.mental conditions
and individual features of the subject. But:Ler's data (122) on the
- frequency specificity of these components aiid on their zeflection of the
tonotopic organization of the auditory cort.e~c also im~1y p~~rticipation of
the auditory cortex in generation of N1-P2. ~i~he complex polyge~zic nature
of componerits N1-P2 is confirmed by Khechinastivili et al. (1973), who
suggested, on the basis of the presence of two "saturata.on" po~nts
accompanying an increase in the ampJ.itiic~~ oi Nl-P2 conr,.ec~ed with g:cowth
in stimulus intensity, existence of two independent sys~ems generating these
components.
And so, most of the data today indi.cate that the a~icii~.ory and associative
(frontal and parietotemporal) cortex perticipates .~i thc~ gener.ation of
N1-P2. These components are cloubtlessly polygenic, and our i.deas about
- their generators will obviously become more refined, and they will undergo
" change. Much less research has been c~nducted on tt,e origin of the rest of
the components falling within this period, ones wh~.~h may vary it.,~ependently
of one another. In s1eeP, f.or_ examp].~ ~ only the arn~~1.i.tnde o:E oscillatxon N2,
74
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200100045-9
FOh JFFICIAL USE ONLY ~
which is interpreted as a reflection of difuse activation of the cortex
(375) in sleep, rises. However, why the rest of the components do not in-
crease concurrently remains unclear. Ritter and Vaughan (1972) believe _
oscillation N250 to be part of an associative potential of the cortex,
inasmuch as it is also observed in the presence of an expected stimulus
(301,375). However, it is unclear as to how much the nature of the negative
oscillation of this period in response to an indifferent stimulus is similar
to that of an oscillation arising in response to omission of an expected
stimulus or the action of signaling stimuli (458).
Goff et al. (246) suggest that oscillations following components N115-P180
in an EP to an indifferent sti.mulus are possibly a manifestation of the
so-called postdischarge. The idea that oscillations P270 and P340 reflect _
a so-called late positive oscillation is believed by Goff et al. to be
less probable, inasmuch as these oscillations exhibit different amplitude
distributions over the sca3p.
Kevanishvili et al. (299) published an article analyzing the origin of a
number of components of tonal EP's. Relying on their own and published
data, the authors concl.uded that components P51, N100, and P192 and the
following negative-positive complex have varying origin. This is implied
by facts such as the difference in the nature of the association expressed
by the first three components and two subsequent ones with stimulus
intensity, the differences exhibited in their ontogenetic changes, the
decline of the former in response to an increase in the latter during
sleep,~the shorter recovery period of the latter, and so on. It is
hypothesized that oscillations P51, N100, and P192 reflect activity of the
associative cortex arising in response to impulsation from nonspecific
- thalamic synchronizing nuclei.
Considering the certain contradiction existing between this hypothesis and
information on modal specificity of components within the first 200 msec
_ following a stimulus, the authors note that the existing notions, based on
animal experiments, that a relationship is necessarily present between
specificity and projection zones cannot be applied with full certainty to
man. They emphasize the known fact of inter- and intramodal specificity of
reactions within the thalamic nonspecific system, and they suggest the
hypothesis that reactions of the associative cortex may be more specific in -
man than in animals.
75
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
I
FOR OFFICIAL USE O~iL;.'
' cxAP~R vzzz -
SOM~ITOSENSORX E?~' S
_ Somatosensory EP's are reactions to stim~~lation of slcin receptors and
proprioceptors. Electrocutaneous stimulation continues ~o be ~hP most
frequently employed stimulation technique when reco.rding somatosensory EP`s,
though natural stimulation techniques aze alsa bein7 utilized with in-
creasing frequency--pricking the skin lightly with a pin, touch zng the s;cin,
_ applying pressure, blowing streams of air, or striking a tendon wii~h a
hammer. The shortcoming of electrocutaneous stimula~ion is i:hat it is "not
physiological," while its advantages include synchrony of stimulation,
_ which produces EP's of greater arnplitude than thos:~ zcquired by oth~r.
stimulation techniques, and the ease of "dosirzg" :he sti;nulus. Incidentally,
the possibility is not excluded that the "nonspecific" acta_on of electro-
cutaneous stimulation--alarm caused by an unusual st.imtz:l_us--plays some
~~~t of r.ole in the high amplitude of responses to electric stimulation:
Recordinq EP's directly from tiie cortex during ~~n oper.ai:.ion, Jasper et al.
(277) noted that the amplitude of the response to moderate "physiological" '
tactile stimulation is greater than a response to a s~rong electric
stimulus.
It should be remembered that the nature of outwardly similar components
recorded with the use of different stimulus techniques may turn out to
be er~tirely different, inasmuch as different :Eorms o� stimulation may
address themselves, in part or in their entiret~~, to dizferent peripheral
receptors, and as a result they may activa�e entirely dziferent cortical
neuron elements, in the same way as has been demonstrat~d for_ di ~ferent
types of visual stimulation.
_ Morphology of EP's to Electrocutaneous Stimi.~lation
Electrocutaneous stimulation is perform~cl most rr.~c,,:;ently in the wrist
region, where the median nerve is located closest the surface, or (more
rarely) in the region of the distal phalanx ~f the i:znd, usuzlly of the
, in3ex finger. Short (0.1 msec and less) square pu:~sc~s of minimum intensity
sufficient to cause muscle contraction in r.esponse t.o stimulation, or of
76
FOR OFFICIAL USE OYLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR i)FFICIAL USE ONLY
near-threshold intensity (on the basis of a subjective report) are used (for _
greater detail see (76)). _
A tremendous number of papers describe EP's to electrocutaneous stimulation,
beginning with the work of Dawson (179), who used averaging for the first
time (81,131,186,243,245,246,353,442,489; Goff et al., 1962; Shagass,
Schwartz, 1963; etc). We can apparently adopt, as our normative data, the
descriptions of somatosensory EP's to electrocutaneous stimulation presented
in a number of works (245,246,489) (figures 14 and 15, Table 3j.
Differences in tne characteristics of early components cited in different
works can be seen in Table 4. One of the possible reasons for the
discrepancies in amplitudes and latent periods of the early components of
EP's to electrostimulation may be differences in the system of filters used ~
to record the EP's (187). -
~J' 0. -
1 B 0
A gg aJ r` Z Ol 1
~ ; ~ Z
~r
P~ r ~ ~ ~
~n ~ ~ . ~
ti y ~ ~i
~ ~ ~
6~ppmsec 61rS m~c
Figure 14. Schematic Somatosensory EP's at Age 9-18 (1) and
19-29 (2) (For Groups of 20 and 40 Persons Respectively):
A--EP as a whole; B--early components (489)
An investigation that isolated two groups of subjects on the basis of
8ifferent characteristics exhibited by the early components of EP's to
electrocutaneous stimulation (243) indicates a possible cause of the
discrepancy in descriptions of early components of somatosensory EP's to
- electrostimulati~n. Components N19, P27, N36, and P45 were recorded in _
one of the subgroups (18 persons), and compon~nts N18, P22, N26, P30, and _
P49 were recorded in the other (seven persons)--that is, another two com- -
ponents were recorded between components P22 and P49 (common with the
first subgroup)--N26 and P30. Giblin suggests that all subjects have all of
- the components, but that in the first subgroup they are summed during the
averaging process in such a way that they are indistinguishable as -
i.ndividual components. In general, positive oscillations with latent
periods of 21-31 msec and 33-50 msec are the early components most frequently
isolated. Oscillation P2 (33-50 msec) is followed by two negative-positive
complexes: N3 (65 � 14 msec) , P3 (85 � 20 msec) , and N4 (135 � 25 msec) ,
P4 (220 � 45 msec) (245). According to Tamura et al. (489) a graup of 40
77
FOR OFFICIAL USE ONLY "
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007/02108: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONI,Y
Table 3. Somatosensory EP Components (From {246;)
f1 ~ 3 ~ MBKCRNy?7 9Yrt11HTV;~H, ::}:9 ~~Jf
- Koxft~~nr n~ p nt ~ i~p_�zo.;arae-
pHOJ(, 11ClK MtJ~1f8XE ~~P / \ ^ -iGC npu~~cn0){^
_ 7a3~~~ 1~/ ACHHC
- (7~i4 14,2�0,9 -2,3 -!,4 ~--11,5 Fi
_ (8) ril5 14,9�1,4 -}-i,9 -}-i,fi -}-4,~s i-i
020 2t,3�2,1 -2,0 -0,9 -3,3 H
- T120 20,9�1,8 ~-2,2 -;-0,7 ;-3,8 H
O?a 24,4�1,9 -}-!,2 -~0,5 r4,3 H
1730 28,6�3,4 -}-2,7 -~-1,2 ;-fi,p ~ H ~
035 34,5�4,5 -4,2 -2,5 --7,3 H -
II45 44,9-�-4,8 -f-4.2 -{-1,8 -~-9,f H
Q55 55,3�6,6 -3,8 -1,7 -�9,2 H
r1&5 65,5�5,2 -{-6,i -}-!,3 -'-91,3 M
070 71,8�6,g -2,3 -0,4 -8,3 N
It80 79,5�6,3 -{-6,3 -}-2,0 -!-i'/,~ H
~ 89,3�G,5 -t,i -}-7,4 -5,5 H -
II100 102�12 -{-13,0 -}-6,5 -}-42,7 M-}-H
Oi40 i40�i3 -8,9 -8,0 -18,2 H
f]I90 192�16 -!-18,! -{-l;1,6 -}-29,2 H
- 0260 256�24 -4,7 -0,1 -12,5 H
II300 295�30 -~-7,6 -}-3,5 -}-l8,1 H
0360 362�~ -9,5 -3,9 - li,3 H
I1420 418�48 -{-2, 7 1-! , 4 8 H
0460 461�52 -5,1 -2,9 -8,'. H
Symbols: H--neurogenic origin; M---myogenic origin _
,
~ Key : .
1. Component 5. Scatter
2. Latent period, msec 6~ Hypothesi_~~d, o~:?_gin
3. Amplitude maximum, Uv '7, N
4. Median g. p -
persons had the following components: N3-50 (8 � 1.. 39 msec) P3 (85.4 � 2.3) ,
_ and N4 (135.9 � 4.0) , P4 (242.3 � 8.8) . In this worY the authors ais~ iso-
lated negetive oscillation N5 with a latent period of 391.8-!-13.5 msec. A
particular oscillation may be absent from any real EP tai~h a lesser or
greater probability. Data on the frequency with whicti individual components
- are recorded may be found in the work of Goff e~ al. (1976) (see L'igure 15) ,
- and in tables 3 and 4.
_ Registration of oscillations from the scalp and abov~~ the cervicothoracic
= division of the spine reflecting activity in the aff~rent pathways and �
nuclei of the spinal cord, the brain stem, and sui~c~~rtical nuclei in
' response to electrocutaneous stimulation has been re,~orted several times in
recent years. Jones (290) distinguishes four negati�~e oscillations with
latent periods of 9, 11, 13, and 14 msec; he believes that they reflect
activity arising in response to stimulation of the zn,~dian nerve in the
78
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
- FOF GFFI(;rAL USE ONLY
Tabl:: 4. Polarity ar;3 Latent Periods of the Components of EP' s to Electro-
stimulation of the Median iJerve or Finger, According to the Data
~ of Different Authors
~ X8p8ICTlpHCTNKH Bit ~1~ I A~'+~P~ r~A
~
~3~i6 oZ0 II25 T131 CI48 065 Goff et af., i962 -
020 II27 038 CI46 Larson, Prevec, 196~�
018 TI25 034 II44 Nakanishi et a[., 19~~~
018 r125 032 II43 05~ II85 0136 I1242 0392 Tamura et al., 1972
IIi7 024 T13i n62 I182 Oii8 II160 Lee et al., 1974
TIi6 019 II24 032 II37 059 Ii88 0125 fIi61 Cracco, 1972 -
- 020 II44 OiE I1110 0168 Ikuta, 1972
~ 020 II29 033 II35 073 TI96 Shagass, 1972
II26 I732 CI42 TI95 0110 II217 Veiasco, Velasco, 1975"
* The named components were isolated on the basis of their own
data and seven works by other authors.
_ Recorded from the exposed cortex of 10 persons. `
. Note: The authors underscored the most stable components.
= Key: -
1. EP characteristics 3. P
2. Author, year 4. N
- brachial plexus, the posterior roots or posterior col~,unns of the spinal
cord, the brain stem, and possi.bly the thalamus.
_ Morphology of EP' s to Tactile Sti.muli
- MechanicaL stimulation was used much more rarely in research on somatosensory
EP's than electrocutaneous stimulation. This is connected with the
methodological difficulties of the former. Nevertheless the possibility
for recording, from the sca].p, EP's in response to light touching of the
skin or short-term touching and presstire with a hammer (226,251,309,340,
353,354; Erenberger et al., 1966; etc.), and to striking a tendon with a
hammer (268 ~ 269) has been undisputably proven as of today (r igure 16) .
Distinct EP' s can be recorded in response to mechanical stimuli whose
- intensity is insufficient for acquisition of re~~r~ducible cumulative nerve
_ action potentials (309). l
An electric vibrator is used most frequently for stimulation; the amount
- the head moves deperids on the voltage applied, which makes it possible to
- fix the moment of stimulation and dose the stimulus with precision to
allow study of the dependence b~tween the ~P and the intensity (226). The
= head of the vibracor is situated next to the pad of the finger (or some
other object of stimulation) such that movements of the head are perpendicular
- to the surface to be stimulated. The nature of stimulation depends on what
i~ attached to the head of the vibrai:or: a soft- or hard-bristled brush, '
79
FOR OFFICZAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL USE ONLY
, P15 N20 P20
- (~i) (i~! ~4-~~ (iif J',o.
r ,1 1 r
- ~:~6:,':.~. � ~ ~ � 0 4 -
? ~ ~ ~ ~ 1
~ o ~ ~ / ~ ~ /
c p 0 0 0~
\ ~ ~ A~/ -
O O ~ O
625 ,~T~` P2 0~~~\ N2 5
. ~ . ~
0
P o o P~ o
~ ~ 9 ~
t ~ ~ ~ ~ ~~.T
. '
p p ~ : p ~ p , p p p
_ \ ~ i ~ / ~ ~
~o~A~ A~ \`0. 0~
o o 0
P45 N55 P65
~
( I 2 ) ~ o- ~ ( ~ 2 ) ; :lac~;~'~;:�. ( 7 ) :'`;:Q;:
� o d o . � o ~ ~2
~ 1 ~ 1 1
L
.o' ~ O :.~,~'.1~
~`A~~~~
O O
Figure 15. Topography of A Somatosensor_y Ep (Goff ~t al., 1976}
80
FOR OFFICIAL USE ONLY
~ ~
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOF ~FFICl"AL USE ONLY
N70 . P80 i PI00
(6) .;~:i~ f ro;; (51 ~~�-II'O.\ (7)
' ~ o � o � C+
- 1 ~ ~ i
1
1 1 ' ' 1
O ~ ~ / \ � j ~ ` ~
~ � O
~p .
~ �p/
~ O
{J140 ~ P190 ~ N260 -
(7) ,.o-~--c~\ (12) ~c-I'~. (I2)
� � : -
;V:: .O. ~
~ ~ ~ `
l ~ ~ ~ -
j ~ �
_ ~ � . / ~ d, ;:f,'�;:
?\~~i
0 0 0
P300 , N360 P420
- (12) .~4-f'3~ ti2) '
i~(f ~ ~ ( ~ 0) ~
:~!f~: o b :;o: ~ -
~ , 'w-~~h
~ ~ ~
. r.
~ ~ O
- Figure 15. (Continued)
81 -
- FOR OFFICIAL USE ONLY -
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
- FOR OFFICIAL USE ONLY
^ N 1SJ u
~ N24S 10 v
PJ/o -
PS9 P20S
6 ~
- uv fladaBnuBayuc ~ 1 ~
- 6 ' ~,q na~~y
y I !S S
/ ~
J I i!0 3
~ V ~ I I S I~ .
100 3S0 ~ 600 BSO Z00 400 600 B00 l000
MA'M/! MKMA ~ 2 ~
~ Figure 16. EP in RF:sponse to Pressure Applied to the Skin of the
J Right Index Finger: A--Typical EP averaged from 128
_ responses. Monopolar recording, with the active
electrode in the primary projection zone; negativeness `
upward. The latent periods are indi.cated in msec;
B--psychophysical cwrve. Subj~ctive assessment of
stimulus intensity in response tc :lifferent forces
- (averaged normalized data for a group of nine persons);
C--Dependence of peak-to-peak amplitudes of different
components of the somatosensory EP of the same subject
on pressure applied to the skin. 1--N25-P59; 2--P206-
N245; 3--N245-P318; 4--N158-P206; 5--N59-P158.
_ Vertical lines indicate standa,rd error of amplitude
determined from 128 responses (Jonson et al., 1972)
- Key :
1. Pressure upon finger 2, um1
= a hammer, or a pin. The action of such stimulators is usually accompanied
- by a short burst of sound, which by itself may elicit auditory EP's (483),
which should be accounted for when analyzing the results. Ir~ order to
preclude "contamination" of the tactile EP's by reactions to sound, the
_ EP's are usually recorded on a noise background that masks the clicks
generated when delivering the tactile sta.muli.
Early reactions to tactile mechanicai stimuli are ,imilar in shape to
reactions to electrostimulation, differing on].y in 't~~vil~g a lower
amplitude and greater_ latent perio3s. EP's to diffcrent types of
samatosensory stimulation were compare3 in a group of 33 persons (354).
EP's were recorded in this study bipolarly with silver needles. Good
= agreement was revealed between the results of these autho~~ and those of
82
FOR OFFICIAL USE ONLY '
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOI. OFFICIAL USE ONLY
other works in which electrostimulation was employed; latent periods were
longer and amplitudes were lower for the components of EP's in response to
mechanical stimulation than those in response to electric stimulation;
moreover the expressiveness of the responses was lower for the former
(measured in terms of the number of subjects from whom this EP component
was recorded): Responses to stimulation by touching with a soft-bristled
brush were recorded from 69 percent of the subjects while responses to ~
electrostimulation were recorded from 100 percent of the subjects.
Differences in the latent periods of EP's to stimulation of fingers and
toes corresponded to the difference in the distance from the cortex and
_ in the rate of conduction of nerve impulses. The latent period and shape
of EP's recorded at symmetrical points on different hemispheres were the
same, while the amplitude could differ significantly. The EP character-
_ istics cited by these authors are close to reactions described in response
to similar stimulation in a number of other works (309,340; Offenloch,
1968; etc.).
Note tnat similarity in EP's to electric and tactile stimulation does not _
mean that the structures generating them are completely identical. Precisely
what fibers are stimulated with electrostimulation is not entirely clear; it
- may be that some part of them are not fibers activated by tactile stimuli.
Proprioceptors, particularly receptors providing information on tendon posi-
tion, or "articulation sense," play a significant role in electrostimulation
(243).
EP's were recorded in response to blowinq a stream of air at the cornea or ~
the nasal mucosa (338). Two oscillations with peak latent periods of 100 and
200 msec were isolated. The authors believe the first oscillation to be a
- m~anifestation of the blinking reflex, and the second they interpret as a
vertex potential--that is, as an analog of the positive oscillation with a
latent period of about 200 msec_ observed in EP`s in response to all stimuli,
~articularly in EP's in response to electrostimulation of the median nerve
and to vibrotactile stimulation. Z'his oscillation was best expressed at
point N3, where its amplitude attained 2-5 uv. The authors believe absence
- of earlier oscillations to be a consequence of low stimulus intensity, the
increase of which is hindered by blinking.
Registration of scalp EP's in response to electrostimulation of tooth pulp -
has been reported (140). The authors interpret the oscillations they recorded
as the fixst objective correlate of acute experimental pain. EP's have been
recorded in response to change in skin temperature (142), and stimulation of
taste receptors (232).
- EP's Arisir.g in Response to Passive Nbvement of a Finger
EP's in response to passive movement of the index finger 45� were recorded
in seven pa~ients in the procentral, postcentral, and prefrontal cortex with
subdu~al ~lectrodes, and with super~icial electrodes in the precentral and post-
central regions and vertex of thesame patients and five healthy individuals (362).
83 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/08: CIA-RDP82-00850R000200100045-9

APPROVED FOR RELEASE: 2007102/08: CIA-RDP82-00850R000200100045-9
FOR OFFICIAL US:. ONLY
All electrodes were positioned in ~h~ right hemisphere. All subjects exhibited
stable reactions in response to passive moveinent of the finger. An initial
i~.csitive oscillation was observed after 36 t 6 msec in the con~tralateral
pre- and postcentral regions of the co-rtex, attazning a peak after 42 � 4
msec. Its amplitude varied from 5 to 25 uv in di~'feren~ p~ople, and it was
qreater in the postcentral r~gion. I~c was followed in the precen~ral r.egion
by a negative oscillation witn a peak latent period oL GS � 1.5 msec, ax~d an
oscillation in the postcentral region with a pea}c laten~ period of 97 � 20
msec; its amplitude varied from 30 to 50 uv. Subsequent oscillztions obscrved -
over a period of 400 msec were stable in the same individuzl but varied signi-
ficantly am~ong different people. Reactions began to occur in the prefroni:al
region after 100 msec, and they were identical for i~si:].ateral and contra-
lateral movemen~s. The EP to ipsilat~ral. movement, .recorded in th~ pre- and
postcentral regions, began with ~ positive asci].lation with a latent period
of 60 msec and more. High cross correlation (0.85) was noted for EP's re-
corded from the same points of contact, and l.ow cross corre].ation (0.25) was -
observed for EP's recorded with differ_cnt poin~ts of contact. Flexing and -
extending the hand did not change the EP, no.r was acti.vity noted in the finger `
EMG in response to passive movement, which is interpr.eted as evidence of re- -
flection, within the EP, of the activit~y of ~Fferent elem~nts in the li.gaments,
and not in the muscles. Additional investigation of EP�s recorded during
short-tenn ischemic anesthesia permitted the authors to conclude that the EP's
they recorded basically reflect acr,ivation oi affereni: fibers in the liga-
mental apparatus. Other characteristics of EP's recorded during passive
_ movements are presented in (46).
Neurogenesis of the Com,ponents of ~P's Arising in Ftesponse to Electrocutaneous
Stimulation
The earliest oscillation recorded irom the back surface of the neclc, N14,
reflectspassage of afferent impulses along the cervical divisi_on of the
spinal cord (Goff et al., 1976). ~e first EP component recorded on the
scalp--P15--is viewed by most researchers to be a reflection of activation
of subcortical structures--the thalamocortical radiation or the ventrobasal -
thalamic nuclei (81,112,246; etc.)~ Opinions conccrni.ng subsequ~nt components
are less unanimous.
Allison et al. (1973) evaluated the negative oscillaL-ion following the first
positive one as a reflection of the transmission of impulses through the
thalamocortical radiation; thc next positive oscillation is interpreted as -
- a reflection of primary positiveness--that is, activaticn of the bodies of
neurons in the primary projection zone of the cortex by an afferent volley.
The next positive oscillation, faliing within the 35-50 msec period, which -
,in distinction to p.revious ones was suppressed by a:~esi~hesia, was eualuated
to be a reflection of the activity of neurons ancl t,,~ associai:ed cortex, as
a manifestation of activating influences by the asc