Biofeedback and Self-Regulation, Vol. 1, No. 1, 1976

Detection of EEG Abnormalities with Feedback Stimulation' Thomas Mulholland Veterans Administration Hospital, Bedford, Massachusetts

Frank Benson i/eterans Administration Hospital, Boston, Massachusetts

A feedback method for testing the reactivity of the occipital-parietal EEG in selected brain-lesioned patients revealed abnormalities of (a) insufficient reactivity, (b) bilateral differences in reactivity, and (3) asynchrony. These abnormalities were more evident during feedback stimulation than in the baseline conditions. The utility of feedback method for detecting EEG abnormalities rests on the increased stability or decreased "'noisy'" variation in the EEG during feedback. The EEG becomes more predictable even to the "'on-line" human observer. This makes it easier to detect aberrations or deviations from normal effects. Some effects can only be seen with feedback such as the bilateral differences which occur when the left side controls the feedback compared to when the right side controls it. The results show that feedback EEG is a useful tool in clinical research and indicate that a clinical diagnostic test could be developed with more research. However, the feedback EEG method is not yet a proven diagnostic technique.

Clinical and experimental studies of the typical human bilateral parietaloccipital EEG have shown that the recordings from the left and right sides are similar, although some mismatching of details always occurs. Both visual observation and measurement schemes which classify the EEG recording into intervals of synchronous and desynchronous activity emphasize gross similarities, while more refined and precise measurements, e.g., phase differences, emphasize the dissimilarity of details. On all levels of observation, a large and recurring bilateral mismatch compared to norms is a pathological sign. 'The assistance of Eric Peper, Sylvia Runnals, Rosemary Billingslea, and Thomas McLaughlin in the EEG recording and data analysis was indispensible to this study. 47 © 1976 Plenum Publishing Corporation, 227 West 17th Street, New Y o r k , N.Y. 10011. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, w i t h o u t w r i t t e n permission o f the publisher.

48

Mulhoiland and Benson

Recordings of occipital alpha rhythms are bilaterally similar during continuous darkness or steady light and also following visual stimulation. The decline (habituation) of EEG response to repeated visual stimulation is also bilaterally represented. The evaluation of the magnitude, bilateral symmetry, and synchrony of the response of the posterior alpha rhythms to stimulation has been an important part of "functional EEG" and has been studied in a wide range of brain disorders (Gastaut, 1949; Bancaud, Hecaen, & Laity, 1955; Fischgold & Mathis, 1959; Cobb, 1963; Hill, 1963). The bilateral absence or marked reduction of alpha blocking by stimuli is an abnormal sign (Liberson, 1944; Blum, 1957; Kooi & Thomas, 1958; Wells, 1962; Hollaway & Parsons, 1971). Unilateral hyporeactivity or unilateral changes of alpha frequency are less commonly reported (Cobb, 1963). Disturbances of normal habituation and a reduction or absence of dishabituation of the alpha suppression response have also been documented for thebrain damaged (Wells, 1962; Holloway & Parsons, 1971).

METHOD

In this study, changes in the occurrence of the occipital alpha rhythms were used as an index of the orienting response in normal subjects and selected brain-lesioned patients. The description of the EEG index of the orienting response presented here is different from the familiar description of the singular alpha "block" or suppression following visual stimulation (Jus & Jus, 1960). The response of the occipital EEG is described here as a disturbance followed by a recovery of the ongoing alternation between alpha and intervals of little or no alpha. These latter intervals are collectively called no-alpha. The time series is called the "alpha-attenuation" cycle (Milstein, Stevens, & Sachdev, 1969). There is no mechanism or physiological process implied by the term; it is simply a name for a familiar EEG phenomenon (Bagchi, 1937). Alpha and no-alpha durations are extremely variable between and within individual records. To reduce variation, we have utilized a biofeedback method. A visual stimulus causes a blocking of the EEG alpha which is connected to the stimulus through an external path of electronic instruments, controlling both its presentation and removal. In this study, alpha caused a visual stimulus to go on; no-alpha caused it to go off. The feedback EEG method markedly reduces unsystematic variation of alpha duration and no-alpha durations after recovery from the initial disturbance, both within and between subjects, and produces a more stable alternation between intervals of alpha and intervals of little or no-alpha (Mulholland, 1968, 1973).

Detection of EEG Abnormalities

49

Though feedback E E G / s a quantitative method, the EEG record can be read by an experienced electroencephalographer to detect abnormal " feed b ack " effects. In the first part of this paper examples were selected which are best seen with feedback stimulation compared to a baseline "resting" EEG. This extension of the EEG as a diagnostic test reflects the increased control over the EEG which feedback offers. The examples are not exhaustive and only salient effects which can be observed on-line in the recording are presented in the first part of this paper. They permit the experienced electroencephalographer to judge the utility of the feedback method for revealing abnormal effects. The second part of this paper presents a computer-generated graphical display for the quantitative analysis of individual response features, e.g., reactivity, which is the initial disturbance and subsequent recovery from the effect of the stimulus and the bilateral similarity of the response. They are intended to be " r e a d " in the same way an EEG record is read, that is, by the experienced electroencephalographer who knows the advantages and limitations of the method. These graphs provide a standardized characterization of the responsivity of the EEG on a trial-by-trial basis. The feedback method is also useful for testing higher mental functions when the proper complex feedback stimuli are used. Since different kinds of feedback stimuli produce different effects on the occipital EEG (Mulholland, 1973), the ability of the person to react differently to different stimuli can be evaluated. An indirect test of stimulus discrimination can be provided at the same time that other features of the EEG are being tested. The feedback methods used here have been described elsewhere (Mulholland & Gascon, 1972; Mulholland, 1973). In all cases, occipital alpha was detected and, by means of external electronic instruments, was associated with a visual stimulus flashed on a screen in front of the subject. The stimulus was either a word, a picture, or an unpatterned spot of light. The visual stimulus was on with alpha and off with no-alpha (Loop 1) or reversed so that the stimulus was off with alpha and on with no-alpha (Loop 2). At the level of on-line monitoring of an EEG during feedback, the experienced examiner can identify the reactivity to feedback stimulation, the recovery (habituation), the stability of the EEG features during feedback after habituation has progressed, the bilateral symmetry during feedback, and the system response when the left hemisphere EEG is controlling the stimulus compared to when the right side is controlling. The latter conditions are called "left-contingent" and "right contingent," respectively. Clinical Material The examples were selected from an ongoing study aimed at developing instrumentation, test protocols, and data analysis for a clinical feedback

50

Mulholland and Benson

EEG method. All cases were from an aphasia therapy unit; all were ambulatory and able to follow simple instructions. For comparison, two EEG typical records and one atypical record from three nonpatient volunteers are presented. All of the patients with brain lesions did not show the effects described here. Also, since recordings were made of the occipital-parietal EEG, abnormal responses in other EEG locations, if they occurred, were missed.

Selected Examples In the illustrations herein, recordings from left (0~-P3) and right (0~-P4) are compared. The response of the alpha detector which controls the feedback stimulus is also marked on the record. In some records only the side which controls the feedback is marked; in other records, alpha is marked for both sides. This mark is called "alpha relay." Unless indicated otherwise, the start and stop of feedback is marked by two arrows. In some illustrations, a continuous epoch is shown; in others, samples which are discontinuous are selected.

AR

/~.R.

t 11~

I

&R I ......

a II , . . . . . . . .

t*tl]la

~

Gd[JA . . . . .

LJlh~ ~

_

_

.

.

~

. , a , , ~ s ~,_. - ~ I I I A

. . . . . . .

o,% A.R. Fig. 1. Normal response to feedback. Note the alternation between bilateral alpha spindles and no-alpha intervals at the beginning (top two tracings). The alternation between alpha and noalpha is more stable after habituation (bottom two tracings). Feedback begins with the first arrow and ends with the second. Calibration is 50 taV and 1 second. A.R. indicates "alpha relay."

51

Detection of EEG Abnormalities

OI-P3 A.R.

_ _ ,

,

,

~

~

~

~

~

02O,% J

A.R.

L

oz-P, OI-P3 A.R.

Oz-P4

OI-P3 F - ~ x . ' . ' . ; * , ~ . ~ . ~ v ~ - ~ - ~ - - ~ ~ ~ A.R

~

-

-

-

J

'---~-~

~-,~

. . . . . .

OzFig. 2. Normal response to feedback. Same format as Figure 1.

The first example (Figure 1) is from a normal subject. The stimulus was a picture of a nude woman. The upper two sets of tracings are continuous and show the response to the first four stimuli; the lower two sets are continuous and show the response to the last 9 of 30 stimuli, after habituation had occurred. When feedback begins, controlled by alpha from the left side, a prominent bilateral disturbance of the alternation between alpha and noalpha (the alpha-attenuation cycle) occurs. After recovery from the disturbance (habituation) a regular alternation between alpha and attenuation with feedback stimulation can be seen. Bilateral symmetry is easy to evaluate because of the enhanced stability of the alpha-attenuation cycle during feedback. A similar pattern is seen in more than 50°7o of normals. The second example (Figure 2) is also from a normal subject. It was selected because, though within normal limits, there is a mixture of alpha and other frequencies and a diminished bilateral symmetry which could be appraised visually. Despite this visual impression, the alpha marker shows a normal response. First, a prominent and prolonged attenuation response to the first two stimuli (nude woman) is shown in the upper two sets of tracings. The response to the last 12 out of 30 stimuli in the lower two sets of tracings shows the characteristic stability of the alternation between alpha and attenuation after habituation. Bilateral symmetry of the alpha marker

52

Mulholland and Benson

OI-P 3 - [ ~ . ~ : ! , A.R. 1-.

Oi-F'3 1 ~

~, .

. . . .

.

~'I,'~

.

~ - . < . , ~ - ~ - ~ Z~'+.,~+,~.

~I~.-+-'~-'~,+~,~,~I~--~I~-~-"-~'~4'. '~-

AR.

OI-P 3 i~.,'~' ....... ~ - ~ , ~ , ~ , ~ : e ~ - ~ < ~ . . . . ~ - ~ - . . ~ , ~ . . . . . AR.-[ . . . . 02-p 4 - . ~ - ~ , , , : . ' , ~ t t , ~ , ~ ~ : ~ - 4 , ~ O ~ - - ~ ~-'~-

AR.

~:. . . . .

$

Fig. 3. Atypical hyporeactivity to feedback in a patient with bilateral carotid occlusion. Same format as Figure 2.

is obvious, though asymmetry does occur occasionally as in all normal subjects. Figure 3 is from the record of a patient with a bilateral carotid occlusion. The only atypical feature is reduced reactivity to the stimuli. Because reactivity is reduced, the period of alternation between alpha and no-alpha is brief. This permits the whole series of 27 stimuli to be shown in Figure 3. Low reactivity at onset of feedback is unusual but can be seen in a normal subject who is habituated or who is stimulated with simple or excessively redundant stimuli. This patient showed a diminished reactivity to stimuli which are usually quite evocative for normals such as pictures of nudes, emotional words such as "bitch" or "raped," and a "real person" stimulus. The generalized low reactivity is definitely atypical, yet stability and bilateral symmetry are normal, i.e., it is a "borderline" record. The next patient with an old aphasic disturbance, gait difficulties including right foot drop, was frankly hyporeactive to a variety of pictorial, verbal, and "real person" stimuli (Figure 4). The feedback stimulus in this example was a picture of a nude woman. Bilateral symmetry of the alpha attenuation and stability are good, but the low reactivity is abnormal for this stimulus. The next case had surgical occlusion of the left carotid artery for ruptured left posteriol: communicating artery aneurysm. He had right hemiparesis, alexia, and anomia. Reactivity is prominent but abnormally asymmetrical at the start of feedback (middle set of tracings of Figure 5). The at-

Detection of EEG Abnormalities

A.R.

~-~

.'----'

--'----

"

,---'

53

-'

~ ' ±

'

'

--

. . . .

= .

=

'-

/

OI-P 3 ~ , , / - ~ ' ~ , ' ~ # ~ - - ~ ~ t ~ ' ~ - ' ~ ' ~ u ~ ' ~

A.R.

~

~

='

--'-

--

.

.

A.R.t, , . . _ , .-_ ,;_--, --, ,_,__, O ~ - p4 ~ # ~ - . ~ . , ' ~ - ~ - ~ ' ~ - ~ , ~ * ~ 4 ~ w ~ - ~ ~ ' ~ ~ ~ 4 ~

.

.

=

.

.

.

.~.~.~..~.-,.-~--~~4~

Fig. 4. Abnormal hyporeactivity to feedback in a left-lesioned patient.

tenuation response, though prolonged on the right side, has much shorter duration on the left. This asymmetry is not seen in the resting "eyes-open" EEG (top set of tracings) and in the EEG after habituation to the feedback stimulus (bottom set of tracings). No abnormal wave forms are evident. This patient's abnormally asymmetrical response to verbal and pictorial stimuli will be illustrated later, using a computer-generated graphical display. I,

A.R. ~ '

'

~

I

I

I,

'

~

.

.

.

.

.

~

~

~

k

,

~

/

02-[34 b ~ ` ~ t ~ ` ~ ` 01-P 3 f

~

A.R. [ ~ ' - ' /

!

~##/~?~.~:~.v#~..~`#:~g~.~/#I~#/.V~L.#~#~.~.~#~..~:~#"..M~.~;.~.~.~w/.`:..`...".~.:#.~..4.`~:~ ~"~, ,

~

~

#

. . . .

~

~

~

-

'

. . . . .

~

w ~

~

.... .

.

.

.

.

~ ' ~ ......" .

/

AR

1' ~ - - ' - - '

,__--r~-~--

,

, ~ _ - . i . - , ~

~,..-,--~.,,~ . . . .

Fig. 5. Abnormal asymmetry of the EEG response to feedback stimulation (middle tracings). Desynchronization persists on the right side while alpha returns sooner on the left. In the resting record (top tracings) or after habituation (bottom tracings) asymmetry is much less and within normal limits. Patient has temporal lobe seizures, posttraumatic.

54

Mulholland and Benson

OI-P3

[,

,

A.R. /

o:P3 A.R.

q-P3 A.R.

o:p. o:P3 A.R. Fig. 6. Abnormal asynchrony at the beginning of feedback stimulation.

The next individual had a diagnosis of posttraumatic temporal lobe seizures and illustrates a different kind of asymmetry which was seen primarily at the beginning of feedback (see Figure 6). The asymmetry due to slow waves (7-8-Hz) in the left recording could be seen in the resting record, but becomes more evident during feedback. The right side shows a normal attenuation of synchronous rhythms in response to the stimulus. This accentuates the asynchrony, due to the slow waves appearing on the left side. After the third presentation of the stimulus (a picture of a nude woman), slow activity (7-8-Hz) appears on the left side, while alpha attenuation continues on the right. This is seen clearly in the second set of tracings. The two sets of tracings show 7-8-Hz continuing on the left along with increasing alpha in the left and right EEG's. The difference between left and right is not obvious after habituation because of increased bilateral alpha. The next two cases illustrate the utility of feedback for "on-line" testing of an hypothesis about asymmetry. In both cases the alpha-attenuation cycle was normal on the right. When the feedback was controlled by the right side, the alpha bursts on the left were either poorly developed or did not occur (see Figure 7). The first case had occlusion of the left posterior cerebral artery from its source in the posterior fossa to the occipital area. The top set of tracings was made during an "eyes-open" condition in the dark before feedback. The relatively lower-amplitude alpha on the left is evident.

55

Detection of E E G Abnormalities

A.R. /

t,

OI-P3 L ~ , ~ , , , ~ ~ ~ , . ~ , . ~

.

.

.

,

.

A.R OI-P3 ~.~

F ~ A.R. Ii ~

~..~...~p~.~

-

....

,.~-~

--

u

02- P4-L,~'~s~',~

. ~ 'l ~ t ~ ' ~ - ,~ , ~ , ~ l ~~~ : I ~ J ~ ' '~ ' ~

Fig. 7. Abnormal asymmetryand asynchronygreatly exaggerated during two kinds of feedback: right side controllingthe stimulus (middle tracings) and left side controlling the stimulus(bottom tracings). Alpha relay (A.R.) marks only the one side controlling the stimulus. Patient has occlusionof the left posterior cerebral artery.

During feedback (a spot of white light) controlled by the right side (A.R. marks alpha in the contingent channel) the stable alpha-attenuation cycle is seen on the right; on the left, minimal alpha occurs. One could hypothesize that even though a bilateral attenuation of alpha frequencies occurred in response to stimulation, when the stimulus was off it took much longer for alpha to recover on the left. The right-sided alpha always would occur sooner and cause a stimulus which again blocked alpha bilaterally. Thus, alpha on the left could not develop. This hypothesis can be tested by connecting the left side into the feedback loop. The bottom set of tracings shows what happens. (Note that 0,-P3 is now the top tracing in the bottom set.) Since the stimulus does not occur until sufficient alpha on the left occurs, alpha bursts can develop on the left. However, since alpha on the right occurs sooner once the stimulus is off, long, prominent asymmetrical alpha spindles develop on the right. By applying the feedback to one side and then the other, the degree and kind of asymmetry can be manipulated to emphasize the pathological asymmetry. A similar effect has not been observed in normals. The next case had suffered a left frontal infarct. In Figure 8, an effect similar to that in Figure 7 is shown (A.R. is marking alpha in the contingent channel only)-. The left recording was abnormal during the "eyes-open" in the darkness prior to feedback, as shown in the top set of tracings. The shift from shorter to longer alpha spindles with a shift from "right-contingent" (middle set of tracings) to "left-contingent" (bottom set of tracings) is

56

Mulbolland and Benson

o~-~F A.R

~

01- P3 ~

~ -

~

-

-

~

~

~

~

'

'

~

-

~

~

~

~

~

o2-P4 A.IR.

c~-~

0,% A.R

o2-P4 Fig. 8. Abnormal asymmetry and asynchrony exaggerated with left feedback and right feedback as in Figure 6. Alpha relay (A.R.) marks only the o n e side controlling the stimulus. Patient diagnosed as left frontal infarct.

A.R

AR

o2-~ AR

~-P3

Fig. 9. Abnormal asynchrony and asymmetry of EEG activity. Subject is a nonpatient.

Detection of EEG Abnormalities

57

evident, though not as clearly as in the previous example. Note that the position of 01-P3 and 02-P, are reversed in the bottom set of tracings. Figure 9 presents a rare asymmetry recorded from a nonpatient several years ago in an unrelated study. During the condition of resting with eyes open in the dark (top set of tracings), the record is bilaterally symmetrical. With the onset of feedback stimulation (spot of white light), the record is markedly asymmetrical. With the cessation of feedback (bottom set of tracings), the delayed return of alpha on the left side is evident. This degree of asymmetry has not been seen in any other nonpatient so far, i.e., it has occurred in less than 1 in 1000 nonpatient volunteers who havebeen tested in various experiments since 1960.

Graphical Display The evaluation of reactivity, habituation, and the bilateral similarity during feedback is facilitated by computer-generated graphical displays of the response of the occipital EEG to feedback stimulation. The response of the EEG to a series of feedback stimuli is a disturbance followed by a recovery of the attenuation between intervals of alpha (Ata) and intervals of no-alpha (Atna), as defined by the feedback control system (Boudrot, 1972). This series is extremely variable with eyes open in the dark, i.e., the "before-feedback" condition. With the initiation of Loop-1 feedback (alpha turns the stimulus on; no-alpha turns it off) the intervals of alpha become brief, much less variable, and quite stable over repeated stimuli; the intervals of no-alpha increase and then decrease irregularly to a level near initial values. Though the variance of no-alpha is also reduced by feedback, this reduction is not as marked as for the alpha durations. In previous studies (Mulholland & Runnals, 1962; Mulholland, 1973) it was shown that the disturbance and recovery of the occipital alpha-attenuation cycle could be described by fitting curves to the series of alpha durations and to the series of no-alpha durations using the method of Least Squares. A straight line can be fitted to the alpha (Ata) and a hyperbola to the no-alphas (Atna) to describe the response during feedback. Each best-fit function has statistical measures which determine it: N is the serial number of alpha and no-alpha events. The scatter of actual data around the best-fit lines is shown by the standard error of estimate SE. As shown below, A, B, C, and D are parameters of the functions.

1. Ata = ( A . N

+B)+_SE

2. Atna- (C/N + D) +_ SE

58

Mulholland and Benson

From these best-fit functions, graphs are computed which we call "alertographs." They show the disturbance and recovery o f alpha and no-alpha durations before feedback, during feedback, and for left and right EEGs. For more details concerning the computerized curve-fitting procedures, see G o o d m a n (1973). The following examples in Figures 10 and 11 were taken from single trials from the patient whose EEG is presented in Figure 5. They are intended to illustrate the graphical description o f an abnormal E E G response to two visual stimuli, the word " b r o o m " and the word " b i t c h . " Normal subjects usually show a definite bilateral reactivity, initially with a reduction in reactivity with repetition over trials, but not the unilateral reduction shown here in the left EEG. Normal subjects usually show a greater response to the word " b i t c h " than to the word " b r o o m " if they can read. In Figures 10 and 11 the curves are the estimated durations of intervals of noalpha (Atna). Alpha best-fit functions are not shown• Before feedback, the estimated no-alphas on the right are greater than on the left, and stable• In Figure 10, the response to the word " b r o o m " is shown. At the start o f feedback the estimated value o f no-alpha definitely increases on the right but only slightly on the left. In Figure 11, the second visual stimulus, the word " b i t c h , " produced some response on the right but none on the left. In

R

SEC" 3Z.~

BROOM

BEFORE

DURING

FEEDBRCK

FEEDBRCK

Z4.0 )

!

e.q :.. • . ...........

:• " "

R

L.

o.... " . . . . . . . . . . . . . . . . . . . . . . . . .

•

L.

"'...

"•''';°l''..

................

. ..............

?

0.0

05

10

15 N

2:0 05 .10 SERIRL. ORDER

15

20

25

Fig. 10. Alertograph for patient shown in Figure 5. Reactivity is definite for the right side but is much less on the left. The stimulus was the word " b r o o m . "

Detection of EEG Abnormalities

59

4T

BITCH

SRC" e

3Z.~

BRFDR£

DURING

F£EDBRCK

FREDBRCK

z4 .o

.14 Ix

e.o.

...°°"

R

L

.0

~

.tO

15

D,I

L

........................

ZO 05 10 Sl;:RI RL, ORDErR

::::::::,,s.

15

...............

20

Z5

Fig. 11. Same as Figure 9 except that the reactivity on the left is nil while on the right it is attenuated. Stimulus is the word " b i t c h . "

Figure 10, an abnormal asymmetry of response is seen; in Figure 11, an abnormal hyporeactivity is illustrated.

DISCUSSION The use of a closed-loop feedback method has advantages for estimating EEG reactivity. First, the alpha-blocking response is made to recur again and again in a relatively stable way. In our terms, visual feedback stabilizes the alpha-attenuation cycle bilaterally. Against the background of a regular recurring response, aberrations from the expected pattern are more observable. For instance, asymmetries of reactivity if they occur can be visually identified because of the regular alternation of alpha and noalpha intervals on the unaffected side. Feedback emphasizes and may exaggerate asymmetries of EEG reactivity and EEG asynchrony. The evaluation of EEG reactivity is more accurate since the response pattern is more predictable because of feedback, and many responses are sampled. Reactivity can be quantitatively evaluated with computer methods, but the experienced electroencephalographer familiar with this method can appraise reactivity from the ink record.

60

Mulholland and Benson

The cases illustrated here were all diagnosed as having focal, localized brain lesions. Such a sample would be expected to show a higher proportion o f bilateral asymmetries. On the other hand, the feedback evaluation involved only the parietal-occipital E E G . We did not use feedback f r o m other E E G locations, which could have increased the n u m b e r and kind o f abnormalities which would be observed. We examined only alpha rhythms. Much more research on feedback f r o m different E E G locations and with E E G rhythms at other frequencies needs to be done. The present report is only a beginning. The full potential for feedback method in clinical E E G is not yet known. The results are encouraging and indicate that further research on this topic is reasonable and likely to yield useful results. The use of this feedback method to test stimulus discrimination was not shown, though such tests could be useful. Also tests of voluntary control o f the response to feedback like those shown with normals (Mulholland, 1973) were not tested. Both s t i m u l u s discrimination and v o l u n t a r y m o d u l a t i o n of the response could be included in a clinical evaluation. More research is required here. For a better estimation o f the a b n o r m a l i t y of feedback effects, normative data is most desirable. However, standardization and reliability of the m e t h o d had to be established before meaningful norms could be established. The present method is sufficiently standardized and reliable so that the collection of normative data could begin, especially if a number o f researchers were to try the method. The effects shown here are not claimed to be diagnostic signs of known reliability. To appraise the frequency of occurrence o f that type of feedback p h e n o m e n a which would assist the diagnostician, m o r e extensive data would be required. However, f r o m the standpoint of clinical E E G , asymmetry of E E G reactivity, asynchrony, and hyporeactivity which are effectively revealed by the feedback method are diagnos,tically relevant.

REFERENCES

Bagchi, B. K. The adaptation and variability of response of the human brain rhythm. Journal of Psychology, 1937, 3, 463-465. Bancaud, J., Hecaen, H., & Lairy, G. C. Modifications de la reactivite E.E.G., troubles des fonctions symboliques et troubles confusionnels dans leslesions hemispheriques localisees. Electroencephalography and Clinical Neurophysiology, 1955, 7, 179. Blum, R. H. Alpha-rhythm responsiveness in normal, schizophrenic and brain-damaged persons. Science, 1957, 126, 749-750. Boudrot, R. An alpha detection and feedback control system. Psychophysiology, 1972, 9, 461-466. Cobb, W. A. The EEG of lesions in general. In D. Hill & G. Parr (Eds.), Electroencephalography. A symposium on its various aspects. New York: MacMillan, 1963.

Detection of EEG Abnormalities

61

Fischg01d, H., & Mathis, P. Obnubilations, Comas et Stupeurs. Paris; Masson et cie, 1959. Gastaut, H. L. L'activite electrique cerebral en relation avec les grande problemes psychologiques. L'Annee Psychologique, 1949, 51, 61-88. Goodman, D. ALFIE: Collection of EEG alpha under feedback control using time series analysis. Psychophysiology, 1973, 10, 437-440. Hill, D. The EEG in psychiatry. In D. Hill & G. Parr (Eds.), Electroencephalography. A symposium on its various aspects. New York: MacMillan, 1963. Holloway, F., & Parsons, O. Habituation of the orienting reflex in brain-damaged patients. Psychophysiology, 1971, 8, 623-634. Jus, A., & Jus, C. Etude de l'extinction par repetition de l'expression EEG du reflexe d'orientation et de l'action du frein externe dur les reactions EEG aux differents stimuli chez l'homme. Electroencephalography and Clinical Neurophysiology, 1960, Suppl. 13, 321-333. Kooi, K. A., & Thomas, M. H. Electronic analysis of cerebral responses to photic stimulation in patients with brain damage. Electroencephalography and Clinical Neurophysiology, 1958, 10, 417-424. Liberson, W. T. Functional electroencephalography in mental disorders. Diseases of the Nervous System, 1944, 5, 357-364. Milstein, V., Stevens, J., & Sachdev, K. Habituation of the alpha attenuation response in children and adults with psychiatric disorders. Electroencephalography and Clinical Neurophysiology, 1969, 26, 12-18. Mulholland, T. Feedback electroencephalography. Activitas Nervosa Superior, (Prague) 1968, 10, 410-438. Reprinted in Barber, T. et al (Eds.), Biofeedback and Self Control: A Reader. Chicago: Aldine, 1970. Mulholland, T. Objective EEG methods for studying covert shifts of visual attention. In F. J. McGuigan & R. A. Schoonover (Eds.). The Psychophysiology of Thinking. New York: Academic Press, 1973. Mulholland, T., & Gascon, G. A. quantitative index of the orienting response in children. Electroencephalography and Clinical Neurophysiology, 1972, 33, 295-301. Mulholland, T., & Runnals, S. Evaluation of attention and alertness with a stimulus-brain feedback loop. Electroencephalography and Clinical Neurophysiology, 1962, 14, 847-852. Wells, C. E. Response of alpha waves to light in neurologic disease. Archives of Neurology, 1962, 6, 478-491. (Received December 15, 1974)