350
Electroencephalography and clinical Neurophysiology, 83 (1992) 350-357 © 1992 Elsevier Scientific Publishers Ireland, Ltd. 0013-4649/92/$05.00
EEG91128
Maturation of the coherence of EEG activity in normal and learning-disabled children Erzs6bet Marosi, Thalla Harmony, Luis Sfinchez, Jacqueline Becker, Jorge Bernal, Alfonso Reyes, Ana Eugenia Diaz de Le6n, Mario Rodrlguez and Thalla Fernfindez National University of Mexico, ENEP, Iztacala (Mexico) (Accepted for publication: 28 July 1992)
Summary The age effect on coherence has been studied in control (98) and learning-disabled (LD, 54) school-aged children (from 6.0 to 16.8 years old). The EEG recordings were made at rest in 15 leads, and 105 pairwise combinations for coherence were calculated (each lead was compared with all the rest) for delta, theta, alpha, beta and total frequency bands. A significant increase of coherence with age was found in both groups, with a different pattern of maturation. In the control group, a significant increase with age was found in the coherences between posterior regions and vertex (Cz). A significant decrease with age in the coherence between frontal areas was observed, especially in the theta band. The LD group showed a different pattern: no significant relation with age was found in the coherence between any lead and vertex. A high effect of age on coherence between temporal regions was observed with a predominance of the left side in comparison with the contralateral and the ipsilateral. No decrease in frontal coherence was found: in the same region where the control group showed negative values with age, the LD groups had no age effect. The results obtained are discussed as differences in brain organization, in myelogenesis and synaptogenesis and an explanation of the etiology of LD is proposed. Key words: EEG coherence; Learning disability; EEG maturation; Coherence development; EEG frequency
Q u a n t i t a t i v e E E G m e t h o d s have p r o v e n to b e useful in r e v e a l i n g f u n c t i o n a l a l t e r a t i o n u n d e r l y i n g l e a r n i n g disabilities (LDs; J o h n et al. 1983; H a r m o n y et al. 1990b). C o h e r e n c e analysis o f E E G is u s e d to study the c o u p l i n g b e t w e e n cortical r e g i o n s a n d to m e a s u r e the c o v a r i a t i o n b e t w e e n two r e c o r d i n g s as a function of frequency. H i g h c o h e r e n c e b e t w e e n E E G signals has b e e n i n t e r p r e t e d as e v i d e n c e o f a s t r u c t u r a l a n d funct i o n a l c o n n e c t i o n b e t w e e n c o r t i c a l a r e a s u n d e r l y i n g the r e c o r d i n g e l e c t r o d e s ( F e i n et al. 1988). A s t h e effect o f age is very s t r o n g on the m e a s u r e m e n t s o f p o w e r (John et al. 1980), we a s s u m e t h a t c o h e r e n c e m e a s u r e m e n t s w o u l d also b e s t r o n g l y a f f e c t e d by age. Previous studies on t h e m a t u r a t i o n of c o h e r e n c e m e a s u r e m e n t s a r e very scarce. G a s s e r et al. (1987) c o m p a r e d n o r m a l a n d mildly r e t a r d e d c h i l d r e n at rest a n d d u r i n g p e r f o r m a n c e o f a visual task. T h e y f o u n d a slight i n c r e a s e in c o h e r e n c e with age for a very n a r r o w r a n g e ( 1 0 - 1 3 years). T h a t c h e r et al. (1986) s t u d i e d t h e E E G c o h e r e n c e o f c h i l d r e n f r o m 5 to 16 y e a r s old, b u t did n o t c a l c u l a t e h e age effect. B o t h a u t h o r s s u g g e s t e d that m e n t a l d a t i o n or l o w e r i n t e l l i g e n c e a r e r e l a t e d to h i g h e r
c o h e r e n c e m e a s u r e m e n t s ( T h a t c h e r a n d W a l k e r 1985; G a s s e r et al. 1987), a finding which a p p a r e n t l y c o n t r a dicts t h e i n c r e a s e o f c o h e r e n c e with age. A s u b s e q u e n t p a p e r o f T h a t c h e r et al. (1987) s t u d i e d t h e age effect o n E E G c o h e r e n c e s from 2 y e a r s old to early adulthood, finding 5 stages of c o h e r e n c e m a t u r a t i o n with a d i f f e r e n t cortical p a t t e r n , w h e r e left f r o n t o - o c c i p i t a l a n d f r o n t o - t e m p o r a l regions p r e c e d e d right h o m o l o gous areas, a l t h o u g h the right f r o n t o p o l a r m a t u r e d b e f o r e t h e left one. H a r m o n y (1984) c a r r i e d out an i n t e r h e m i s p h e r i c c o r r e l a t i o n coefficiency study on 110 c h i l d r e n f r o m 5 to 12 y e a r s old, as well as on adults. She f o u n d a lower m e a n o f the c o r r e l a t i o n values b e t w e e n right a n d left d e r i v a t i o n s for the c h i l d r e n t h a n for adults. T h e first goal o f this p a p e r is to quantify t h e age effect on c o h e r e n c e b e t w e e n d i f f e r e n t cortical a r e a s in c h i l d r e n f r o m 6.0 to 16.8 y e a r s old. T h e s e c o n d goal is to see if L D c h i l d r e n have d i f f e r e n t E E G m a t u r a t i o n p a t t e r n s t h a n n o r m a l children, as m e a s u r e d by c o h e r ence.
Subjects and method Erzs6bet Marosi, Apartado Postal 82, "qdo de M6xico, C.P. 52971, Zaragoza
This study was c a r r i e d out with s c h o o l - a g e c h i l d r e n r a n g i n g in age f r o m 6.0 to 16.8 years. T h e c o n t r o l
EEG COHERENCE
MATURATION
group consisted of 98 children, while 54 had learning disabilities. Learning-disabled children were sent for this study from a behavioral clinic and learning disability was confirmed by the following criteria: (a) a questionnaire was filled out by the children's parents providing information on each child's academic achievement; (b) almost all learning-disabled children had repeated at least 2 academic years; (c) their parents and teachers considered they had learning problems. Neurological testing exploration was carried out, in some cases finding a mild deficit in visual or auditory acuity. These children, as well as children with epileptic seizures or other neurological problems, were eliminated from the study. The participating subjects were healthy, with normal results from neurological examination, and free of medication at the time of the study. All children selected had IQ values higher than 85 as measured by the Wechsler Intelligence Scale. The distribution of age by sex for the two groups was balanced. In the control group we had 44 boys and 54 girls with mean ages of 9.7 and 10.3 years respectively, while the LD group consisted of 34 boys and 20 girls with mean ages of 9.9 and 10.2 years. The EEGs were recorded in a shielded, dimly lit room. The recordings were made at rest with eyes closed and with subjects awake and in a sitting position, with the head supported by the back of a comfortable chair. All E E G s were made without sedation, even those of the younger LD children. The E E G s were visually monitored and segments with artifact were eliminated. Recordings were obtained from 15 monopolar leads with silver-silver chloride electrodes positioned at F3, F4, C3, C4, P3, P4, O1, 02, F7, F8, T3, T4, T5, T6, and Cz with linked earlobes as reference (10-20 international). Electrode resistances were below 5 kO. E O G activity was constantly monitored. The E E G s were recorded by a MEDICID-03 computer system. The amplifiers had a bandwidth of 0.5-30 Hz. Twenty-four segments of 2.56 sec, that is 61.44 sec of artifact-free E E G were obtained for each subject, with a sampling rate of 100 Hz. The E E G data were submitted to the Fast Fourier Transform (FFT) to calculate the cross-spectral matrix between all leads for each E E G segment. Average values were used to compute coherence for the following frequency bands: delta, 1.5-3.5 Hz; theta, 3.5-7.5 Hz; alpha, 7.5-12.5 Hz; beta, 12.5-19 Hz (Alvarez et al. 1987; Harmony et al. 1990). Coherence was defined as: C°h(f) 2 =
IG×y(f) 12 Gx~(f) Gyy(f)
where G~x and Gyy were the power spectral density functions of x(t) and y(t) respectively, and Gxy was the cross-spectral density function between x(t) and y(t) (Ba§ar 1980).
351
Coherence was computed for all pairwise combinations of the 15 leads, resulting in 105 combinations for each frequency band. The data obtained was z-transformed by the Fisher transformation: z = 0.5 In (1 + ~ ) / ( 1 - ~)
where ~ was the coherence value. The transformation was made in order to ensure the gaussian distribution of the measurements of coherence. There were some doubts when it came to choosing the statistical model for the regression against age. Plotting the z values against age generally gave a clear linear distribution, although in some cases the decision between quadratic or linear distribution was difficult. For this reason both linear and quadratic regressions against age for each pairwise combination of z values of coherence were calculated in all bands.
Results
Comparing the results of the linear and the quadratic regressions, it was observed that the linear model explained a higher percentage of the variance in almost all areas compared. Fig. 1 shows the data distribution of the z-transformed coherence values for the C4-O2 comparison in the delta band for the control group, where highly significant age effect was found by the linear but not by the quadratic regression. A marked increase of the coherence values with increase of age could be observed. In Fig. 2 the distribution of z values plotted against age are shown for P3-P4 in the alpha band for the LD group. No significant age effect was found by the two types of regression. In this figure a very slight tendency towards increase of coherence with age could be observed. As the quadratic regression better explained the variance of the data only in 5 of 525 comparisons, we considered the linear model to be more adequate than the quadratic one, and we describe below the results obtained by the linear regression. Fig. 3 illustrates the effect of age calculated by linear regression in the delta band for the control and LD groups, in all significant pairwise combinations. The upper figure shows the significant age effects seen in the control group, while the lower figure shows the age affects that were observed in the LD group. In both groups coherence increased with age, except in the comparisons F7-F4 and F8-C3 in the control group. The dotted lines show the significant age effect observed in both groups, and the continuous lines show the group-specific age effect that was seen only in one but not in the other group. There were many more group peculiarities than similarities in all bands. This
352
E. M A R O S I E T A L
CONTROL
0.z F ]
0,6~ r
C 0 H E " E
N
0,5 I 0,4~
,
• -
0,3
E
....
t
,
FZ..........-r3
• "
"
.
.'.
"
"
..
0.1 i O]II
6
~ lilil
II
7
:': ll~ll
8
•
. II
9
I~ il
.'." Ill
10
II
IlllLilll
IIIIt
11 12 AGE
18
II111
14
'lit
15
III
%:T4
Ill~
16
17
Fig. 1. Data distribution of the z values of coherence plotted against age for C4-O2 in the delta band and in the control group, where highly significant ( P = 0.000l) age effect was found by linear but not by quadratic regression.
figure shows that in the control group, the vertex (Cz) shared high age influence with the parietal and occipital regions in a rather symmetrical form. In the LD group, the vertex compared with other regions had no age effect on coherence, but the left temporal and occipital areas both showed high age influence, especially at interhemispheric comparisons. In the control group, negative slopes for two frontal comparisons are seen, in contrast to the LD group where all slopes are positive. In Fig. 4 the effect of age in both groups for the theta band is illustrated. We can observe that in the theta band age has less influence on coherence values than it does in the delta band. In the same form as in the delta band, coherence increased with age, except in some frontal comparisons (F7-O1, F3-F4, F3-T4, F8C3) in the control group where coherence decreased with age. Here too, the similarities are less than the
LD
0,7
Fig. 3. Significant age effect on coherences in the delta band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
0,6
C 0 H E R E
0,5 0,4
.:"i."
N C E
.-
0,2
.'.-
.
. "..
0,1 0
I l l l l l l l l [ l l l l / l l l l l l l l l l l l J l l l f l l l l t l l l l [ I ? l l l l
7
8
9
10
11
12
18
14
IIIItll
16
16
17
AGE
Fig. 2. Data distribution of the z values of coherence plotted against age for P 3 - P 4 in the alpha band and in the LD group, where no significant age effect was found either by linear or quadratic regressions.
group-specific influences. In the control group the vertex-centered and in the LD groups the left temporalcentered distribution is evident. Fig. 5 shows the effect of age in the alpha band. We can observe the symmetrical age effect and the strong influence of age in the right parietal comparisons in the control group. Also, a negative age effect between F8 and T4 and abundant interhemispheric comparisons were found. The influence of age is reflected more in the delta and alpha bands than in the other bands. In the LD group the same left-temporal and occipital orientation of the age influences, as observed in the
EEG COHERENCE MATURATION
353
other bands, was seen. The L D group manifested more numerous frontal coherences affected by age than did the control group. Fig. 6 shows the significant effect of age in the beta band. Although in lesser quantity, the same differential age effects are observed as in the other bands: in the control group the effect of age on coherences is centered around Cz in a symmetrical way, while in the LD group the left-temporal and occipital comparisons are affected. In addition, negative slope is observed in frontal F8-T4 comparison in the control group, while in
CONTROL
F7
F3
°'I'.,°°•, "% • O,o ..% ".........
:.F4
I1 T3 ............,C3 "".. CZ'~C4F-!::~"" T4 ................ [ ..... W
CONTROL
''': .
:
'%
." "'.%.
F7
F3~
-
_
F8
{ ".. "-. ~ "" --"
-.
°"
.,"
F4 / F 8
LD
0,
P3
LD
"%'°''...,...°....
T3 lc %;. i
~ CZ
F4
F8
C4~
T4
P41/6
01 Fig. 4. Significant age effect on coherences in the theta band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
P4'
!ii6
........ ..o..°." °°'"°"°""
Fig. 5. Significant age effect on coherences in the alpha band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
the LD group abundant frontal age influences with positive slopes can be seen. The only comparison where age affected coherence in a similar way for both groups is between T3 and O1. Fig. 7 shows the effect of age for the total band. We observe few similarities in age effects between both groups and a large number of group-specific age effects, as in the case of the other bands. The control group shows a vertex-centered symmetrical orientation,
354
E. MAROSI ET AL.
while in the L D group left-temporal, i n t e r h e m i s p h e r i c c o m p a r i s o n s reflect the effect of age. S u m m i n g up the results of all the previous figures, we can state that a different m a t u r a t i o n a l p a t t e r n for the two groups is present. T h e control g r o u p has a C z - c e n t e r e d symmetrically a r r a n g e d increase of coherence values with increase in age. W e can also observe high age effects on the i n t e r h e m i s p h e r i c P3-P4 coherence a n d where P4 is included, as well as negative
CONTROL
F7,,,
F,3-,,,
F4
T 3 ~ ~
F7
F3
T3 "%,.
C3
CZ
T4
T5 \ P3~,,/×\ ~ P 4 i
CONTROL
F4
F8
C4
T4
Fi
~T6
I'
LD
%.
T5 '-...
76
8 T4
T5 ~
LD
P3
P4 i ~
01
O2
T6
Fig. 7. Significant age effect on coherences in the total band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
F3,
F•k
7F4
F8
IS "''~'.01v Fig. 6. Significant age effect on coherences in the beta band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
slopes with age in frontal i n t e r r e l a t i o n s , mainly in the theta band. O n the o t h e r h a n d , in the L D g r o u p there is n o age effect in any of those c o h e r e n c e s which i n c l u d e vertex. L D c h i l d r e n are c h a r a c t e r i z e d by a m a t u r a t i o n a l process c e n t e r e d o n the left t e m p o r a l a n d occipital areas, resulting in a m a r k e d asymmetrical m a t u r a t i o n of coh e r e n c e s a n d high age effect o n frontal coherences, a process that is n o t seen in the control group. W h e n c o m p a r i n g the two groups, we can observe that age affected some regions in a similar way for b o t h groups,
EEG COHERENCE MATURATION although there were more group-specific differences than similarities, and in all bands the pattern of maturation was completely different for the two groups.
Discussion
It is generally considered that E E G coherence is a measurement of cortical interconnectivity or coupling. Thatcher et al. (1986) reviewed several anatomical analyses in relation to coherence, estimating that no more than 1% of cortical fibers arise in the thalamus, 2 - 4 % in the contralateral cortex and approximately 95% in the ipsilateral cortex. These cortico-cortical association fibers have an average length of several centimeters, while the intercolumnar connections are of short range and are less than 1 mm in length. These authors propose that gross E E G cortical phenomena are produced by long cortico-cortical fibers. To explain the coherence they present a 2-compartment model based on the fact that the neocortex contains two types of cell: long-axoned piramidal or Golgi type I cells and short-axoned Golgi type II cells. If the density of Golgi type II cells is 10-100 times higher than that of Golgi type I ceils, a major contribution of the short axonal connections to coherence is expected for short scalp electrode distance. At long electrode separation the contribution of short axonal connections to coherence should be minimal, and an increase in the coherence would be produced by long axonal connections. Since coherence values reflect this cortical anisotropy, we may say that the coherence reflects cortical connectivity. Theoretically, an increase in coherence may be the result of two cortical areas directly coupled or by a third cortical or subcortical source related to both, or by a cascade of regions functionally interrelated with the recorded areas. Another cause for high coherence value might be the volumen conduction producing a smooth decrease of coherence with increased electrode separation, as coherence decays as an exponential function of distance. The propagation of E E G rhythms also depends on the spatial properties of the cortical neuronal networks, as the cortex is not a homogeneous medium. Different systems of intracortical fibers, which involve much larger distances than the diameter of a column, assure the interconnectivity of several columns (Lopes da Silva 1987). All these factors are reflected in the coherence of E E G and, therefore, we may say that coherence is a measure of the functional organization of E E G activity, which reveals structural characteristics of the brain. On the other hand developmental changes in the brain are mainly determined by myelination and synaptogenesis. Myelination produces more efficient interconnections between different cerebral regions, and consequently increased coherence values may be ex-
355 pected. Synaptogenesis means a greater quantity of synapses and better interaction between different structures, which may be reflected in increased coherence values. The goal of our paper was to look for the influence of age on coherence values, based on the belief that similar age effect on coherences for the control and LD groups would reveal similar brain organizational and maturational processes. But this was not the case. Considering the age effect as observed in our investigation, we can conclude that the "normal" maturational process (seen in our 98 control children) seems to have 3 important features: (a) A marked increase in coherence with age in central, temporal and posterior interconnections, especially in the delta and alpha bands. The increase of coherence with age may be the consequence of myelination in process, as myelogenesis is not completely finished until the second decade of life, and our subjects are in the age range of late childhood, when these changes have been shown to occur (Yakovlev and Lecours 1967). (b) A symmetrically distributed age effect on coherence, with abundant involvement of Cz in all bands. Our recordings are made in awake, relaxed children with the eyes closed, so maximum alpha activity is expected. These symmetrical age influences related to the vertex may be due to the developmental increase of coupling in the thalamo-cortical projections, as the thalamo-reticular nucleus has a very important role in the production and synchronization of alpha rhythm (Steriade et al. 1990). These findings may also be due to the development of cortico-cortical association fibers, including the corpus callosum. (c) A decrease in coherence with the increase of age in frontal comparisons, especially in the theta band. Decreased frontal coherences may be related to increased cortical differentiation (Thatcher et al. 1986). In 1979, Huttenlocker found that neuronal and synaptic density is at its maximum during the first 2 years of life and then gradually declines between the 2nd and 16th years, in all cortical areas. Our subjects are 6-16 years old, and for this reason the frontal negativity observed may be due to the fact that our study was carried out at the exact time that this reduction in frontal synaptic density occurs. Nevertheless, the resuits obtained by Goldman-Rakic (1987) contradict this assumption, since she did not find a different time-table or rate of synaptic elimination for different cortical areas. Available data on human synaptogenesis are limited and differential measurements on cortical maturation are quite scarce as well as controversial. In the present state of neuroscience, it is impossible to determine exactly what this frontal negativity in coherence values is due to. How does coherence relate to behavior, and why
356
does this pattern of coherence maturation reflect learning disability? This question is very difficult to answer, as we know very little about the basic cerebral mechanism involved in learning disability. Correlations between intelligence and brain development say little about real links between the brain and behavioral development because they are too general (Fisher 1987). Obviously, brain development is related to cognitive functioning. Rapid and periodic behavioral changes occur between the ages of 6 and 16. These behavioral changes are not related to the maximum synaptic and neuronal density in the brain, and this fact implies that there are other basic maturational processes, such as dendritic growth or improved synaptic efficiency, which permit the achievement of complete behavioral competence. There are different hypotheses that seek to explain LD from the neurophysiological point of view. One theory states that LD may be due to a maturational lag, since at older ages academic achievement may become normal. Considering the spectral parameters of EEG, John et al. (1983) found two types of electrical activity in an LD sample. A group of children had normal maturation of EEG, but it corresponded to the parameters of younger children. Another group of children had abnormal spectral measurements and maturational processes that one assumes to be different than those in the normal children. Our results definitely demonstrate that the LD children have a completely different pattern of brain maturation, and that they differ from normal children more in a qualitative way than in a quantitative one. Thus learning disability may be due to a maturational abnormality or a variance of the normal processes, which may cause all of these cognitive problems to which we give the common name of "learning disability." Another theory on the etiology of learning disability states that inverted h e m i s p h e r i c lateralization (Geschwind and Galaburda 1985) or a failure in the interhemispheric transfer (Gross-Glenn and Rothenberg 1984) may be the cause of this problem. In fact, the way in which the coherence values of our LD children are affected by age shows signs of a different hemispheric arrangement: our normal children show symmetrically distributed central and posterior maturational effect, while the LD children have the age effect centered on the left temporal areas. The different age effect observed in the frontal region (negative slope in the control group and abundant frontal age effect in the alpha band with positive slope in the LD group) may suggest massive frontal functional differences between the two groups. Motivational and attentional processes, so important for adequate academic achievement, may be deactivated by this anomalous functioning of the frontal cortex and prevent LD children from achieving a steady and continuous perfect
E. MAROSI ET AL.
execution. Direct frontal and limbic interaction is demonstrated by Mesulam (1990). All these speculations continue to be dubious so long as the nature of learning disability itself is not ~letermined. For the moment, we can state only that a different developmental mechanism is obvious in our LD group. Before concluding, we must make it clear that there is another important factor which has a well-demonstrated influence on electroencephalographic measurements: gender (Harmony et al. 1990a). Looking for gender effect in this study, we found very similar means, deviations and variances of age according to sex for the two groups, so we may infer that the sex variable was controlled and affected our results in a balanced way, and that these differences in coherences by groups are in fact due mainly to age. This project was supported by Grant Number IN205689 from DGAPA, UNAM, and by funds granted by the Government of the State of Mexico. Our thanks are due to Mr. Sandor John for the valuable corrections of this paper.
References Alvarez, A., Vald6s, P. and Pascual, R. EEG developmental equations confirmed for Cuban schoolchildren. Electroenceph. clin. Neurophysiol., 1987, 67: 330-332. Ba§ar, E. EEG Brain Dynamics. Relation between EEG and Brain Evoked Potentials. Elsevier/North-Holland Biomedical Press, Amsterdam, 1980: 67. Fein, G., Raz, J., Brown, F.F. and Merrin, E.L. Common reference coherence data are confounded by power and phase effects. Electroenceph. clin. Neurophysiol., 1988, 69: 581-584. Fisher, K.W. Relation between brain and cognitive development. Child Dev., 1987, 58: 623-632. Gasser, T., Jennen-Steinmetz, C. and Verleger, R. EEG coherence at rest and during a visual task in two groups of children. Electroenceph. clin. Neurophysiol., 1987, 67: 151-158. Geschwind, N. and Galaburda, A.M. Cerebral lateralization. Biological mechanism, associations and pathology. Arch. Neurol., 1985, 42: 634-654. Goldman-Rakic, P.S. Development of cortical circuitry and cognitive function. Child Dev., 1987, 58: 601-622. Gross-Glenn, K. and Rothenberg, S. Evidence for deficit in interhemispheric transfer of information in dyslexic boys. Int. J. Neurosci., 1984, 24: 23-25. Harmony, T. Neurometric Assessment of Brain Dysfunction in Neurological Patients. Lawrence Erlbaum, Hillsdale, NJ, 1984: 363369. Harmony, T., Marosi, E., Diaz de Le6n, A., Becker, J. and Fernandez, T. Effect of sex, psychosocial disadvantages and biological risk factors on EEG maturation. Electroenceph. clin. Neurophysiol., 1990a, 75: 482-491. Harmony, T., Marosi, E., Hinojosa, G., Becker, J., Rodrlguez, M., Reyes, A. and Rocha, C. Correlation between EEG spectral parameters and an educational evaluation. Int. J. Neurosci., 1990b, 54: 147-155. Huttenlocker, P.R. Synaptic density in human frontal cortex. Developmental changes and effects of aging. Brain Res., 1979, 163: 195-205.
EEG COHERENCE MATURATION John, E.R., Ahn, H., Prichep, L., Trepetin, M., Brown, D. and Kaye, H. Developmental equations for the electroencephalogram. Science, 1980, 210: 1255-1258. John, E.R., Prichep, L., Ahn, H., Easton, P., Friedman, J. and Kaye, H. Neurometric evaluation of cognitive dysfunctions and neurological disorders in children. Prog. Neurobiol., 1983, 21: 239-290. Lopes da Silva, F. Dynamics of EEGs as Signals of Neuronal Populations: Models and Theoretical Consideration. In: E. Niedermeyer and F. Lopes da Silva (Eds.), EEG: Basic Principles, Clinical Applications and Related Fields. Urban and Schwarzenberg, Baltimore, MD, 1987: 15-29. Mesulam, M.M. Large-scale neurocognitive networks and distributed processing for attention, language and memory. Ann. Neurol., 1990, 28: 597-613. Steriade, M., Gloor, P., Llin~s, R.R., Lopes da Silva, F.M. and
357 Mesulam, M.M. Basic mechanism of cerebral rhythmic activities. Electroenceph. clin. Neurophysiol., 1990, 76: 481-508. Thatcher, R.W. and Walker, R.A. EEG coherence and intelligence in children. Electroenceph. clin. Neurophysiol., 1985, 61: S161. Thatcher, R.W., Krause, P.J. and Hrybyk, M. Cortico-cortical associations and EEG coherence: a two-compartmental model. Electroenceph, clin. Neurophysiol., 1986, 64: 123-143. Thatcher, R.W., Walker, R.A. and Giudice, S. Human cerebral hemispheric development at different rates and ages. Science, 1987, 236: 1110-1113. Yakovlev, P.I. and Lecours, A.R. The myelogenetic cycles of regional maturation of the brain. In: A. Minkowski (Ed.), Regional Development of the Brain in Early Life. Blackwell Scientific, Oxford, 1967: 3-70.
Electroencephalography and clinical Neurophysiology, 83 (1992) 350-357 © 1992 Elsevier Scientific Publishers Ireland, Ltd. 0013-4649/92/$05.00
EEG91128
Maturation of the coherence of EEG activity in normal and learning-disabled children Erzs6bet Marosi, Thalla Harmony, Luis Sfinchez, Jacqueline Becker, Jorge Bernal, Alfonso Reyes, Ana Eugenia Diaz de Le6n, Mario Rodrlguez and Thalla Fernfindez National University of Mexico, ENEP, Iztacala (Mexico) (Accepted for publication: 28 July 1992)
Summary The age effect on coherence has been studied in control (98) and learning-disabled (LD, 54) school-aged children (from 6.0 to 16.8 years old). The EEG recordings were made at rest in 15 leads, and 105 pairwise combinations for coherence were calculated (each lead was compared with all the rest) for delta, theta, alpha, beta and total frequency bands. A significant increase of coherence with age was found in both groups, with a different pattern of maturation. In the control group, a significant increase with age was found in the coherences between posterior regions and vertex (Cz). A significant decrease with age in the coherence between frontal areas was observed, especially in the theta band. The LD group showed a different pattern: no significant relation with age was found in the coherence between any lead and vertex. A high effect of age on coherence between temporal regions was observed with a predominance of the left side in comparison with the contralateral and the ipsilateral. No decrease in frontal coherence was found: in the same region where the control group showed negative values with age, the LD groups had no age effect. The results obtained are discussed as differences in brain organization, in myelogenesis and synaptogenesis and an explanation of the etiology of LD is proposed. Key words: EEG coherence; Learning disability; EEG maturation; Coherence development; EEG frequency
Q u a n t i t a t i v e E E G m e t h o d s have p r o v e n to b e useful in r e v e a l i n g f u n c t i o n a l a l t e r a t i o n u n d e r l y i n g l e a r n i n g disabilities (LDs; J o h n et al. 1983; H a r m o n y et al. 1990b). C o h e r e n c e analysis o f E E G is u s e d to study the c o u p l i n g b e t w e e n cortical r e g i o n s a n d to m e a s u r e the c o v a r i a t i o n b e t w e e n two r e c o r d i n g s as a function of frequency. H i g h c o h e r e n c e b e t w e e n E E G signals has b e e n i n t e r p r e t e d as e v i d e n c e o f a s t r u c t u r a l a n d funct i o n a l c o n n e c t i o n b e t w e e n c o r t i c a l a r e a s u n d e r l y i n g the r e c o r d i n g e l e c t r o d e s ( F e i n et al. 1988). A s t h e effect o f age is very s t r o n g on the m e a s u r e m e n t s o f p o w e r (John et al. 1980), we a s s u m e t h a t c o h e r e n c e m e a s u r e m e n t s w o u l d also b e s t r o n g l y a f f e c t e d by age. Previous studies on t h e m a t u r a t i o n of c o h e r e n c e m e a s u r e m e n t s a r e very scarce. G a s s e r et al. (1987) c o m p a r e d n o r m a l a n d mildly r e t a r d e d c h i l d r e n at rest a n d d u r i n g p e r f o r m a n c e o f a visual task. T h e y f o u n d a slight i n c r e a s e in c o h e r e n c e with age for a very n a r r o w r a n g e ( 1 0 - 1 3 years). T h a t c h e r et al. (1986) s t u d i e d t h e E E G c o h e r e n c e o f c h i l d r e n f r o m 5 to 16 y e a r s old, b u t did n o t c a l c u l a t e h e age effect. B o t h a u t h o r s s u g g e s t e d that m e n t a l d a t i o n or l o w e r i n t e l l i g e n c e a r e r e l a t e d to h i g h e r
c o h e r e n c e m e a s u r e m e n t s ( T h a t c h e r a n d W a l k e r 1985; G a s s e r et al. 1987), a finding which a p p a r e n t l y c o n t r a dicts t h e i n c r e a s e o f c o h e r e n c e with age. A s u b s e q u e n t p a p e r o f T h a t c h e r et al. (1987) s t u d i e d t h e age effect o n E E G c o h e r e n c e s from 2 y e a r s old to early adulthood, finding 5 stages of c o h e r e n c e m a t u r a t i o n with a d i f f e r e n t cortical p a t t e r n , w h e r e left f r o n t o - o c c i p i t a l a n d f r o n t o - t e m p o r a l regions p r e c e d e d right h o m o l o gous areas, a l t h o u g h the right f r o n t o p o l a r m a t u r e d b e f o r e t h e left one. H a r m o n y (1984) c a r r i e d out an i n t e r h e m i s p h e r i c c o r r e l a t i o n coefficiency study on 110 c h i l d r e n f r o m 5 to 12 y e a r s old, as well as on adults. She f o u n d a lower m e a n o f the c o r r e l a t i o n values b e t w e e n right a n d left d e r i v a t i o n s for the c h i l d r e n t h a n for adults. T h e first goal o f this p a p e r is to quantify t h e age effect on c o h e r e n c e b e t w e e n d i f f e r e n t cortical a r e a s in c h i l d r e n f r o m 6.0 to 16.8 y e a r s old. T h e s e c o n d goal is to see if L D c h i l d r e n have d i f f e r e n t E E G m a t u r a t i o n p a t t e r n s t h a n n o r m a l children, as m e a s u r e d by c o h e r ence.
Subjects and method Erzs6bet Marosi, Apartado Postal 82, "qdo de M6xico, C.P. 52971, Zaragoza
This study was c a r r i e d out with s c h o o l - a g e c h i l d r e n r a n g i n g in age f r o m 6.0 to 16.8 years. T h e c o n t r o l
EEG COHERENCE
MATURATION
group consisted of 98 children, while 54 had learning disabilities. Learning-disabled children were sent for this study from a behavioral clinic and learning disability was confirmed by the following criteria: (a) a questionnaire was filled out by the children's parents providing information on each child's academic achievement; (b) almost all learning-disabled children had repeated at least 2 academic years; (c) their parents and teachers considered they had learning problems. Neurological testing exploration was carried out, in some cases finding a mild deficit in visual or auditory acuity. These children, as well as children with epileptic seizures or other neurological problems, were eliminated from the study. The participating subjects were healthy, with normal results from neurological examination, and free of medication at the time of the study. All children selected had IQ values higher than 85 as measured by the Wechsler Intelligence Scale. The distribution of age by sex for the two groups was balanced. In the control group we had 44 boys and 54 girls with mean ages of 9.7 and 10.3 years respectively, while the LD group consisted of 34 boys and 20 girls with mean ages of 9.9 and 10.2 years. The EEGs were recorded in a shielded, dimly lit room. The recordings were made at rest with eyes closed and with subjects awake and in a sitting position, with the head supported by the back of a comfortable chair. All E E G s were made without sedation, even those of the younger LD children. The E E G s were visually monitored and segments with artifact were eliminated. Recordings were obtained from 15 monopolar leads with silver-silver chloride electrodes positioned at F3, F4, C3, C4, P3, P4, O1, 02, F7, F8, T3, T4, T5, T6, and Cz with linked earlobes as reference (10-20 international). Electrode resistances were below 5 kO. E O G activity was constantly monitored. The E E G s were recorded by a MEDICID-03 computer system. The amplifiers had a bandwidth of 0.5-30 Hz. Twenty-four segments of 2.56 sec, that is 61.44 sec of artifact-free E E G were obtained for each subject, with a sampling rate of 100 Hz. The E E G data were submitted to the Fast Fourier Transform (FFT) to calculate the cross-spectral matrix between all leads for each E E G segment. Average values were used to compute coherence for the following frequency bands: delta, 1.5-3.5 Hz; theta, 3.5-7.5 Hz; alpha, 7.5-12.5 Hz; beta, 12.5-19 Hz (Alvarez et al. 1987; Harmony et al. 1990). Coherence was defined as: C°h(f) 2 =
IG×y(f) 12 Gx~(f) Gyy(f)
where G~x and Gyy were the power spectral density functions of x(t) and y(t) respectively, and Gxy was the cross-spectral density function between x(t) and y(t) (Ba§ar 1980).
351
Coherence was computed for all pairwise combinations of the 15 leads, resulting in 105 combinations for each frequency band. The data obtained was z-transformed by the Fisher transformation: z = 0.5 In (1 + ~ ) / ( 1 - ~)
where ~ was the coherence value. The transformation was made in order to ensure the gaussian distribution of the measurements of coherence. There were some doubts when it came to choosing the statistical model for the regression against age. Plotting the z values against age generally gave a clear linear distribution, although in some cases the decision between quadratic or linear distribution was difficult. For this reason both linear and quadratic regressions against age for each pairwise combination of z values of coherence were calculated in all bands.
Results
Comparing the results of the linear and the quadratic regressions, it was observed that the linear model explained a higher percentage of the variance in almost all areas compared. Fig. 1 shows the data distribution of the z-transformed coherence values for the C4-O2 comparison in the delta band for the control group, where highly significant age effect was found by the linear but not by the quadratic regression. A marked increase of the coherence values with increase of age could be observed. In Fig. 2 the distribution of z values plotted against age are shown for P3-P4 in the alpha band for the LD group. No significant age effect was found by the two types of regression. In this figure a very slight tendency towards increase of coherence with age could be observed. As the quadratic regression better explained the variance of the data only in 5 of 525 comparisons, we considered the linear model to be more adequate than the quadratic one, and we describe below the results obtained by the linear regression. Fig. 3 illustrates the effect of age calculated by linear regression in the delta band for the control and LD groups, in all significant pairwise combinations. The upper figure shows the significant age effects seen in the control group, while the lower figure shows the age affects that were observed in the LD group. In both groups coherence increased with age, except in the comparisons F7-F4 and F8-C3 in the control group. The dotted lines show the significant age effect observed in both groups, and the continuous lines show the group-specific age effect that was seen only in one but not in the other group. There were many more group peculiarities than similarities in all bands. This
352
E. M A R O S I E T A L
CONTROL
0.z F ]
0,6~ r
C 0 H E " E
N
0,5 I 0,4~
,
• -
0,3
E
....
t
,
FZ..........-r3
• "
"
.
.'.
"
"
..
0.1 i O]II
6
~ lilil
II
7
:': ll~ll
8
•
. II
9
I~ il
.'." Ill
10
II
IlllLilll
IIIIt
11 12 AGE
18
II111
14
'lit
15
III
%:T4
Ill~
16
17
Fig. 1. Data distribution of the z values of coherence plotted against age for C4-O2 in the delta band and in the control group, where highly significant ( P = 0.000l) age effect was found by linear but not by quadratic regression.
figure shows that in the control group, the vertex (Cz) shared high age influence with the parietal and occipital regions in a rather symmetrical form. In the LD group, the vertex compared with other regions had no age effect on coherence, but the left temporal and occipital areas both showed high age influence, especially at interhemispheric comparisons. In the control group, negative slopes for two frontal comparisons are seen, in contrast to the LD group where all slopes are positive. In Fig. 4 the effect of age in both groups for the theta band is illustrated. We can observe that in the theta band age has less influence on coherence values than it does in the delta band. In the same form as in the delta band, coherence increased with age, except in some frontal comparisons (F7-O1, F3-F4, F3-T4, F8C3) in the control group where coherence decreased with age. Here too, the similarities are less than the
LD
0,7
Fig. 3. Significant age effect on coherences in the delta band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
0,6
C 0 H E R E
0,5 0,4
.:"i."
N C E
.-
0,2
.'.-
.
. "..
0,1 0
I l l l l l l l l [ l l l l / l l l l l l l l l l l l J l l l f l l l l t l l l l [ I ? l l l l
7
8
9
10
11
12
18
14
IIIItll
16
16
17
AGE
Fig. 2. Data distribution of the z values of coherence plotted against age for P 3 - P 4 in the alpha band and in the LD group, where no significant age effect was found either by linear or quadratic regressions.
group-specific influences. In the control group the vertex-centered and in the LD groups the left temporalcentered distribution is evident. Fig. 5 shows the effect of age in the alpha band. We can observe the symmetrical age effect and the strong influence of age in the right parietal comparisons in the control group. Also, a negative age effect between F8 and T4 and abundant interhemispheric comparisons were found. The influence of age is reflected more in the delta and alpha bands than in the other bands. In the LD group the same left-temporal and occipital orientation of the age influences, as observed in the
EEG COHERENCE MATURATION
353
other bands, was seen. The L D group manifested more numerous frontal coherences affected by age than did the control group. Fig. 6 shows the significant effect of age in the beta band. Although in lesser quantity, the same differential age effects are observed as in the other bands: in the control group the effect of age on coherences is centered around Cz in a symmetrical way, while in the LD group the left-temporal and occipital comparisons are affected. In addition, negative slope is observed in frontal F8-T4 comparison in the control group, while in
CONTROL
F7
F3
°'I'.,°°•, "% • O,o ..% ".........
:.F4
I1 T3 ............,C3 "".. CZ'~C4F-!::~"" T4 ................ [ ..... W
CONTROL
''': .
:
'%
." "'.%.
F7
F3~
-
_
F8
{ ".. "-. ~ "" --"
-.
°"
.,"
F4 / F 8
LD
0,
P3
LD
"%'°''...,...°....
T3 lc %;. i
~ CZ
F4
F8
C4~
T4
P41/6
01 Fig. 4. Significant age effect on coherences in the theta band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
P4'
!ii6
........ ..o..°." °°'"°"°""
Fig. 5. Significant age effect on coherences in the alpha band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
the LD group abundant frontal age influences with positive slopes can be seen. The only comparison where age affected coherence in a similar way for both groups is between T3 and O1. Fig. 7 shows the effect of age for the total band. We observe few similarities in age effects between both groups and a large number of group-specific age effects, as in the case of the other bands. The control group shows a vertex-centered symmetrical orientation,
354
E. MAROSI ET AL.
while in the L D group left-temporal, i n t e r h e m i s p h e r i c c o m p a r i s o n s reflect the effect of age. S u m m i n g up the results of all the previous figures, we can state that a different m a t u r a t i o n a l p a t t e r n for the two groups is present. T h e control g r o u p has a C z - c e n t e r e d symmetrically a r r a n g e d increase of coherence values with increase in age. W e can also observe high age effects on the i n t e r h e m i s p h e r i c P3-P4 coherence a n d where P4 is included, as well as negative
CONTROL
F7,,,
F,3-,,,
F4
T 3 ~ ~
F7
F3
T3 "%,.
C3
CZ
T4
T5 \ P3~,,/×\ ~ P 4 i
CONTROL
F4
F8
C4
T4
Fi
~T6
I'
LD
%.
T5 '-...
76
8 T4
T5 ~
LD
P3
P4 i ~
01
O2
T6
Fig. 7. Significant age effect on coherences in the total band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
F3,
F•k
7F4
F8
IS "''~'.01v Fig. 6. Significant age effect on coherences in the beta band. Dotted lines show the significant age effect seen in both groups, and continuous lines show the age effect seen only in one group.
slopes with age in frontal i n t e r r e l a t i o n s , mainly in the theta band. O n the o t h e r h a n d , in the L D g r o u p there is n o age effect in any of those c o h e r e n c e s which i n c l u d e vertex. L D c h i l d r e n are c h a r a c t e r i z e d by a m a t u r a t i o n a l process c e n t e r e d o n the left t e m p o r a l a n d occipital areas, resulting in a m a r k e d asymmetrical m a t u r a t i o n of coh e r e n c e s a n d high age effect o n frontal coherences, a process that is n o t seen in the control group. W h e n c o m p a r i n g the two groups, we can observe that age affected some regions in a similar way for b o t h groups,
EEG COHERENCE MATURATION although there were more group-specific differences than similarities, and in all bands the pattern of maturation was completely different for the two groups.
Discussion
It is generally considered that E E G coherence is a measurement of cortical interconnectivity or coupling. Thatcher et al. (1986) reviewed several anatomical analyses in relation to coherence, estimating that no more than 1% of cortical fibers arise in the thalamus, 2 - 4 % in the contralateral cortex and approximately 95% in the ipsilateral cortex. These cortico-cortical association fibers have an average length of several centimeters, while the intercolumnar connections are of short range and are less than 1 mm in length. These authors propose that gross E E G cortical phenomena are produced by long cortico-cortical fibers. To explain the coherence they present a 2-compartment model based on the fact that the neocortex contains two types of cell: long-axoned piramidal or Golgi type I cells and short-axoned Golgi type II cells. If the density of Golgi type II cells is 10-100 times higher than that of Golgi type I ceils, a major contribution of the short axonal connections to coherence is expected for short scalp electrode distance. At long electrode separation the contribution of short axonal connections to coherence should be minimal, and an increase in the coherence would be produced by long axonal connections. Since coherence values reflect this cortical anisotropy, we may say that the coherence reflects cortical connectivity. Theoretically, an increase in coherence may be the result of two cortical areas directly coupled or by a third cortical or subcortical source related to both, or by a cascade of regions functionally interrelated with the recorded areas. Another cause for high coherence value might be the volumen conduction producing a smooth decrease of coherence with increased electrode separation, as coherence decays as an exponential function of distance. The propagation of E E G rhythms also depends on the spatial properties of the cortical neuronal networks, as the cortex is not a homogeneous medium. Different systems of intracortical fibers, which involve much larger distances than the diameter of a column, assure the interconnectivity of several columns (Lopes da Silva 1987). All these factors are reflected in the coherence of E E G and, therefore, we may say that coherence is a measure of the functional organization of E E G activity, which reveals structural characteristics of the brain. On the other hand developmental changes in the brain are mainly determined by myelination and synaptogenesis. Myelination produces more efficient interconnections between different cerebral regions, and consequently increased coherence values may be ex-
355 pected. Synaptogenesis means a greater quantity of synapses and better interaction between different structures, which may be reflected in increased coherence values. The goal of our paper was to look for the influence of age on coherence values, based on the belief that similar age effect on coherences for the control and LD groups would reveal similar brain organizational and maturational processes. But this was not the case. Considering the age effect as observed in our investigation, we can conclude that the "normal" maturational process (seen in our 98 control children) seems to have 3 important features: (a) A marked increase in coherence with age in central, temporal and posterior interconnections, especially in the delta and alpha bands. The increase of coherence with age may be the consequence of myelination in process, as myelogenesis is not completely finished until the second decade of life, and our subjects are in the age range of late childhood, when these changes have been shown to occur (Yakovlev and Lecours 1967). (b) A symmetrically distributed age effect on coherence, with abundant involvement of Cz in all bands. Our recordings are made in awake, relaxed children with the eyes closed, so maximum alpha activity is expected. These symmetrical age influences related to the vertex may be due to the developmental increase of coupling in the thalamo-cortical projections, as the thalamo-reticular nucleus has a very important role in the production and synchronization of alpha rhythm (Steriade et al. 1990). These findings may also be due to the development of cortico-cortical association fibers, including the corpus callosum. (c) A decrease in coherence with the increase of age in frontal comparisons, especially in the theta band. Decreased frontal coherences may be related to increased cortical differentiation (Thatcher et al. 1986). In 1979, Huttenlocker found that neuronal and synaptic density is at its maximum during the first 2 years of life and then gradually declines between the 2nd and 16th years, in all cortical areas. Our subjects are 6-16 years old, and for this reason the frontal negativity observed may be due to the fact that our study was carried out at the exact time that this reduction in frontal synaptic density occurs. Nevertheless, the resuits obtained by Goldman-Rakic (1987) contradict this assumption, since she did not find a different time-table or rate of synaptic elimination for different cortical areas. Available data on human synaptogenesis are limited and differential measurements on cortical maturation are quite scarce as well as controversial. In the present state of neuroscience, it is impossible to determine exactly what this frontal negativity in coherence values is due to. How does coherence relate to behavior, and why
356
does this pattern of coherence maturation reflect learning disability? This question is very difficult to answer, as we know very little about the basic cerebral mechanism involved in learning disability. Correlations between intelligence and brain development say little about real links between the brain and behavioral development because they are too general (Fisher 1987). Obviously, brain development is related to cognitive functioning. Rapid and periodic behavioral changes occur between the ages of 6 and 16. These behavioral changes are not related to the maximum synaptic and neuronal density in the brain, and this fact implies that there are other basic maturational processes, such as dendritic growth or improved synaptic efficiency, which permit the achievement of complete behavioral competence. There are different hypotheses that seek to explain LD from the neurophysiological point of view. One theory states that LD may be due to a maturational lag, since at older ages academic achievement may become normal. Considering the spectral parameters of EEG, John et al. (1983) found two types of electrical activity in an LD sample. A group of children had normal maturation of EEG, but it corresponded to the parameters of younger children. Another group of children had abnormal spectral measurements and maturational processes that one assumes to be different than those in the normal children. Our results definitely demonstrate that the LD children have a completely different pattern of brain maturation, and that they differ from normal children more in a qualitative way than in a quantitative one. Thus learning disability may be due to a maturational abnormality or a variance of the normal processes, which may cause all of these cognitive problems to which we give the common name of "learning disability." Another theory on the etiology of learning disability states that inverted h e m i s p h e r i c lateralization (Geschwind and Galaburda 1985) or a failure in the interhemispheric transfer (Gross-Glenn and Rothenberg 1984) may be the cause of this problem. In fact, the way in which the coherence values of our LD children are affected by age shows signs of a different hemispheric arrangement: our normal children show symmetrically distributed central and posterior maturational effect, while the LD children have the age effect centered on the left temporal areas. The different age effect observed in the frontal region (negative slope in the control group and abundant frontal age effect in the alpha band with positive slope in the LD group) may suggest massive frontal functional differences between the two groups. Motivational and attentional processes, so important for adequate academic achievement, may be deactivated by this anomalous functioning of the frontal cortex and prevent LD children from achieving a steady and continuous perfect
E. MAROSI ET AL.
execution. Direct frontal and limbic interaction is demonstrated by Mesulam (1990). All these speculations continue to be dubious so long as the nature of learning disability itself is not ~letermined. For the moment, we can state only that a different developmental mechanism is obvious in our LD group. Before concluding, we must make it clear that there is another important factor which has a well-demonstrated influence on electroencephalographic measurements: gender (Harmony et al. 1990a). Looking for gender effect in this study, we found very similar means, deviations and variances of age according to sex for the two groups, so we may infer that the sex variable was controlled and affected our results in a balanced way, and that these differences in coherences by groups are in fact due mainly to age. This project was supported by Grant Number IN205689 from DGAPA, UNAM, and by funds granted by the Government of the State of Mexico. Our thanks are due to Mr. Sandor John for the valuable corrections of this paper.
References Alvarez, A., Vald6s, P. and Pascual, R. EEG developmental equations confirmed for Cuban schoolchildren. Electroenceph. clin. Neurophysiol., 1987, 67: 330-332. Ba§ar, E. EEG Brain Dynamics. Relation between EEG and Brain Evoked Potentials. Elsevier/North-Holland Biomedical Press, Amsterdam, 1980: 67. Fein, G., Raz, J., Brown, F.F. and Merrin, E.L. Common reference coherence data are confounded by power and phase effects. Electroenceph. clin. Neurophysiol., 1988, 69: 581-584. Fisher, K.W. Relation between brain and cognitive development. Child Dev., 1987, 58: 623-632. Gasser, T., Jennen-Steinmetz, C. and Verleger, R. EEG coherence at rest and during a visual task in two groups of children. Electroenceph. clin. Neurophysiol., 1987, 67: 151-158. Geschwind, N. and Galaburda, A.M. Cerebral lateralization. Biological mechanism, associations and pathology. Arch. Neurol., 1985, 42: 634-654. Goldman-Rakic, P.S. Development of cortical circuitry and cognitive function. Child Dev., 1987, 58: 601-622. Gross-Glenn, K. and Rothenberg, S. Evidence for deficit in interhemispheric transfer of information in dyslexic boys. Int. J. Neurosci., 1984, 24: 23-25. Harmony, T. Neurometric Assessment of Brain Dysfunction in Neurological Patients. Lawrence Erlbaum, Hillsdale, NJ, 1984: 363369. Harmony, T., Marosi, E., Diaz de Le6n, A., Becker, J. and Fernandez, T. Effect of sex, psychosocial disadvantages and biological risk factors on EEG maturation. Electroenceph. clin. Neurophysiol., 1990a, 75: 482-491. Harmony, T., Marosi, E., Hinojosa, G., Becker, J., Rodrlguez, M., Reyes, A. and Rocha, C. Correlation between EEG spectral parameters and an educational evaluation. Int. J. Neurosci., 1990b, 54: 147-155. Huttenlocker, P.R. Synaptic density in human frontal cortex. Developmental changes and effects of aging. Brain Res., 1979, 163: 195-205.
EEG COHERENCE MATURATION John, E.R., Ahn, H., Prichep, L., Trepetin, M., Brown, D. and Kaye, H. Developmental equations for the electroencephalogram. Science, 1980, 210: 1255-1258. John, E.R., Prichep, L., Ahn, H., Easton, P., Friedman, J. and Kaye, H. Neurometric evaluation of cognitive dysfunctions and neurological disorders in children. Prog. Neurobiol., 1983, 21: 239-290. Lopes da Silva, F. Dynamics of EEGs as Signals of Neuronal Populations: Models and Theoretical Consideration. In: E. Niedermeyer and F. Lopes da Silva (Eds.), EEG: Basic Principles, Clinical Applications and Related Fields. Urban and Schwarzenberg, Baltimore, MD, 1987: 15-29. Mesulam, M.M. Large-scale neurocognitive networks and distributed processing for attention, language and memory. Ann. Neurol., 1990, 28: 597-613. Steriade, M., Gloor, P., Llin~s, R.R., Lopes da Silva, F.M. and
357 Mesulam, M.M. Basic mechanism of cerebral rhythmic activities. Electroenceph. clin. Neurophysiol., 1990, 76: 481-508. Thatcher, R.W. and Walker, R.A. EEG coherence and intelligence in children. Electroenceph. clin. Neurophysiol., 1985, 61: S161. Thatcher, R.W., Krause, P.J. and Hrybyk, M. Cortico-cortical associations and EEG coherence: a two-compartmental model. Electroenceph, clin. Neurophysiol., 1986, 64: 123-143. Thatcher, R.W., Walker, R.A. and Giudice, S. Human cerebral hemispheric development at different rates and ages. Science, 1987, 236: 1110-1113. Yakovlev, P.I. and Lecours, A.R. The myelogenetic cycles of regional maturation of the brain. In: A. Minkowski (Ed.), Regional Development of the Brain in Early Life. Blackwell Scientific, Oxford, 1967: 3-70.
