Developmental changes in cortical processing as reflected by visually evoked potential variability


Material Information

Developmental changes in cortical processing as reflected by visually evoked potential variability
Physical Description:
viii, 52 leaves : ; 28 cm.
Street, Wilma Jeanne, 1952-
Publication Date:


Subjects / Keywords:
Human information processing in children   ( lcsh )
Variability (Psychometrics)   ( lcsh )
Visual evoked response   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis--University of Florida.
Includes bibliographical references (leaves 47-51).
Statement of Responsibility:
by W. Jeanne Street.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000209976
notis - AAX6795
oclc - 04164108
System ID:

Full Text








I would like to express grateful appreciation, respect,

and affection to Dr. Nathan W. Perry, Jr., for his firm,

but unintrusive, support and guidance in all of my graduate

education endeavors. Many thanks to Dr. Robert L. Isaacson

for continually whetting my research taste buds, and to

Drs. Wiley Rasbury and Vernon Van De Riet for their personal

interest in me as a human being. Thanks to Dr. Donald

Childers, who pointed out to me the difference between a

fact and a piece of data.

I am extremely grateful to Drs. Bill Cleveland and

Dee Ramm, of the Duke University Medical Center, for their

unselfishness with their time and statistical and computer

expertise. I would also like to thank Denise Dixon for her

patience and promptness in typing the manuscript, and Helen

Corless for the figure drawings.

Family and friends provided immeasurable emotional

support over the years; most particularly, I am grateful

to my parents, Joan Duer, Peggy Brooks, and Clif Dopson for

their encouragement of me to finish without ever pushing

too hard.




LIST OF TABLES . .. .. iv

ABSTRACT . ... . v




Subjects . 13
VER Testing Procedure . 14
Preliminary Data Processing. .17
Data Analysis .. 22


Control Data . 23
Whole VER Variability. . 23
Component VER Variability. .. 24
Whole VER Variability by Sex .. .30


REFERENCES . . .. .. 47






1. Significant SD Changes Over Time for Whole VERs 25

2. Mean SDs and Directions of Change Over Time for
Whole VERs 26

3. Significant SD Changes Over Time for Component
VERs 28

4. Mean SDs and Directions of Change Over Time for
Component VERs 29

5. Significant SD Changes for Year 1 for Component
VERs 31

6. Significant SD Changes Over Time for Whole VERs
by Sex 32

7. Mean SDs and Directions of Change Over Time for
Whole VERs by Sex 33

Abstract of Dissertation Presented to the
Graduate Council of the University of Florida
in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy



W. Jeanne Street

March 1978

Chairman: Nathan W. Perry, Jr.
Major Department: Clinical Psychology

There is general agreement that the variability, or

flexibility, seen in perceptual and intellectual response

patterns is important to the understanding and predict-

ability of behavior. In this study the variability of

cortical processing over time is investigated using

visually evoked response (VER) standard deviations as a

measure of variable cortical processing. Since VERs are

felt to reflect the sequence of neural events that occur

as a visual stimulus is perceived and processed by the

brain, VER variability is felt to reflect the variability

of cortical processing.

The subject population in this study consisted of

76 children who participated in the study for three years

starting when they were in kindergarten. All the children

had normal visual acuity. The sample was homogeneous with

respect to socio-economic status. VERs were collected for

each subject in response to four different stimulus

conditions and recorded from three or four scalp locations.

The variability measure derived was the standard deviation

(SD) statistical measure. This was derived for each whole

VER response, and one each for the early, middle, and late

components of each VER.

Upon determining that the control SD data were

significantly different from the other SD data, SD changes

over the years were examined across the whole VER for each

stimulus condition and location. Significant changes

were found from year 1 to year 3, and year 2 to year 3.

Most changes seen were a decrease in variability over time,

except for an increase from year 2 to year 3 at the parietal

lobe location. The overall analysis of change in the

variability of the VER components over time was not statis-

tically significant, but the pattern was the same as that

for the whole VERs of a decrease in variability over time

except for an increase at the parietal lobe location from

year 2 to year 3. There were differences between the

components within years, but the differences were con-

sistent over time. Changes in variability of the whole

VERs over time as a function of sex were examined. The

overall effect was not statistically significant, but the

females, in almost all conditions, demonstrated more VER

variability than the males. Also, the females demonstrated

a pattern similar to that of the total sample in that a

steady decrease in variability was seen across years,

except for the parietal lobe which showed an increase

from year 2 to year 3. However, the males showed an

increase in variability from year 1 to year 2 and a

decrease from year 2 to year 3 at all electrode locations.

The major finding of a consistent change in variability

over time in the total sample as measured by whole VER SD

indicated a significant decrease in variability of cortical

processing from age five to age seven and age six to age

seven, except for the cortical functioning detected at

the parietal lobe which indicated an increase in vari-

ability. These findings are consistent with, and extend,

the previous findings in the VER literature. The findings

are related to the literature on child cognitive develop-

ment. It is suggested that the decrease in VER variability

is indicative of cognitive processing that is mature or

practiced and that the increase in variability at the

parietal lobe is indicative of a currently active or

developing cognitive function. Similarly, a difference

in rate of growth of different cortical processes or

basic differences in cortical processing itself are hypo-

thesized in reference to the sex differences found. The

data provide the beginning of baseline data for normal


changes in cognitive flexibility which may eventually

be used to help assess abnormal child growth and




The central nervous system is characterized by

ongoing intrinsic electrical activity which is described

by the electroencephalogram (EEG). When this activity

is measured in response to a large number of presentations

of light flashes averaged together, the response change

produced is called a visual evoked response (VER). The

response to an individual flash is small and cannot

ordinarily be visually detected apart from the ongoing

activity; therefore, the responses to a large number

(50-100) of light flash presentations are averaged

together by computer. The averaging process results in

an electrical response that is easily identified. Widely

accepted today is the thought that neural structure and

function underlie mental processes. The described VER

has been proposed and used as an indicator of neural

functioning and mental processing; the VER is felt to

reflect in part the sequence of cortical events that

occur as a visual stimulus is perceived and processed

by the brain (Perry & Childers, 1969).

Several parameters of the VER, usually amplitude and

latency, have been derived and used to investigate brain

functioning that occurs during visual stimulation.

Developing from these investigations has been interest

in another parameter, variability, as an indicator of

cortical functioning, specifically cortical processing

variability (Brazier, 1964; Callaway, 1975; Ellingson,

1970). Although much of the theory and research on the

development of specific cortical functions concentrates

on the maturation of stable, constant response patterns,

there is general agreement that the variability seen in

response patterns is also important to development

(Callaway, 1975). Essential to the understanding and

predictability of behavior, in this instance cortical

processing, is knowing its patterns and changes over time.

The existence of critical points of brain and behavior

change may be detected not only through patterns of

continuity and stability, but through patterns of

instability and flexibility. Thus, VER investigations,

including ones measuring some form of variability, have

been done in order to better understand the cognitive

organization of brain-behavior relationships.

VER investigations have progressed on the assumption

that the patterns produced may be more accurate reflections

of cognitive substrates than currently used performance

measures of mental processing. An investigation of VER

data of five year olds using a factor analysis procedure

indicated that the VER is multidimensional in nature

(Street, Perry, & Cunningham, 1976). That is, several

different aspects of VER are functioning together at any

one time to contribute to the final detected signal.

Thus, if the resulting VER actually consists of a variety

of ongoing functions, investigations of the variability

of cognitive processing may lead to more accurate deter-

minations of the functional contribution of the different

aspects of the VER over time. Since measures of VER

variability are assumed to reflect variability in cognitive

processing, they may be used to more accurately correlate

neurophysiological processes with cognitive behavioral


There is strong evidence to indicate that aspects of

the VER are correlated with cortical processing. Early

work using the VER as an electrophysiological measure of

cognitive ability found high IQ scores to be associated

with short VER latency in later components (Chalke & Ertl,

1965). This finding has subsequently been replicated

several times: Bigum, Dustman, and Beck (1970) found

later component latencies to be significantly later

(p<.05) in mongoloids than in normal subjects, and

Shucard and Horn (1972) found correlations from -.15 to

-.32 between latency and cognitive abilities. Several

studies have found higher amplitudes in brighter children

and adults than in less bright people (Bigum et al.,

1970; Rhodes, Dustman, & Beck, 1969). Amplitude and

latency of VER components are the characteristics

usually used to make correlations with intelligence,

but it has been suggested that VER variability might be

an important phenomenon of evoked potentials (Brazier,

1964; Callaway, 1975).

Several VER studies, not all of which are develop-

mental in nature, have investigated VER variability,

although the variability examined is often differently

defined and measured across studies. Among 20 normal

adults the greatest VER variability was found between

subjects, next between areas of the head, and least

across time within an individual (Werre & Smith, 1964).

In another study on 20 normal adults, Ciganek (1969)

saw a decrease in amplitude variability after 80 msec

and an increase in variability in early latency waves.

He also found the amplitude variability of the EP

(evoked potential) to be negligible between the averaged

responses of individual subjects, although high across

the whole group of subjects. In a study of five adults,

Robinson (1975) demonstrated that attentive viewing as

compared to passive viewing produced a decrease in

amplitude variance of a particular VER.

Although most of the studies with psychiatric

patients do not control for level of intellect, the

evoked response findings suggest a correlation between

variability and thought disorders. Among psychiatric

patients, psychotic depressives showed greater variability

compared to other subgroups (Borge, 1973). Schizophrenics

as compared to normals had more variable VERs in response

to different stimulus intensities (Rappaport, Hopkins,

Hall, Belleza, & Hall, 1975). Schizophrenic adults,

some of whom were said to have a thought-process disorder,

showed greater variability of auditory evoked potentials

(based on correlations between AERs to different stimuli)

than normal subjects (Callaway, Jones, & Donchin, 1970).

Callaway and Jones conclude from another study that

variable evoked potentials are correlated with variable

and unstable cognitive functioning (1975). Lifshitz

(1969) found that schizophrenics showed more variability

than normals to simple (visual and compound (visual and

auditory) stimuli. Among schizophrenics, EP variability

(also based on between-EP correlations) is greatest in

those who show inaccurate and variable perceptual per-

formances (Inderbitzen, Buchsbaum, & Silverman, 1970).

Brazier (1964) was the first to look at variability

of the separate responses that are usually summed to

form a VER. She found variability to be greater in the

first 30 responses in a train of 300 than the last 30,

when habituation was said to have occurred. Barnet and

Lodge (1967) examined the unaveraged individual auditory

evoked responses of 15 mongoloid and 55 normal subjects

under 14 months of age and saw great variability in the

amplitudes, but found the mongoloids to have many more

extremely large responses and to show less decline in

response stimulation with repetitive stimulation than

the normal subjects. This suggests differences in brain

mechanisms governing sensory input.

Ellingson (1970) has done much work in the area on

infant VER variability and, overall, has found neonatal

VER variability to be high. Although he does not explain

mathematically how he measures variability, he states that

individual neonatal VERs, in contrast to the findings in

adults, were often variable in latency and amplitude

during a single recording session. In one study the

mean latencies of components of averaged auditory evoked

potentials were compared between sleeping children;

younger children's AERs were found to be more variable

than older children's (Barnet, Ohlrich, Weiss, & Shanks,


Callaway (1975) hypothesized that if evoked potentials

reflect cognitive processes, then habitually irregular or

unstable modes of cognitive processing should be accompanied

by variable evoked potentials. In a 1969 study (Callaway

& Stone) samples were described in which normal adults

showed low variabiliy VERs, schizophrenics had inter-

mediate values, and children (aged nine) had high variability.

Data on the children showed that lower evoked response

variability tended to be correlated with higher scores on

a visual-motor integration test. In another developmental

study done with visual and auditory evoked potentials,

variability of the EPs of 119 children from ages six

to 15 were investigated (Callaway & Halliday, 1973).

Variability decreased with increasing age. Further

investigations with the Beery Visual-Motor Integration

test (1967) led Callaway to conclude that EP stability is

sufficient, but not necessary, for good cognitive task

performance (1975). In several studies with normal

adults, Callaway (1975) found high EP variability to be

correlated with low verbal IQ. He hypothesized that,

in general, decreasing variability with increasing age

might be due to an increasing stability of cognitive

functioning with age.

Thus, VER variability has been used as a measure

of change, sometimes between VERs across subjects and

conditions, and sometimes between responses making up a

VER within a subject. Another way to consider VER

variability is to study the differences between early

and late components of the VER. The VER waveform is

considered to be composed of a number of components, or

positive and negative amplitude deflections. It has been

postulated that early components represent primary

sensory system activity and processing of the physical

parameters of the stimulus (John, Ruchkin, & Villegas,

1964; Ertl, 1969). Ertl (1969) states that the late

components of the AEP are sensitive to changes in

stimulus parameters involving decision-making (Sutton,
Braun, Zubin, & John, 1965), pattern recognition,

attention, and problem solving (Beinhocker, Brooks,

Anfenger, & Copenhaver, 1966; Callaway, 1966; Chapman,

& Bragdon, 1964; Uttal, 1965), drug-inducing changes in

levels of alertness (Allison, Goff, Abrahamian, & Rosner,

1963; Brazier, 1963; and Garcia-Austt, 1963), and

generally the informational content of the stimulus.

Buchsbaum feels that, because later components of EPs

(after 200 msec) change more than earlier ones as a

function of attention, arousal, and expectancy, greater

experiential effects might be expected in late components

(1974). He also suggests that early components may be

more stable and genetically determined. Callaway and

Halliday (1973) found the variability of late EP components

(100 msec or later) decreasing with increasing age more

than the variability of earlier components. Their findings

suggest that by age six a child's sensory processing is

stable, but higher perceptual information processing is

still stabilizing. Barnet et al. (1975) found that

decreases in the latencies of the various components

proceeded at different rates and felt that this suggested

that the components reflect independent neural substrates.

They also noted that the components of shortest latency

displayed the weakest relationship to age. Apparently,

not only the variability of the whole VER, but the

variability of its components gives an indication of

changing cortical processing.

As was previously noted, studies in the area of

VER variability are sometimes difficult to compare since

variability may be conceived of in several different

ways. For example, one is measurement, or experimental,

variability which, in most studies, is assumed to be

controlled for and, therefore, not contributing in a

large part to detected variability. Variability in

cortical processing may also be examined. This may be

thought of as consisting of two parts: one is a

structural variability in that the structure, or physical

parameters, of the sensory system and/or the stimulus

itself may be variable; the other part is-a functional

variability referring to variability, or flexibility, in

cortical processing of input--such changes are felt to

reflect processing variability at a level higher than the
primary sensory level. Developmental variability is also

examined and may also be thought of as consisting of two

parts: variability may be apparent as basic structures

develop, and when functioning develops after the physical

structure is clear.

Of major interest here is VER variability reflecting

cortical reorganization as a function of developmental

processes. It has been long established that most physical

and structural changes have occurred by this age range

(five to seven years) (Yakovlev & Lecours, 1967); therefore,

variability changes seen will be investigated as

indicators of cortical processing changes and/or

functional developmental changes in processing.

The age range chosen is an important one to

investigate not only because of the scarcity of VER

variability data on this group, but also because of the

important cognitive changes reportedly occurring at this

time. White (1965) has postulated that the five to

seven age range is a significant one for changes in

particular learning paradigms. Piaget (1962) stressed

the increasing flexibility of thinking and the resultant

acquisition of certain forms of conservation from the

ages four to seven. In general, ideas and research in

the developmental area indicate that response patterns

go through a period of increasing flexibility for proper

adaptation to the changing environment, but at some

point the flexibility either decreases or becomes more

selective. If, as expected, the VER variability measure

is useful as an indicator of normal growth and develop-

ment of cognitive processing, the timely significance

of the study becomes apparent. As a predictor and

diagnostic tool the measure may be useful in the detection

of and therapeutic intervention in problems of child

development, such as hyperactivity (Halliday, Rosenthal,

Naylor, & Callaway, 1976) and specific reading disability

(Preston, Guthrie, Kirsch, Gertman, & Childs, 1977).

It is felt that the variability measure provides a

picture of changing cognitive functioning and an index

to the developmental stages of differentially maturing

functions of the central nervous system.

In order to assess developmental changes in cortical

reorganization during this important age range, VER

variability measures were collected in a longitudinal

study of children over a three year span, variability

is expected to show a change in direction indicative

of functional changes in cortical and developmental

processing. Variability measures are derived on VERs

collected over the three years under several stimulus

conditions and from several cortical locations. VERs

in response to different frequencies of stimulation and

to different stimulus characteristics (pattern or

diffuse) are expected to show differences in variability,

as the brain may be expected to respond more or less

variably depending on the complexity of the stimulus.

Variability derived from different brain locations is

expected to differ as some are more purely perceptual

and may be at different developmental stages over the

three years. VER variability of early, middle, and late

components is expected to differ as these components

are thought to represent different stages and functions

of cortical processing. Since early components are

thought to represent simple sensory processes which

reach full development relatively early, they are expected

to show less VER variability over time than the later

components, which are thought to represent processing at

a higher level of complexity not yet completely

developed. The change in variability over time will

also be compared between males and females to investigate

possible sex differences, since, at this age range,

sex differences in specific cognitive abilities have

been found (Mussen, 1970).



Letters describing the study and asking for partici-

pation were sent to parents of all children entering

kindergarten in 1973 in the city of Gainesville, Flordia.

Those parents who responded were contacted and their child

was accepted in the study if (a) the family had no plans

to move within three years and (b) the child had no

unusual medical or developmental history. The subject

sample for this study consisted of 76 white children

(37 females, 39 males) with complete data for three years.

Attrition from the original sample of 98 was nine and

another 13 had incomplete data due to occasional equipment

and/or collection difficulties. During the first year of

data collection the mean age was 67.3 months (SD=2.9,

range=62-75 months), 79.2 months (SD=2.6, range=74-85 months)

during the second year, and 91.8 months (SD=2.7, range

86-100 months) during the third year. The mean full

score IQ (WPPSI) for the first year was 118.8 (SD=9.5,

range=94-144), the mean full score IQ (WISC-R) for the

second year was 114.2 (SD=11.4, range=84-141), and

the mean full score IQ (WISC-R) for the third year was

120 (SD=11.4, range=83-142). All children had normal

visual acuity in each eye (Snellen "E") and normal

stereopsis (Titmus "Fly"). The sample was relatively

homogeneous and upper middle-class with respect to

socio-economic status, with, for example, 16.6% of the

mothers and 65% of the fathers holding advanced degrees

and most families having an income in the range of

10-20 thousand dollars.

VER Testing Procedure
Silver-silver chloride cup electrodes and paste

(Beckman) were placed on the scalp at locations C3, C4,

0z, and Pz of the international 10-20 electrode system

(Jasper, 1958). The central locations (C3 and C4),

sampling left and right hemispheres, were referenced to

linked ear lobes. The occipital location was a bipolar

derivation between O0 and Pz, on the midsaggital plane.

After the first year a fourth location over the left

parietal (P3) was added. Impedances of approximately

2 Kv were obtained between electrodes, as measured with

a d.c. impedance tester (IMA Electronics).

Each subject was seated in an adjustable ophthalmic

chair in an electrically shielded and light-proof room

(ACE). Ventilating blowers produced a steady 62 dB noise

level (General Radio 1551-A) to mask equipment and

extraneous noise. Solid state d.c.-powered differential

amplifiers were placed inside the shielded room near the

subject, enabling the use of very short electrode leads

(Microdot). Following amplification, electrical activity

was filtered (1.0 to 50 Hz, Krohn-Hite 330 BR) and

simultaneously routed to an FM tape recorder (Sanborn

7000) and in year 1 averaged by a CAT computer and

subsequently by a Nicolet MED-80 averaging computer.

Frequency response of the complete recording system was

relatively flat from 2.0 Hz to 30 Hz, and was 50 percent

attenuated at 1 Hz and 50 Hz. A 5 uV calibrate signal

(Medistor C-1A) was introduced through the amplifiers

for each subject and processed in the same manner as the

VERs. Each VER was the result of sixty 500-msec sweeps

of the computer.

Throughout the VER session of approximately 25 min.,

subjects viewed binocularly a 60 (visual angle) circle

of achromatic light, continuously illuminated to provide

a background level of 0.1 log ft.-L. (SEI photometer).

The continuous background and the stimuli were provided

by four projectors (Viewlex V-120), mounted outside the

shielded room and projecting through a double plexiglas

conductive window (Tecknit) just above and behind the

subject's head. The stimuli were projected onto an

aluminized screen located eight feet in front of the

subject. Electronic shutters (Gerbrands) attached to

each projector determined stimulus durations. During

stimulation, eye fixation of subjects was aided by a chin

rest and monitored by an experimenter seated obliquely

in front of the subject. Fixation upon the center of

the circle was required for 30 sec for each trial.

Loss of fixation (greater than about 20) resulted in

aborting and repeating the VER averaging.

Three different stimulation conditions were used,

in order to obtain VERs reflecting presumably different

types of cortical processing. For the first stimulus

condition, 2 Hz diffuse, light flashes were presented

at a rate of 2/sec for 30 sec (giving 60 flashes). The

flashes, of 50 msec duration, were superimposed on the

60 background circle, and were functionally diffuse

since, other than the dim background illumination, the

room was dark.

The second stimulus condition, 6 Hz diffuse, consisted

of a 15 msec-duration flash superimposed on the 60 back-

ground at a rate of 6/sec for 30 sec (giving a total of

180 flashes, with the responses averaged over 60 sweeps

of 500 msec each, as in the 2 Hz diffuse condition).

Illuminance of the individual flashes in both the 2 Hz

diffuse and 6 Hz diffuse conditions was 2.35 log ft.-L.

There is evidence that diffuse light, as used in

the 2 Hz diffuse and 6 Hz diffuse conditions, is processed

differently from pattern stimulation by the cortex (Hubel

& Wiesel, 1962; Perry & Childers, 1969). To see what

effect pattern stimulation might have on the variability

of cortical processing the third stimulus condition

was the presentation of 2 Hz pattern. The pattern was a

6 "sunburst," with a dark 14' center from which an equal

number of light and dark rays extended, each of which

subtended 30' at the circumference. The pattern was

superimposed on the background circle at an illuminance

of 2.2 log ft.-L., and was alternated at a rate of 2/sec

with a diffuse 60 circle adjusted to the same apparent

brightness. The pattern appeared for 50 msec and disap-

peared for 450 msec, for a total of 30 sec (60 times),

without any apparent change in the brightness of the


In addition to the three stimulus conditions, a

control condition was used in which the cortical activity

was averaged in the same manner as for the 6 Hz diffuse

condition, but with the flashing light occluded so that

the subject viewed only the continuously-lit background


Ten 60-sweep VERs (trials) of 500 msec each were

collected from each subject, with approximately 2 min

between trials. The order of stimulus presentation for

the trials was as follows: (1) control; (2) (3) (4)

6 Hz diffuse; (5) 2 Hz pattern; (6) 6 Hz diffuse; (7) 2

Hz pattern; (8) 2 Hz diffuse; (9) 2 Hz pattern; (10) 2

Hz diffuse. This procedure yielded a total of 30 VERs

(10 trials by 3 electrode locations) for each subject.

Preliminary Data Processing

VERs for all subjects were first normalized relative

to the amplitude of a 5 &.V calibration sent through the

amplifiers and averaged in the same manner. Pearson

product-moment correlations were then performed between

common VERs measured from the same electrode location to

the same stimulus condition (that is, trials (2) (3) (4)

and (6) all at 6 Hz diffuse from the occipital location

were correlated, then from the left hemisphere location,

then the right hemisphere location, then the left parietal

location; trials (5) (7) and (9) all at 2 Hz pattern from

the occipital location were correlated, then from each of

the other locations; and trials (8) and (10) at 2 Hz

diffuse from each of the locations were correlated). VERs

from the same location to the same stimulus condition were

then pooled for each subject, reducing the 30 VERs to 12

or 16 (3 or 4 locations by 4 conditions including the control

trial). As a result, the control VER represents 60 sweeps,

each 6 Hz diffuse VER represents 240 sweeps, each 2 Hz

pattern VER represents 180 sweeps, and each 2 Hz diffuse

VER represents 120 sweeps. Based on these pooled VERs the

computer (Nicolet MED-80) was programmed to provide amplitude

measures at 128 data points across each VER waveform.

Fortunately, the data as described up to this point had

been collected as part of a larger longitudinal study

conducted by Nathan W. Perry, Jr., Ph.D. The available

amplitude data were then used in this study to derive

standard deviation (SD) measures, also using the Nicolet

MED-80 computer. The SD program resulted in 32 SDs

(one at every fourth point of the 128 data points) for

each VER. The 32 SDs were condensed to one SD by con-

verting the 32 SDs back to variances, getting an average

variance, and then taking the square root of that value.

The VER was also divided into early, middle, and late

components based on msec passed. The early component

consisted of the VER from 40 to 150 msec, the first 40

msec being dropped as the variance, there is felt to reflect

experimental error (Perry & Childers, 1969). The middle

component was the VER from 151 to 275 msec and the late

component was the VER from 276 to 400 msec. The last 100

msec were dropped for the same reason as cited above.

A single SD was derived for each of the components of each

of the VERs by the averaging process described above, so

that there were subsequently four SD measures for each

VER: one each for the whole, early, middle, and late VER.

The figure illustrates the steps taken to derive the SD

measures from the summated amplitude values of the VER


It has been suggested (Callaway, 1975) that the

amplitude of VERs is highly correlated with the standard

deviation(s) of a particular VER; therefore, standard

deviations may need to be divided by the VER peak-to-peak

Figure. Pictorial representation of derivation
of SDs. (A demonstrates a typical VER
waveform at a given location and stimulus
condition with summated waves apparent.
B demonstrates the computer data points
at which the SD measures were derived
from the contributing 120-240 available
amplitude measures at every fourth
point; the 32 resulting SDs were then
averaged to get one SD measure per VER
waveform. C demonstrates the division
of the VER waveform into an early,
middle, and late component; the SDs
derived from the whole waveform were
then averaged within each component
to result in three more SD measures
per waveform.)




Data Points

Middle I Late

I I 1 I

4 I I I
4 812

Data Points


I I i fI .

I I 12
4 812





amplitude in a normalizing process. In order to test

this hypothesis, the SDs on the first year of data were

compared to the SDs divided by the peak-to-peak amplitudes

on the first year of data by using a Pearson means

correlation procedure. The two variables were found to

be largely uncorrelated across the 12 VERs (values

ranged from .25 to -.15) suggesting that standard devia-

tions do not need to be divided by the VER amplitudes

in order to discuss standard deviation changes. Based

on this finding, the rest of these analyses are based on

simple standard deviations, not SDs divided by peak-to-

peak amplitudes.

Data Analysis

Subsequent analyses were done using the General

Linear Models procedure from the SAS statistical program

package (1976). This procedure was chosen over other

possibilities due to its known reliability and its

flexibility in testing procedures. It is basically a

linear regression procedure using a least-squares fitting

of univariate and multivariate models of regression.


Control Data

Essential to further analyses is a finding of a

significant difference between the SD data collected

on the control trial and the SD data collected on the

experimental trials. As was noted earlier, VERs are

felt to reflect cortical activity above and beyond the

ongoing intrinsic electrical activity reflected by the

EEG. If the SD data from the control trial, in which

light was continuous rather than flashing, was not

significantly different from the other SD data, then the

SD data could be felt to reflect nothing more than the

variability of intrinsic processing and methodology.

The control SD data were compared to the VER SD data on

the first year of data collected at the occipital location.

The control data were significantly different (F(3,73)=

88.19, p < .0001) from the SD data derived from the 6 Hz

diffuse occipital condition, the 2 Hz pattern occipital

condition, and the 2 Hz diffuse occipital condition.

Whole VER Variability

In order to assess changes in variability over time

as evidenced through whole VER SDs the data were examined

by use of the SAS GLM procedure. Changes over the three

years were examined separately for each stimulus condition

and location and an overall effect was assessed. As can

be seen in Table 1, the overall effect was significant at

the p <.0003 level (F(21,55)=3.22). The individual

analyses indicated no significant changes in variability

from year 1 to year 2, whereas there were changes from

year 2 to year 3 and from year 1 to year 3, particularly

for the occipital brain location. There were significant

changes at the right lobe location for 6 Hz and 2 Hz

pattern from year 1 to year 3, but the significant change

was at the left lobe location for 2 Hz diffuse. Table 2

shows the mean SD at each year for each VER and the

direction of change. Except in a few cases, most of the

changes were a steady decrease in variability over time.

The changes in the parietal location for all three

stimulus conditions were increases from year 2 to year

3 (these changes were not statistically significant).

The SD changes at the occipital location for the 2 Hz

diffuse condition were the only ones that did not show

a linear change; there was an increase from year 1 to

year 2 (not a statistically significant change) and a

decrease from year 2 to year 3 (this was significant, as

was the overall decrease from year 1 to year 3).

Component VER Variability

In order to detect any significant changes in the

variability of the early, middle, and late components of

Table 1
Significant SD Changes Over Time for Whole VERs

Condition 1-3 2-3

6Hz diffuse, occipital F(1,75)=16.53 F(1,75)=24.00
location (6HzO) P < .0001 p <.0001

6Hz diffuse, left lobe
location (6HzL)

6Hz diffuse, right lobe F(1,75)=4.48
location (6HzR) p < .0377

6Hz diffuse, parietal
location (6HzP)

2Hz pattern, occipital F(1,75)=29.83 F(1,75)=25.68
location (2HzP,0) p < .0001 p <.0001

2Hz pattern, left lobe
location (2HzP,L)

2Hz pattern, right lobe F(1,75)=7.62
location (2HzP,R) ~ < .0073

2Hz pattern, parietal
location (2HzP,P)

2Hz diffuse, occipital F(1,75)=9.67 F(1,75)=14.15
location (2HzD,0) p < .0037 p <.0003

2Hz diffuse, left lobe F(1,75)=4.42
location (2HzD,L) J < .0388

2Hz diffuse, right lobe
location (2HzD,R)

2Hz diffuse, parietal
location (2HzD,P)

*overall F(21,55)=3.22, p < .0003

Table 2

Mean SDs and Directions of Change Over Time for Whole VERs


Condition 1 2 3 Direction Significant

6HzO 544.5 537.6 485.1 Decrease yrs. 1-3,2-3

6HzL 496.6 491.4 483.3 D

6HzR 495.2 483.5 474.8 D yrs. 1-3

6HzP 487.7 499.9 Increase

2HzP,O 614.8 599.4 534.4 D yrs. 1-3,2-3

2HzP,L 501.1 496.7 484.5 D

2HzP,R 505.7 488.9 476.7 D yrs. 1-3

2HzP,P 507.3 512.5 Increase

2HzD,O 520.8 527.1 479.5 I-D yrs. 1-3,2-3

2HzD,L 511.2 506.2 496.1 D yrs. 1-3

2HzD,R 509.5 498.0 496.3 D

2HzD,P 498.6 507.1 Increase

Note. The unit of measure for the numbers is uV.

the VER over time, and significant differences between

the components, the SDs derived on the components were

compared using the SAS GLM procedure. In this case,

differences between early and middle components were

compared between year 1 and year 2 and between year 2

and year 3, and differences between middle and late

components were compared between year 1 and year 2 and

between year 2 and year 3. First of all, the overall

tests of significance done in this procedure indicated

no significant differences; therefore, even though some

significant differences were found on the separate

analyses, conclusions based on these findings must be

conservative. Table 3 gives the findings based on

this analysis. As can be seen, most of the changes in

the SDs of the different components over time occur

between the second and third years in the 2 Hz pattern

conditions and the changes are evident across all three

components. Significant changes from year 1 to year 2

were found in the 2 Hz pattern left lobe and the 2 Hz

diffuse right lobe conditions. Table 4 shows the mean

SDs over time and the directions of change. Similarly

to the analysis of the whole VERs, most of the changes

are a linear decrease over time, except for the parietal

lobe components which show an increase from year 2 to

year 3. Also the 2 Hz diffuse occipital condition shows

an overall decrease, but an initial increase from year

Table 3
Significant SD Changes Over Time for Component VERs

Early-Middle Middle-Late

tion 1-2 2-3 1-2 2-3





2HzP,O F(1,75)=4.79
p <.0317

2HzP,L F(1,75)=4.87 F(1,75)=9.52 F(1,75)=4.97
S<.0304 p <.0028 p < .0288

2HzP,R F(1.75)=9.40
p <.0030

2HzP,P F(1,75)=8.26
____ 0053


2HzD,L F(1,75)=3.81
R < .0545

2HzD,R F(1,75)=7.79
S. < .0067


F(42,34)=1.18, p. < .3095


Table 4

Mean SDs and Directions of Change Over Time for Component VERs


Condition Component 1 2 3 Direction

6HzO E 544.3 535.9 483.5 Decrease
M 543.4 538.4 484.4 D
L 546.6 539.2 486.8 D
6HzL E 495.8 488.7 482.2 D
M 494.7 492.7 482.3 D
L 498.3 491.3 486.0 D
6HzR E 495.4 482.2 475.1 D
M 497.0 481.9 473.1 D
L 494.3 484.3 474.5 D
6HzP E 485.6 498.0 Increase
M 486.8 499.0 I
L 487.2 501.7 I
2HzP,0 E 625.3 613.0 539.0 Decrease
M 615.8 596.1 532.3 D
L 603.7 587.0 519.2 D
2HzP,L E 501.2 500.2 483.6.. D
M 499.9 491.1 485.4 D
L 499.7 494.8 482.0 D
2HzP,R E 506.9 491.7 475.3 D
M 503.8 484.3 477.2 D
L 503.1 486.6 474.4 D
2HzP,P E 508.8 513.9 Increase
M -499.8 511.5 I
L 508.8 509.1 I
2HzDO E 529.7 536.3 486.1 I-Decrease
M 525.8 528.5 482.6 I-Decrease
L 513.7 519.5 472.3 I-Decrease
2HzD,L E 503.4 503.7 491.6 D
M 511.9 510.3 501.9 D
L 517.7 507.7 496.4 D
2HzD,R E 502.9 496.0 492.5 D
M 509.7 504.0 500.9 D
L 513.9 497.7 496.7 D
2HzD,P E 499.5 507.2 Increase
M -499.4 509.5 I
L 501.1 508.7 I

Note. The unit of measure for the numbers is


1 to year 2. Since the first analysis on the components

demonstrated differences between them that were consistent

over time, it was felt appropriate to look at the change

in component variability within a given year. Table 5

gives the results of an analysis comparing the SDs of the

early components to the SDs of the middle components and

the SDs of the middle components to the SDs of the late

components within year l's data. Looking at the data in

this fashion, similar findings to those in Table 3 are

evident in that no significant differences in the SDs of

the components of the 6 Hz VERs were obtained.

Whole VER Variability by Sex

Changes in variability over time as a function of

sex were examined in an analysis of the whole VER SDs.

Changes over the three years were examined separately

for males and females at each stimulus condition and

location, and an overall effect was assessed. As can

be seen in Table 6, the overall effect was not statisti-

cally significant (F(18,58)=2.04, p '.2565). The

individual analyses indicated statistically significant

differences in variability change between males and females

from year 1 to year 2 and from year 1 to year 3 at the

right lobe location for the 6 Hz stimulus condition and

at the occipital location for the 2 Hz diffuse stimulus

location. Table 7 shows the mean SD for the sexes at

each year for each VER and the direction of change.

Table 5

Significant SD Changes for Year 1

for Component VERs

Condition Early-Middle Middle-Late



F(1,75)=4.34 F(1,75)=8.79
2HzP,0 -
2p 4 .0407 p <.0041

2HzD,0 F(1,75)=8.98
Sp <.0037
2 L ~F(1,75)=5.00
zl p < .0283
2H DR F(1,75)=4.07
2HzD,R .0473
p < .0473

F(18,58)=2.04, p < .0213


Table 6

Significant SD Changes Over Time for Whole VERs by Sex

Condition 1-2 1-3



6HzR F(1,74)=6.24 F(1,74)=4.35
p_ < .0147 p <.0404






2HzD,O F(1,74)=4.07 F(1,74)=4.78
p < .0473 p.< .0320




*overall F(21,54)=1.24, p <.2565

Table 7
Mean SDs and Directions of Change Over Time
for Whole VERs by Sex




3 Direction By Sex

6HzO Male 512.0 527.4 473.3 Decrease
Female 578.8 548.4 497.6 D
6HzL M 481.0 485.4 474.5 I-D
F 513.1 497.8 492.6 D
6HzR M 468.8 482.1 467.5 I-D yrs. 1-2,
F 523.1 485.1 482.5 D 1-3
6HzP M 485.1 473.6 D i
F 490.5 527.7 I
2HzP,0 M 586.2 577.1 506.8 D
F 644;9 622.9 563.5 D
2HzP,L M 483.6 490.1 469.9 I-D
F 519.5 503.6 499.8 D
2HzP,R M 487.7 487.2 465.3 D
F 524.7 490.7 488.7 D
2HzP,P M 499.9 477.9 D
F 515.1 549.0 I
2HzD,0 M 481.3 515.1 467.6 I-D yrs. 1-2,
F 562.3 539.8 492.0 D 1-3
2HzD,L M 501.0 505.1 484.4 I-D
F 522.0 507.9 508.6 D-I
2HzDR M 492.9 498.2 482.3 I-D
F 527.0 497.9 510.9 D-I
2HzD,P M 499.0 477.2 D
____ F 498.2 538.7 I

Note. The unit of measure for the numbers is .L-V.

For the males, there was usually an increase in

variability from year 1 to year 2 and a decrease from

year 2 to year 3; in a few cases there was a consistent

decrease across the three years. For the females, in

most cases a steady decrease in variability was seen

across years, except for the parietal lobe data, in

which case an increase was seen from year 2 to year 3.

This is in contrast to the males who demonstrated a

decrease in variability from year 2 to year 3 at the

parietal lobe location. Overall, in all conditions and

years except two, the females showed more VER variability

than the males.


The results of this longitudinal study indicate

significant changes in the variability (as measured

by standard deviations) of VERs over time in 76 children

from age five to age seven. Overall, change was

characterized by a consistent, progressive decrease in

variability over time with the major shifts between the

ages of five and seven, and six and seven. VER vari-

ability changed between the ages of five and six, but

not to a statistically significant degree, suggesting

that changes in cortical processing as a function of

age are not significant from age five to age six.

The major finding of a consistent change in

variability as a function of time is positive with

respect to the developmental research previously done

in the VER area and with the predictions of this study.

Callaway found in one study (Callaway & Halliday, 1973)

that variability decreased with increasing age and later

suggested that the change might be due to an increasing

stability of cognitive functioning with age (Callaway,

1975). Other authors also found more variable EPs in

children as compared to adults (Ellingson, 1970; Barnet

et al., 1975). The research leads to the prediction of

a consistent change, most likely a decrease, in VER

variability over time, possibly as a reflection of

maturing cortical processing, and this prediction was

supported by the findings in this study.

The single exception to the finding of decreasing

variability as a function of age was in the parietal

lobe location data which were obtained at ages six and

seven only. At this electrode location VER variability

increased from ages six to seven rather than decreased

raising the interesting speculation that this area of

brain functioning might be moving toward increasing

flexibility or may not yet be matured. Studies in related

areas of research provide possible partial explanations

for this finding. From the results of animal studies,

Lynch, Mountcastle, Talbot, and Yin (1977) conclude with

the hypothesis that the parietal lobe performs a matching

function between the neural signals of the nature of objects

and the internal drive state of the organism, and also

contains a neural apparatus for the direction of visual

attention to objects of interest and for shifting attention.

Perhaps parietal lobe functioning is more complex and

involves a discriminatory function which either matures

after the age of seven or fluctuates with development.

Luria (1966) described the functions of different areas

of the parietal lobe and they all involve some kind of

complex visual discrimination or integration. Thus,

parietal lobe variability in this group of children

appeared to be increasing, most likely due to an earlier

stage of development of certain visually-related cognitive


Preston et al. (1977) found VERs from the left

parietal lobe location to be important in differentiating

between groups of normal and disabled reading adults.

Normal readers showed larger VER amplitude differences

between a work and a flash condition at the left parietal

lobe location as compared to disabled readers. The

study basically supports earlier findings of decreased

VER amplitude at the left parietal lobe location in

disabled readers (Conners, 1970; Preston, Guthrie, &

Childs, 1974). The evidence linking reading functions

with VER measures from the left parietal lobe is compatible

with the VER variability data derived from the left

parietal lobe in this study. The increase in the measure

of variability from age six to seven is suggestive of a

currently active and developing process, such as reading,

that is not yet stable and habitual in its pattern of


Although the overall significance level for the

analysis of change in whole VER variability over time

by sex was statistically low, the apparent differences

are striking. In most cases, the females not only

demonstrated more variability within a given year at a

given stimulus location and condition, but their vari-

ability change from year to year was more than that of

the males. The change in the females followed the pattern

of decreasing variability over time, whereas the males

often demonstrated a non-linear change of an increase

from age five to age six and then a decrease from age

six to age seven. Apparently, the variability contributed

by the females to the total sample masked, somewhat, the

pattern of change in the males which differed substantially

from the females. That the females demonstrated more VER

variability than males in cortical processing overall

possibly indicates more flexibility in processing at this

age range, or it may be that a basic sex difference in

the development of cortical processing exists. The dif-

ference between the sexes in the direction of the change

in variability may reflect a difference in maturational

stage of the cortical functions detected, possibly related

to a basic sex difference in cortical function development.

The males may be simply demonstrating a lag in which they

begin to demonstrate similar cognitive processing by age


The difference between the sexes in the change in

variability at the parietal lobe location from age six

to seven is of interest. The decrease in variability

from age six to seven for the males parallels the pattern

of decrease the males showed at the other locations, but

the increase in variability for the females is unique

for them. As hypothesized previously, this increase may

reflect an earlier stage of development of the cognitive

functions detected at this location.

Also consistent in the data was a highly significant

change in variability from age six to seven in the occipital

location VERs. Although the visual system has supposedly

reached physical maturity by this age, even the primary

sensory system appears to be undergoing marked signal

processing changes. However, a portion of the change

seen at the occipital location may be more sensory in

nature, since brain processing at this location is so

closely linked to the visual sensory system.

It is somewhat difficult to explain that significant

variability decreases were seen at the right lobe

location for the 6 Hz diffuse and 2 Hz pattern conditions,

whereas the significant changes for the 2 Hz diffuse

condition were at the left lobe location. This may have

to do with the known differences in functioning of the

opposite hemispheres combined with the unique characteristics

of the different stimulus conditions. The 6 Hz diffuse

and 2 Hz pattern stimulus conditions may be thought of as

slightly more complex in nature than the 2 Hz diffuse.

Apparently, cognitive functioning in the right lobe

became significantly less variable in response to complex

stimuli, while the left lobe's functioning became less

variable to a simpler stimulus.

The lack of overall statistically significant change

in differences in the variability of the early, middle,

and late components of the VER over time indicates that,

whatever the differences in cognitive functioning are

as reflected by the different components, they do not

reflect a significant difference in the flexibility of

processing over time. There were differences within a

year which may represent an artifact and/or contribute

to difficulty in detecting differences between years.

As might be expected, the variability over time within

components follows the same general pattern as the

overall VER variability changes (decreasing in all

conditions except at the parietal lobe location). How-

ever, a consistent pattern of variability change across

components within conditions is not apparent; in some

cases variability increases, in some it decreases, and

in some it is non-linear. Therefore, the hypothesis

that early components would show less variability change

over time than late components was not upheld. Evidently,

at least for this sample, the variability of the components

of any particular VER response does not reflect an easily

detectable pattern of change in cognitive functioning.

The major finding in this study of a consistent,

linear decrease in the variability of the whole VER.

response amplitude and its components complements and

extends thought and work done in the area of cognitive

development. As mentioned previously, thought in the

developmental area points to changes in flexibility of

cortical processing for proper adaptation to the

environment. White (1965) discusses the nature of

changes in children's learning processes which are known

to take place during the range of five to seven years of

age. He presents evidence for the idea of temporal

stacking in several kinds of learning. The five to seven

age period is perhaps a time when maturation inhibits a

broad spectrum of lower level functioning in favor of a

higher level of functioning. Literature is cited on

several learning paradigms in children which stress the

importance of the five to seven age range for changes in

learning and thinking processes.

Piaget also stressed important changes from four to

seven years of age (1962). During the intuitive phase of

the preoperational period (two to seven years of age) the

child begins to learn the concept of conservation. With

maturity, the child learns to respond more flexibly and

focuses on the more relevant aspects of an object or

situation in order to attain conservation skills.

Some theories of individual cognitive style are also

pertinent to the topic of response flexibility. Kagan's

reflectivity-impulsivity dimension (Kagan, Rosman, Day,

Albert, & Phillips, 1964) addresses the cognitive style

differences between children who take the time and

opportunity to try different alternatives and solutions

to a task situation and those who impulsively respond

after testing a limited number of hypotheses. In this

context, a reflective, flexible cognitive style generally

results in more accurate responding than a less varied

one. Witkin's field dependence and independence (Witkin,

Dyk, Faterson, Goodenough, & Karp, 1962) refers to the

tendency to perceive the perceptual field as undifferentiated

versus the tendency to analyze the constituents of the

field and perceive the different parts as separate from

the field. Children are apparently more field dependent

and this levels off by the teen years. The field inde-

pendent cognitive style implies a more flexible mode of

responding in that many parts of the picture are detected

rather than the whole picture being detected as one


White's (1965) and Piaget's (1962) theories may be

seen as conducive to a hypothesis of change, particularly

one toward decreasing variability, in cortical functioning.

Kagan et al. (1964) and Witkin et al. (1962) present ideas

that are more conducive to a hypothesis of increasing

change in cortical functioning variability with age.

Scott (1957) blends the two directions and suggests that

higher organisms balance a tendency toward behaving

variably and one toward behaving predictably. In a

typical learning situation an individual needs to display

a certain amount of variability in responding for proper

adjustment, but later behaves more predictably when

learning has occurred. Werner takes almost the opposite

viewpoint (Langer, 1970) stating that the child first

responds rigidly with reflex actions and then develops

more flexibility in his worldly interactions.

The developmental literature on the developing

stability or instability of cognitive organization is

extensive. In general, authors agree that cognitive

processing goes through some specific changes as the child

matures, but they seem to be equally split as to whether

increasing stability or flexibility is essential and

primary to normal cognitive functioning. Of course,

situations, conceptual constructs, and critical age

periods vary across theories making comparison somewhat

difficult. In this study, an obvious change in cortical

processing was noted and at most stimulus conditions

and locations variability was seen to decrease as a

function of age. The change toward less flexibility

implies that the cognitive functions reflected by the

measure utilized are becoming more narrow and efficient

in nature, possibly due to a practice effect. The change

may reflect a cognitive function that reached maturity

at an earlier age and repeated use has shaped into an

increasingly stable function. However, as discussed

earlier, the VER data also reflected increasing vari-

ability in the parietal lobe location suggesting that

functions in this area may be newly developing and, con-

sequently, increasing in flexibility. Significant to

the findings is the differing contribution by each sex

to the overall variability measures. The differences

in variability changes between males and females suggests

a striking and stable difference in the development of

cognitive functions at this age range.

The fact that the average IQ of these children was

well above the average may also have influenced the per-

ceived variability changes. Perhaps this group was

somewhat advanced in the development of their overall

cognitive functioning contributing to the detection of

decreasing variability in most conditions. An investi-

gation of a group of children with a lower average IQ,

and one with the parietal lobe location measured for

more than two years should certainly help clarify some

of these issues.

In conclusion, this study demonstrated significant

changes in the variability of cognitive processing as a

function of time as reflected by a measure of variability

of the VER. Evidence is provided for the developmental

hypothesis that brain functioning changes in flexibility

as children grow older. Overall, the changes appear to

be linear based on these data, although the changes

decrease or increase depending on the brain location of

collection of the VER. Patterns of variability change

were also substantially different depending on sex of the

subject. This finding has important indications for

understanding child growth and development, and for under-

standing and managing specific development difficulties

that appear to be sex-related, such as specific reading

disability, which is more prevalent in males than in

females (Wender, 1971).

The variability measure has provided a view of an

aspect of child development that is often neglected: that

of the flexibility of responding, which must necessarily

change over time for appropriate growth and adaptation to

the environment. Consistent patterns of change in vari-

ability were found implying that flexibility is an important

and, possibly inherent, dimension of the normal growth and

development of cognitive functioning. The finding of

different directions of change depending on brain location

supports the notion that different brain areas develop

at different rates and that this is reflected in behavior.

The implications for the difference in parietal lobe data,

depending on sex, to data relevant to the problem of

disabled readers indicates the possible utility of this

measure in the early detection of problems with cognitive

functions that develop at different rates. The

longitudinal data examined here provide the beginnings

of needed baseline data for the overall normal changes

in cognitive flexibility which may eventually be used

to assess abnormal growth and related difficulties in

the cognitive development of children.


Allison, T., Goff, W. R., Abrahamian, H. A., & Rosner, B. S.
The effects of barbiturate anesthesia upon human somato-
sensory evoked responses. EEG, 1963, Suppl. 24, 68-75.

Barnet, A. B., & Lodge, A. Click evoked EEG responses in
normal and developmentally retarded infants. Nature,
1967, 214, 252-255.

Barnet, A. B., Ohlrich, E. S., Weiss, I. P., & Shanks, B.
Auditory EPs during sleep in normal children from ten
days to three years of age. EEG, 1975, 39, 29-41.

Barr, A. J., Goodnight, J. H., Sall, J. P., & Helwig, J. T.
A user's guide to SAS 76. Raleigh, N.C.: Sparks Press,

Beery, K. E. Developmental test of visual-motor integration.
Chicago: Follett, 1967.

Beinhocker, G. D., Brooks, P. R., Anfenger, E., & Copenhaver,
R. M. Electroperimetry. IEEE Transactions in Biomedical
Engineering, 1966, 13, 11-18.

Bigum, H. A., Dustman, R. E., & Beck, E. C. Visual and
somatosensory evoked responses from mongoloid and normal
children. EEG, 1970, 28, 576-585.

Borge, G. F. Perceptual modulation and variability in
psychiatric patients. Archives of General Psychiatry,
1973, 29, 760-763.

Brazier, M. A. B. Information carrying characteristics of
brain responses. EEG, 1963, Suppl. 24, 55-67.

Brazier, M. A. B. A study of the variability of response
to flicker: The influence of other sensory stimuli.
In H. E. Henkes & L. H. van der Tweel (Eds.), Flicker.
The Hague: Dr. W. Junk Publishers, 1964, 221-237.

Buchsbaum, M. S. Average evoked response and stimulus
intensity in identical and fraternal twins. Physiological
Psychology, 1974, 2(3A), 365-370.

Callaway, E. Averaged evoked responses in psychiatry.
Journal of Nervous and Mental Diseases, 1966, 143, 80-92.

Callaway, E. Brain electrical potentials and individual
psychological differences. New York: Grune & Stratton,

Callaway, E., & Halliday, R. A. Evoked potential vari-
ability: Effects of age, amplitude and methods of
measurement. EEG, 1973, 34, 125-133.

Callaway, E., & Jones, R. T. Evoked responses for the study
of complex cognitive functions. In M. Kietzman, J. Zubin
& S. Sutton (Eds.), Experimental approaches to psycho-
pathology. New York: Academic Press, 1975.

Callaway, E., Jones, R. T., & Donchin, E. Auditory evoked
potential variability in schizophrenia. EEG, 1970,
29, 421-428.

Callaway, E. & Stone, G. C.- Evoked response methods for the
study of intelligence. Agressologie, 1969, 10(1), 1-5.

Chalke, F. C. R., & Ertl, J. Evoked potentials and intel-
ligence. Life Sciences, 1965, 4, 1319-1322.

Chapman, R. M., & Bragdon, H. R. Evoked responses to
numerical and nonnumerical visual stimuli while problem
solving. Nature, 1964, 203, 1155-1157.

Ciganek, L. Variability of the human visual evoked potential:
Normative data. EEG, 1969, 27, 35-42.

Conners, C. K. Cortical evoked response in children with
learning disorders. Psychophysiology, 1970, 7, 418-428-.

Ellingson, R. J. Variability of VERs in the human newborn.
EEG, 1970, 29, 10-19.

Ertl, J. Evoked potential, neural efficiency, & IQ. In
L. D. Proctor (Ed.), Biocybernetics of the central
nervous system. Boston: Little & Brown, 1969, 419-428.

Garcia-Austt, E. Influence of the states of awareness upon
sensory evoked potentials. EEG, 1963, Suppl. 24, 76-89.

Halliday, R., Rosenthal, J. H., Naylor, H., & Callaway, E.
Averaged evoked potential predictors of clinical improve-
ment in hyperactive children treated with methylphenidate:
An initial study and replication. Psychophysiology,
1976, 13, 429-439.

Hubel, D. H., & Wiesel, T. Receptive fields, binocular
interaction and functional architecture in the cat's
visual cortex. Journal of Physiology, 1962, 160,

Inderbitzen, L. B., Buchsbaum, M., & Silverman, J. EEG-
averaged evoked responses and perceptual variability
in schizophrenia. Archives of General Psychiatry, 1970,
23, 438-444.

Jasper, H. H. The ten-twenty electrode system of the
international federation. EEG, 1958, 10, 370-375.

John, E. R., Ruchkin, D. S., & Villegas, J. Experimental
background: Signal analysis and behavior correlates of
evoked potential configurations in cats. Annals of the
New York Academy of Science, 1964, 112, 362-420.

Kagan, J., Rosman, B. L., Day, D., Albert, J. & Phillips,
W. Information processing in the child: Significance
of analytic and reflective attitudes. Psychological
Monographs, 1964 (Whole No. 578).

Langer, J. Werner's comparative organismic theory. In P.
Mussen (Ed.), Carmichael's manual of child psychology,
Vol. 1 (3rd ed.). New York: Wiley, 1970.

Lifshitz, K. An examination of EPs as indicators of informa-
tion processing in normal and schizophrenic subjects.
In E. Donchin & D. B. Lindsley (Eds.), Average evoked
potentials: Methods, results, and evaluations. Wash-
ington, D. C.: NASA SP-191, 1969, 357-362.

Luria, A. R. Higher cortical functions in man. New York:
Basic Books, 1966, 52.

Lynch, J. C., Mountcastle, V. B., Talbot, W. H., & Yin,
T. C. T. Parietal lobe mechanisms for directed visual
attention. Journal of Neurophysiology, 1977, 40(2),

Mussen, P. H. (Ed.) Carmichael's manual of child psychology,
Vol. 1 (3rd ed.). New York: Wiley, 1970.

Perry, N. W., Jr., & Childers, D. G. The human visual
evoked response: Method and theory. Springfield:
Thomas, 1969.

Piaget, J. The stages of the intellectual development of
the child. Bulletin of the Menninger Clinic, 1962,
26(3), 120-145.

Preston, M. S., Guthrie, J. T., & Childs, B. Visual evoked
responses (VERs) in normal and disabled readers.
Psychophysiology, 1974, 11, 452-457.

Preston, M. S., Guthrie, J. T., Kirsch, I., Gertman, D., &
Childs, B. VERs in normal and disabled adult readers.
Psychophysiology, 1977, 14, 8-14.

Rappaport, M., Hopkins, H. K., Hall, K., Belleza, T., &
Hall, R. A. Schizophrenia and evoked potentials:
Maximum amplitude, frequency of peaks, variability and
phenothiazine effects. Psychophysiology, 1975, 12,

Rhodes, L. E., Dustman, R. E., & Beck, E. C. The VER: A
comparison of bright and dull children. EEG, 1969,
27, 364-372.

Robinson, D. N. Statistical features of brain responses
under conditions of attention and habituation.
Psychological Reports, 1975, 32, 287-292.

Scott, J. P. The genetic and environmental differential
of behavior. In D. B. Harris (Ed.), The concept of
development. Minnesota: University of Minnesota
Press, 1957.

Shucard, D. W. & Horn, J. L. Evoked cortical potentials
and measurement of human abilities. Journal of Com-
parative and Physiological Psychology, 1972, 78, 59-68.

Street, W. J., Perry, N. W., Jr., & Cunningham, W. R. A
factor analysis of visual evoked responses. Psycho-
physiology, 1976, 13, 352-356.

Sutton, S., Braun, M., Zubin, J., & John, E. R. Evoked
potential correlates of stimulus uncertainty. Science,
1965, 150, 1187-1188.

Uttal, W. R. Do compound evoked potentials reflect psycho-
logical codes? Psychological Bulletin, 1965, 64, 377-392.

Wechsler, D. Manual for Wechsler preschool and primary
scale of intelligence. New York: The Psychological
Corporation, 1967.

Wechsler, D. Manual for Wechsler intelligence scale for
children--Revised. New York: The Psychological
Corporation, 1974.

Wender, P. H. Minimal brain dysfunction in children. New
York: Wiley, 1971.

Werre, P. F., & Smith, C. J. Variability of responses
evoked by flashes in man. EEG, 1964, 17, 644-652.

White, S. H. Evidence for a hierarchical arrangement of
learning processes. In L. P. Lipsitt & C. C. Spiker
(Eds.), Advances in child development and behavior,
Vol. 2. New York: Academic Press, 1965, 187-220.

Witkin, H. A., Dyk, R. B., Faterson, H. F., Goodenough,
D. R., & Karp, S. A. Psychological differentiation.
New York: Wiley, 1962.


Yakovlev, P. I., & Lecours, A. The myelogenetic cycles
of regional maturation of the brain. In A. Minkowski
(Ed.), Regional development of the brain in early
life. Philadelphia: F. A. Davis Company, 1967,


Name: W. Jeanne Street

I was born on October 1, 1952, in Lexington, Virginia,

in what was once the home of General Stonewall Jackson, was

at the time of my birth a military hospital, and is now a

museum to General Jackson. As my father was in the military

service my family, which consisted of my parents, two older

sisters, an older brother, and myself, moved and traveled

quite a bit. I lived in Ohio, Alabama, West Germany, and

Massachusetts before settling in Florida upon my father's

retirement from the service in 1965. The moving and tra-

veling had been fun and stimulating, but it was a welcome

change to settle in one place.

I graduated from Clearwater High School, Clearwater,

Florida, in 1970 after being involved in academics, student

government, the school newspaper, and numerous community

activities. Upon entering the University of Florida I

concentrated my efforts on my education in psychology and

received my B.A. in 1973, my M.A. in 1975, and now my Ph.D.

in 1978 in clinical psychology. I did my clinical psychology

internship at Duke University Medical Center, Durham, North

Carolina, and am currently employed there in the Center for

the Study of Aging and Human Development and the Division

of Medical Psychology.

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of
Doctor of Philosophy.

Nathan W. Perry, Jr., Chdirman
Professor of Clinical Psychology

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of
Doctor of Philosophy.

Robert L. Isaacson
Professor of Physiological

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of
Doctor of Philosophy.

Donald Childers
Acting Chairman
Professor of Electrical

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of
Doctor of Philosophy.

Wiley asbury
Associate Professor of Clinical

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of
Doctor of Philosophy.

Vernon Van 'DVRiet
Associate Professor of Clinical

This dissertation was submitted to the Graduate
Faculty of the Department of Clinical Psychology in
the College of Arts and Sciences and to the Graduate
Council, and was accepted as partial fulfillment of
the requirements for the degree of Doctor of Philosophy.

March 1978

Dean, Graduate School

3 1111111 262 08553 6364111111 1111111111111
3 1262 08553 6364

A u2