EFFECTS OF SOUND TRANSMISSION IN AN AQUEOUS AND AIR
ENVIRONMENT ON INFANT MOTOR BEHAVIOR
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THEE REQUIREMENTS FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
To Judy and my parents
The people that I have to thank occupy a special
place in my life. They, for that reason, are more and
mean more to me than I am able to describe.
First, Dr. John Newell, who as my chairman helped me
through school and the research and the writing. Above
that he helped me as a teacher and a person by serving as
a source of supreme stability.
Next, Dr. William Ware, who taught me about research
design and statistics. He, along with his wife Andrea
and children Emily, Matt and Erin, is among the people
that define true friendship for me.
Dr. William Wolking, who helped me realize the importance
of the experimental analysis of behavior and always was
there with the probing questions about my work.
To Arthur Newman, a friend and someone that helped
me put a different perspective in some of my thinking.
The six families of the infants that served as subjects
gave of their time and energy. In return, I became very
close to several of them by teaching their infants to swim.
To my parents, their constant support and belief in me
helped in a way I can't describe.
Finally, to my wife Judy, her help, support and love
were and will always be deeply appreciated.
TABLE OF CONTENTS
ACKNOWLEDGMENTS . . . . . . . . . iii
LIST OF TABLES . . . . . . . ... v
LIST OF FIGURES . . . . . . . . .. vi
ABSTRACT . . . . . . . . . .. . vii
I. INTRODUCTION . . . . . . . . 1
II. REVIEW OF LITERATURE . . . . . . 6
III. METHODS AND PROCEDURES . . . . .. 35
IV. RESULTS . . . . . . . ... 47
V. DISCUSSION AND CONCLUSION . . . .. 51
APPENDIX . . . . . . . . . .. . 59
BIBLIOGRAPHY . . . . . . . . .. 68
BIOGRAPHICAL SKETCH . . . . . . . .. 76
LIST OF TABLES
1. Schedule of Presentations of
Sourd Conditions . . . . . . .. 38
2. Condition Mean Coefficients for Five
Planned Non-Orthogonal Comparisons ... 45
3. Mea= Kick Rate per Second for Six
Conditions and Calculated Observed
Differences . . . . . . ... 49
LIST OF FIGURES
1. Underwater Speaker Unit . . . . .. 64
2. Surface Speaker Arrangement . . . .. 65
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment
of the Requirerents for the Degree of
Doctor of Philosophy
EFFECTS OF SOUND TRANSMISSION IN AN AQUEOUS AND AIR
ENVIRONMENT O I:;FANT MOTOR BEHAVIOR
Chairman: Jnhn M. Newell
Major Department: Foundations of Education
The purpose of this study was to examine the differences
in leg kicking rates for six infants ages ten and one half
months old as they heard five seconds of a heartbeat sound
while underwater as compared to hearing the heartbeat sound
in air for five seconds.
A comparison was also made between the heartbeat sound
presented underwater and three other sounds presented under-
water. Those three other underwater sound presentations
were rhythmic music, white noise, and human speech each
being presented for five seconds. A total of six sounds
were presented to each infant each day, five underwater
treatments and one treatment on land.
Each trial was filmed with an underwater and surface
movie camera for later coding of the ten continuous days of
data collection. The data were analyzed by a two way ANOVA
and a Dunn's planned comparison among the means. Significantly
less kicking was observed for the heartbeat heard in air
condition than when the heartbeat sound was present-ed to
the infant underwater.
When the underwater heartbeat sound condition -_,ch
was a tape made from a human female heartbeat recorfd- in
utero was presented to the infant subjects, they kiz=sd
significantly less than when they heard the other uierwater
sound conditions. The first underwater sound cond:=.n,
the heartbeat condition was compared to a control c:dtition
for any rhythmical sound as represented for this sta-y by
Tchaikovsky's Sleeping Beauty.
The second underwater sound was designed to ser,- as
a control for any sound being heard underwater. Thas con-
dition was presented by playing "white noise" or fres-encies
between 20 and 20,000 c.p.s. simultaneously.
The final underwater condition was designed to -cntrol
for a familiar sound that the subjects heard previously.
That condition was created by having each subject's nether
record herself reading the introduction to Alice in
A comparison of those three underwater sound c- ditions
mentioned above to the underwater heartbeat showed -te
infants kicked significantly less to the heartbeat z:und
than to the other underwater sounds.
The results suggest that bodily reactions of tte infant
may be influenced not only by what they hear, but the
mechanisms used to perceive the sound. Since undervater
sounds transduce the entire body, this type of sound
transmission takes on tactual properties. It is noted
that the fetus hears sounds such as the mother's heartbeat
by this mechanism. Possibly our earliest learning involves
perception of sound by tactual means.
The results of this study raise questions about prior
research on infant conditioning and suggestions are offered
for future studies.
What is the first thing human beings learn? Previous
research on conditioning of the fetus has reported that the
fetus can react to a number of stimuli and that conditioning
can sometimes be demonstrated. A number of studies have
attempted to demonstrate similar kinds of conditioning later
in infancy but the results were not consistent. In some
cases, later conditioning has been demonstrated but in the
majority of studies conditioning established in the fetus
has not been demonstrated later in infancy. One of the con-
clusions reached from such studies was that the fetal con-
ditioning was not actually established since it could not
be demonstrated later in infancy.
An alternative explanation, which provides the basis
for the present study, is that the "environment" in which
the conditioning took place was different for the fetus and
the infant. The fetus was conditioned in the womb sur-
rounded by amniotic fluid and could be viewed as conditioning
in water while the infant is being exposed to the stimulus
in an air environment. It may be that this difference in
the environments where conditioning took place could be a
significant factor in explaining the discrepant results
reported on fetal and neonatal conditioning.
In light of the confusion in the literature concerning
fetal conditioning, and the as yet not totally accepted idea
that a heartbeat sound as a calming effect on infants, it
appears reasonable to test two aspects of this problem. One
question needing further study is whether or not there is
a difference between air and water sound transmission for
infants where the sound is a heartbeat? In order to examine
this question, two additional fields of research need to be
mentioned. Does the heartbeat sound have a calming effect
in air, and what is the utility of using leg kicking as a
dependent measure for these parameters? The other aspect
concerning the original question is whether or not there is
a calming effect for infants when a heartbeat is presented
underwater when compared to other sounds presented under-
The purpose of the present study is to examine the
differences in a leg kick response in infants to sounds
transmitted in air and through water. The air environ-
ment is the "normal" environment for infants and the water
environment of a swimming pool is an attempt to simulate
the fetal womb environment. If differences can be demon-
strated, they would suggest that the prior research in this
area must take into account the environment within which
the conditioning occurred. This study should also add to
an understanding of how early learning occurs and provide
additional information about the early growth and develop-
ment of sensory capacities of infants and young children.
Some of the conceptual assumptions for the present
study involve ideas put forward by Lee Salk (1966) and H.
MKurka (1974). They have contended that the heart beat
so-nd, because it is heard in utero and therefore associated
with that environment, has a soothing or calming effect on
the individual after birth. Research reporting results
which disagree with those of Salk and Murooka will be
reviewed in Chapter II along with a more complete review of
Salk's and Murooka's work.
Another assumption for the present study involves the
mediums of water and air and their differences as far as
sc-d transmission through them is concerned. There is
considerable evidence that the physical properties of sound
transmitted through water and air are different. The
transmission of sound is faster underwater than in air due
to the greater density of water and, for the same reason,
mere energy is required to transmit sound of equal amplitude
through water than in air. In situations where humans are
hearing underwater sounds, different factors involving audi-
tion become apparent. Humans are able to localize sounds
in air more easily than in water because the sound may be
"blocked" by the head. It is difficult to localize sound
runerwater because the sound is not blocked by other body
parts and in fact the entire body can be transduced by
scand in this medium. If the sounds transmitted underwater
are of sufficient amplitude, sound perception may become
more tactual in nature. The tactual nature of underwater
sound may be an important variable when one considers that
the fetus experiences various sounds while totally immersed
in amniotic fluid and that this experience may be as much
tactual as auditory.
The use of sound as a stimulus for conditioning has
much to recommend it when one considers the sensory equip-
ment of the fetus. The sense of vision even in the neonate
is not well developed. Visual acuity is somewhere between
20/350 and 20/450 (Gorman, et al., 1957). While little is
known about the sense of taste in the fetus, there is evi-
dence to suggest that the sense of taste in new born infants
is not well developed and that the neonate probably can
distinguish only the taste of salt and sweet. While the
sense of touch is apparently quite well developed, there
are difficulties in actually touching the fetus while it
is still in the womb. There is evidence that the fetus does
touch himself in the womb and several dramatic photographs
attest to this. The neonate emits reflexes to touch that
are stable enough to suggest that the nervous system of the
neonate is well established for the sense of touch.
Babies at birth can differentiate between certain low
frequencies and some reflexes can be elicited by sound of
different amplitudes. The evidence suggests that the senses
of hearing and touch are probably the best developed in
the fetus and should serve as the modalities to be stimulated
for conditioning purposes.
If, as suggested earlier, the sensing of sound in the
fetus is tactual as well as auditory, the combined senses
of sound and touch may be contributing to the conditioning
of the fetus. In the present study, the air and water con-
ditions will provide differential environments in which
only the sense of hearing will be operative in the air
environment while the combined senses of hearing and touch
could possibly be operative in the water environment. If
there are differences in responding to sounds heard by an
infant in the water environment, when compared to the air
environment, it may mean that our understanding of hearing
and how it relates to early perception of the environment
in the infant is incomplete unless we include the tactual
component of sound experienced by the fetus.
In order to investigate the differences between an air
transmitted heartbeat sound and an underwater heartbeat
sound, six infants were presented with these conditions.
An additional four underwater sound conditions were presented
as controls for various properties of a heartbeat sound.
REVIEW OF LITERATURE
This study examined the differences in leg kicking
activity of infants exposed to a heartbeat sound presented
in air when compared to hearing the same heartbeat sound
presented underwater. It also examined the differences
between leg kicking responses as the infants heard the
underwater heartbeat sound and four other underwater sound
conditions. These other underwater sound conditions in-
cluded: a rhythmical sound, a familiar sound, a hissing
sound and silence.
This study builds upon conclusions derived from several
different and seemingly unrelated areas of research. Given
this study's goal of investigating the differences mentioned
above, three major areas of literature were seen as being
relevant for review: developmental aspects of the fetus and
infants, methodological problems encountered in research
involving infants and the research on fetal and infant con-
The literature review which focuses upon the develop-
mental concept, will include research on fetal movement and
hearing capacities, as well as the relative effectiveness
of the other fetal and neonatal sensory capacities. Other
studies were reported from this area if they provided in-
formation about sensory acuity, habituation to certain
stimuli and other developmentally related systems of a
fetus, neonate or infant. Studies using infants and
particularly infants underwater as subjects have certain
methodological differences from studies not using infants.
Among the studies reviewed in this area were those concern_,
with reliable observation of infant rotor behaviors. The
advantage of filming these infant activities to provide a
permanent record that could be re-evaluated later was
documented in the review of literature. Within this methon:-
logical area the literature concerning different neonatal
and infants measurements of activity including heart rate
and motor activity was reviewed. Finally, the literature
concerned with learning in the fetus and infant and im-
printing of sound stimuli on the fetus is presented.
Watson in 1919b made statements based on his observa-
tions of infant behavior in water. McGraw cited Watson
(1919b) and herself (1935) as the only two who used actual
experimental evidence to describe swimming-like behaviors
in infants and newborns. Watson tested newborns immediatE-
after birth in water at 98.6 degrees Farenheit. He observrd
three babies placed on their backs and supported by the
hands of the experimenter, which allowed the subjects'
faces to remain above the surface of the water. Watson
reported (1919a), "Violent expression of fear--a cry,
checking of breathing, followed by deeper inspiration and
rapid, entirely uncoordinated, slashing of hands and feet
were all that could be observed" (p. 243). On the basis of
these observations, Watson rejected the evidence of swimming
in the behavioral repertoire of the newborn.
McGraw (1939a) used notes she made while directly
observing and cinemagraphically recording the aquatic be-
haviors of forty-two different infants ranging in age from
eleven days to thirty months. She made a total of 445 observa-
tions and out of all those observations, 164 were motion
picture records while the remaining 281 were written notes
taken during the observations. The number of repeated
observations on the same infant varied from two to 51 with
a median of ten. McGraw pointed out a limitation of her
study in the area of external validity. The data did not
justify a longitudinal analysis. On the other hand, the
limited number of different age brackets did not support a
purely cross-sectional analysis. In spite of the methodo-
locical limitations of her study, the notes and film she
took indicated a swimming behavior pattern characteristic
of newborns, babies, four months to twenty-four months, and
children older than twenty-four months. McGraw stated
that if Watson had used a prone and not a supine position
for his subjects, he would have noted as she did that,
"Definitive rhythmical associated flexor-extensor nove-
ments in upper and lower extremities together with a lateral
flexion of the trunk corresponding to the flexor phase of
the lower extremity are usually manifested" (p. 486). She
further noted that these behaviors were more pronounced
when the baby was free of support while underwater.
It appears that where Watson, with his subjects on
their back and able to breathe, observed a violent expres-
sion of fear, McGraw, by letting them float face down,
observed no such drastic behaviors. She also noted activity
so consistent that she cited these behaviors in water as
indicative of developmental changes or maturational stages
in the infant.
Several studies relating auditory stimulation where
the dependent variable was the change in the neonates or
infant's heart rate have been reported. Studies in this
area have been represented by Kearsley (1973) who recorded
head movements, and Steinschneider (1968) who found that
sound intensity had to be above 70 db. to get much reaction
from either cardiac or respiratory activities. The in-
fluence of acoustic sound on habituation for the newborn
and infant populations is well documented in the literature.
Representative studies in this area have been done by Pomerleau-
Malcuit and Clifton (1973); Berg (1972); Lewis (1971);
McCall and Nelson (1970); Roberts and Campbell (1967);
Smith (1967); Steinschneider, Lipton and Richmond (1966);
and by Keen, Chase and Graham (1965).
First to report the phenomenon of the effect of sound
on infants however, was probably Plato when he reported his
observations concerning this in Laws Book VII, Section 790
(380 B.C.); "You know, when mothers want to put fractious
babies to sleep, the remedy they exhibited is not still-
ness, but its very opposite, movement--they regularly rock
the infants in their arms--and not silence, but a tune of
some kind" (p. 41). From Biehler (1976) comes an observation
that Francis Warner (1887-1888) mentioned how sounds soothe
small children in his book The Children: How to Study Them.
Studies where body movements, among other things, were
measured in response to auditory stimulation are represented
by: Stubbs (1934); Ashton (1973); Moreau et al. (1970);
Ling (1972); Heriot (1970); Hoversten (1969); Ashton (1971);
Pratt (1934); and Eisenberg (1965).
Stubbs (1934) cited Silvo Canestrini as being the
first to use objective recordings of infant responses to
sounds in 1913. She further noted that in Pratt (1934),
a stabilimeter polygraph was used to record the infant's
changes in activity as a function of hearing certain sounds.
Stubb's (1934) research showed that bodily reactions were
effected better by duration and intensity than by intensity
alone. She used durations of three, five, and fifteen
seconds and found the last two to be acceptable. Using
intensities of 30, 50, 70, and 85 db. she found the last
two again to be conditions that would predict greater bodily
Weiss (1934) used pure tones of 50 and 75 db. with a
duration of five minutes and found the louder tones had
greater quieting effects on infants. Similar findings
were reported by Birns, Blank, Bridger, and Escalona
Weiss (1934) also carried out experiments in which
the sounds were presented in both a dark and a light environ-
ment. She reported that, "It has been found that light alone
is quieting and that sound alone (in darkness) is quieting"
(p. 66). This question of sound and the effect it has on
the neonate was re-examined by Lee Salk (1960) when he
reported that infants hearing a heart beat sound at 85 db.
and at a rate of 72 beats per minute continuously for four
days gained more weight, cried less, and were more quiet
than the controls who were cared for in a quiet room. In
trying to replicate this study, Tulloch et al. (1964)
failed to get the same results reported by Salk, but they
exposed their experimental group to a simulated heart
beat sound at 45 db. That intensity was probably too low
for neonatal reaction as Brackbill et al. (1966) stated,
"Their failure to replicate Salk's results is probably
attributable to the choice of stimulus intensity level
that is not clearly audible to neonates" (p. 179).
Later, Salk (1961) played a heartbeat at 72 beats
per minute, a metronome ticking at 72 beats per minute,
and lullabies. Of note, he found the heartbeat sound to
be twice as efficient at putting the children to sleep
than silence, metronome and lullaby conditions.
Ling (1972) tested 160 neonates hearing a pulsing
narrow band noise stimulus and found that gross body move-
ments increased as the duration of the stimulus became
longer. Interestingly, he also reported that continuous
stimulation increased bodily reaction better than inter-
rupted stimulation of the same duration. Pratt (1934)
reported that the physiological state (i.e.: alert vs
sleepy) of the infant was more powerful than auditory
stimulation to predict activity in the newborn. Ashton
(1971) found that neonates were differentially responsive
to changes in frequency only during quiet sleep. Moreau,
Birch and Turkewitz (1970) compared the effects of habitua-
tion to an acoustic stimulus in newborns by their autonomic
and a musculoskeletal response. They found that the
musculoskeletal response habituated faster, was independent
of habituation in the autonomic response, and was also more
stable. Behavioral changes were recorded by observers in
a study by Hoversten and Moncur (1969) where they reported
that a pulsed voice stimulus had the largest response
percentages for both three month and eight month old infants.
In a study by Foreschels and Beebe (1946) involving bodily
movements and facial expressions, 33 infants were observed
making responses to auditory stimulation. Whistles had
the greatest effect with a tuning fork having little or
A study looking at newborn hearing capacity was reported
also by Aldrich (1928) and, by its title, "A New Test for
Hearing in the Newborn: the Conditioned Reflex," gives
the impression of being the first of its kind which
examined hearing for this population. The reflexive be-
haviors resulting from a newborn hearing a two pound iron
weight falling three feet onto a hard wood surface were
filmed and reported in an article by Clark (1939). Ob-
viously, normal hearing newborns ey-xbited the Moro reflex
but, as they became older, the typography of the response
began to resemble a startle reaction.
A multitude of studies have been done on hearing acuity,
sound localization, adaption, habituation, discrimination
and other aspects of an infant hearing sounds. With
respect to recognizing sounds, Vanderplas and Blake (1949)
reported that a valued and familiar stimuli may actually
decrease the recognition threshold far that stimulus. While
that isn't too surprising, a study by Hutt et al. (1968)
reported that the most effective stimuli for eliciting
behavioral responses in neonates were fundamental fre-
quencies found in the range of the hb.man voice. Hutt sug-
gested that there may be survival vmlue in this because the
human cochlea is structurally mature at birth and that may
mean early hearing acuity plays a zart in the affectional
child-parent bond. Vanderplas and BLake's (1949) familiarity
and value selectivity for auditory stimulation may have
reported results that were more physiologically based when
applied to infant populations. Kasatkin and Levikova (1935)
reported three stages of conditioning auditory reflexes
in infants. They were indifference, inhibition of move-
ments where the first signs of conditioned reacti=zs began,
and the stage of clearly conditioned reaction. It might
be that a sound heard early or before birth may h5 e im-
printing properties as postulated by Salk (1960, ::il, 1962).
Bernard and Sontag (1947) reported that the human fatus in
utero is capable of perceiving a wide range of toits after
their experimental data were analyzed.
Grimwade, Walker and Wood (1970) worked with eaimals
and humans in examining several properties of fetal environ-
ment. By inserting a microphone into pregnant women they
found the fetus did not receive high frequency so=ds,
sound loss was appreciable, and sound stimulation :- the
fetus changed its heart rate and muscular activity- Kellogg
(1941) used auditory stimulation with the fetus wtile
recording movements of the fetus by mechanical te=-iiques.
Among other results Kellogg reported that after 2-1 days
the fetal movement provided evidence of response i: auditory
stimulation. Forbes and Forbes (1927) reported *itt the
fetus will react to a loud sudden noise 30 days before
delivery, but they only hypothesized that the fet. was
hearing or employing tactual modalities to perceive auditory
stimulation. The manner in which the fetus devel:rs and
how the sensory apparatus becomes mature may supp--t the
ideas put forth by McGraw (1939a).
If McGraw (1939a) was correct with her postulation
about infant aquatic behaviors reflecting the phylogenic
origin of man, Schevill (1953) added support with a study
showing that the bottlenosed porpoise has a frequency range
that approximates the bone-conducted thresholds of man's
hearing. The proximodistal, cephalocaudal ontogenetic
development of aquatic behavior exhibited by McGraw's (1939a)
subjects supports the differential activity of the opposing
ends of the fetus recorded by Kellogg (1941) where the
caudal end is never as active as the head end of the fetus.
If the proximodistal trend of development for the whole
body is valid for the cochlea in the human ear that Hutt
et al. (1968) described as being fully mature at birth,
then Dishoeck's (1948) observation that the outer hair
cells of the cochlea (last to develop) are sensitive for
air conduction and the inner hair cells (first to develop)
are stimulated by compressive tones found in mediums like
bone and fluids are valid. A fetus may have a greater
capacity to hear compressive tones while in the amniotic
fluid. This ability exhibited by the fetus to hear compres-
sive tones would have the same frequency limits as those
of the bottlenosed porpoise researched by Schevill (1953).
This may support McGraw's (1939a contention concerning
the phylogency of man cited earlier.
The auditory capacity of humans underwater changed
because as terrestrial animals we typically hear sounds
transmitted through air. Research has been performed on
humans while underwater. However, such underwater research
has been performed on adult divers and not infants.
The above studies probed the relationship of sounds
heard underwater by adult humans. Research into the in
utero sounds heard by the human fetus as it floats totally
immersed in amniotic fluid was completed by Salk (1966, 1962,
1961, 1960). Salk felt that these early sounds and particularly
the heartbeat sound heard by the fetus took on the characteris-
tics of an imprinted stimulus.
The research which has been cited concerning the develop-
ment of various capacities in the fetus, neonate and infant
suggest the following conclusions. There is evidence that
the fetus reacts to sound stimuli after the seventh month
of gestation when a vibro-tactile stimulus is used. The
sense of hearing is more fully developed in the fetus than
are the other senses. There are suggestions that a simulated
heartbeat sound has a soothing effect on infants when that
sound is presented at a level of 65 db. or higher. There
is conjecture about the effect of the duration of sound
presentation on the behavior of the fetus. Stubbs (1934)
argues for a duration of five to fifteen seconds.
It appears that the sense of hearing is an appropriate
sense to use in conditioning of a fetus or infant and that
a heartbeat sound will produce a relaxed or lowered fre-
quency of response in the infant. The present study uses
a heartbeat sound as one stimulus for conditioning a leg-
Cinemagraphic data collection means the behaviors under
investigation were filmed while cinemanalysis usually refers
to filming the activity and using that film exclusively as
the observations to be coded or scored.
McGraw considered the aquatic behaviors she filmed and
directly observed to be of evolutionary importance. To
that end, she performed a comparative study to show that this
swimming reflex apparent in her infant subjects may be at-
tributable to some earlier amphibian existence of homo
The comparative study by McGraw (1939b) involved filming
young quadrupeds in water; notably an opossum, a rat, a
kitten, a guinea pig, monkeys, and a two-month-old chimpanzee.
The purpose of that study was to see if the newborn swimming
reflex, ". . bears striking testimony to the phylogenesis
of man (p. 485). She felt that the swimming reflex was
related to man's evolution because in interpreting her study
she stated, "It is interesting that developmental changes
in swimming behavior correspond in chronological order to
the major phases of other behavior patterns which also appear
to be of phylogenetic origin" (p. 490).
McGraw also noted that the four-month old to twenty-
four month old human subjects usually rotated to a dorsal
position when submerged in a prone position and the rhythmical
aquatic behavior observed earlier had become what she called
the disorganized phase. Her later research into this area
was also oriented toward physiological development of
neural brain centers in the infant in an attempt to explain
the phases of aquatic behavior she filmed and observed in
1939. Cinemagraphic techniques for infant behaviors started
with Watson's (1919b)480 foot silent movie. It was heralded
in the literature as being of importance both historically
and experimentally. As noted previously, Watson filmed an
absence of swimming movements but his film included grasping
and Babinski reflexes, defensive movements, reaching,
crawling and some basic emotional reactions of the infant.
Since Watson's early work, cinemagraphic data collection
grew in popularity in the late thirties and early forties.
From the mid sixties until present day, the introduction
of the transistor and light weight television cameras have
all but replaced film with video tape.
Gesell (1932) cited cinemagraphic techniques as being
ideally suited to record morphological development and
particularly complex movements. Gesell went so far as to
develop normative data on a variety of infant capacities
that were filmed. He concluded that this type of data col-
lection allowed systematic recording and experimental
analysis. Several researchers have used film to record a
bide variety of neonate and infant behaviors.
McGraw (1941) filmed and compared mass responses in
newborns and specific responses in older babies to a pin
prick stimulus probing various body parts. Ames (1944)
filmed the visual and motor patterns of two normal infants
from their eighth to fifty-sixth week of life.
Ames (1939), had previously used cinemanalysis to
describe ascent postures and the body orientations beginning
with prone progression and ending with stair climbing. In
addition, Ames found that although quantative analysis of
the filmed behaviors of his twelve subjects between one and
three years of age differed, there were patterns within a
single subject that were constant.
Gesell (1932) cited several benefits of cinemanalysis
and Ames (1939), having filmed body orientations for com-
parisons of babies, published a study (Gesell and A-es, 1940)
in which the data were taken exclusively from filmed trials.
In the same year, Birkmayer and Goll (1940) used a film
method to analyze the development of walking in young
children. Their study is of importance in that they were
able to describe dynamic functions such as gait, center of
gravity shifts, stepping phase and other actions difficult
to examine except by stopping the action of motion picture
photography. Probably the most detailed study of rotor
activity using film was produced by Ames, (1942). A descrip-
tion of leg, foot, and toe postures was written after the
film had been coded of infants in a supine posture.
Gilmer (1933) also used cinemanalysis to investigate
unique limb movement patterns in a series of infant behaviors.
Cinemanalysis included both the filming of the activity as
well as an analysis of the activity based upon observations
by the experimenter. Two major types of observational
systems are used; an "open" and a "closed" system. In a
discussion of "open" and closedd" observation systems,
Gordon and Jester (1973) stated that,
Specimen description r "s detailed sequential
recording of all that :s taking place in a
situation. This may be accomplished by audio-
or-video-tape recording :r by continuous note-
taking. Following the rrrording, a coding or
scoring scheme is impose- on the record as
contrasted to coding during the initial collec-
tion of data. Statist;r2l analysis was done
only after the original records had been coded.
Wright (1960) called th:s an open method because
the coding follows the rc:ent-to-moment recording.
A distinguishing characteristic of open and
closed systems is that a-r open system preserved
the raw data so that it was available for other
coding systems while the closed system imposed
a coding scheme at the :rre of initial data col-
lection. In the latter, the observer simply
notes specific behaviors as they occur. Examples
of closed methods are t:-e and event sampling
and trait rating. In t-e closed system, ef-
ficiency of basic data recording is gained at
the loss of raw data. right's classification of
Time Sampling can be further divided into sign
and category systems. Sign systems preserve
some elements of the raw data since behaviors
are listed in the observer's record and he
usually makes some sort of frequency count of
the occurrence of specific behaviors. (pp. 175-176)
In a study investigating research procedures involving
auditory perception and resulting behavior in the human neo-
nates, Eisenberg (1965) compared cinemanalysis and direct
observation. She concluded that,
Cinetechniques are val-able in that even fleeting
changes in behavior ca.n he captured, and with
the aid of sophisticate- accessories and data
reducing procedures, ana-.lzed quantitatively.
Moreover, no ancillary measures arc required
since observer judgments can be validated
statistically by including no stirus_ trials
in a silent film record. Motion pi,-:re
photography is prohibitively expens_-- for most
purposes, however, and it poses te-rc al
problems of sizable dimensions. It -: recom-
mended only for special purpose res-zzh that
is otherwise impossible. (p. 170)
She further stated that using obser-rs with certain
parameters of sound was risky. For instc-ces, when coding
intensity, she reported: "For 45 to 55 :. three independent
examiners reported less than one-third cf -teir observations
with confidence ." (p. 164) and fcr duration and fre-
quency parameters, "although optimal consr-ts for clinical
and research purposes remain to be define, durations in the
range between one to five seconds seem t: ie reasonably
satisfactory for present use with wide-b-zd stimuli" (p. 165).
While many investigators in the 193L's and 1940's
employed film as a recording device for mncor activities,
Escalona and Leitch (1950) edited a large amount of film
to produce a 42 minute film examining rei-cnses made of the
prolonged stimulation of eight infants. "-eir study sug-
gested that prolonged perceptual stimulat-:n may generate
psychological tension in infants.
Possibly one of the most interesting studies based upon
data taken from film was done by Hooker 1>39). Reflexive
movements of aborted human fetuses to ta=-1e stimulation
from seven to 25 weeks of menstrual age wre filmed. Hooker
found that at between 11 to 14 weeks of ne:strual age most
of the neonatal reflexes were represented azd after 14 weeks
all the movements observed gained in amplitude after closely
observing what he filmed.
Film provides a permanent record that can be coded by
others to enhance coder reliability in the data. It is
important to keep in mind, however, that other factors ncre
basic to research design and not just precise measurement
need to be considered. Ling, Ling and Doehring (1970)
found that an observer's judgments about infant behavior
may be significantly influenced by knowing the presentation
schedule for the stimuli. Their study involved observer;
whole-body movements as the infants were presented with
narrow band acoustic stimulation.
The use of visual methods of recording infant behavior,
beginning with movie cameras and more recently television
recordings, has provided a useful means of providing a
permanent record of the behavior in question. Gordon
and Jester (1973) review the assets and limitations of
two observational systems. The "open" system provides for
the recording of behavior as it occurs with an analysis or
coding of the behaviors being done at a later time. A
"closed" system imposes a coding scheme on the behavior
as it is recorded. For the present study, there was no
logical or theoretical grounds for employing a coding scheme
prior to recording the infant's behavior. As a result, an
"open" recording system was employed.
Learning and Imprinting Studies
In an attempt to differentiate between learning and
imprinting, Klopfer and Hailman (1967) stated that the,
"responses may be learned very rapidly in newborn animals
by the process of imprinting. Imorinting, unlike conditioning
and other common forms of learning, is rapid, long lasting
and apparently unrewarded" (p. 52). Lee Salk (1966) takes
that definition further by stating that the human fetus is
imprinted to the sound of the human heart beat heard in
utero and this has profound influences on our later adult
Smart and Smart in Children (1972) interpreted the
fetal imprinting to the heart beat sound reported by Salk
Because the sound was found soothing to infants
and children, it seems worthwhile to consider
what meaning it may have prenatally. The heart-
beat sound has the criteria of an imprinting
stimulus being intermittent and repetitious and
occurring early in the life of the organism,
before it experiences fear. Later, when the
organism is exposed to fear, it is reassured
by the stimulus to which it has been imprinted.
Thus it seems quite possible that the unborn baby
interacts with his environment in such a way
that the rhythms of sounds and pressures prepare
him for coping with some of the difficulties he
will encounter after he is born. Seeking to
comfort his distress and express their love, his
parents and other people will hold him in their
arms and walk with him or rock and jiggle him.
The resulting sounds and pressures will reassure
him perhaps because he was imprinted prenatally
to such stimuli. (pp. 31, 32)
Grier, Counter and Shearer (1967) did work with pre-
natal auditory imprinting in chickens by exposing the
experimental group to a patterned sound continuously from
the twelfth to the eighteenth day of incubation while the
control group incubated in quiet. The experimental group
creeped toward a stationary sound source as did the con-
trols but the experimental group preferred the sound heard
before they hatched. The experimental group also followed
a moving object longer if it emitted the "imprinted" sound
versus any other sound condition.
Smart and Smart (1972) put forth other implications
of the fetus hearing the mother's heartbeat by reporting
some of the results of Salk's (1962) study where he increased
the rate of the heartbeat to which the infants were exposed.
The newborn infants who had quieted to the sound
of a normal heartbeat showed an immediate
increase in crying and disturbance to these
unusual heartbeats and also to a hissing sound
which the machine accidentally developed. A
frightened or disturbed pregnant mother could
be expected to show variations from the normal
pattern of heartbeat and breathing. Her baby
then might be disturbed by the resulting tactile,
auditory, and kinesthetic stimuli. Either by
themselves, or added to the effects of the
endocrines in the blood, these stimuli could
be significant prenatal influence. (p. 33)
Dr. Akira Sugiyama (personal communication), claimed
that hearing these sounds has a "calming" influence on the
infant and reported that 100 percent of 403 crying newborns
were calmed (stopped crying) with a mean of 41 seconds
after onset of the heartbeat recording. This calming effect
may be due to a classical conditioning process completed
before birth, where the mother's heartbeat served as the
CS and the UCS was the stimuli associated with the womb
environment itself. The fetus may respond by becoming in-
active. This appears feasible since Murooka (1974) re-
ported the neonate was calmed by essentially the same
sound heard in utero, which included the heartbeat.
Contrasting Salk's postulations and Murooka's as yet
unpublished work, Brackbill, Adams, Crowell and Gray (1966)
contended that activity levels in infants decreased under
continuous auditory stimulation which was also reported in
an earlier study by Weiss (1934). Brackbill, et al. (1966)
compared no sound versus any sound including heartbeat
and found that the heartbeat condition did not quiet the
infants more than any other prolonged auditory stimulus.
Brackbill (1970) varied the quality of the acoustic
stimulus presented to infants and measured the level of
arousal by state, respiration, heart rate and motor activity.
The aforementioned qualities included no sound, intermittent
sound, or continuous sound. She reported that level of
arousal was decreased by continuous sound and increased by
intermittent sound when compared to the no sound condition.
Keen, Chase and Graham (1965) reported that a ten
second moderately intense auditory stimulus not only ac-
celerated the neonate's heartbeat, but was a condition to
which the neonate readily habituated. However, a brief
acceleration in heart rate showed no habituation effect in
30 trials over two days. While the ten second condition
hardly qualified as an intermittent auditory stimulus,
when Salk (1966) "played" a heartbeat sound for four con-
tinuous days, it was more effective than the two second
stimulation in accelerating heart rate. The ten second
condition was also habituated to much more readily. The
intensity of sound on infant cardiac and motor behavior
was investigated by Steinschneider, Lipton and Rich-ond
(1966). They presented five second white noise bursts of
55, 70, 85, and 100 db. and a silence condition. All nine
infants responded to a stimulus intensity of 70 db. Other
results reported were that motor responses increased and
their latency decreased with increasing intensity of stimula-
tion. The cardiac measure showed that as the sound intensity
increased, the heartbeat latency decreased and the dura-
tion of the faster heartbeat was longer.
Several researchers in early learning processes have
considered a developing entity that is completely immersed
in amniotic fluid, that is the human fetus. Fetal activity
was probably first reliably recorded by Ray (1932) by using
three tambors taped to the mother's abdomen and masked for
her respiratory responses. A majority of these types of
studies have some purpose other than just recording when
a fetus moves. Sontag and Wallace (1935) reported that a
reliable increase in fetal movement followed a 120 cps.
sound applied to the mother's abdomen. The latency for
this fetal movement was sometimes as long as one minute.
The authors reported that this phenomenon appeared at
the thirty-first week of intra-uterine life and increased
in magnitude as birth approached. Newberry (1941) studied
fetal mcvement by mechanical means and from mothers reports.
Slow sq-irming, quick jerks and a rhythmic series of quick
movements were distinguishable and appeared to follow a
pattern within five lunar months. Newberry (1941) reported
that squirming type motions increased while kicking
decreased rapidly in the last two prenatal months.
Fitzgerald and Kindle (1942) found that if the amniotic
sac was tapped lightly, reflexes of the trunk and appendages
were apparent. They observed this reaction in three of fifteen
fetuses seven to eight weeks of menstrual age when the
sac was intact and still contained fluid but was surgically
removed from the mother. In addition, the fetuses had to
have their circulation unimpaired and as the inevitable
anoxia progressed the reactions rapidly ceased. Similar
studies have been performed by Bridgman and Carmichael
(1935) and Rose (1941) with consistent results for various
menstrual ages of the fetuses' they tested. Probably the
most complete compilation of fetal activity studies and
their implications was published by Hooker (1952). Among
some very physiologically based observations, a chronological
account of the fetuses responding to tactile stimulation
was contained within his text titled, The Prenatal Oricin
of Behavior, (1952).
Other researchers have studies the relationship of
fetal behaviors to a host of other factors. For example,
Harris and Harris (1946) found mot-ers can reliably
report fetal movements and that there was a rapid increase
in strength and number during the period when first
noticed. Interestingly, their data showed the mother's
fatigue did not predict fetal acti-Tty but that the fetus
did tend to be more active at the =lose of the day.
Margoshes and Collins (1965) note the maternal heartbeat
and its function explained dominant= of right-handedness.
They hypothesized that an evolutionary advantage is gained
by right handed mothers who more naturally hold their infant
on the left side to free their dor.-ant right hand. But
78 percent of left handed mothers also hold their babies
on the left side. It is reasoned tat the biological
advantage evolved because mothers vto held neonates on
their left side had calmer babies. They were being held
on the same side as the mother's heart and "the sound of
the mother's heartbeat is nurturanz to the neonate, who
has been prenatally conditioned to her heartbeat" (p. 443).
Richards. Newberry and Fallga=rer (1938) studied
activity of the fetus and the mother's basal metabolic
rate and found very low correlatices. They did report
that, "there is a peak of activity during the eighth and
ninth lunar month, with a dropping; off at the tenth month"
(p. 78). In a study by Richards and Newberry (1938),
they investigated the power of fetal activity being reported
by the mother as a predictor of performance on test items
administered six months postnatally. The items were 46
Gesell Schedule items given at six months plus or minus
four days. After casually observing that their rcst
active subjects as fetuses were also precocious c- certain
motor tasks and that the converse was also fairly constant,
they took their more statistically treatable data and
analyzed it. They reported that the correlation between
fetal activity and Gesell performance, isn't hich. They
stated, "However although they are based on only a few cases,
they are consistently positive, and large enough to be
reliable" (p. 68). Based on a study by the authors and
Ruth Fallgatter (1938) including these results they continue,
". there would seem to be a factor or group of factors
common throughout fetal life, which are still present at
six months of age" (p. 76).
The relationship of fetal movement to a sou- stimulus
outside the mother has been studied by several investigators
and extensions of that relationship have been reported by
Ray (1932) and Spelt (1938), (1948). Ray and Spe-t were
involved with conditioning fetal movements to a ribrotactile
stimulus which served as the CS and a loud sudfan noise
served as the UCS. Ray (1932) applied the CS for five
seconds by tambors taped upside down to the mother's ab-
domen. This was followed by the onset of the UCS delivered
by triggering a modified rat trap in a wooden bcx near the
Although Ray's attempts at conditioning were not very
successful and he used only one subject, he felt strongly
enough about his apparatus and data to conclude that ac-
curate records could be made of fetal movements made in
response to sounds emitted from outside the mother. Spelt
(1938) explored the possibility of classically conditioning
a fetus in utero but his most conclusive work was reported
Spelt (1948) reported in the summary of this study,
"Utilizing a vibrotactile CS and a loud noise as UCS, it
was possible to establish a CR in the human fetus in utero
during the last two months of gestation. Some 15-20 paired
stimulations were required to establish the response to
the point at which one could anticipate three or four suc-
cessive responses to CS alone, but additional practice made
possible as many as eleven successive CR's. Experimental
extinction, spontaneous recovery, and retention of the
response over a three week interval were demonstrated"
(pp. 345, 346).
Spelt (1948) reported that leg activity was much
greater than any other body part of the fetus that was
recorded. Leg movement in infants has been noted as a
response that is easily repeatable and occurs naturally
(Deitz, 1972). In operant conditioning paradigms where the
S is active, responses must meet certain requirements as
Sheppard (1969) noted. He stated, "The leg kick in infants
has a great deal to recommend it. The response is rot
subject to multiple sources of control; infants can perform
it rapidly with little effort, and no apparent fatigue;
the operant level is above zero and the maximum response
rate is quite high . ." (p. 48). Operant conditioning
of leg movements in infants where an auditory reinforcer
was used has been demonstrated by Rovee and Rovee (1969)
using ten week old infants.
For their CS, both Ray (1932) and Spelt (1938, 1948)
employed a vibrotactile stimulus for their studies in
classical conditioning. The use of vibrotactile stimula-
tion as a reinforcer for infants was reported by Schaefer
(1960). A seven month old was observed prolonging contact
with an electric hair cutter by turning his head and
making postural adjustments. The electric hair cutter was
eventually replaced by an electric tooth brush. When ap-
plied to any part of the child's body he would show a
preference for contact with his head. When this subject
was 17 months old, the vibratory stimulus was effective
as a secondary reinforcer in shaping medicine taking be-
havior for a temporary childhood illness. Schaefer in
reporting this phenomenonstated, "It would be interesting
to establish whether the application of vibration is more
effective when applied locally or generally (as when a
vibrator is attached to the crib) and whether its effective-
ness is limited to certain individuals. So far nobody has
pointed out harmful effects of vibration on children" (P. 160).
Vibration of an auditory sort also has been researched
from the standpoint of an imprinted stimulus. The stimulus
prenatally imprinted is the sound of the heartbeat and
according to Salk (1966), it has influenced our art, music,
dance, handedness and several other factors. He has ap-
proached the question experimentally, anthropologically,
and comparatively. He was not concerned with the associate
type learning in either fetuses or infants and he clearly
defined his use of the term.
Imprinting is distinguished from associative
learning in that the conditions under which
imprinting occurs do not provide any reward
in association with their stimulus. The ef-
fects of associative learning can be altered
but in the case of imprinting, the effects
are irreversible. Furthermore in association
learning the spacing of practice trials rather
than massing is most effective, in imprinting
the converse is true. (p. 295)
Salk (1966) further makes his point on human imprinting
and its manifestations in our music, dance and various
rhythms associated with the heartbeat as a fetus.
A question still remains in connection with
imprinting in the human. It was said earlier
that aspects of adult behaviour are influenced
by an imprinted stimulus even though the
pattern of responding to that stimulus may not
have been developed at the time of imprinting.
We must ask ourselves, "What elements are there
in adult behavior that might have been in-
fluenced by imprinting?" To answer this question
we merely have to look at some very obvious
facts. In most, if not all, cultures throughout
the world man engages in dance and musical ex-
pression. The universality of this behavior,
in spite of cultural differences, provides
fairly clear evidence that it is related to man's
biological nature. The fact that the rhythm of
the heartbeat, the rhythm of breathing and
the rhythm of walking, cannot be overlooked.
The sensory experiences provided by music
and dance are indeed similar to those sensory
experiences received during prenatal life.
Music and dance are the result of i-printing
and are created and experienced by ran in his
attempt to remain in proximity with imprinted
stimuli. (p. 302)
Salk has been cited earlier in studies involving handed-
ness (Margoshes& Collins, 1965) but Salk's 1966 data involved
some research into art and biological tendencies for survival.
In examining paintings and pieces of sculpture
created during the past few hundred years which
involve a child being held by an adult, it was
observed that among 466 such works of art, 373
or 80 percent were constructed so that the child
was on the side closest to the adult's heart.
The automatic tendency of the huLan mother
to hold her child close to the heart requires
additional explanation. It is not the character
of nature to provide living organisms with
biological tendencies unless such tendencies
have survival value. (p. 302)
Salk (1966) cited several other experimental type
studies lending support to his hypothesis and he concluded
his argument concerning imprinting with a simple explanation
of how it operates and what value it has for other types
This discussion of experimental evidence together
with evidence based on naturalistic observations
concerning the problem of imprinting, will be
incomplete unless we deal with it in its most
basic sense. By this is meant, the role imprinting
plays in the development of living organisms.
To begin with, under natural conditions the im-
printed stimuli serve to bring the developing
organism into proximity with conditions that
enhance its survival. There is not only direct
protective value resulting from this at a time
when the developing organism is helpless, but it
serves as the base for all later learning.
Imprinting compels the organism to seek con-
tinued sensory stimulation by coming into
contact with its environment and by so doing
enhances the development of behavior patterns
that have adaptive value, through associative
learning. (p. 303)
The work of Salk (1962) demonstrated the relationship
between heartbeat and infant behaviors. while e there remains
some questions concerning the effect of heartbeat on
specific behaviors, the evidence from the studies just re-
viewed support the position that heartbeat serves as a
powerful stimulus for conditioning in the fetus.
The choice of leg kick as a response to be conditioned
is supported by the work of both Sheppard (1969) and Deitz
(1972). The leg kick is easily conditioned and is easy to
measure. The leg kick will be the dependent variable for
the present study.
Given what has been reviewed in the relevant literature
concerning development aspects, differences in methodology
and the confusion in the learning research, the author of
the present study saw a need to investigate the differences
in leg kicking activity of infants as they were exposed to
a heartbeat sound presented in air compared to hearing the
same sound underwater. Additionally, the differences between
the leg kicking activity for the underwater heartbeat condi-
tion and four other underwater control sound conditions were
investigated. The procedures for collecting the data and
treating it are described in the next chapter.
METHOD ;-DO PROCEDURE
In order to investigate the differences in infant leg
kicking, infants were expos_- to a heartbeat sound in air
and the same sound underwatr-. Also, four other underwater
sound conditions were used is controls. The procedures
to achieve this are present in this chapter.
The procedures used it the present study were designed
to yield data to insure appropriate testing of the following
two hypotheses. Stated in :zll form, the two hypotheses
for the present study were:
There will be no difference in mean kicking
rates for an underwater audio taped human
heartbeat when compare: to the same tape
recording transmitted t-ough air.
There will be no difference in mean kicking
rates for an underwater audio taped human
heartbeat when compare- to underwater presenta-
tions of rhythmic musir, white noise and human
With those hypotheses in mind, the following procedures
were followed. Each infanr -as brought to the swimming pool
at the same time each aftercon in a bathing suit used
exclusively for this experiment. The infants were not sup-
posed to eat anything for r-c hours before they were brought
to the pool.
The assistant loaded all the audio c=a=ette tapes in
the proper sequence, tested all remaining apparatus and
prepared to begin the sequence of sound cr:ditions for
that day. The E was blind to when a souni condition was
going to be presented except for the surface heartbeat
condition. On the five days when the surface heartbeat
condition was first in the sequence, the =ild was placed
on its back on a towel between the two surface speakers.
Following that condition, they were brouch: into the water
and adapted to the water by the E before sie underwater
Prior to the first underwater trial =- the first day of
the experiment each infant was operantly nditioned to
close the mouth while underwater. This was accomplished
by employing air as a primary reinforcer ;-d a squeeze to
the child's chest by the E as a secondary reinforcer. All
six subjects came under the stimulus contr-l of being
underwater within four shaping trials. rD-ing the five
days when the underwater trials were firs,. the infant
was adapted to the water before the trials began. For the
underwater conditions, the infant was released by the E
between the two underwater speaker units. The assistant
would choose if this event was to be a trial. If it was a
trial, the movie camera was operated which also activated
the tape player to present the sound condition. In this
manner, the E was blinded to the conditic: presentation.
When the trial was over the assistant wou:I signal the E
and the child would be picked up.
During the five days when the surface heartbeat
condition was last, the infant was placed between the
surface speakers, the event was filmed, and then the
infant was taken home. The mothers were instructed not
to put drops in the ears after the lesson ncr was any ot-er
special treatment to be given to the infants.
Conditions were as follows: (1) Silence; (2) Mother's
Voice; (3) "Sleeping Beauty;" (4) White Noise; (5) Surface
Heartbeat; and (6) Underwater Heartbeat.
Each condition had a duration of five seconds althc-zt
the tapes were made for seven seconds duration. The silence
condition was employed as a pure control while the white
noise condition served as a control for the presentation =f
all frequencies between 20 and 20,000 cps. The mother's
voice condition was recorded prior to the trials for day
one. Each mother read into a tape player after a sound
level had been taken for her voice level. They each
recorded the following passage:
Alice was beginning to get very tired of sitting
by her sister on the bank and of having nothing
to do; once or twice she had peeped ints the book
her sister was reading, but it had not pictures
or conversations in it. "And what is the use of a
book;" thought Alice, "without pictures or
This condition was used as a control for familiar sound
that is their mother's own voice.
Tchaikovsky's "Sleeping Beauty" (Panora-a) was used as
a control for a rhythmic, soothing presentation. The
heartbeat condition was recorded once and the same tape
was used for both the surface and underwater sound condi-
tions. This condition comprised the experimental condition
for the hypotheses being tested in this study.
Schedule of Presentations of Sound Conditions
1 2 3 4 5 6 7 8 9 10
C 5 1 5 3 5 4 5 6 5 6
0 1 6 3 1 4 6 2 3 4 1
D 2 3 2 4 3 2 1 4 3 3
S3 4 1 6 2 1 6 1 2 4
I 4 2 6 2 1 3 4 2 6 2
S6 5 4 5 6 5 3 5 1 5
1) Silence 4) White Noise
2) Mother's Voice 5) Heartbeat,Surface
3) "Sleeping Beauty" 6) Heartbeat,Water
The order of presentation of the six conditions to each
subject over a ten day period is summarized in Table 1. The
six conditions were assigned at random except for condition
number five. Since the infant was out of the water for
this presentation, it was decided to give the condition
either first or last. Giving this condition first everyday
when the infant was "fresh" might confound the results. The
same would be true for giving this condition last everyday.
It was decided that on odd nu-bered days the heartbeat heard
at the surface would be first and on even days it would be
presented last. An explanation of how these conditions were
recorded is found in the apparatus section of this paper.
The experimental space for this study was the shallow
end of a rectangular swimming pool. The hardware included
a cassette tape player, a stereo amplifier, two surface
speakers, two underwater speaker units and a movie camera
in an underwater housing. A more complete description of
the apparatus including technical characteristics is in-
cluded in the Appendix.
The underwater speakers were two five inch circular
speakers rated at 70-15,000 hz. at ten watts. The speakers
were mounted in a fiberglass shell that was waterproofed
with a plastic film and plastic tape covering. The film
and tape combination allowed the face of the shell to act
as a diaphragm to transduce the acoustic vibration from
air to water. Three such speakers were constructed although
only two were used. All three speakers were balanced and
calibrated with the two most similar speakers being used.
Balancing and calibration was achieved by submerging a
dynamic microphone two feet away from the speaker unit as
it emitted white noise at a present decibel level. The shells
were bolted to four foot boards that had 20 pound weights
which served to anchor the speaker units in position. Each
unit was positioned four feet apart and facing each other
with the speaker four inches underwater (See Figures 1 and
The surface speakers were calibrated and balanced in
air by the same procedure as the underwater speakers. The
left channel speaker appeared somewhat stronger so ap-
propriate compensations were made with the amplifier to
equalize the surface speakers. In their dark oak cabinets,
these units physically resembled the underwater units.
Both the surface and underwater speaker units were connected
to the amplifier by braided audio cable.
The cassette tape player was a battery operated Sony
Cassette Corder connected to the amplifier with a standard
audio jack. The magnetic recording tape used to record
the sound conditions was "Memorex" TM 60.
The mother's voice condition was recorded before the
first session and retaped so the decibel level in air was
115 db for each mother's presentation. Three of the other
sound conditions were recorded directly from a stereo record
via recording jacks. Those three sound conditions were
Sleeping Beauty"and the two heartbeat conditions. The
silence condition was presented with all the audio equipment
turned off. The white noise condition was recorded from a
white noise generator. All underwater presentations were
recorded so that reasonable fidelity could be achieved when
these sound conditions were boosted 50 db to 115 db as
calibrated in air. This increase was necessary to make the
db level equal for air (65) and underwater conditions (115).
A single eight movie camera in an underwater housing
was used to film the underwater and surface trials. The
camera was operated by a direct pressure trigger very
similar to a mechanical cable release used in various photo-
graphic situations. Linked to the trigger was a toggle
switch leading to the cassette player. When the camera
was operated, sound conditions were automatically presented
except in the silence condition. The film used was ASA 25
and ASA 200 with appropriate filters and lenses used for
appropriate film documentation. The exposed film was sent
to a professional processing laboratory and returned in 50
foot reels. Immediate inspection revealed all the film
was properly focused and exposed suitably for coding.
All batteries used to power the camera and cassette
tape player were removed and recharged at the end of each
The water temperature in the pool was constantly
monitored with a photographic thermometer and recorded for
each child as they entered the water. The water chemistry
was checked each day twenty minutes before the first subject
arrived. Only pH and chlorine levels were tested so ap-
propriate levels could be maintained.
Two female and four male infants aged ten and one half
months plus or minus twelve days served as subjects for
this study. Each subject received all six conditions of
the experiment each day. Three of the subjects were obtained
from lists maintained by a local birth study group and the
other three were referred to the E by one of the original
three mothers. The parentss were informed about all facets
of the experiment an: told that, after the experiment was
over, the E would teitc all of the infants involved to
swim. All parents consented to have their infant participate
in the study and sir-d the standard consent form required
of researchers using human subjects at the University of
Each infant's ha-ring was normal as tested by a
pediatrician as part of a regular checkup, and reported to
the E by the mother :f each child. They were not supposed
to be fed during the rto hours prior to the sessions each
day to prevent vomit ng of water they may have ingested.
This would unduly fatgue them and could have possibly in-
fluenced the data. 5cne of the infants vomited water or
food during the exper-aental trials. Each infant came in
a consistent bathing suit but one of the female subjects
started the experiment in a suit which was too small in the
arms and had to switr to a larger sized suit. It should
be pointed out that t-e suit did not constrict the move-
ments of her legs or lower body in any way.
Each mother reported that her infant tended to sleep
more in an afternoon r;ap and that the infant's appetites
seemed to improve. Five of the six mothers continued after
the experiment to bring their infants for swimming lessons.
At fourteen months of age they are finally proficient
swimmers. They can tick and stroke with their arms to
propel themselves ani roll over onto their backs for air
when needed. One of the female s'-ects swam fifty meters
unassisted at fourteen months of a:e. During that swim,
she rolled over on her back to brezane twenty-one times and
flipped over on her sto-ach to conte.ue swimming.
Initially the films were played on a regular single
eight movie projector to assess thn:r quality which was
deemed by the E to be acceptable fir coding. Coding was
accomplished independently by the C and an assistant familiar
with counting infant aquatic behav:=rs. The subjects on
each roll of film were identified Ezd the sound condition
being filmed was also identified. "his listing of subjects
and conditions became the raw data sheet.
Following this inspection and identification the film
was cut into 90-frame segments. IZzety frames on single
eight film is equal to five seconds of real time. Each
ninety frame segment was projected and each time the subject
returned the foot that was initiall- closer to the surface
at the beginning of the film trial rt was counted as one
kick underwater. The quality of d;th of both feet at the
beginning of the trial was assessed by use of a projection
screen that had one inch grids draz on it.
The surface condition footage v-s coded by simply
counting any right leg extension t-e movement as a kick.
Since the segments were cut into urLts known to be five
seconds long, it was possible for t-z coders to use the stop
action and slow motion projectors to code each sez-ent
more carefully. The E had eleven years experience with
counting infant aquatic behaviors and the assistant had
five years experience in the same area. After each had
coded independently it was found that perfect agreement
existed among their records of the segments they coded.
Rates of kicks per second were cc-pleted by dividing
the number of kicks by five. These rates became t-e
dependent variable in a two way analysis of variance with
the ten days and six conditions serving as the independent
The design of the experiment involved each cf the six
subjects receiving all six sound conditions each day for
ten days. The subject's were not matched and the order of
administration of the sound conditions was not ra-don each
day for each subject. Each subject received the sound
conditions in the order presented in Table 1. The design
best suited to interpret the data as collected by these
procedures was a randomized block factorial desic= (Kirk,
1968, p. 237).
In addition to those two major hypotheses stated
earlier, three other comparisons of reans were performed.
The coefficients for all five planned nonorthogoral compari-
sons are shown in Table 2. These coefficients are arranged
for each comparison so that when each is multiplied by its
respective mean, the sum of the rows must equal zero if
the means are equal.
Condition Mean Coefficients for Five
Planned Non-Orthogonal Comparisons
Mother's Sleeping White beat Heartbeat
Silence Voice Beauty Noise in Air Underwater
O 1 3 -1 -1 -1 0 0
P 2 0 1 0 0 0 -1
R 3 0 0 1 0 0 -1
S 4 0 0 0 0 1 -1
N 5 0 1 1 1 0 -3
For example in comparison number 1, the silence condi-
tion is being compared to three other conditions. To equate
the comparison to zero, the mean for the silence condition
is multiplied by three and the mean for each of the other
three conditions is multiplied by minus one.
Comparison number one indicated the differences between
the means for the silence condition underwater and the
three other underwater control conditions: mother's voice,
"SleepingBeauty"and white noise. Comparison number two
examined differences between the mother's voice condition
and the underwater heartbeat presentation. Comparison
number three showed if any differences exist between the
underwater control condition for rhythmical properties
represented by the Sleeping Beauty"presentation and the
underwater heartbeat condition. Comparison number four
was designed to show any differences in the underwater
heartbeat condition and the heartbeat transmitted in air
condition. That comparison was the test for hypothesis
number one. Comparison number five tested the differences
between the means for the mother's voice, 'sleeping Beauty,"
and white noise which were all presented underwater, and
the heartbeat underwater condition. That comparison was
the test for hypothesis number two. Complete results for
all of the above comparisons are reported in Chapter IV.
This study examined the differences between the means
for infant leg kicking behavior for infants exposed to an
underwater heartbeat sound when compared to a heartbeat
sound presented in air. This study also examined the dif-
ferences between the means for the underwater sound condi-
tions that served as controls and the underwater heartbeat
There were three other comparisons made as described
in Chapter III, but the two hypotheses that were tested
formed the research problem stated above. In order to
analyze the data collected using the procedures described
earlier in Chapter III, the data were submitted on cards
to a computer at the North East Regional Data Center in
Gainesville, Florida. The program used was under release
76.4 of the Statistical Analysis System to the University
of Florida. An analysis of variance was performed by the
computer. The program was adapted to relate any interaction
between days and conditions. In addition, the means for
subjects, days, conditions and conditions by days were
demanded of the program.
Prior to proceeding with the planned comparisons,
a test of the day by condition interaction was conducted
and the results indicated no interaction (F (45,295) = 1.01,
p < .01).
Since the condition effect was found to be consistent
across days, the condition data were pooled over days and
tests of the main effect were conducted as planned. The
results of these tests are summarized in Table 3. To
demonstrate whether a significant difference existed between
the means in a planned non-orthogonal comparison,
Dunn's procedure was used. Five planned comparisons were
made among the six conditions.
For comparisons number 2, 3, and 4 in Table 3 where the
alpha was set at .01, the critical difference was calculated
to be .1365. For comparisons number 1 and 5 in Table 3,
where the alpha was set at .01, the critical difference was
calculated to be .3345.
By comparing the observed differences in Table 3 with
the appropriate critical difference as reported above,
comparison number 2, 3, 4 and 5 in Table 3 were significant
at the p< .01 level.
The hypothesis stating that there would not be a
significant difference in mean kicking rates for an infant
underwater when presented with an audio taped human heart-
beat when compared to the same tape recording transmitted
through air was rejected. That comparison is related as
number 4 in Table 3. The hypothesis stating that there would
not be a significant difference in mean kicking rates for
Mean Kick Rate per Second for Six Conditions and
Calculated Observed Differences
1 2 3 4 5 6
M11 11tel ( 01-I.II.q Uill 1I IBasi 1IL)- f Ca 11. I ia-
Bilenoe Voioe Beauty Noise Air Water
.74 .68 .80 .75 .10 .34
Observed Differences (Dunn)
(1) 1 vs 2,3,4
(2) 2 vs 6
(3) 3 vs 6
(4) 5 vs 6
(5) 2,3,4 vs 6
#Example: (.74 x 3) + (.68 x -1) + (.80 x -1) + (.75 x -1) = -.01
*Sig. at .01 level.
an infant underwater when presented with an audio taped
human heartbeat when compared to underwater transmission
conditions of rhythmic music, white noise and human speech
was rejected. That ccmparis:n is related as number 5 in
The implications for t-ese results as well as the
relationship of the other comparisons performed on the mean
kicking rates under various sound conditions will be dis-
cussed in the next chapter.
DISCUSSION AND CONCLUS:':
Some of the results of previous research cited in
Chapter II are related to the results of this study as
reported in Chapter IV. For example, the research into
hearing in the neonates and infants showed that they require
an elevated sound intensity threshold. Generally, it must
be sixty-five decibels or higher in order t= be effective.
The effectiveness of sound stimulation as ia independent
variable with regard to a number of dependent measures such
as bodily activity, cardiac and respiratory activity, and
the effects of habituation are well represented in the
literature. Studies where the intensity le-.els were sixty-
five decibels or higher yielded some significant results or
at least did not incur criticism from later researchers con-
cerning the sound intensity parameter. In criticizing
Tulloch et al. (1964), Brackbill et al. (1i6) stated that
their failure to replicate Salk's (1962) findings was due
to the lower level of the sound stimulus (43 db.). This is
roughly 20 db. below the effective threshold for neonates
and infants. The decibel level for the transmission of
sound through air in the present study was 65 db. The
change in amplitude of the sounds transmitted through
water was available from Hamilton (1957) where the intensity
had to be raised 45 db. to 60 db. above the threshold
established in air for equal underwater perception. The
decibel level as calibrated in air for the underwater
presentations for the present study was 115.
The duration of the sound presentation for each trial
followed the parameters researched by Stubbs (1934). hat
research showed that bodily reactions were effected better
by durations of five and fifteen seconds for infants hearing
pounds at 70 and 95 db. In the present study, even a ten
second trial underwater would have been too long. Since
Stubbs found the five second duration acceptable and the
concern of this study's author for infant breath holding
capacity underwater, the five second duration was used.
It was also felt that a sound duration of five seconds
had less chance of being habituated to although research
by Moreau et al. (1970) may disagree. It should be noted
that previous research into the influences of habituatizc
on sound parameters with infants has involved acoustic
and not underwater hearing.
One of the underwater sound conditions involved hearing
the mother's voice. Research by Hutt et al. (1968) shci-d
neonates are behaviorally most responsive to fundamental
frequencies found in the range of the human voice. Again, Hutt
et al. conducted the research in the air while the present
study had both an air and water condition. This "familiar"
sound" was not as effective in calming the bodily activity
as recorded by leg kicking of the infant, as the heart-
beat sound heard underwater.
Other properties of the physical environment have
been studied concerning this calming effect on infants.
Weiss (1934) showed that light alone had a calming effect
and sound alone in darkness also had a calming effect.
This present study was conducted at the same time each
day in an outdoor pool. The light intensity of the sun was
consistent as there were very few clouds in the sky during
the ten day period of the data collection. If light
intensity was a factor in calming the infants, it was a
consistent factor for all ten days and all six subjects in
the study. The number of days of observation for this study
agree with the median of observations made by McGraw (1939a).
She observed forty-two different infants with a median of
ten observations per infant. A lack of control over the
infants and their parents seems to be a cause for a rather
large range but rather small total number of observations
cited in the research.
Research by Grimwade, Walker and wood (1970) examined
the sound properties of the womb and also reported how
sound stimulation of the fetus influenced activity. The
present study with the underwater sound conditions and
particularly the underwater heart beat condition to some
degree simulated a womb-type environment. If fetal condi-
tioning or at least fetal imprinting to sounds such as a
heartbeat heard in utero is possible, then at least the
hearing capacities of the fetus must be delineated. Kellogg
(1941) and Forbes and Forbes, (1927) showed that the fetus
can hear fairly well from about the seventh month of
gestation on. Hearing sounds as a fetus may be one thing
but conditioning a fetus to react to a sound produced out-
side the mother's abdomen was the work of Ray (1932),
Sontag and Wallace (1935) and Spelt (1938, 1948). The
present study did not investigate any relationship of fetal
movement or conditioning per se but rather, some of the
sounds that the infant could have heard as a fetus and how
those sounds differentially influenced kicking behavior in
infants. The present study showed there was a significant
difference between the simulated womb environment heart-
beat condition and any other sound heard while in this
environment. That finding agrees with Salk (1966) where
he stated that the heartbeat sound heard by the fetus was
imprinted and associated with the womb environment. Murooka
(1974) said this heartbeat sound had a calming influence in
newborns and that too agrees with the findings of this
Since the heartbeat has been reported to be imprinted
on the fetus (Salk, 1966), any comparison involving that
sound would require appropriate control conditions. The
heartbeat as a sound condition would have to have controls
for its rhythmical sound properties, controls for its
frequencies and, because it is heard by the fetus in the
womb and the infant as it is fed, it must also have a
control that is also a familiar sound. Those controls
were represented in this study by Tchaikcvsky's"Sleerinf
Beauty," white noise and the motherly voice respectively.
Salk (1966) however, goes further than anyone else by
stating that this early imprinted heart sound serves as
the base for all later learning.
If the environment where the heartbeat is heard in
utero is simulated, that condition should have greater in-
fluence because of the site of the imprinting than if the
sound is heard in a dissimilar environment. As reported in
the present study, there was a significant difference in
the means of the heartbeat underwater condition and the con-
dition where the heartbeat was heard in air.
While it is true that sound has different characteristics
in water than in air, it may be that the way sound is heard
by a fetus in the womb and an infant underwater are
drastically different. The infant may be adapting air
hearing techniques rather than reverting back to fetal
hearing techniques. It is possible that those transforma-
tions rather than the sounds themselves may account for
some of the differences in the air versus underwater heart-
beat comparisons. Where the means for some sound conditions
presented underwater were compared with the underwater heart-
beat condition and a significant difference was found, it
could be the nature of any pulsing sound would have produced
these results. In that case, the imprinted stimulus Salk
(1966) cites as having greater calming value would break
TWe author of the present study feels that the heart-
beat s:und may be imprinted but researchers must simulate
the e--ironment in which the imprinting took place if
their research is to properly show any relationship con-
cernin- that sound and any later behavior.
"Te scope of the present study was deliberately limited
and designed to test the two hypotheses. Due to the small
N in ~Be present study and the uniqueness of the procedures,
care should be taken in generalizing these results to
other areass of research. However, this study does generate
some interesting areas for further investigation. For
instance, if a fetuswereexposed to a 1,000 cps. tone
for a period of time each day after the sixth month of
gestation, what would the influence of that sound condition
have = that individual as an infant if it were one of
the uierwater sound conditions as described in the present
study? It would also be interesting to have several sub-
jects participate in this type study and to test them at
birth and at three ronth intervals until they are two-years-
old. Would the responses, if they were different initially,
dissipate over time? Would the differential responding to
the heartbeat as reported in this study change over time?
Good research doesn't prove, it probes relationships. For
the implications bolily put forty by Salk (1966) and
others this area of investigation needs more probing of
relationships and less proving of implications.
The present stur- examined the differences in leg
kicking rates for six infants ages ten and one half months
old as they heard fire seconds of a heartbeat sound while
underwater as compared to hearing the heartbeat sound in
air for five seconds.
A comparison was also made between the heartbeat sound
presented underwater and three other sounds presented
underwater. Those three other underwater sound presenta-
tions were rhythmic rusic, white noise and human speech
all being presented fzr five seconds.
Significantly less kicking was observed for the
heartbeat heard in air condition than when the heartbeat
sound was presented a the infant underwater.
When the underwa:er heartbeat sound condition, which
was a tape made from a human female heartbeat recorded in
utero, was presented to the infant subjects, they kicked
significantly less t-2n when they heard the other under-
water sound conditions. The first underwater sound
condition, the heart-eat condition, was compared to a
control condition for any rhythmical sound as represented
for this study by TcZhikovsky's "Sleeping Beauty."
Another underwater sound was.!esigned to serve as
a control for any sound being he-. underwater. This
condition was presented by playing "white noise" or fre-
quencies between 20 and 20,000 cps. simultaneously.
The final underwater conditi=: was designed to
control for a familiar sound that -te subjects heard
previously. That condition was created by having each
subject's mother record herself rieaing the introduction
to Alice in Wonderland.
A comparison of those three =derwater sound conditions
mentioned above to the underwater heartbeat showed the
infants kicked significantly less :o the heartbeat sound
than to the other underwater souni-.
The present study has demonstrated the crucial importance
of replicating the environmental meditions when comparing
early and later learning in the fe-is and neonate. The
results of the present study appear to be supportive of an
imprinting explanation of early learning. Present theories
of learning in children and adults do not emphasize the role
of imprinting. The development of educational models of
how learning occurs should examine the role imprinting plays
in early learning. A clearer explimation of the role of
imprinting in early learning may contribute to our better
understanding of learning in and Dz of the classroom through
associative learning procedures.
Technical information concerning the apparatus
described in Chapter III is presented here. The construc-
tion of the underwater speaker units is descriLed followed
by relevant information concerning the surface s-eakers,
the tape player and tapes. The video components used in
the experiments are described and include the mo-ie camera,
the underwater case for it and the film that was used.
Finally, the technical information concerning the electrical
systems employed in the data collection are described.
The underwater speakers were two circular five inch
air suspension speakers (40-1909) manufactured hy Tandy
Corporation. They were rated at 70-15,000 Hertz with an
impedance of 8 ohms. Each of these speakers were suspended
via monofilament line inside a fiberglass shell. The face
of each shell was covered with polyurethane filn produced
by Dow Chemical Company. Further water proofing was ac-
complished by water-proof plastic tape produced ty the 3M
Company. The film and tape combination allowed the face
of the shell to act as a diaphragm to transduce the acoustic
vibration from air to water. Although only two speakers
were used underwater, three were constructed with the third
being a back-up unit. All three underwater specers were
balanced and calibrated by means of a VU meter. A dynamic
microphone (Realistic 33-1034) with sensitivity in the
150-10,000 Hertz range and rated at 200 ohms impedance.
This microphone was submerged two feet away from the
speaker as the speaker emitted white noise at a preset decibel
level. The microphone was plugged into a Sony Stereocorder
"Reel to Reel" and the sensitive VU meter in this unit
served as the calibration instrument. It was found that
two of the speakers were virtually identical with the third
requiring only slight modification to become equivalent
to the other two underwater speakers. Two bolts were im-
bedded in the back of the fiberglass shells so the units
could be bolted to the support frames. The frames were
made out of 1' x 12' cypress shelving boards with a twenty
pound weight attached to the bottom. The weights served
to securely anchor the speaker units in position. Both
units were positioned facing each other four feet apart
with the center of each speaker six inches underwater.
The two surface speakers were eight ohm, ten watt units
manufactured by Mitsubishi Electrical Corporation. They
were calibrated and balanced in air by the same procedures
as the underwater speakers. The left channel speaker
measured stronger, so appropriate compensations were made
to the speaker and the amplifier to equalize the two sur-
face units. In their dark oak cabinets, these units
physically resembled the underwater units. Both the sur-
face and the underwater speakers were connected to the MGA
model SA-10 twenty-two watt amplifier by braided audio cable.
The cassette tape player was a Sony Cassette-Corder
manufactured by Sony Inc. It operated on four GE 1.25
volt Rechargeable Nickel-Cadmium Batteries that were
recharged daily. The tape player was connected to the
amplifier by means of a standard audio jack linking the
female monitor plug on the cassette player and the "play"
receptor of the amplifier unit.
The magnetic recording tape used to record the sound
conditions was "Memorex" TM 60 audio cassette tape manu-
factured by Audio Magnetic Corporation. The mother's voice
condition was recorded before the first session and they
were retaped so the decibel level in air was 115 db. for
each mother's presentation. Three of the other sound condi-
tions were recorded directly from the Capitol St-11421 stereo
album Lullaby from the Womb via recording jacks. The white
noise condition was recorded from a white noise generator.
All underwater presentations were recorded so that reason-
able fidelity could be achieved when these sound conditions
were boosted 50 db. to 115 db. as calibrated in air. This
increase was necessary to make the decibel levels equal
for air and underwater conditions. The silence condition
was presented with all the audio equipment turned off.
The water temperature was constantly monitored by a
Kodak color temperature thermometer and recorded each day
just before the first subject arrived and immediately after
the final subject left the experimental space.
The water chemistry was checked each day about 20
minutes before the sessions began. Only pH and chlorine
levels were tested for with an HTHI standard chemistry kit
available at most pool supply stores.
A Fujica AX100 single eight movie camera in a Marine-8
AX 100 underwater housing was used to film the underwater
and surface trials. The camera was operated by a direct
pressure trigger very similar to a mechanical cable release
used in various photographic situations. The film used
was Fujichrome R25 and due to local unavailability of that
film during the experiment, Fujichrome R200 was used with
a neutral density 4S filter counted in front of the Fujinon
1.1 lens used with the camera referred to above.
The exposed film was sent to a professional processing
lab and returned by mail six days later. Immediate
inspection revealed all the film was properly focused and
exposed suitably for coding.
Since electrical equipment was employed in the
apparatus for this study appropriate safety precautions
were taken. All fuses were checked where 110 A.C. was
used but only battery powered D.C. equipment using low
voltages and amperes was used near and in the water. The
power sources for movie-camera and cassette tape player
were six 1.25 volt batteries.
The AA batteries used to power the camera were G.E.
Rechargeable Nickel-Cadmium Batteries rated at 1.25 volts.
They were removed each day and charged for twelve hours
in a General Electric battery recharger model BC-3 along
with the four batreries from the audio cassette tape
A plastic tape
B plastic film
D securing lines
E fiberglass shell
F audio cable
G securing bolts
H wing nuts
I frame board
J brace screws
K anchor weight
M frame board
Underwater Speaker Unit
A tape player
C audio cable
D audio cable
E water level
F underwater speaker un:i
G underwater speaker un::
H anchor weight
Surface Speaker Arrangement
Rates of Kicking per Second for Each Subject
by Condition and Days
Subject # 1 Jason
Condition 1 1 Silence
2 Mother's voice
3 Sleeping Beauty
4 White Noise
5 Surface Heartbeat
6 Underwater Heartbeat
1 2 3 4 5 6
.4 .2 .2 .2 .4 .2
1.0 .8 .6 .2 1.2 .2
1.0 1.2 .8 .2 1.2 0
1.0 1.2 .8 .4 1.2 .6
.8 .8 .8 .2 1.4 .6
.6 .4 .4 .2 .6 0
1.2 .8 1.0 .2 1.2 .2
.4 .4 .4 0 .4 .2
1.4 1.0 1.0 .4 .8 .6
.8 1.0 .6 .4 1.0 .8
1.2 1.2 .8 .2 1.0 .4
.2 0 0 .4 0 0
.2 .2 0 .2 0 0
1.0 1.0 1.0 .2 .8 .8
.8 .6 .6 0 .8 .4
1.0 1.0 1.2 .2 1.0 .4
.6 .4 .4 0 .8 .2
1.0 1.0 .8 .4 .8 .8
1.2 1.0 1.2 .2 .8 1.0
1.0 1.0 .8 0 1.0 .6
1.0 .8 .8 .2 1.0 .6
.4 .2 .2 0 .6 0
.6 .6 .8 0 .6 .2
0 0 0 0 0 0
0 0 0 .2 0 .2
.8 .6 .6 .2 1.0 .8
.8 .8 .6 .8 .8 .6
.6 .8 .6 .8 .6 .4
.8 .6 .8 .8 .8 .8
.4 .4 .4 .4 .2 .4
Rates of Kicking per Second for Each Subject
by Condition and Days (cont.)
Day Conditions 1 2 3 4 5 6
6 4 .4 .6 .8 .8 .6 .6
6 .2 0 .2 .4 .6 0
2 .4 .4 .6 .6 .8 .2
1 .6 .6 .6 .7 .6 .2
3 .6 .6 .8 .6 .8 .2
5 0 0 0 .2 .2 0
7 5 0 .4 .2 0 .2 0
2 .8 .6 .6 1.0 .4 .2
1 1.0 .8 .8 .8 .6 .2
6 .4 .2 .6 .4 0 .6
4 .6 .6 .8 1.0 .4 .2
3 1.0 1.0 .8 .8 .6 .2
8 6 .4 0 .4 .4 .4 .2
3 .8 .4 .8 .8 .6 .4
4 .8 .8 .6 .6 .6 .4
1 .8 .6 .6 .6 .8 .6
2 .6 .4 .6 .8 .6 .6
5 0 .2 0 .2 .2 0
9 5 .2 .2 0 0 0 0
4 .8 1.0 1.2 1.2 .8 .4
3 1.0 1.2 1.2 1.2 1.2 .4
2 .6 1.2 1.4 1.4 1.2 1.2
6 .6 .6 .8 .6 .8 .6
1 .6 1.0 1.4 1.2 1.0 .8
10 6 .4 0 .2 0 .4 0
1 1.0 .6 .8 1.0 .6 .4
3 1.0 .6 1.2 1.2 .6 .4
4 1.4 .6 1.4 1.4 .8 .2
2 1.0 .6 1.0 1.4 .8 0
5 .2 0 0 0 .2 .2
Abrahamson, D., Brackbill, Y.E., Fitzgerald, H.E.
Interaction of stimulus and response in infant
conditioning. Psychosomatic Medicine, 1970, 32 (3),
Aldrich, C.A. A new test for hearing in the newborn: The
conditioned reflex. American Journal of Diseases of
Children, 1928, 35, 36-37.
Ames, L.B. Some relationships between stair climbing and
prone progression. Journal of Genetic Psychology,
1939, 54, 313-325.
Ames, L.B. Motor correlates of infant crying. Journal
of Genetic Psychology, 1941, 59, 239-247.
Ames, L.B. Supine foot and leg postures in the human
infant in the first year of life. Journal of Genetic
Psychology, 1942, 61, 87-107.
Ames, L.B. Early individual differences in visual and motor
behavior patterns. A comparative study of two normal
infants by the method of cinemanalysis. Journal of
Genetic Psychology, 1944, 65, 219-226.
Ashton, R. State and the auditory reactivity of the human
neonate. Journal of Experimental Child Psychology,
1971, 12 (3), 339-346.
Ashton, R. The influence of state and prandial condition
upon the reactivity of the newborn to auditory
stimulation. Journal of Experimental Child Psychology,
1973, 15 (2), 315-337.
Berg, K. Habituation and dishabituation of cardiac
responses in 4-month-old, alert infants. Journal of
Experimental Child Psychology, 1972, 14 (1), 92-107.
Bernard, J., & Sontag, L.W. Fetal reactivity to tonal
stimulation: a preliminary report. Journal of
Genetic Psychology, 1947, 70, 205-210.
Biehler, R.F. Child development: An introduction,
Boston: Houghton Mifflin Company, 1976.
Birkmayer, W., & Goll, H. (The development of walking.
First communication.) Dtsch. Z. Nervenhei:k., 1940,
151, 237-253. (Psychological Abstracts, 1941, 15,
Birns, B., Blank, M., Bridger, W. H. & Escalora, S.K.
Behavioral inhibition in neonates produced by
auditory stimuli. Child Development, 1965, 36
Brackbill, Y. Acoustic variation and arousal level in
infants, Psychophysiology, 1970, 6 (5), 517-526.
Brackbill, Y., Adams, G., Crowell, D., & Gray, M. Arousal
level in neonates and preschool children =n4er con-
tinuous auditory stimulation. Journal of Exoerimental
Child Psychology, 1966, 4 (2), 178-188.
Brandt, J., & Hollien, H. Underwater hearing thresholds
in man as a function of water depth. Journal of the
Acoustical Society of America, 1969, 46 (4), 893-894.
Bridgman, C.S., & Carmichael, L. The reflex character of
first fetal movements. Psychological Bulletin, 1935,
32, 564. (Abstract).
Brogden, W. Sensory conditioning measured by the facilition
of auditory acuity. Journal of Experimental Psycholoev,
1950, 40, 512-519.
Carol, L. Alice in wonderland and through the looking glass.
New York: Grosset and Dunlap, 1946.
Clark, F.M. A developmental study of the bodily reactions
of infants to an auditory startle stimulus. Journal
of Genetic Psychology, 1939, 55, 415-427.
Deitz, S.M. Operant behavior in the human infant:
Differentiation and discrimination. Dissertation
Abstracts International, 1972, 32 (11-A), 6187-6188.
Dishoeck, H.A.E.v. The detailed audiogram in air-and-bone-
conduction. Acta Oto-Laryng., Stockh., 194S, 36,
441-446. (Psychological Abstracts 1949, 23, :o. 6002.)
Eisenberg, R.B. Auditory behavior in the human neonate: 1.
Methodologic problems and the logical design of
research procedures. Journal of Auditory Research,
1965, 5 (2), 159-177.
Eisenberg, R., Hunter, M.A., Griffin, E.J. & Coursin, D.B.
Auditory behavior in the human neonate: A preliminary
report. Journal of Speech & Hearing Research, 1964,
7 (3), 245-269.
Escalona, S., & Leitch, M. Eight infants: Tension nani-
festations in response to prolonged stimulatio-.
Topeka, Kansas: Meninger Foundation, 1950. (Film)
Fitzgerald, J.E., & Windle, W.F. Some observations on
early human fetal movements. Journal of Comoarative
Neurology, 1942, 76, 159-167.
Forbes, H.S., & Forbes, H.B. Fetal sense reaction: Hearing.
Journal of Comparative Psychology, 1927, 7, 353-355.
Froeschels, E., & Beebe, H. Testing the hearing of newborn
infants. Arch. Otolaryncj., Chicago, 1946, 44, 710-
714. (Psychological Abstracts, 1947, 21, No. 1775).
Gesell, A. The developmental morphology of infant behavior
pattern. Proceedings of the National Academy of
Sciences, Washington, 1932, 18, 139-143.
Gesell, A. Life begins, New York: Erpi Classroom Films
Gesell, A., & Ames, L.B. The ontogenetic organization of
prone behavior in human infancy. Journal of Genetic
Psychology, 1940, 56, 247-263.
Gilmer, B.V.H. An analysis of the spontaneous responses
of the newborn infant. Journal of Genetic Psychology.
1933, 42, 392-405.
Gordon, I.J., Jester, E. (R.M. Travers, (Eds.) Second
handbook of research on teaching: A project of the
American Educational Research Association. Chicago;
Rand McNally, 1973.
Gorman, J.J., Cogen, D.G., and Gellis. Apparatus for
grading the visual acuity of infants on the basis of
opticokinetic nystagmus. Pediatrics, 1957, 19, 1088-1092.
Grier, J.B., Counter, S.A., & Shearer, W.M. Prenatal
auditory imprinting in chickens. Science, 1967, 155,
Grimwade, J.C., Walker, D.W., & Wood, C. Sensory stimula-
tion of the human fetus. Australian Journal of Mental
Retardation, 1971, 1 (2), 63-64. (Psychological Abstracts,
1971, 46, No. 6466).
Hamilton, P.M. Underwater hearing thresholds. Journal
of the Acoustical Society of America, 1957, 29,
Harris, D.B., & Harris, E.S. A study of human fetal move-
ments in relation to mother's activity. Human
Biology, 1946, 18, 221-237.
Herit, J.T. Two measures of general motor activity and
their relationship in acoustically stimulated versus
unstimulated human neonates. Journal of Auditory
Research, 1970, 10 (4), 288-291.
Hocker, D. Fetal behavior, Res. Publ. Ass. nerv. ment. Dis.,
1939, 19, 237-243. (Psychological Abstracts, 1940,
14, No. 824).
Booker, D. The prenatal origin of behavior. Lawrence,
Kansas: University of Kansas Press, 1952.
Hoa--sten, G.H., & Moncur, J.P. Stimuli and intensity
factors in testing infants. Journal of Speech &
Hearing Research, 1969, 12 (4), 687-702.
Hurlock, E.B. Experimental studies of the newborn. Child
Development, 1933, 4, 148-163.
Hutt, S.J., et al. Auditory responsivity in the human
neonate. Nature, 1969, 218 (5144), 888-
Jones, H.E. Tests of motor functions. Berkeley:
University of California, 1936. (Film)
Kasatkin, N.I., & Levikova, A.M. On the development of
early conditioned reflexes and differentiations of
auditory stimuli in infants. Journal of Experimental
Psychology, 1935, 18, 1-19.
Kearsley, R.B. The newborn's response to auditory
stimulation: A demonstration of orienting and defensive
behavior. Child Development, 1973, 44 (3), 582-591.
Kee=, R.E., Chase, H.H., 5 Graham, F.K. Twenty-four hour
retention by neonates of an habituated heart rate
response. Psychonomic Science, 1965, 2 (9), 265-266.
Kellcgg, W.N. A method for recording the activity of the
human fetus in utero, with specimen results. Journal
of Genetic Psychology, 1941, 58, 307-326.
Kirk. R.E. Experimental Design: Procedures for the
behavioral sciences. Belmont, California: Brooks/
Klopfer, P.H. & Esilman, J.P. An introduction to animal
behavior: r-ology's first century. Englewood
Cliffs, New -ersey: Prentice-Hall Inc., 1967.
Lewis, M.A. A de-ielpmental study of the cardiac responses
to stimulus :eset and offset during the first year
of life. Pf;-:ophysiology, 1971, 8 (6), 689-698.
Ling, D. Acoustic stimulus duration in relation to
behavioral r-sponses of newborn infants. Journal
of Speech & -iaring Research, 1972, 15 (3), 567-571.
Ling, D., Ling, A-.., & Doehring, D.C. Stimulus, response,
and observer variables in the auditory screening of
newborn infaz:s. Journal of Speech & Hearing
Research, 157:, 13 (1), 9-18.
Margoshes, A., & CZllins, G. Right-handedness as a
function of eternal heartbeat. Perceptual & Motor
Skills, 1965. 20, (2), 442-444.
McCall, R.B., & talson, W.H. Amount of short term
familiarizatn:n and the response to auditory dis-
crepancies. b7ild Development, 1970, 41 (3), 861-869.
McGraw, M.B. Gr^tSh, a study of Johnny and Jimmy. New
York: Applean-Century, 1935.
McGraw, M.B. Swiiming behavior of the human infant.
Journal of p7diatrics, 1939a, 15, 485-490.
McGraw, M.B. P.eflx swimming movements in the newborn
of different species. New York: Columbia Medical
Center, 1931:. (Film)
McGraw, M.B., Gr---h: A study of Johnny and Jimmy. New
York: Wari- & Gilbert, Psychological Lab.,
Columbia Unrlrrsity, 1941a. (Film)
McGraw, M.B., Re -:ions of the infant to pin rick. New
York: warm=- & Gilbert, Psychological Lab.,
Columbia -Un:-rsity, 1941b. (Film)
McGraw, M.B., & ---eeze, K.W. Quantative studies in the
development :f erect locomotion. Child Development,
1941, 12, 2f--303.
Moreau, T., Birc-. H.G., & Turkewitz, G. Ease of
habituation to repeated auditory and somesthetic
stimulation ma the human newborn. Journal of
Experimental Zhild Psychology, 1970, 9 (2), 193-207.
Murooka. Lullaby from the Womb. Hollywood, California:
1974, no. ST-11421 Toshiba-EMI limited (Stereo
record and album jacket).
Newberry, H. Studies in fetal behavior: IV. The measure-
ment of three types of fetal activity. Journal of
Comparative Psychology, 1941, 32, 521-530.
Plato. The Laws, translation with an introduction by A.
E. Taylor. London: Dent, 1960.
Pomerleau-Malcuit, A., & Clifton. R.K. Neonatal heartrate
response to tactile, auditory, and vestibular
stimulation in different states. Child Development,
1973, 44 (3), 485-496.
Pratt, K.C. The effects of repeated auditory stimulation
upon the general activity of newborn infants.
Journal of Genetic Psychocl;y, 1934, 44, 96-116.
Ray, W.S. A preliminary report on a studi of fetal
conditioning. Child Development, 1932, 3, 175-177.
Richards, T.W., & Newberry, H. Studies in fetal behavior:
III. Can performance on test items at six months
postnatally be predicted on the basis of fetal
activity? Child Developme-t, 1938, 9, 79-86.
Richards, T.W., Newberry, H., & Fallgatter, R. Studies
in fetal behavior: II. Activity of the human fetus
in utero and its relation to the other prenatal
conditions, particularly the mother's basal rate.
Child Development, 1938, 9, 69-68.
Roberts, B., & Campbell, D. Activity in newborns and the
sound of a human heart. Psychonomic Science, 1967,
9 (6), 339-340.
Rose, D. Experimentally induced changes in fetal behavior.
Psychological Bulletin, 1941, 38, 733. (Abstract)
Rovee, C.K, & Rovee, D.T. Conjugate reinforcement of
infant exploratory behavior. Journal of Experimental
Child Psychology, 1969, 8, 33-39.
Salk, L. The effects of the normal heartbeat sound on the
behavior of the newborn infant: Implications for
mental health. World Mental Health, 1960, 12, 168-175.
Salk, L. The importance of the heartbeat to h=an nature:
Theoretical, clinical, and experimental c:servations.
Proc. Third World Contr. Psychiatr. Montreal:
McGill University Press, 1961, 1, 740-74:.
Salk, L. Mother's heartbeat as an imprinting soi-ulus.
Trans. New York Acad. Sci., 1962, 24, 753--53.
(Brackbill, Y., Adams, G., Crowell, D., ia:, M.
Arousal level in neonates and preschool C:.ldren
under continuous auditory stimulation. !:-rnal of
Experimental Child Psychology, 1966, 4 (2 178-188).
Salk, L. Thoughts on the concept of imprintin- and its
place in early human development. Canadia: Psychiatric
Association Journal, 1966, 11, 295-305.
Schaefer, H.H. Vibration as reinforcer for infa-t children.
Journal of the Experimental Analysis of r-:avior,
1960, 3, 160.
Schevill, W.E., & Lawrence, B. High frequenc-- aditory
response of a bottlenosed porpoise Tursi:;
Truncatus (Montagu) Journal of the Acoust5.al Society
of America, 1953, 25, 1016-1017.
Sheppard, W.C. Operant control of infant vocal and motor
behavior. Journal of Experimental Chile --.-chology,
1969, 7 (1), 36-51.
Simons, G. Comparisons of incipient music resconses among
very young twins and singletons. Journal :f Research
in Music Education, 1964, 12 (3), 212-221.
Sivian, L. J. On hearing in water vs. hearing in air.
Journal of the Acoustical Society of AmernO a, 1947,
Smart, M.S., & Smart, R.C. Children (2nd ed.). New York:
Smith, K.J. Habituation of the orienting resp:=se to
auditory-stimulus sequences in the human -:wtorn.
Conditional Reflex, 1967, 2 (2), 160-161.
Sontag, L.W., & Wallace, R.F. The movement re0 cnse of
the human fetus to sound stimuli. Child :evelopment,
1935, 6, 253-258.
Spelt, D.K. Conditioned responses in the hrn-m fetus in
utero. Psychological Bulletin, 1938, 35, 12-713.
Spelt, D.K. The conditioning of the human fetus in utero.
Journal of Experimental Psychology, 1948, 38, 338-346.
Steinschneider, A., Lipton, E.L., & Richmond, J.B.
Auditory sensitivity in the infant: Effect of
intensity on cardiac and motor responsivity. Child
Development, 1966, 37 (2), 233-252.
Steinschneider, A. Sound intensity and respiratory
responses in the neonate. Psychosomatic Medicine,
1968, 30 (5), 534-541.
Stubbs, E.M. The effect of the factors of duration,
intensity, andpitch of sound stimuli on the responses
of newborn infants. University of Iowa Studies in
Child Welfare, 1934, 9, No. 4, 75-135.
Sugiyama, A. Personal communication, July 21, 1976.
Tulloch, J.D., Brown, B.S., Jacobs, H.L., Prugh, D.G.,
& Greene, W.A. Normal heartbeat sound and the behavior
of newborn infants: A replication study.
Psychosomatic Medicine, 1964, 26 (6), 661-670.
Vanderplas, J.M., & Blake, R.R. Selective sensitization
in auditory perception. Journal of Personality.
1949, 18, 252-266.
Watson, J. Psychology from the standpoint of the
behaviorist. Philadelphia: J.B. Lippincott, 1919a.
Watson, J. Experimental investigation of babies. Chicago:
Stoeling, 1919b. (Film)
Weiss, L.A. Differential variations in the amount of
activity of newborn infants under continuous light
and sound stimulation. University of Iowa Studies
in Child Welfare, 1934, 9, No. 4.
Wright, H.F. Observational child study. In P.H. Mussen
(Ed.) Handbook of research methods in child study.
New York: Wiley, 1960.
Harvey Barnett was born at March Air Force Base near
Riverside,California, on March 4, 1948. He was one of four
children in the family of a career military officer. As is
common with service children, he traveled around the world
several times and attended twenty-one different schools
before graduating from high school in Frankfurt, Germany, in
1966. He attended Brevard Junior College and received an
AA degree with honors from there in 1969. He attended the
University of Florida and graduated with honors from there
in 1971 receiving a BA in psychology.
He took a faculty position at East Tennessee State
University from 1972 to 1974. He finished a Master of
Science in Teaching (Psychology) from the University of
Florida in 1973. He has acted as Director of Research for
Infant Swimming Research since 1970. He received his
Ph.D. in March, 1978.
Mr. Barnett is married to the former Judith Carlene
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.
John M. Newell, Chairman
Professor of Education
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.
William B. Ware
Professor of Education
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.
Arthur J/ Newman
Professor of Education
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.
William D. Wolking
Professor of Educat on
This dissertation was submitted to the Graduate Faculty
of the Department of Foundations of Education in the
College of Education and to the Graduate Council, and was
accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
Chairman, Foundations of Education
Dean, G/ ute Sc
Dean. Graduate School
UNIVERSITY OF FLORIDA
3 1262 08552 9310