Departments of Obstetrics Gynecology

ORIGINAL ARTICLE

Persistence of fetal memory into neonatal life

N.L. GONZALEZ-GONZALEZ 1 , M.N. SUAREZ

1 , B. PEREZ-PIÑERO

1 , H. ARMAS

1 ,

E. DOMENECH 1

& J.L. BARTHA 2

1 Departments of Obstetrics Gynecology and Paediatrics, University Hospital of the Canary Islands, Tenerife,

2 Departments of

Obstetrics Gynecology and Paediatrics, University Hospital ‘‘Puerta del Mar’’, Cádiz, Spain

Abstract Objectives. To test the hypothesis that fetal memory persists into the neonatal period. Study design. Forty-one newborns that had been repeatedly stimulated by using fetal vibroacoustic stimulation were compared with 31 controls. The same vibroacoustic stimulator was used for both fetal and neonatal stimulation tests. For the neonatal test the stimulus was applied against the mastoid of the newborn with the interposition of a specifically designed solid�liquid interface to simulate intrauterine conditions. Student’s t -test was used. Results. Neonatal habituation rate (the number of consecutive stimuli applied before a baby stopped responding) was significantly higher in those newborns who had not participate in the fetal habituation study (7.09/5.4 stimuli) than in those who had (4.19/4.1 stimuli), p�/0.01. Conclusions. Newborns who were stimulated in utero habituated earlier than those who had not previously experienced the stimulation. These results suggest that fetal memory persists into neonatal life.

Key words: Fetal habituation, fetal learning, fetal memory, vibroacoustic stimulation, fetal behavior

Habituation is a decrease in responsiveness to a

stimulus when that stimulus is presented repeatedly

or for a prolonged time (1). This process is a basic

form of learning and a normal pattern may be one

indication of intact central nervous system function

(2).

Fetal habituation to repeated vibroacoustic stimuli

has been used to study fetal memory (3) and it was

concluded that fetuses were able to learn, had a

short-term memory of at least 10 min, and a long-

term memory of at least 24 h.

Little attention has been paid to the possible

functions of fetal memory. It has been proposed

that it can be involved in recognition of and attach-

ment to the mother, which may influence promotion

of breastfeeding and language acquisition (4). More-

over, potentially fetal learning and memory may have

a clinical interest if they could be positively influ-

enced.

In spite of these potential benefits, there are only a

few studies focusing on the persistence of fetal

memory into neonatal life (5) and the vast majority

of them did not use an adequate methodology to test

this hypothesis. One of the main methodological

problems of studying fetal learning during the

postnatal period is finding a fetal stimulus that can

be reproduced postnatally, thus able to be remem-

bered by the newborn.

We have developed a neonatal mechanism that

reproduces fetal vibroacoustic stimulation and we

have used that system to test the hypothesis that fetal

memory persists into the neonatal period. This was

the aim of the present study.

Subjects and methods

Forty-one pregnant women who attended the an-

tenatal clinic between September 2001 and August

2002 were studied. The inclusion criteria were as

follows: healthy single pregnancies with absence of

both medical and obstetric disorders, no toxic

habits, cephalic presentation, gestational age be-

tween 38 and 40 weeks, and normal fetal growth

assessed by ultrasound.

Correspondence: N.L. Gonzalez Gonzalez, Avda de la Universidad no. 27, 38208 La Laguna, Tenerife, Canary Islands, Spain. E-mail: ngonzalezg@sego.es

Acta Obstetricia et Gynecologica. 2006; 85: 1160�1164

(Received 18 April 2006; accepted 5 June 2006)

ISSN 0001-6349 print/ISSN 1600-0412 online # 2006 Taylor & Francis

DOI: 10.1080/00016340600855854

Maternal weight was 67.69/7.2 kg, maternal

height 162.49/5.7 cm, ultrasound-assessed maternal

abdominal adiposity 16.09/6.1 mm, and gestational

age at the study 39.59/0.3 weeks. Amniotic fluid

index was normal in all cases.

The study was approved by the Local Ethical

Committee and all women gave their written consent

form for the study.

After checking amniotic fluid index, fetal presen-

tation, fetal growth, and maternal adiposity, the

ultrasound probe was placed to observe totally or

partially the fetal face, the upper extremities, and the

upper part of fetal trunk on the same view. In all

cases fetal movements and fetal heart rate (FHR)

were monitored (HP 8041 A) at the same time until

the registered pattern corresponded to a fetal active

state (F2). Then an ultrasound observation was

started and after checking the fetus really was in

active state for longer than 10 min, the study was

initiated. The fetuses were stimulated by using an

artificial larynx as a fetal vibroacoustic stimulator

(calibration made by the manufacturer by checking a

sound level 85�110 dB and audible sound 100� 9,000 Hz a meter away from the sound transmitter;

‘‘SERVOX INTON’’, Servox Medizintechnik, Mer-

heim).

This stimulus was repeatedly applied to the

maternal abdomen above the fetal head for a period

of 1 s every minute till a maximum of 24 stimuli.

A general movement of the fetal trunk, the upper

extremities, the head or the eyes starting in the first

second post-stimulus was considered as a positive

response. If the fetus was moving before stimulation,

then a change in the pattern of the previous move-

ments within 1 s after the application of the stimulus

was considered as a positive response. Some fetuses

remained very active after stimulation but no cases

required prolonging the interval between stimuli for

longer than 3 min to evaluate the reflex response as it

was always possible to find an interval to evaluate the

response to a new stimulus.

Habituation was defined as the lack of response to

four consecutive stimuli. Habituation rate was de-

fined as the number of consecutive stimuli applied

before a fetus stopped responding.

Habituation tests were repeated every 48�72 h until delivery. Eighteen fetuses undertook three

habituation tests, 14 had two tests, and in the

remaining 9 only one test was possible before

delivery.

All habituation tests were done in the same room

by the same operator between 9:00 and 13:00 in the

morning. Neonatal habituation was compared be-

tween the 41 neonates who had undergone fetal

habituation tests and a control group composed of

41 newborns who had not participated in the

fetal habituation study and who had not received any

kind of prenatal vibroacoustic stimulation test. This

group was selected by taking the consecutive healthy

newborn after one of the babies participating in the

fetal habituation study was born. Inclusion criteria

for the control group were healthy singleton preg-

nancy at term and a birth weight between 2,500 and

4,000 g. Ten neonates of the control group were

further excluded from the study due to loss of

follow-up (mothers first accepted but subsequently

decided to resign from the study). There were no

statistically significant differences in gestational age

at delivery, analgesia in labor, mode of delivery,

Apgar scores, and umbilical pH between the studied

groups.

All neonatal habituation tests were done by the

same observer who was blinded to the studied groups.

Neonatal habituation assessment took place

within 1�2 days after delivery before an otoacoustic emission test was performed. A normal result of this

test demonstrates the integrity of the audit system

and it has been implemented in our hospital as a

universal newborn hearing-screening program. For

the habituation test the newborns were placed into a

soundproof cot in a quiet room, until a maximum of

60 min after a feed between 10 and 12 in the

morning.

All babies were observed for 10 min and the

operator waited until the newborn was quiet (neither

crying nor having intense general movements). Then

the same vibroacoustic stimulator that had been

used for the fetal habituation study was applied

against the mastoid of the newborn with the inter-

position of a solid�liquid interface composed of muscular animal tissue (2 cm thick) plus a plastic

bag of 50 ml of warm saline in order to simulate the

intrauterine conditions by which the stimulus

reached the fetus.

Newborn stimulation was done using the same

protocol as used in the fetal study and applying the

same criteria to define a positive response, habitua-

tion, and habituation rate.

Sample size was calculated by using previously

reported data (6�8). Statistical analyses were per- formed by SPSS 11.0.1. (SPSS Inc Corp, Chicago,

IL, USA). Distributions were checked by histograms

and the Kolmogorov�Smirnov test. Results are presented as mean and SD and qualitative variables

as number and percentage. Groups were compared

using an unpaired Student’s t -test. Comparisons

between the first two measurements of habituation

were made by using the paired Student’s t -test.

When available (n�/18), comparisons among the three measurements of habituation were made by

Persistence of fetal memory into neonatal life 1161

using the ANOVA test. Proportions were compared

by using the x 2 -test. For correlations, Spearman’s

correlation coefficient was used. The significance

level was previously set at the 95% level (p�/0.05).

Results

All the fetuses showed positive responses to vibroa-

coustic stimulation. Thirty-nine of the 41 stimulated

fetuses in the first test (95.1%), 30 of those 32 who

undertook the second one (93.7%), and the all the

18 fetuses who received the third test (100%)

habituated (Table I). The two fetuses who did not

habituate during the first test were the same who did

not do it during the second one. One of them was

born before having a third procedure. The other one

underwent a third stimulation test and habituated

after receiving the maximum number of stimuli.

The habituation rate was 7.09/4.6 stimuli. When

two procedures were performed (n�/32), the habitua- tion rate was significantly lower in the second one

(Table II). The habituation rate was even lower in the

third one but this could be only studied in those

fetuses who underwent three tests (n�/18) and the differences were only of borderline significance (p�/ 0.1), probably due to the small sample size (Table II).

The percentage of newborns who showed a

positive response to vibroacoustic stimulation was

similar in both of the studied groups: n�/35 (85.3%) versus n�/26 (83.8%) for the studied and the control groups respectively, p�/0.86. The postnatal habitua- tion rate was significantly higher in those newborns

who had not participated in the fetal habituation

study than in those who had (7.09/5.4 stimuli versus

4.19/4.1 stimuli) (p�/0.01) (Figure 1). There was a significant positive correlation be-

tween the last prenatal habituation rate and the

postnatal one (r�/0.42, p�/0.01).

Discussion

Habituation, the ability of an organism to cease

responding to a repeated stimulus, has been pro-

posed as a simple form of non-associative learning.

Unfortunately habituation studies have used very

different methodology and there is a wide range of

definitions of habituation (Table III). We have used

an artificial larynx to generate a vibroacoustic

stimulus which is able to provoke a reflex motor

response in the fetus that is also associated to a

transient acceleration in the FHR. This effect has

been the rationalization of its use in clinical obstetric

practice (13�16). We chose 24 as the number of stimuli to perform

this study to be able to compare our results with

those from previous reports, the majority of them

using between 20 and 24 stimuli (range 8�50) (Table III). In comparison, we found a high percen-

tage of fetuses that habituated although the number

of stimuli needed to habituate (habituation rate) was

slightly lower than those previously reported (Table

IV). These differences may be explained by the

special characteristics of the sound we used for

stimulation in terms of quality (novel sound for the

fetuses), frequency, intensity, and a longer interval

between stimuli in order to perfectly identify the

response that caused the stimuli.

We found that the more the number of habituation

procedures the fetuses underwent, the lower the

habituation rates. This decrease in habituation rates

has been previously reported by Van Heteren et al. (3),

who demonstrated that the responses to a repeated

vibroacoustic stimulation could reappear as soon as

10 min after starting the procedure, which suggests

this fact is a result of fetal habituation rather than fetal

exhaustion. However, few studies of human fetal or

newborn habituation have included dishabituation

procedures (i.e. assessment of the reemergence of a

habituated response) to determine if response decre-

ments are the result of reevaluation of information (a

brain process) or fatigue of peripheral receptors.

Sandman et al. (18) have found that the 32-week

human fetus is capable of detecting, habituating, and

dishabituating to an external vibroacoustic stimulus

and this supports the premise that areas of the human

fetal central nervous system critical for detecting and

discriminating information and for learning and

memory have developed by the early third trimester.

Table I. Fetal habituation in each one of the performed tests

Habituation No habituation

N n % n %

Test 1 41 39 95.10 2 4.80

Test 2 32 30 93.75 2* 6.25

Test 3 18 18 100.00 � �

*These two fetuses were the same ones that did not habituate

during the first test.

Table II. Habituation rates according to the number of habitua-

tion tests

Number of

fetuses

Test 1

Mean9/SD

Test 2

Mean9/SD

Test 3

Mean9/SD

41 7.09/4.6

32 7.89/4.9 6.49/5.1*

18 7.49/4.7 6.09/5.1 5.39/4.0**

SD, standard deviation. *p�/0.02, **p�/0.1.

1162 N.L. Gonzalez-Gonzalez et al.

Fetal learning and memory have been studied

using habituation testing but only a few studies have

focussed on the persistence of fetal memory into

neonatal life. Most of them did not use a reliable

methodology and their results were difficult to

interpret.

De Casper and Spencer (9) found significant

differences in the neonatal sucking responses be-

tween a group of newborns whose mothers had read

out loud selected passages twice daily for 6 weeks of

pregnancy and a group of controls. Another study

showed that neonatal sucking responses varied when

newborns were exposed to recordings of either

maternal voice or other female voices (10). Along

the same lines, other studies have reported neonatal

ability to recognize maternal voice (19,20), musical

sounds (21�25), or TV tunes (26,27). Several of these studies suggest the existence of fetal learning

and memory as well as the persistence of learning

into neonatal life. However, they illustrate that this

area of research is very difficult, with many of the

studies having methodological limitations.

James et al. (5) performed a prospective rando-

mized study using music played on the maternal

abdomen in 10 pregnant women who were com-

pared with another group of 10 women who did not

undergo the test. They studied FHR and neonatal

behavioral states. In spite of the small sample size

they found that prenatally exposed neonates showed

more state transitions and spent more time in

awakened states. The authors concluded that pre-

natal music exposure alters neonatal behavioral

states, suggesting a simple form of fetal learning.

In the present study we designed a simple experi-

mental model by using a real ‘‘novel’’ stimulus for

both fetuses and newborns, as they had never been in

contact with a similar sound. We used a solid�liquid interface in the newborns for simulating as closely as

possible the intrauterine conditions through which

the sound reached the fetuses. Thus habituation

rates were evaluated by using the same procedure in

both fetuses and neonates. The habituation rate of

those newborns who had been stimulated in utero was

significantly lower than that of controls. In other

words, they habituated earlier, suggesting some kind

of persistence of fetal learning and memory into

neonatal life. The neonatal habituation rate of

controls (7.09/5.4 stimuli) was virtually the same as

the fetal habituation rate (7.09/4.6 stimuli), suggest-

ing an excellent fetal�neonatal reproducibility of the test in individuals first-time exposed to that stimulus.

In addition, the existence of a positive correlation

between the prenatal and postnatal habituation rates

suggests that this ability to learn and memorize is not

the same in all the babies and the more stimuli

needed in the fetal period to habituate, the more

stimuli needed in the postnatal period for this.

New born group

H a

b itu

a tio

n R

a te

–1

1

3

5

7

9

11

13

15

Study Control

p= 0.015

Figure 1. Habituation rate to vibroacoustic stimulation in the

neonatal period.

Table III. Criteria for defining fetal habituation

Fetal response Number of stimuli

Reference Stimulus Motor/FHR Maximum Frequency Criterium Thabituation

Leader, 1982 (2) Electric brush Motor 50 � No response to 5 consecutive stimuli Van Heteren, 2000 (3) Vibrator 74 dB,

20 �9,000 Hz Motor 24 1/30 s No response to 4 consecutive stimuli

Madison LS,1986 (6) Vibrator Motor 40 10 s Less than 5% movements to 5

consecutive movements

Johansson 1992 (7) Vibrator 110 dB

and 3,000 Hz

Motor and FHR 20 1/min Change in FHR B/15 s poststimulus.

Absence of fetal movements

Doherty, 2000 (8) Artificial larynx Motor 40 2/5 s No response to 5 consecutive stimuli

Groome, 1994 (11) Artificial larynx Motor and FHR 8 10 s Decreased fetal movements. No FHR

accelerations

Shalev, 1986 (12) Door bell Motor and FHR 50 2/10 s No response to 5 consecutive stimuli

Present study Artificial larynx Motor 24 1 min No response to 4 consecutive stimuli

FHR, fetal heart rate.

Persistence of fetal memory into neonatal life 1163

Safety of habituation studies for both fetuses and

newborns may be questioned. Three previous studies

have evaluated hearing function in infants who were

exposed to intrauterine stimulation with artificial

larynxes and no effect could be demonstrated (28� 30). To our knowledge, there are no other studies

specifically designed to evaluate side effects of habi-

tuation tests. In addition, we have followed the

methodology proposed by Van Heteren et al. (3,17),

who did not report any adverse effect of habituation

tests. Our interval between stimuli was longer (1 min)

than that used in their study (30 s), to give the fetus

enough time to recover, so we could more clearly

evaluate the response to the following stimulus.

In conclusion, our results suggest that those new-

borns who were stimulated in utero recognized the

stimulus and habituated earlier than those who had

not previously experienced the stimulation. This

suggests that those babies who were stimulated

during fetal life were able to learn and memorize

and both capabilities persisted into their neonatal life.

References

1. Thompson RF, Spencer WA. Habituation: a model phenom-

enon for the study of neuronal substrates of behavior. Psychol

Rev. 1966;/73:/16 �43. 2. Leader LR, Baillie P, Martin B, Vermeulen E. The assessment

and significance of habituation to a repeated stimulus by the

human fetus. Early Hum Dev. 1982;/7:/211 �9. 3. Van Heteren CF, Boekkooi PF, Jongsma HW, Nijhuis JG.

Fetal learning and memory. Lancet. 2000;/356:/1169 �70. 4. Hepper PG. Fetal memory: Does it exist? What does it do?

Acta Paediatr Suppl. 1996;/85:/16 �20. 5. James DK, Spencer CJ, Stepsis BW. Fetal learning: a

prospective randomized controlled study. Ultrasound Obstet

Gynecol 2002; 431 �8. 6. Madison LS, Adubato SA, Madison JK, Nelson RW, Ander-

son JC, Erickson J, et al. Fetal response decrement true

habituation? Dev Behav Pediatr. 1986;/7:/14 �20. 7. Johansson B, Wedenberg E, Westin B. Fetal heart rate

response to acoustic stimulation in relation to fetal develop-

ment and hearing impairment. Acta Obstet Gynecol Scand.

1992;/71:/610 �5.

8. Doherty NN, Hepper PH. Habituation in fetuses of diabetic

mothers. Early Hum Dev. 2000;/59:/85 �93. 9. De Casper DJ, Spencer MJ. Prenatal maternal speech

influences newborn’s perception of speech sound. Infant

Behav Dev. 1986;/9:/133 �50. 10. Fifer WP, Moon C. Psychobiology of newborn auditory

preferences. Semin Perinatol. 1989;/13:/430 �3. 11. Groome LJ, Watson JE, Dykman RA. Heart rate changes

following habituation testing of the motor response in normal

human fetuses. Early Hum Dev. 1994;/36:/69 �77. 12. Shalev E, Bennett MJ, Wallace RM, Zuckerman H. Fetal

habituation to sound stimulus in various arousal states. In:

Schenker JD, Weinstein D, editors. The Intrauterine Life-

management and Therapy. New York: Elsevier Science

Publishers BV (Biomedical Division); 1986.

13. Read JA, Miller FC. Fetal heart rate acceleration in response

to acoustic stimulation as a measure of fetal well being. Am J

Obstet Gynecol. 1977;/129:/512 �7. 14. Clark SL, Sabey P, Jolley K. Nonstress testing with acoustic

stimulation and amniotic fluid volume assessment: 5973 tests

without unexpected fetal death. Am J Obstet Gynecol. 1989;/

160:/694 �7. 15. Smith CV. Vibroacoustic stimulation. Clin Obstet Gynecol.

1995;/38:/68 �77. 16. Tan KH, Smyth R. Fetal vibroacoustic stimulation for

facilitation of tests of fetal wellbeing. Cochrane Database

Syst Rev. 2001; (1): CD002963.

17. Van Heteren DF, Boekkooi PF, Schiphorst RH, Jonsgma HW,

Nijhuis JD. Fetal habituation to vibroacoustic stimulation in

uncomplicated post-term pregnancies. Eur J Obstet Gynecol

Reprod Biol. 2001;/97:/178 �82. 18. Sandman CA, Wadhwa P, Hetrick W, Porto M, Peeke HV.

Human fetal heart rate dishabituation between thirty and

thirty-two weeks gestation. Child Dev. 1997;/68:/1031 �40. 19. De Casper DJ, Fifer WP. Of human bonding: newborns prefer

their mothers; Hepper PG, Scott D, Shahidullah S. Newborn

and fetal response to maternal voice. J Reprod Infant Psychol.

1993;/11:/147 �53. 20. Damastra-Wijmenga SM. The memory of the newborn baby.

Midwives Chron. 1991;/104:/66 �9. 21. Zimmer EZ, Divon MY, Vilensky A, Sarna Z, Petretz BA,

Paldie E. Maternal exposure to music and fetal activity. Eur J

Obstet Gynecol Reprod Biol. 1982;/13:/209 �13. 22. Chen DG, Huang YF, Zhang JY, Qi GP. Influence of prenatal

music and touch-enrichment on the IQ motor development

and behaviour infants. Chin J Psychol. 1994;/8:/148�51. 23. Shetler DE. The inquiry into prenatal music experience: a

report of the Eastman Project 1980 �87. Pre-Peri-Natal Psych J. 1989;/3:/171�89.

24. Arabin B. Music during pregnancy. Ultrasound Obstet

Gynecol. 2002;/20:/425�30. 25. Lamont A, Dibben N. Motivic structure and the perception

of similarity. Music Perception. 2001;/18:/245 �74. 26. Hepper PG. Fetal soap addiction. Lancet. 1988;/1:/1347 �8. 27. Hepper P. An examination of fetal learning before and after

birth. Irish J Psychol. 1991;/12:/95 �107. 28. Arulkumaran S, Skurr B, Tong H, Kek LP, Yeoh KH, Ratnam

SS. No evidence of hearing loss due to fetal acoustic

stimulation test. Obstet Gynecol. 1991;/78(2):/283 �5. 29. Ohel G, Horowitz E, Linder N, Sohmer H. Neonatal auditory

acuty following in utero vibratory acoustic stimulation. Am J

Obstet Gynecol. 1987;/157:/440 �1. 30. Nyman M, Barr M, Westgren M. A four-year follow-up of

hearing and development in children exposed in utero to

vibro-acoustic stimulation. Br J Obstet Gynaecol. 1992;/99:/

685 �8.

Table IV. Fetal habituation to repeated vibroacoustic stimulation

Habituation

Reference

Sample size

(n ) (%) Rate

Leader, 1982 (2) 40 85 10 �50 Madison, 1986 (6) 42 96 18 �22 Johansson, 1992 (7) 34 64 �/3 and B/21

Doherty, 2000 (8) 34 � 12 Shalev, 1986 (12) � 95 20 �30 Van Heteren, 2001 (17) 37 81 9.5

Present study 41 95 6.97

1164 N.L. Gonzalez-Gonzalez et al.