Fetal memory: Does it exist? What does it do?

Acta Pædiatrica Supplement, Volume 416: Pages 16-20, 1996.

PG Hepper

Foetal Behaviour Research Centre, School of Psychology, The Queen's University of Belfast, Belfast, Northern Ireland, UK

Whether fetal memory exists has attracted interest for many thousands of years. The following review draws on recent experimental evidence to consider two questions: does the fetus have a memory? And if so, what functions(s) does it serve? Evidence from fetal learning paradigms of classical conditioning, habituation and exposure learning reveal that the fetus does have a memory. Possible functions discussed are: practice, recognition of and attachment to the mother, promotion of breastfeeding, and language acquisition. It is concluded that the fetus does possess a memory but that more attention to the functions of fetal memory will guide future studies of fetal memories abilities.  Attachment, breastfeeding, classical conditioning, fetus, habituation, language acquisition, learning, memory, recognition

PG Hepper, Foetal Behavior Research Centre, School of Psychology, The Queen's University of Belfast, Belfast BT7 1NN, Northern Ireland, UK

Memory is something we all take for granted. It is difficult to imagine functioning in the world without a memory. The effects of Alzheimer's disease (1,2), or reports of people with brain lesions that affect specific aspects of memory (3-5) all vividly illustrate the crucial role that memory plays in our lives. For example, our ability to learn, retain information for short- or long-term periods, recognise individuals, objects, and to recall previous events or learning experiences, all depend on memory. For many years, it was thought the newborn infant did not possess a functioning memory, but rather memory developed over the months and years following birth. Study of newborns, and premature infants (6,7), in recent years, however, has changed this view. Newborns have been shown, via a variety of learning paradigms, habituation (8,9), classical conditioning (10,11), associative learning (12,13) and imitation (14,15), to possess a functioning memory.

It would be entirely possible that the memory abilities start at the moment of birth. Some event during the birth process triggers, along with other changes required for life outside the womb, activity in the central nervous system and hence memory begins. However this is unlikely (16). It is more likely that memory begins prenatally and the period of birth merely marks a transition from memory functioning in utero to memory functioning ex utero. This is not to suggest that memory when it begins to function is capable of all the feats of an adult's memory. Rather memory, at its developmental origin in the prenatal period probably functions in some rudimentary form and develops, both quantitatively and qualitatively, as the individual matures.

It is the aim of this review to discuss memory in utero. It is not intended to comprehensively review all studies in the area but rather illustrate salient points and issues. Two general areas will be considered. First, is there evidence of memory before birth? And second, if memory exists, does it have a function(s)?

Is their evidence of memory in utero?

Just as studies exploring memory in the newborn infant have used learning paradigms, similar paradigms have been applied to the fetus to explore its memory. Paradigms of "exposure learning", classical conditioning and habituation have been used to examine fetal learning and fetal memory.

Studies of classical conditioning of the fetus, although few, have a long history, dating back to the 1930s. Ray (17) paired a vibration (the CS) with a loud noise (the UCS). Although using only a single subject and reporting no data, Ray concludes the subject suffered "no ill effects from her prenatal education"(p.177), suggesting the fetus was successfully conditioned. Using a similar procedure Spelt (18) paired a vibration (the CS) with a loud noise (the UCS). He reported that after 15-20 pairings, most fetuses, in the last 2 months of gestation, responded to the vibration (CS) alone. More recently Feijoo (19,20) paired maternal relaxation (the UCS) with music (the CS) and examined individuals as fetuses and after birth. After 24 pairings of the stimuli, when the music was played to the fetus in the last weeks of pregnancy, individuals began moving and when played to the new-born, these babies stopped crying, opened their eyes and exhibited fewer clonic movements.

Recently I have replicated Spelt's study (18). I used a pure tone as the CS and a vibro-acoustic stimulus as the UCS. Fetuses were aged between 32 and 39 weeks of gestation at the time of testing. Approximately 50% of the fetuses (10/19) showed successful conditioning after 10-20 trials. Successful conditioning was unrelated to gestational age, or sex. It remains to be determined exactly why some fetuses exhibit conditioning and others do not.

To date the earliest observations of successful classical conditioning are in fetuses at 32 weeks of gestation (18, pers. obs.). However, it must be noted that attempting to determine the first appearance of particular fetal abilities is problematical (21). Whilst successful classical conditioning (or other paradigms) may indicate, if properly conducted, the existence of memory. Poor paradigms, inappropriate stimuli, or response measures may all contribute to unsuccessful conditioning rather than the absence of memory. Hence descriptions of the earliest occurrence of behaviours must be treated cautiously.

One of the most successful paradigms used to explore newborn memory and perceptual abilities has been that of habituation. Habituation can be defined as the decrement in response to stimuli following repeated presentation of the same stimulus (22). Whilst the majority of habituation studies in newborn infants have used visual stimuli, studies examining habituation in the fetus have used auditory stimuli. This preference is largely due to the ease of presenting auditory stimuli to the fetus and the difficulty of presenting visual stimuli. Despite many reported studies of habituation in the fetus, poor experimental paradigms in many of these reduce their usefulness in elucidating underlying memory processes (23).

However, the fetus has been demonstrated to habituate to auditory stimuli. For example (24) using movement as the response, fetuses at 36 weeks of gestation were shown to habituate to a 250 Hz pure tone; to respond again, to the presentation of a novel stimulus, a 500 Hz tone and following repeated presentation of this tone to habituate, finally on re-presentation of this tone to habituate; finally on re-presentation of the original tone, 250 Hz, the fetus initially responded (dishabituated) but then habituated quickly, faster than on its original presentation. Habituation of heart-rate response has also been demonstrated (e.g. 25,26).

The earliest habituation response has been demonstrated at 22-23 weeks of gestational age and seems to occur earlier in females than in males (27). Interestingly, the onset of auditory habituation corresponds to the onset of fetal auditory abilities (28). It may be that habituation is present earlier than this however as the fetus is unable to respond to auditory stimuli before this time, habituation to auditory stimuli is unable to be evidenced earlier than 22 weeks gestation. Stimulation using other sensory modalities, e.g. olfaction/taste, whih are functional at earlier gestational ages (16,21), may reveal habituation at even earlier gestational ages.

The final paradigm used to investigate fetal learning and memory is that of "exposure" learning. This paradigm has the potential to be a powerful tool in the exploitation of fetal memory abilities given the control available to the experimenter over the presentation of the stimulus and subsequent testing.

One study looked at the ability of the fetus to learn a TV theme tune, "Neighbours", frequently heard by the mother during her pregnancy (29,30). In the first experiment newborn infants (2-4 days of age) of mothers who watched Neighbours during pregnancy (and heard the theme tune) became alert stopped moving and their heart rate decreased (orienting) upon hearing the tune. These same individuals showed no such reaction to other, unfamiliar tunes. Newborns of mothers who did not watch the TV programme during pregnancy showed no reaction to the tune. There was no exposure to the TV tune after birth, the last exposure occurring before birth. Hence individuals must have learned the information about the tune prenatally and retained it 2-4 days until tested postnatally. A second experiment examined when learning and memory could first be evidenced before birth. Individuals were observed using ultrasound and their response to the theme tune noted. Individuals exposed to the theme tune showed a significant increase in movements at 37 weeks of gestation but not at 30 weeks of gestation. Fetuses not previously exposed to the tune showed no response at either age. This suggests that the ability to recognise familiar stimuli commences between 30 and 37 weeks of gestation (although see comments above). A third experiment looked at the duration of this memory by examining the response of individuals, prenatally exposed to the tune, 21 days after birth. Whilst the previous study showed newborns at 2-4 days of age "recognized" the tune there was no response when individuals were tested at 21 days of age. This may indicate that any recognition of memory is lost by 21 day of age in the absence of any postnatal exposure. Despite making inroads into the development of fetal memory much remains unanswered. Exactly how this memory is acquired, simple exposure leading to recognition or some form of classical conditioning, is unknown. Similarly, prenatal exposure to stimuli in this study was relatively uncontrolled thus exactly how much exposure to establish a preference is yet to be determined. The study does indicate that the fetus is able to learn and remember familiar auditory stimuli in the womb, retain this information over the birth period and that this learning is specific to the familiar stimulus. Other studies have confirmed the ability of the fetus to learn familiar audtory stimuli in utero (31,32)

The above studies, using different paradigms, have all demonstrated successful learning and the presence of memory abilities in the fetus. The use of identical paradigms as used in newborns, infants and adults may suggest identical memory processes are used in solving these tasks. However great care must be exercised here. Similar learning abilities have been observed in a variety of fetal animal, including invertebrates (33-37). Similar behavioural demonstrations of memory need not necessarily indicate similar memory processes or underlying neural mediation, or even continuity with the later memory processes of the individual.

Function(s) of fetal memory

One important question that has frequently been overlooked is the study of fetal memory is what function(s) does it serve? The function(s) of fetal memory are of central importance. A better understanding of fetal memory function may enable researchers to design more appropriate paradigms to investigate memory in utero. Whilst there is much that is unknown about the development of fetal behavior, and much more to be found out in the future, one things that I can be sure of is that fetal memory does not exist to enable the fetus to learn about TV theme tunes!

It is therefore important to consider the potential function(s) memory in utero may serve. There is little direct evidence to support the following speculations at present but they form a useful guide for future investigations of fetal memory.

It may be that fetal memory serves no function. Memory has to start at sometime. A time probably determined by the maturational status of the central nervous system. If the central nervous system develops to the extent required to sustain or support memory function prenatally, then memory will also start prenatally.

Fetal memory may serve a "practice function". There is much evidence of other behaviours starting before the time they are needed. Fetal breathing is one example of this. These movements beginning at 10-11 weeks of gestation (38), are similar to those exhibited after birth to enable breathing (39). Yet there is no air in the womb. A similar example is presented by eye movements; coordinated eye movements occur in the womb in the absence of all but the most limited visual stimuli (40). It has been considered that behaviours that are crucial to a certain stage of development are practiced prior to this time to ensure they functioned efficiently when needed. Such interpretations have often been considered only with respect to behaviours serving direct physiological need, e.g. breathing. The same however may hold true for psychological functions.

Memory is essential for normal functioning and it is not too surprising that such an important psychological function is "practiced" before birth in some form. It has been well-established that the structure of the central nervous system is partly under the control of activity within the system (41-44). It may well be that activity generated within the central nervous system from the fetus's use of memory also exerts an influence on the structural development of those parts of the CNS involved in memory.

Fetal memory may serve a number of specific functions, dependent upon the learning of particular stimuli prenatally.

Prenatal memory may be important for the development of attachment and other maternal recognition. There is much evidence that the fetus learns the speech characteristics of its mother prenatally and prefers its mother's voice to other female voices after birth (12,13, 45-47). It may be that by learning to recognise its mother prenatally the newborn infant has a "familiar" stimulus in its environment after birth to respond to. As well as enabling recognition of the mother this may also mark the beginnings of attachment. It should be noted that it is unlikely that the newborn recognizes this familiar stimulus as its mother, but rather a familiar auditory stimulus. However this prenatal priming may ensure and promote both the recognition of the individual's mother and the development of attachment. Although further evidence is required it may be that the individual learns about its mother's smell prenatally (48) and this too would similarly promote maternal recognition and attachment.

A memory in utero may be important for the establishment of breast feeding. The mother's diet flavours both the amniotic fluid and her breast milk (49-51). The fetus may learn about the flavour of the amniotic fluid via its swallowing of this fluid which begins at 12 weeks of gestation (38). When presented to the breast for the first time, the newborn recognises the colostrums as familiar due to the presence of the same tastes that have been present in the amniotic fluid. This may enhance the individual's willingness to suck and promote breast feeding. It should be noted that in today's society, alternatives to breast milk are readily available and infants who do not breast feed can be readily supplied with alternatives. However, in evolutionary terms when the mother-newborn system was evolving, any newborn which failed to feed would die due to lack of food. In such an environment it may be expected that conditions would be established to ensure the infant did breastfeed. Prenatal familiarisation with the taste of milk would contribute to ensuring breastfeeding was successfully established. Recent evidence suggests the fetus can learn tastes experienced only prenatally (52) and through such learning acquires a preference for these tastes (53). Furthermore, preliminary observations at my lab indicate those mothers who experience the greatest change in diet between before and after birth have the greatest difficulty in establishing breastfeeding.

One final area where fetal memory may be important, is in the acquisition of language. I have noted above the ability of the fetus to learn its mother's voice (12, 13, 45-47). Recordings the fetal uterine auditory environment, reveal the prosodic nature of speech can be clearly heard inside the womb (54, 55). Further, the fetus has been shown to be able to differentiate between different speech sounds in the womb (54, 55). Further, the fetus has been shown to be able to differentiate between different speech sounds in the womb (56-58). Newborns also seem to have a preference for their mothers' native language (59). It may be that experience of speech prenatally begins the process of acquiring language.


Thus there are a number of possible reasons why the fetus should have a functioning memory, not perhaps of the complexity of the adult or even infant, but sufficient to ease its progress in its new world after birth. Further studies are required to examine whether these proposed functions, or indeed others, do actually exist. They may also serve to direct future research into fetal memory, the existence of which, in some form, is no longer in doubt.


  1. Båchman L., editor Memory functioning in dementia. Amsterdam: North-Holland. 1992.
  2. Parasuramun R. Martin A. Cognition in Alzheimer's disease: disorders of attention and semantic knowledge. Curr Opin Neurobiol 1994;4:237-44
  3. Mays AR. Memory and amnesia. Behav Brain Res 1995;65:29-36
  4. Badderly A. Human memory. Theory and practice. Hove: Lawerence Erlbaum Associates. 1990
  5. Milner B. Disorders of learning and memory after temporal lesions in man. Clin Neurosurg 1972;19:421-46
  6. Thomas EB, Ingersoll EW. Learning in premature infants. Devel Psychol 1993;29:692-700
  7. Gekoski MJ, Fagen JW, Pearlman MA. Early learning and memory in the preterm infant. Infant Behav Devel 1984;7:267-76
  8. Slater A, Murison V, Rose D. Habituation in the newborn. Infant Behav Devel 1984;7:183-200
  9. Slater A, Morison V, Town C, Rose D. Br. Devel Psychol 1985;3:211-220
  10. Blass EM, Ganchrow Jr, Steiner JE. Classical conditioning in newborn humans 2-48 hours of age
  11. Crowell DH, Blurion LB, Kobayashi LR, McFarland JL, Young RK. Studies in early infant learning: classical conditioning of the neonatal heart rate. Devel Psychol 1976;12:373-97
  12. DeCasper AJ, Fifer WP. Of human bondage: Newborns prefer their mothers' voices. Science 1980;208:1174-6
  13. Fifer TM, Moon C. Psychobiology of newborn auditory preferences. Seminars in Perinatology. 1989;13:430-3
  14. Field TM, Woodson R, Greenberg R, Cohen D. Discrimination and imitation of facial expressions by neonates. Science 1982;218:179-81
  15. Anisfield M. Neonatal initiation. Devel Rev. 1991:1160-97
  16. Hepper PG. The beginnings of the mind-evidence from the behaviour of the fetus. J Reprod Infant Psychol 1994;12:143-144.
  17. Ray WS. A preliminary report on a study of the foetal conditioning. Child Devel. 1932:3;175-7
  18. Spelt DK. The conditioning of the foetus in utero. J Exp Psychol 1948;38:338-46
  19. Feihoo J. Ut conscientia Noscatue. Cahier de Soprologie 1975;13:14-20
  20. Feihoo J. Le foetus Pierre et le loup ...ou une approche originale de l'audition prenatale humanine. In: Herbinet E. Busnell M.C. editors L'aube des sens. Paris: Stock, 1981.
  21. Hepper PG. Fetal psychology: an embryonic science. In: Nijhuis JG, editors. Fetal Behavior: developmental and perinatal aspects. Oxford: Oxford University press. 1992
  22. Thompson RF. Spencer WA. Habituation: A model for the study of neuronal substrates of behavior. Psycho Rev 1966;73:16-43
  23. Hepper PG. Fetal habituation - another pandora's box? Arch Dis Childhood, in press.
  24. Hepper PG. Shadidullah S. Habituation in normal and Down Syndrome fetuses. Quart J Exp Psychol 1992;44:305-17
  25. Goldkrand JW, Litvack BL. Demonstration of fetal habituation and patterns of fetal heart response to vibroacoustic stimulation in normal and high risk pregnancies. J Perinatology 1991; 11;25-9
  26. Leader LR, Baille P, Martin B, Molteno C. Fetal responses to vibrotactile stimuli: a possible predictor of fetal and neonatal outcome. Aust NZ J Obstet Gynecol 1984;24:251-6
  27. Leader LR, Baille P, Martin B, Vermeulen E. The assessment and significance of habituation in normal and high risk pregnancies. J Perinatalogy 1991;11:25-9
  28. Hepper PG, Shahdullah S. Development of fetal hearing. Arch Dis Childhood 1994; 71:F81-F87
  29. Hepper PG. Foetal "soap" addiction. Lancet 1988; 11th June:1347-8
  30. Hepper PG. (1991) an examination of fetal learning before and after birth. Irish J Psychol 1991;12:95-107
  31. Wilkin PE. Prenatal and postnatal responses to music and sound stimuli. In: Blum T, editor. Prenatal perception learning and bonding. Berlin. Leonardo, 1993
  32. Damstra-Wijmenga SMI. Fetal soap addiction. Lancet 1988; July 23:223
  33. Smotherman WP, Robinson SR. Habituation in the rat fetus. Quart J Exp Psychol 1992;44B:215-30
  34. Smotherman WP. Odor aversion learning by the rat fetus. Physiol Behav 1982;29:769-71
  35. Lickliter R, Stoumbos J. Modification of prenatal auditory experience alters postnatal auditory preferences of bobwhite quail chicks. Quart J Exp Psychol 1992;44B:199-214
  36. Hepper PG, Waldman B. Embryionic olfactory learning in frogs. Quart J Exp Psychol 1992;44B:179-97
  37. Cauber Y, Jaison P, Lenoir A. (1992) Preimaginal induction of adult behaviour in insects. Quart J Exp Psychol 1992;44B:165-178
  38. de Vries JIP, Visser GHA, Prectl IIFR. The emergence of fetal behaviour II. Quantitative aspects. Early Human Devel 1985;12:99-120
  39. Kozuma S, Nemoto A, Okai T, Mizuno M. Maturational sequence of fetal breathing movements. Biol Neonate 1991;60;36-40
  40. Horimoto N, Hepper PG, Shahidullah S, Koyanagi T. 1993. Fetal eye movements. Ultrasound in Obstet Gynecol 1993;3:362-9
  41. Purves D, Riddle DR, White LE, Gutierrez-Ospina G. Neural activity and the development of of the somatic sensory system. Curr Opin Biol 1994;4:120-3
  42. Blakemore C, Cooper GF. Development of the brain depends on the visual environment. Nature 1970;228;477-8
  43. Weisel TN, Hubel DH. Effects of visual deprivation on morphology and physiology of cells in the cat's lateral geniculate body. J Neurophysiol 1963;26:978-93
  44. Miller KD. Models of activity-dependent neural development. Progr Brain Res 1994;102:303-18
  45. Hepper PG, Scott D, Shahdullah S. Newborn and fetal response to maternal voice. J Reproduct Infant Psychol. 1993;11:147-53
  46. DeCasper AJ, Spence MJ. Prenatal maternal speech influences newborns' perception of speech sound. Infant Behav Devel 1986;9:133-50
  47. Hepper PG, Scott D, Shahdullah S. Newborn and fetal response to maternal voice. J Reproduct Infant Psychol 1993;11:147-53
  48. Porter RH. Mutual mother-infant recognition in humans. In: Hepper PG, editor Kin recognition. Cambridge: Cambridge University Press. 1991
  49. Mennella JA, Beauchamp GK. Maternal diet alters the sensory qualities of human milk and the nursling's behavior. Pediatrics 1991;88:737-744
  50. Mennella JA, Beauchamp GK. The transfer of alcohol to human milk: effects on flavor and the infant's behavior. New Engl J Med 1991;325:981-85
  51. Hepper PG. Adaptive fetal learning: prenatal exposure to garlic affects postnatal preference. Animal Behav 1988;36:935-6
  52. Schaal B, Orgeur P. Olfaction in utero: can the rodent model be generalized? Q J Exp Psychol Med 1992;44B:245-78
  53. Hepper PG. Human fetal "olfactory" learning. Int J Prenatal Perinatal Psychol Med 1995;7:153-9
  54. Querleu D, Renard X, Versyp F, Paris-Delrue L, Crepin G. Fetal hearing. Eur J Obstet Gynecol Reprod Biol 1988;29:191-212
  55. Hepper PG, Shadullah S. Noise and the fetus: a critical review of the literature. Sudbury Norfolk: HSE Books, 1994
  56. Lecanuet J-P, Granier-Deferre C, DeCasper AJ, Maugeais R, Andrieu A-J, Busnel M-C. Perception et discrimination foetale de stimuli langagiers, mise en evidence a partir de la reactivite cardiaque. Resultats preliminaries. C-R de l'Acad Sci Paris. Serie III 1987;305:161-64
  57. Lecanuet J-P, Granier-Deferre C, Jacquet AY, Capponi I. Ledru L. Prenatal discrimination of male and female voice uttering the same sentence. Early Devel Parenting 1993;2:217-28
  58. Shahdullah S, Hepper PG. Frequency discrimination by the fetus. Early Human Devel 1994;36:13-26
  59. Moon C, Cooper RP, Fifer WP. Two-day-olds prefer their native language. Infant Behav Devel 1991;16:495-500

© Scandinavian University Press 1996 ISSN 0803-5326


The Circumcision Information and Resource Pages are a not-for-profit educational resource and library. IntactiWiki hosts this website but is not responsible for the content of this site. CIRP makes documents available without charge, for informational purposes only. The contents of this site are not intended to replace the professional medical or legal advice of a licensed practitioner.

Top  © CIRP.org 1996-2024 | Filetree | Please visit our sponsor and host:  External link IntactiWiki.