Guide 2 · The evidence
What science says, and doesn't
5 min read · Updated 2026-10-12
People sometimes say that 'science has proved garbh sanskar'. That is not quite true — and it doesn't need to be. What research has shown is fascinating in its own right: babies hear in the womb, remember what they hear, and arrive already tuned to their mother's voice. It also shows that a mother's wellbeing matters. Here is what the studies actually say, and where tradition goes further than the evidence can follow.
When babies begin to hear
The structures of the ear form early, but the cochlea and the nerve pathways that carry sound to the brain become functional around the middle of pregnancy. In a well-known study, Hepper and Shahidullah (1994) played sounds to babies in the womb and recorded their movements on ultrasound; the earliest responses appeared at about 19 weeks, starting with low-pitched sounds, with the range of frequencies widening over the following weeks.
So a fair summary is that babies start to respond to sound somewhere between about 18 and 24 weeks, and hearing keeps maturing until birth. The womb is not silent: the baby hears the mother's heartbeat, her digestion, and above all her voice, which travels both through the air and through her own body. Outside voices and music reach the baby too, but more muffled — lower tones come through better than high ones.
Newborns prefer their mother's voice
In 1980, Anthony DeCasper and William Fifer published a short paper in the journal Science titled 'Of human bonding: newborns prefer their mothers' voices'. They gave babies only a day or two old a special dummy: sucking in one pattern played a recording of their own mother reading, and sucking in another played a different woman's voice. The newborns changed how they sucked in order to hear their mother. They had had almost no time to learn her voice after birth — the familiarity came from the womb.
Later work by Barbara Kisilevsky and colleagues (2003) looked at babies before birth. At around 38 weeks, the babies' heart rates rose when they heard a recording of their own mother reading a poem and fell when they heard a stranger reading the same poem. Recognition of the mother's voice, in other words, is already present before the baby is born.
The 'Cat in the Hat' study: DeCasper & Spence, 1986
The study most often linked with garbh sanskar is DeCasper and Spence (1986), published in Infant Behavior and Development. Pregnant women read one particular children's story aloud — passages including Dr Seuss's The Cat in the Hat — twice a day during roughly the last six weeks of pregnancy. After birth, the newborns were tested with the same dummy method. They sucked in the way that brought them the familiar story rather than an unfamiliar one, and this held even when the story was read by a different woman.
The finding is remarkable: babies seemed to have learned something about the rhythm and sound pattern of a specific story before birth. It is worth being careful about what it does not show. It does not show that the babies understood the story, that they learned its meaning, or that hearing stories before birth makes children cleverer later. It shows familiarity — that sound heard often in the womb is recognised after birth. Other studies have found similar effects for music (for example, Partanen and colleagues, 2013) and for the vowel sounds of the mother's language (Moon, Lagercrantz and Kuhl, 2013).
Taste and smell, too
Hearing is not the only sense at work. Mennella, Jagnow and Beauchamp (2001) found that babies whose mothers drank carrot juice during late pregnancy were more accepting of carrot-flavoured cereal months later. Flavours from the mother's diet reach the amniotic fluid, so a varied, familiar home diet may gently introduce your baby to the tastes of your kitchen.
Maternal stress and wellbeing
A large body of research links high, ongoing stress, anxiety and depression in pregnancy with outcomes such as preterm birth and lower birth weight, and with some differences in children's emotional development. For example, O'Connor and colleagues (2002), using data from the large ALSPAC study in England, found that high maternal anxiety late in pregnancy was associated with more behavioural and emotional difficulties in children at four years. Reviews by Vivette Glover (2014) and by Christine Dunkel Schetter (2011) summarise this area.
Three important cautions. First, these are associations across large groups; they do not mean that a hard day, a cry or a worry will harm your baby. Ordinary stress is part of life. Second, many factors — support, health, money, relationships — shape both stress and outcomes. Third, the most helpful response to serious anxiety or low mood is real support, including from your doctor, not more pressure to 'stay positive'.
Studies of calming practices in pregnancy — mindfulness, relaxation, yoga, music — generally find modest improvements in how mothers feel, such as lower reported stress or anxiety. That is a sensible, evidence-friendly reason for a daily calm ritual: it is good for you.
Where tradition goes beyond the evidence
Some popular garbh sanskar claims have no scientific support, and it is kinder to say so plainly:
- That prenatal practices raise a child's IQ or guarantee a particular personality, talent or character.
- That a baby can learn the meaning of stories, scriptures or skills in the womb — familiarity with sound is not understanding.
- That rituals, foods or mantras can choose the baby's sex or skin colour. Practices aimed at this are not only unsupported; sex selection is illegal in India.
- That a mother's 'negative thoughts' will damage her baby. This idea causes needless guilt; feelings are normal and help is available.
- That any practice can replace antenatal check-ups, scans, supplements or medical treatment.
So what is a reasonable way to think about it?
The evidence supports a gentle version of garbh sanskar very well. Your baby hears you, learns your voice, and may come to know the stories and songs you repeat. Your wellbeing matters, and calm routines can help you feel better. Talking, reading and singing to your bump are free, safe and lovely ways for parents to begin bonding.
The stories of Abhimanyu and Prahlad are best enjoyed as what they are: treasured stories that carry a true intuition — that a baby is already listening — wrapped in the wonder of legend. You can hold both the science and the tradition at once.
Sources mentioned
- DeCasper, A. J. & Fifer, W. P. (1980). Of human bonding: newborns prefer their mothers' voices. Science.
- DeCasper, A. J. & Spence, M. J. (1986). Prenatal maternal speech influences newborns' perception of speech sounds. Infant Behavior and Development.
- Hepper, P. G. & Shahidullah, B. S. (1994). Development of fetal hearing. Archives of Disease in Childhood.
- Kisilevsky, B. S. et al. (2003). Effects of experience on fetal voice recognition. Psychological Science.
- Partanen, E. et al. (2013). Prenatal music exposure induces long-term neural effects. PLOS ONE.
- Moon, C., Lagercrantz, H. & Kuhl, P. K. (2013). Language experienced in utero affects vowel perception after birth. Acta Paediatrica.
- Mennella, J. A., Jagnow, C. P. & Beauchamp, G. K. (2001). Prenatal and postnatal flavor learning by human infants. Pediatrics.
- O'Connor, T. G. et al. (2002). Maternal antenatal anxiety and children's behavioural/emotional problems at 4 years. British Journal of Psychiatry.
- Dunkel Schetter, C. (2011). Psychological science on pregnancy. Annual Review of Psychology.
- Glover, V. (2014). Maternal depression, anxiety and stress during pregnancy and child outcome. Best Practice & Research Clinical Obstetrics & Gynaecology.
Frequently asked questions
When can my baby hear my voice?
Babies begin to respond to sound at roughly 18–24 weeks (Hepper & Shahidullah, 1994), and hearing keeps maturing until birth. Your own voice reaches them most clearly.
What did the Cat in the Hat study find?
DeCasper & Spence (1986) found that newborns preferred a story their mothers had read aloud during the last weeks of pregnancy over an unfamiliar story — evidence of learning sound patterns before birth, not of understanding.
Does stress in pregnancy harm the baby?
Ordinary everyday stress is normal. Research links severe or long-lasting stress, anxiety and depression with some harder outcomes, which is why support matters. If you are struggling, talk to your doctor.