Hi Rob, As you correctly point, in gestation the external canal and middle ear are not air-filled in the way we usually assume for postnatal hearing, so the classic air-conduction middle-ear transformer is not really operating in its standard form. That’s exactly why it is a bit misleading to talk as if the fetus “hears” in the same mechanical sense as a term newborn or adult. In practice, I would say that when we try to estimate fetal “hearing” in this context, I understand from the relevant literature that we usually don’t build an explicit “fluid-on-both-sides” middle-ear model. Instead we either measure the intrauterine vibroacoustic field directly with hydrophones near the fetus (e.g., Benzaquen et al., 1990; Richards et al., 1992), or model transmission from the external environment into the uterus using anatomical/finite-element models (e.g., Gélat et al., 2025). A review by Gerhardt and Abrams (2000) summarizes quite well how these measurements translate into fetal exposures. Therefore, in my understanding, functional cochlear activation should be treated as being driven mainly by bone- and tissue-conducted, strongly low-pass filtered vibroacoustic energy (dispersed in the whole of the fetus body), with the unusual mechanics of the fetal ear (including the fluid filling) effectively folded into those empirical transfer functions and into the elevated behavioral/physiological thresholds relative to postnatal air-conducted hearing. So when we talk about “hearing in the womb,” it is really shorthand for auditory pathway activation under these nonstandard mechanical conditions, rather than the usual air-conduction case. I hope I somewhat answered your answer. All the best,
Efthymios Refs: Benzaquen, S., Gagnon, R., Hunse, C., & Foreman, J. (1990). The intrauterine sound environment of the human fetus during labor. American Journal of Obstetrics and Gynecology, 163(2), 484–490. Richards, D. S., Frentzen, B., Gerhardt, K. J., McCann, M. E., & Abrams, R. M. (1992). Sound levels in the human uterus. Obstetrics & Gynecology, 80(2), 186–190. Gélat, P., Van’t Wout, E., Haqshenas, R., Melbourne, A., David, A. L., Mufti, N., Henriques, J., Thibaut de Maisières, A., & Jauniaux, E. (2025). Evaluation of fetal exposure to environmental noise using a computer-generated model. Nature Communications, 16, 3916. Gerhardt, K. J., & Abrams, R. M. (2000). Fetal exposures to sound and vibroacoustic stimulation. Journal of Perinatology, 20(8, Suppl. 2), S21–S30 On 6 Dec 2025, at 05:12, Maher, Rob <rmaher@xxxxxxxxxxx> wrote:
Sorry for a completely novice question, but I have been curious about fetal auditory function: presumably, the ear canal and middle ear are full of amniotic fluid, so any sound perception would need to be modeled with fluid on both sides of the eardrum… How is this effect taken into account in estimating “hearing” in the womb? Curious, Rob Maher **External Sender** Dear Jonathan, Thanks for sending this around – it is a really stimulating idea. Personally speaking, I am not aware of any empirical work that has actually measured maternal footsteps - or any other rhythmic movement induced sound - in utero. The studies I know of on the intrauterine soundscape (I am not a specialist in this field as I work mostly with premature infants- so do not take my word for it) give global levels and spectra, but not event-specific waveforms, so I do not think we currently have data that would let you specify a “real” amplitude envelope for footsteps in a strong, evidence-based way. Conceptually, I would be inclined to treat gait-related input as slow, low-frequency modulations of an already noisy vibroacoustic field, tightly coupled to vestibular input, rather than as discrete, sharply defined acoustic impulses. In any given case SNR for any ‘“auditory” related source (even if assumed; considering also that the intrauterine acoustic neurosystemic function is not the same as ex-utero) will be too low considering the already messy sound environment in utero. I think that one way to make this useful in your modeling would be to show that your results are robust across a family of plausible envelopes: e.g., band-limited low-frequency “bursts” with different rise times, durations, and dynamic ranges, all occurring at gait rate and co-occurring with vestibular cycles. That lets you argue that what really matters is the regular, low-frequency temporal structure and multi-sensory coupling, not the precise acoustic fine structure of each step. I’d be very happy to discuss further or look at a draft/preprint if that would be useful. All best,
Efthymios Professor Efthymios (Timos) Papatzikis
PhD, PhD, FHEA, Cert.Neuro.
- Neuroscience and Medical Psychology, School of Health Sciences and Psychology, Canadian University Dubai, U.A.E.
- NICU Senior Research Fellow, ‘Panagiotis and Aglaia Kyriakou’ Children’s Hospital, Athens, Greece
- World Health Organization roster of consultants for Brain Health
- World Health Organization certified specialist for Social Prescribing
- Associate Editor - Frontiers in Pediatric Psychology Journal
I may write emails at unusual hours. Please don’t feel obligated to reply until your normal working time. Our personal time and moments with family are limited and precious; safeguarding them helps us bring our best to our work.
I've done some modeling arguing that the audio/vestibular pairing of maternal gait inside the womb could scaffold the development of beat perception and the urge to move to music. To respond to reviewers, I need to argue that maternal footsteps are indeed audible inside the womb and give some sense of what the amplitude envelope of the sound would look like. Does anyone have any thoughts about how to go beyond basic assumptions (I've been assuming audible, low-pass-filtered impulses) to get a better sense of what maternal footsteps sound like inside the womb? (Reply to me directly -- doesn't seem worth starting a long chain about this) Thanks, Jonathan
|