Neural discrimination of emotional prosody in natural sentences in 2- to 3-month-old infants using OPM-MEG
- 1University of Birmingham
Abstract
From birth, infants are sensitive to the emotional prosody of speech, yet it remains unclear when the developing brain discriminates emotions conveyed by natural spoken sentences. Here, we recorded brain activity with optically pumped magnetometer magnetoencephalography (OPM-MEG) in 15 infants aged 2-3 months while they listened to semantically neutral sentences spoken with happy or angry prosody. Natural sentences elicited robust auditory evoked responses, with prominent components at 88, 216, 380 and 1,356 ms after sentence onset and sources centred on bilateral superior and middle temporal cortex. Time-resolved multivariate pattern analysis revealed that distributed neural activity discriminated recordings spoken with happy from angry prosody 346-454 ms after the onset of the prosodically prominent word (the first word carrying a pitch peak), despite no difference in global field power between conditions. This discrimination was transient, emerged before the end of the sentence, and was carried primarily by phase-locked delta-band activity, followed by later differences in induced alpha power. Angry prosody also delayed the late auditory response by approximately 15 ms with no detectable changes in amplitude. Our findings show that cortical responses to happy and angry prosody in natural speech differ by 2-3 months of age, and demonstrate that OPM-MEG combined with time-resolved decoding can reveal the temporal, spatial and spectral organisation of early emotional-prosody processing.
Related
- Aging and Hearing Loss in Emotional Prosody Processing: Divergent Neural Adaptation and Behavioral Performance
- An exploratory fNIRS study on directional asymmetry in segmental discrimination: Lateral–rhotic (/la–ra/) and labial–dorsal (/ba–ga/) contrasts tested in 5- and 9-month-old Japanese infants
- Edge structure influences infants' and toddlers' engagement with natural scenes.
- Spoken language comprehension relies on context-dependent neural weighting of syntax and prosody
- Affective prosody and facial expression cues differentially augment spoken sentence and discourse comprehension for acute left and right hemisphere stroke survivors
BibTeX
@misc{yang2026neural,
title = {{Neural discrimination of emotional prosody in natural sentences in 2- to 3-month-old infants using OPM-MEG}},
author = {J. Yang and V. M. Leija and A. Waitt and A. Kowalczyk and A. Bremner and B. Pomiechowska and G. Orioli and O. Jensen and K.-m. An},
howpublished = {bioRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.03.763724},
url = {https://www.biorxiv.org/content/10.64898/2026.11.03.763724v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Neural discrimination of emotional prosody in natural sentences in 2- to 3-month-old infants using OPM-MEG
AU - J. Yang
AU - V. M. Leija
AU - A. Waitt
AU - A. Kowalczyk
AU - A. Bremner
AU - B. Pomiechowska
AU - G. Orioli
AU - O. Jensen
AU - K.-m. An
PY - 2026
DA - 2026/11/03
PB - bioRxiv
DO - 10.64898/2026.11.03.763724
UR - https://www.biorxiv.org/content/10.64898/2026.11.03.763724v1
AB - From birth, infants are sensitive to the emotional prosody of speech, yet it remains unclear when the developing brain discriminates emotions conveyed by natural spoken sentences. Here, we recorded brain activity with optically pumped magnetometer magnetoencephalography (OPM-MEG) in 15 infants aged 2-3 months while they listened to semantically neutral sentences spoken with happy or angry prosody. Natural sentences elicited robust auditory evoked responses, with prominent components at 88, 216, 380 and 1,356 ms after sentence onset and sources centred on bilateral superior and middle temporal cortex. Time-resolved multivariate pattern analysis revealed that distributed neural activity discriminated recordings spoken with happy from angry prosody 346-454 ms after the onset of the prosodically prominent word (the first word carrying a pitch peak), despite no difference in global field power between conditions. This discrimination was transient, emerged before the end of the sentence, and was carried primarily by phase-locked delta-band activity, followed by later differences in induced alpha power. Angry prosody also delayed the late auditory response by approximately 15 ms with no detectable changes in amplitude. Our findings show that cortical responses to happy and angry prosody in natural speech differ by 2-3 months of age, and demonstrate that OPM-MEG combined with time-resolved decoding can reveal the temporal, spatial and spectral organisation of early emotional-prosody processing.
ER -