Infants’ neural responses to emotional prosodies (Kao & Zhang, 2025)

Purpose: This study aimed to investigate infants’ neural responses to changes in emotional prosody in spoken words. The focus was on understanding developmental changes and potential sex differences, aspects that were not consistently observed in previous behavioral studies.

Method: A modified multifeature oddball paradigm was used with emotional deviants (angry, happy, and sad) presented against neutral prosody (standard) within varying spoken words during a single electroencephalography recording session. The reported data included 34 infants (18 males, 16 females; age range: 3–12 months, average age: 7 months 26 days).

Results: Infants exhibited distinct patterns of mismatch responses (MMRs) to different emotional prosodies in both early (100–200 ms) and late (300–500 ms) time windows following the speech onset. While both happy and angry prosodies elicited more negative early MMRs than the sad prosody across all infants, older infants showed more negative early MMRs than their younger counterparts. The distinction between early MMRs to angry and sad prosodies was more pronounced in younger infants. In the late time window, angry prosody elicited a more negative late MMR than the sad prosody, with younger infants showing more distinct late MMRs to sad and angry prosodies compared to older infants. Additionally, a sex effect was observed as male infants displayed more negative early MMRs compared to females.

Conclusions: These findings demonstrate the feasibility of the modified multifeature oddball protocol in studying neural sensitivities to emotional speech in infancy. The observed age and sex effects on infants’ auditory neural responses to vocal emotions underscore the need for further research to distinguish between acoustic and emotional processing and to understand their roles in early socioemotional and language development.

Supplemental Material S1. Additional tables showing the summary of the linear mixed-effect model using the amplitudes of early MMR as the dependent variable (Table S1) and late MMR as the dependent variable (Table S2); additional acoustic properties of each emotional prosody (Table S3); and fundamental frequency contours and intensity contours across the four emotional prosodies (angry, happy, neutral, and sad) using the words “Keep” and “Make” as examples (Table S4). Additional figures including violin plots that show the distributions of the acoustic measures (fundamental frequency, word duration, intensity variation, harmonic-to-noise ratio, and spectral centroid) in each emotion (Figure S1) and event-related potentials (ERPs) to the neutral voice in odd-number trials and even-number trials (Figure S2).

Kao, C., & Zhang, Y. (2025). Age and sex differences in infants’ neural sensitivity to emotional prosodies in spoken words: A multifeature oddball study. Journal of Speech, Language, and Hearing Research, 68(1), 332–348. https://doi.org/10.1044/2024_JSLHR-23-00820