Understanding how infants process speech in noisy environments is critical, as they frequently encounter complex communication settings from early infancy. This study measured cortical responses of typically hearing infants, aged 6 to 8 months, to varying levels of background noise using high-density diffuse optical tomography (HD-DOT).

We adapted nursery rhyme videos used in a well-established social behaviour paradigm to have different levels of intelligibility, for different stimuli. Infants watched audio-visual videos of women singing nursery rhymes with hand gestures, presented in three different competing babble noise (3 male and 3 female speakers) conditions: no background noise, +8 dB signal-to-noise ratio (SNR), and +4 dB SNR. Our intention was to create an objective psychometric performance function of speech engagement/perception as a function of SNR.

In the no background noise condition (high accessibility of speech), we observed significant bilateral activation in the Inferior Parietal Lobule (IPL), a region crucial for integrating audio-visual social cues. Additionally, the right dorsolateral prefrontal cortex (DLPFC) showed significant activation, suggesting that higher-order cognitive processing is engaged when speech cues are clear and undisturbed. Conversely, in the +8 dB SNR (medium speech access) and +4 dB SNR (low speech access) conditions, DLPFC activation was absent, indicating that this region’s engagement may depend on the clarity of auditory signals.

The IPL continued to show bilateral responses across all noise levels, although the strength of the hemodynamic response was inversely related to the amount of background noise, suggesting that while the IPL is sensitive to auditory clarity, it can still respond to lower speech clarity input. Furthermore, the high-accessibility-of speech condition elicited significant activation in the left superior temporal gyrus (STG) and middle temporal gyrus (MTG), regions associated with language processing. This activation diminished with increased background noise, suggesting that these areas are likely more dependent on clear auditory input for optimal engagement.

The robust HD-DOT cortical responses highlight its potential as a tool for psychometric assessments in early childhood. Unlike EEG, HD-DOT is not susceptible to electrical interference, making it particularly suitable for evaluating infants using hearing aids or cochlear implants who are unable to provide reliable feedback. These findings pave the way for HD-DOT to be used in early audiological evaluations, offering insights into infants’ auditory processing capabilities and aiding in the development of interventions during a critical period of speech and language acquisition and social development.