Evidence for temporal-coherence-based segregation of complex auditory scenes in the newborn human brain

Detecting a target sound within a mixture of sounds (referred to as auditory stream segregation) is crucial for perception in natural environments, a skill humans and animals excel at. This study investigates the role of temporal coherence in auditory stream segregation in human newborns using high-density EEG recordings. Sleeping newborns were exposed to temporally coherent auditory tone sequences embedded in random background tones, and their event-related responses were analyzed. The results indicate that newborns can segregate auditory streams based on temporal coherence, suggesting that this stream segregation is driven by automatic mechanisms from birth, as evidenced by brain responses resembling the object-related negativity (ORN) event-related potential (ERP) component. However, discriminating among different signal-to-noise ratios requires further fine-tuning, as evidenced by delayed latencies in neonates compared to adults. These findings indicate that temporal coherence aids in detecting and orienting toward salient stimuli, thereby laying the foundation for the development of abilities such as selective attention and speech perception.