Neural Mechanisms of Auditory Perception and Voice Processing

The ability to decode complex acoustic signals underpins communication across species. In mammals, sound information ascends from the cochlea through brainstem nuclei and the medial geniculate body of the thalamus before reaching primary auditory cortex in Heschl’s gyrus. Beyond these core regions, processing diverges into parallel ventral and dorsal streams facilitating sound identification and spatial localisation, respectively. Along the ventral pathway, specialised patches within the superior temporal gyrus and sulcus—known as temporal voice areas—exhibit heightened sensitivity to vocal signals, supporting speaker identification, emotion analysis and speech-invariant representations. Convergent inputs from temporal and occipital areas integrate auditory and visual cues, enabling robust face–voice recognition. Higher-order frontal and parietal areas modulate attention, working memory and decision processes during complex listening tasks. Recent advances in neuroimaging and electrophysiology have refined our understanding of these hierarchical and network interactions, offering insights into developmental trajectories, individual variability and potential targets for auditory prostheses or rehabilitative interventions.

Research from Nature Portfolio

Recent imaging in non-human primates using ultra-high-field fMRI has detailed a distributed network responsive to combined facial and vocal stimuli, revealing that multisensory integration within occipito-temporal and auditory ‘voice’ patches extends into parietal, prefrontal and cingulate cortices, suggesting adaptive mechanisms for social communication. Complementary high-resolution vocalisation studies in marmosets have identified a caudal-to-rostral cortical gradient, with anterior temporal regions near the temporal pole showing the strongest preference for conspecific sounds, mirroring patterns seen in macaques. In human participants, rigorous acoustic control experiments have demonstrated that temporal voice areas remain selectively tuned to natural human voices even when low-level spectral and temporal cues are matched, indicating genuine voice-specific processing beyond basic acoustics.

Neural Mechanisms of Auditory Perception and Voice Processing publication trend

The graph below shows the total number of articles in neural mechanisms of auditory perception and voice processing across all publications each year (not limited to Nature Index journals).

Technical terms

Temporal Voice Areas (TVAs): Bilateral regions along the superior temporal gyrus/sulcus specialising in vocal sound processing.

Functional connectivity: The statistical correlation between neural activity in distinct brain regions over time.

Superior temporal sulcus (STS): A cortical fissure in the temporal lobe implicated in processing socially relevant auditory and visual cues.

Ventral auditory stream: A pathway from primary auditory cortex involved in sound identification and semantic processing.

Multisensory integration: The neural combination of inputs from different sensory modalities to form coherent percepts.

References

1. Mapping of facial and vocal processing in common marmosets with ultra-high field fMRI. Communications Biology (2024).
2. High-field functional magnetic resonance imaging of vocalization processing in marmosets. Scientific Reports (2015).
3. Functional MRI of the vocalization-processing network in the macaque brain. Frontiers in Neuroscience (2015).
4. Voice selectivity in the temporal voice area despite matched low-level acoustic cues. Scientific Reports (2017).
5. The Temporal Voice Areas are not “just” Speech Areas. Frontiers in Neuroscience (2023).
6. Functional connectivity within the voice perception network and its behavioural relevance. NeuroImage (2018).
7. People-selectivity, audiovisual integration and heteromodality in the superior temporal sulcus. Cortex (2013).