Selective Attention Shapes Neural Representations of Complex Auditory Scenes: The Roles of Object Identity and Scene...

Area of Science:

• Neuroscience
• Auditory Perception
• Cognitive Psychology

Background:

• Everyday auditory environments are complex, with overlapping sounds requiring selective attention to isolate relevant auditory objects from background noise.
• The auditory cortex (AC) plays a crucial role in processing complex sound scenes, but how selective attention shapes neural representations within the AC is not fully understood.

Purpose of the Study:

• To investigate how selective attention influences neural representations of complex sound scenes in the auditory cortex.
• To examine whether attention modulates neural gain or affects scene decomposition and analysis based on context and object type.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to record brain activity in participants (N=20) listening to complex sound scenes.
• Scenes comprised three overlapping sounds, constructed either with objects from different categories (speech, animal, instrument) or from the same category.
• Participants attended to a designated object within each scene, while brain activity patterns were analyzed using spatial pattern analysis.

Main Results:

• Attending to speech sounds enhanced activity in lateral AC subfields, whereas attention to animal or instrument sounds modulated medial AC subfields.
• Spatial pattern analysis revealed that the attended object dominated AC activation patterns, with effects varying by object type and scene composition.
• In category-diverse scenes, attention effects were dominated by category-level processing; in category-similar scenes, attention shifted to exemplar-level processing.

Conclusions:

• Selective attention dynamically prioritizes features offering maximal contrast within a given auditory context.
• Top-down attention signals not only modulate neural gain but also influence auditory scene decomposition and stream segregation in a context-dependent manner.
• These findings support models where attention adaptively shapes auditory processing for effective scene analysis in complex environments.