The problem
At a busy lunch, several voices reach your ears together, yet you can try to follow one person. Following that person as they move is different from listening to whoever speaks on your left. How do sound cues and your listening goal separate voices into streams and keep the chosen voice distinct?[1] [2]
A listener can follow a voice within a mixture even as the speaker’s position changes. How do cues in the sounds and the listener’s goal jointly form and maintain that stream? In a 2026 controlled scene, listeners detected pitch changes while attending to a speaker, a location or the whole mixture. Following the speaker performed better in that task, and electrical responses recorded at the scalp (EEG) differed between modes. Features that made a voice stand out helped its initial separation more than its subsequent tracking after attention was directed to it.[1]
Scope
Human auditory stream formation and selection: the interaction of source-based attention, acoustic-feature attention and bottom-up salience.
What remains open
Attention to a person and attention to a location behave differently. Their coordination is not fully explained.[1]
The 2026 speaker/location/global comparison finds distinct EEG and behavioral effects. Bottom-up salience helps initial segregation more than subsequent attended tracking; the integrated circuitry remains incomplete.[1]
Why it matters
Following a conversation involves recovering information from a particular source, rather than merely registering the mixture. The speaker-versus-location comparison shows why the unit selected by attention matters when the scene changes. Laboratory pitch detection identifies part of that problem without explaining every demand of ordinary conversation.[1] [2]
Definitions
2 terms
auditory stream
Sounds grouped by perception into a continuing source or sequence. Hearing a stream as separate and choosing to follow it are related but distinguishable processes.[1] [2]
bottom-up salience
How strongly a sound stands out because of its acoustic features. Its role in forming a stream can differ from its role in tracking an already attended source.[1]
Sources
- Perception of dynamic multi-speaker auditory scenes under different modes of attention
Notes on Perception of dynamic multi-speaker auditory scenes under different modes of attention
Controlled moving multi-speaker study distinguishes following a source from attending to a location or mixture. Behavior and EEG identify task-specific differences and salience effects on initial grouping versus tracking, without explaining every demand of ordinary conversation.
- Development of auditory scene analysis: a mini-review
Notes on Development of auditory scene analysis: a mini-review
Review distinguishes forming sound streams from selecting one and discusses the limits of simplified developmental paradigms. It supplies background for the everyday listening illustration, not a complete mechanism or universal developmental schedule.