Published in PLOS Biology on July 16, 2026, a study from Trinity College Dublin used scalp electroencephalography (EEG) to track what happens inside the brain during a common but poorly understood process called attention switching — the moment you stop listening to one person and start listening to another.
Twenty-four normal-hearing adults sat in a laboratory designed to simulate a busy, multi-talker environment. Over loudspeakers, they heard two competing speech streams mixed with background babble. Every 15 to 30 seconds, a visual cue told them to switch their attention from one speaker to the other. High-resolution computational parsing of the EEG data allowed the researchers to isolate the neural signature of each speech stream.
The key finding: during the transition period, the brain's cortical activity showed simultaneous tracking of both speakers. Neural oscillations — the rhythmic electrical patterns that reflect how the brain locks onto speech — encoded the voice the participant was switching to as well as the voice they were switching away from. The overlap lasted only a few hundred milliseconds, but it was reproducible and measurable across all participants.
This challenges the classic "spotlight" model of attention, which assumes that shifting focus is a sequential process of disengaging from one stream and then engaging with another. Instead, the brain appears to briefly hold two streams in parallel, creating a temporary neural bridge that may explain how you can follow a conversation even when interrupted, or pick up a thread after someone else has been talking for a few seconds.
Knowledge takeaway: EEG recordings during multi-talker experiments show the human brain briefly encodes two competing speech streams at the same time during attention switches — challenging the sequential spotlight model of auditory attention and revealing a neural mechanism for keeping up with fast-paced conversations.