Korva is a scroll of the internet’s best ideas.
Your brain briefly tracks two voices
The two-talk experiment
Neuroscientists from an international team fitted participants with EEG harnesses, which record neural activity, and played them two TED Talks at the same time amid background noise. The participants were tasked with swapping their attention between the two speakers while the harnesses tracked what their brains were doing.
The overlap in attention
The swap was never a clean cut. Their brains began engaging with the new speaker before they had fully disengaged from the old one, which left a brief period of overlap where attention was split between both. That overlap showed up as a unique signature on the EEG readings.
Who is better at it
The study, published in PLOS Biology, found that everyone has this ability to track multiple conversations at once, even in noisy surroundings, and that some people are better at it than others. Those people may naturally be better multitaskers, able to explore what is happening around them without immediately losing focus on their current conversation. That could help explain why some people seem especially good at navigating busy social environments, and why some seem more skilled at eavesdropping than others.
Why do we hear our own name across a room?
In the 1950s, the British scientist Colin Cherry coined the term cocktail party problem for how people follow one voice among many. Later work by Neville Moray, published in 1959, played messages to people through headphones and told them to ignore one ear. About a third of them still noticed their own name when it was spoken in the ignored ear. Familiar, personally meaningful words seem to break through even when attention is pointed elsewhere.
How does EEG pick out which voice someone follows?
Activity in the brain's auditory regions rises and falls in step with the loudness pattern of speech it is paying attention to. A 2012 study in Nature by Nima Mesgarani and Edward Chang recorded from the brain surface of patients and found that the attended speaker's voice could be reconstructed from the signals. Scalp EEG is far coarser, but researchers have used the same kind of tracking to tell which of two voices a listener was following. The voice being ignored is tracked much more weakly.
Why does switching attention take any time at all?
Experiments on task switching, such as those by Robert Rogers and Stephen Monsell in 1995, found that people respond more slowly right after changing tasks than when repeating one. The delay is called a switch cost. It appears to come partly from reconfiguring the brain's settings for the new task and partly from lingering activity left over from the old one. Switch costs show up even when people know in advance that a switch is coming.
Why does background noise make following speech harder?
Researchers distinguish two kinds of masking in a crowded room. Energetic masking happens when competing sounds occupy the same frequencies as the voice you want, so parts of that voice are physically drowned out. Informational masking happens when the competing sound is itself speech, because the brain has to separate meaningful words from meaningful words. Studies of speech in noise have found that a competing talker is usually more disruptive than steady noise of the same loudness.