The mind can process only so much at once, so attention works like a bottleneck. Broadbent argued the filter sorts incoming sound by physical properties, pitch, volume, which ear it arrives in, and discards the rest before meaning is reached. Colin Cherry had already shown how much the physical cues matter. Then someone tested one particular word.
Your own name. In Moray’s nineteen fifty-nine experiment, participants repeated aloud a message played into one ear while their name was slipped into the other. They caught it in roughly a third of trials. On a filter that sorts by physical cues alone, that should be impossible, because recognizing a name means knowing what the sound means.
Selective attention is the business of directing awareness at what matters and tuning out the rest, and psychologists model the limit as a bottleneck restricting the flow of information. Broadbent’s puts an internal filter at that bottleneck, selecting stimuli by their physical properties and blocking the rest, which then decays quickly.
Colin Cherry began the systematic work in nineteen fifty-three, asking how a listener isolates one voice in a chaotic room and what becomes of the rest. Presenting two spoken messages at once, he found separation leaned heavily on physical differences: the speaker’s gender, the volume, the direction the sound came from. Strip those differences out, the same voice at the same volume in both ears, and listeners could not separate the messages by meaning at all. In tasks they noticed a male voice turning female in the ignored ear. They missed a change of language, and they missed speech played backwards.
Moray’s nineteen fifty-nine experiment broke the pattern. Participants a message in one ear while their own name was inserted into the unattended message in the other, and the name was detected in about a third of trials. Something had to be analyzing the ignored channel for meaning, which is what the phrase cocktail party phenomenon now names: holding one conversation while still catching a significant word from the noise.
Two further results pushed the same way. Gray and Wedderburn played Dear, seven, Jane into the right ear and nine, Aunt, six into the left, and participants reported hearing Dear Aunt Jane, jumping between channels to follow the sense. And Von Wright, Anderson and Stenman conditioned words with mild shocks, then played them unattended, and measured a rise in to those words and their synonyms. Semantic analysis was happening below awareness, so the bottleneck was never absolute.
wat. You are not listening to the other side of the room. Something in you is, just enough to catch your name.
wat wat
It forced the filter to be leaky. Something must analyze the channel you are not attending to, at least far enough to spot what matters to you.
How do we know?
A cognitive psychology reference article on theories of selective attention, which sets out Cherry’s shadowing and dichotic listening experiments, Moray’s own-name result, and the model revisions that followed.
What is still unknown
A teaching reference rather than the primary reports, and the detection rate is an average across trials. The article itself notes methodological criticisms of this literature.
Learn something?
Where to next
Does treating hearing loss slow cognitive decline?hearing
Why is it so hard to say what color a word is printed in?attention
Would you notice if the person you were talking to was swapped for someone else?attention