Kevin Austin has started this thread with his 8/23 posting describing how it is possible to teach many of his listeners to hear out the note “D” in a 10-item chord by presenting the note in isolation as well as a component in the intact chord. He interpreted his observations as representing both a refinement of memory and an improvement of perceptual ability. He asked whether listeners would be able to do this with other sounds.
If Kevin means to ask if listeners can use memory to separate a sound into components, the answer is yes. For example, if listeners are presented with a 1/3-octave band of noise for a few seconds followed by broadband noise they can “hear” the narrow band noise continue as a component for tens of seconds. A linguistic example of the use of long-established memory to “hear out” sounds in noise is afforded by phonemic restoration in which listeners “perceive” missing phonemes or even entire syllables (e.g., the “gis” in legislatures) that have been replaced by noise. Our lab has studied the use of memory to “hear” contextually appropriate sounds by segregating portions of another sound using the rubric “auditory induction.”
“Without the senses there is no memory, and without memory there is no mind.”
-Voltaire in his short essay “Memory’s Adventure”