Detecting Target Words While Monitoring Multiple Auditory Inputs
Abstract
This thesis examines how can humans best detect an auditory input while monitoring several inputs simultaneously. Two separate experiments were conducted, using a divided attention paradigm, to determine what factors influence target word detectability. Results from both experiments show an advantage in target detection if a person listens to one input in one ear and the other in the other ear (stereo) versus listening to both inputs in both ears (mono). Target detection was unaffected by variations in presentation rate in the range of 0.5 to 1.5 seconds. In the second experiment number of voices was examined as a factor. When both inputs were presented to each ear (mono) there was a clear advantage when listening to inputs that were recorded using two different voices (female and male) versus using only one voice (male only). However, the addition of a second voice did not improve target detection in the stereo condition. This latter finding may represent a limit on the effects of channel separation in target detection situations. Both a recency and primacy effect in terms of target detection as a function of the targets serial position in the list was found in both experiments. Results of these experiments have direct implications for practical applications, such as communication systems used by airline pilots. Keywords: Auditory perception; Psychoacoustics; Information processing; Speech; Word recognition; Theses.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1990
- Accession Number
- ADA224687
Entities
People
- John M. Barker Jr.
Organizations
- Air Force Institute of Technology