Description:
<jats:p>This study examined the importance of a priori knowledge about the location of a target in a multitalker environment. On each trial, three equal-level sentences from the Coordinate Response Measure test were presented from different loudspeakers separated by 60° azimuth. The sentences—target and 2 maskers—had the structure ‘‘Ready [callsign] go to [color] [number] now.’’ The task was to repeat the color and number associated with a specific callsign. The target location was chosen among the 3 loudspeakers on every trial. The main experimental variables were the probability of occurrence of the target at each speaker location and whether the listener was provided with the callsign of the target before or after stimulus presentation. Performance ranged from near perfect when the target location was completely certain to about 33% correct when the location was random and the callsign was not known in advance. When target location was random, performance improved from about 33% to 70% correct when the callsign was known in advance. Overall, these results support the view that knowing where to focus attention provides a great advantage in speech recognition in complex and uncertain environments. [Supported by NIH/NIDCD.]</jats:p>