Abstract:

Magnitude estimation and paired-comparison scaling methods were applied to the problem of measuring perceived auditory distance. Stimuli consisted of binaural recordings made in a reverberant hallway at distances of 1, 2, 4, 8, 16, 32, and 64 ft. A loudspeaker produced broadband noise burst, 100 ms in duration, served as the distal stimulus. Insert-style microphones were placed in the ears of a single individual, with measurements made from two directions, extending in distance either from the individual's front or left side. Construction of virtual sound source stimuli in this fashion insured that the stimulation reaching the listener was naturalistic and therefore possessed the acoustic cues thought to be important in the real-world perception of auditory distance. Five listeners were presented these stimuli in several variants of both the magnitude estimation and paired-comparison paradigms. Though relatively large individual variability exists, the resulting scales are all compressed when compared to the corresponding scale of physical distance. The amount of compression is shown to be approximately constant across scaling method, thereby providing validation for each method's measurement of a similar underlying psychological process. [Work supported by NASA.]