Binaural loudness summation for speech and tones presented via earphones and loudspeakers

Abstract

Objectives: Literature reviews of binaural loudness summation assume nearly perfect summation (i.e., a tone presented binaurally is assumed to be twice as loud as the same tone presented monaurally). However, some recent data and classroom demonstrations of this phenomenon using speech stimuli from a familiar visually present talker yield much less summation. Therefore, the following two hypotheses were tested using a preliminary procedure that controlled some, but not all, variables. First, the amount of binaural loudness summation is less for speech from a visually present talker than for recorded speech or tones. Second, the amount of binaural loudness summation is less when sounds are presented via loudspeakers than when sounds are presented via earphones.

Design: Three types of stimuli (monitored live-voice [MLV] spondees, recorded spondees, and tones) were presented monaurally and binaurally across a wide range of levels. The same stimuli were presented via earphones and loudspeakers in an audiometric test booth. Eight young listeners with normal hearing judged the loudness of the stimuli using magnitude estimation.

Results: The amount of binaural loudness summation was significantly less for MLV spondees than for tones or recorded spondees. Binaural loudness summation was also significantly less for loudspeaker presentation than for earphone presentation. Binaural loudness summation was found to be less than perfect (i.e., a sound presented binaurally is less than twice as loud as the same sound presented monaurally) for all conditions. The amount of binaural loudness summation was the least for MLV spondees presented via loudspeakers.

Conclusions: The present results support both hypotheses and indicate that binaural loudness summation in the loudspeaker conditions is significantly less than binaural loudness summation in typical laboratory test conditions using earphones. There may be a subjective effect resulting from expectations about loudness of a familiar, visually present talker, termed here as "Binaural Loudness Constancy."