Binaural loudness summation for directional sounds

In an earlier study (1), the loudness of directional sounds presented in an anechoic chamber was investigated by having listeners make loudness matches to a frontal reference source. The listeners' individual head-related transfer functions (HRTFs) were also measured, in an attempt to model the binaural loudness summation of the at-ear signals. Even though the effects of HRTFs were taken into account, considerable individual differences in the binaural summation of loudness remained. In order to create conditions in which the directional at-ear changes were identical for all participants, the present experiment employed 'generic' HRTFs to create directional sounds via binaural synthesis. When inspecting the results of the listening tests, however, large individual differences were still evident, as in the earlier study. The generality of this finding was further corroborated by running an independent, inexperienced sample of ten participants exclusively being exposed to the present generic HRTFs. Despite the individual differences, the average results suggest a relatively simple rule for combining the binaural input when carrying out acoustical measurements using an artificial head. That is particularly useful when attempting to predict loudness for dichotic listening situations, as they typically result from real sound sources positioned in space.