Absolute magnitude estimation of loudness by adults and children

Twenty-three subjects performed absolute magnitude estimation of the lengths of a series of lines and the loudness of a series of tones as well as cross-modality matching between loudness and perceived line length. The results support the hypothesis that subjects judge stimuli on an absolute scale. Specifically, for 9 out of 12 adults and 9 out of 11 children, lines and tones assigned the same number in magnitude estimation were judged to be subjectively equal in cross-modality matching. A correction procedure was employed to eliminate the effects of idiosyncratic number usage from the magnitude estimations of loudness. This correction procedure, consisting of dividing the loudness exponent by the line length exponent, produced power function exponents for loudness that were virtually identical for magnitude estimation and cross-modality matching. Implications of the results for clinical measurement of loudness are discussed.