Robert A. Lutfi
Eunmi Oh
Dept. of Commun. Disord. and Psychol. Dept., Univ. of Wisconsin, Madison, WI 53706
A method is described for measuring the strategies listeners employ in the determination of simple resonant sources. To demonstrate this method the principles of theoretical acoustics were applied to reconstruct the sound-pressure waveform at the ear as would be generated by a common tuning fork. The result is an inharmonic sum of damped sinusoids whose acoustic parameters covary in lawful relation to the physical dynamics of the source. Listeners were asked to discriminate these tuning fork sounds from a set of nontuning fork sounds for which the lawful covariation among parameters was randomly perturbed from trial to trial. The lawful covariation had to be detected to perform significantly above chance. Finally, individual correlations were computed between the listener's response and the values of the acoustic parameters across trials to determine precisely which aspects of lawful covariation listeners use to identify the source. Discrimination performances was found to be quite good for a tuning fork whose physical dimensions remained constant across trials (driving force varies), though it was exceedingly poor when just one physical dimension was allowed to vary from trial to trial. Individual correlations revealed that the covariation exerting the greatest influence often involved the one spectral component having the most acoustic energy in the signal. [Research supported by NIDCD.]