ASA 127th Meeting M.I.T. 1994 June 6-10

4pPP20. Some evidence that two separate mechanisms encode the F0's of resolved and unresolved harmonics.

Robert P. Carlyon

Trevor M. Shackleton

Exp. Psychol., Univ. of Sussex, Brighton BN1 9QG, England

Sensitivity (d') was measured to differences in fundamental frequency ((Delta)F0's) between two groups of frequency modulated harmonics, presented simultaneously, where each group was filtered into either a LOW (125--625 Hz), MID (1375--1875 Hz), or HIGH (3900--5400 Hz) frequency region. The first set of experiments measured sensitivity to (Delta)F0's between either the LOW and MID groups or the MID and HIGH groups, for baseline F0's of 88 and 250 Hz, and also measured sensitivity to (Delta)F0's between sequentially presented tokens of each individual group. Sensitivity to simultaneous (Delta)F0's was best when the combination of frequency region and F0 caused the two groups to be either both resolved or both unresolved by the peripheral auditory system, and worst when the groups different in ``resolvability.'' Additional experiments showed that the high sensitivity to (Delta)F0's between two groups of unresolved harmonics could be attributed to the pitch pulse asynchronies that invariably arise from a (Delta)F0. A model based on optimum combination of information was applied to the results of the first set of experiments, and succeeded in predicting sensitivity to simultaneous (Delta)F0's between two groups of resolved components from that to successive (Delta)F0's on the individual groups. However, the model overestimated performance when one group was resolved and the other unresolved. The results suggest that the F0's of resolved and unresolved harmonics are processed by separate mechanisms, in contrast to the predictions of ``autocorrelation'' models of F0 encoding.