Subject: Re: Discussion about "place pitch" concept From: Colette McKAY <colette(at)UNIMELB.EDU.AU> Date: Fri, 19 Mar 2004 10:51:30 +1100Uwe Baumann wrote: * d) Concerning autocorelation based models of pitch perception: * "With electrical stimulation, identical temporal information can be provided to different channels by independently stimulating basal and apical electrodes. The summed autocorrelation function should be identical between basal and apical stimulation, predicting an independence of rate pitch on the place of stimulation. This prediction is clearly inconsistent with the observed joint dependence * of pitch perception on rate and place [by pulsatile electrical stimulation]." (Zeng, 2002) * However, there are many studies which have consistently demonstrated the independence of percepts related to rate of stimulation and place of stimulation. The earliest of these was Tong et al. (Tong, Y. C., P. J. Blamey, et al. (1983). "Psychophysical studies evaluating the feasibility of a speech processing strategy for a multiple-channel cochlear implant." Journal of the Acoustical Society of America 74(1): 73-80.) who used a multidimensional scaling task and showed a 2-dimensional stimulus space relating to rate and electrode position. A more recent study was by McKay, C., H. McDermott, & Carlyon. (2000). "Place and temporal cues in pitch perception: are they truly independent?" Acoustic Research Letters Online 1(1): 25-30. We showed that subjects could use independent information about rate and place changes in a difference-limen task. Combined changes which were inconsistent (e.g. increasing rate and apicalward place shift) were equally discriminable to changes which were consistent and this was inconsistent with the idea that such information is combined into a single 'pitch' percept or. The latter study is inconsistent with studies such as Zeng (2002) in which the both place and rate seem to influence the 'pitch' estimate of the subject. My personal view on the reason for this inconsistency is that, when subjects are asked to rate the 'pitch' of sounds which differ both in pitch and timbre (as in the Zeng experiment) they are forced to make decisions based on two different percepts at once, and thus are forced to respond based on some weighted average of the two percepts. Also, an implantee (or even a normal hearing person) is likely to rank timbre as a pitch difference if played two sounds differing only in timbre. Experiments like this one tend to suggest that you can adjust the rate on one electrode to make an equal pitch to stimulation on another electrode at a different rate. However, if these two stimuli were directly compared by a subject they would immediately tell you they are very different sounds, with different pitches, and they would choose one of the place or rate percepts to tell you which one had the higher pitch. Colette McKay *********************************************************************** A/Prof Colette McKay NHMRC Principal Research Fellow Department of Otolaryngology, The University of Melbourne 384-388 Albert St. East Melbourne, Australia Phone +613 9667 7506 Fax +613 9667 7518 email: colette(at)unimelb.edu.au **********************************************************************