van den Brink, G. (1965). Pitch shift of the residue by masking. Intern. Audiol. 4, 183-186
The results reported here are part of the results of a study on pitch perception and diplacusis.
In a study on diplacusis the most interesting subjects are those who have an unilateral hearing loss. This kind of subjects however is scarce. Therefore we looked for a method to cause temporal unilateral hearing loss. This was realized by periodical stimulation of one ear with loud sine wave tone pulses. Diplacusis was measured during the intervals between the presentation of these loud tone pulses.
Two remarks should be made here: firstly that this is not really masking but rather fatiguing or, what some people call satiation, and secondly that the results can not be supposed representative for what happens in the case of real hearing loss. Nevertheless we got some results, which we think may be interesting.
A survey of the sequence of presentation of the masking and test tones is given in Fig.1. Loud tone pulses (in the present case 500 cps, 110 dB) were presented during 1.6 sec with intervals of 1.6 sec to the left ear. During the intervals test and reference tones were presented alternatively in the left and the right ear, respectively. The duration of the tone pulses was 0.4 sec, so that during each interval two pairs of tone pulses were given. The intensities of test and reference tones were 50 dB in the experiment described here. Diplacusis was determined by adjusting the frequency of the reference tone (presented in the right ear) in order to match the pitch of the reference tone with that of the test tone, as a function of the frequency of the test tone. Because the pitch of the test tone changes very rapidly immediately after the masking tone ceases, diplacusis of the second pair of tone pulses is determined, as indicatedinf fig.1.
In fig.2 diplacusis is plotted as the relative frequency difference between test and reference tone, expressed in percents. A positive diplacusis means that the frequency of the reference tone (right ear) was higher than that of the test tone (left ear) for equal pitch. This means that a tone is perceived higher in the test ear if we present tone pulses with equal frequencies in both ears. The upper curve in fig.2 was measured without masking (masking level zero) and represents normal diplacusis as a function of the frequency of the test tone. It shows an irregular pattern which appears to be different for different subjects. The lower curve in fig.2 shows diplacusis of the same subject, with a masking tone (500 cps, 110 dB) presented. The pattern looks more regular (notice the minima at multiples of 500 cps), and the frequency differences are much larger. For other frequencies of the masking tone andfor several other subjects this is not always as clear as it is in this case, however, it is usually possible to show consistent regularities in these cases. The next step was to determine the influence of a masking or fatiguing tone on the pitch of the residue. The regularities of the masked diplacusis pattern provide an easy way to do that with an amplitude modulated signal. The carrier and modulation frequencies were chosen in such a way that the carrier and side band frequencies coincided with the frequencies of either three maxima or three minima of the diplacusis pattern as accurately as possible, the carrier frequency being exactly a multiple of the modulation frequency. If, for instance, three maxima occur at 1600, 2000, and 2400 cps, a residue was composed with a carrier frequency of 2000 cps and a modulation frequency of 400 cps. It was possible to make such combinations of carrier and modulation frequencies for all succeeding trios of maxima and minima, respectively, between about 700 cps and 3500 cps in the masked diplacusis pattern, but not for all of them in the unmasked diplacusis pattern, because of the fact that this pattern was less regular.
The pitch of the reference tone was matched with the pitch of the residue. In fig.2 the crosses represent the difference between reference and modulation frequency, plotted as a function of the carrier frequency for combinations of three successive maxima, the circles for combinations of three minima.
It is clear from these results that the difference between reference and modulation frequencies for equal pitches follows qualitatively the diplacusis pattern over a wide range of frequencies in case of masking, and also in the case that no masking tone is presented, as far as it was possible to make the described combinations. None such an agreement at all is found when these data are plotted as a function of the modulation frequency.
In our opinion, these findings give another indication that the residue is perceived at that place on the basilar membrane, where its components are perceived. It appears obviously to explain the described phenomenon of pitch shifting in terms of fatigue of the peripheral parts of the ear, in a similar way as diplacusis is usually explained. This, however, becomes unlikely since Ritsma  introduced his model of pitch perception supposing that a "timing" mechanism exists, which analyses certain details of a complex sound such as the time between the occurrence of maxima of the amplitude (see also De Boer ). Ritsma concluded that this mechanism can not be located in the mechanical part of the hearing organ but rather in the nervous parts or even in the central nervous system.
The possibility of influencing the pitch of the residue by fatigue shows that somewhere between basilar membrane and sensation, the time-coded signal must be recoded in such a way that a shift of the pitch sensation is possible. This is supported by the fact, that the Broca phenomenon (pitch depending upon intensity) exists, not only for sine wave tones but probably also for the residue.