The Role of Pitch Memory in Pitch Discrimination

and Pitch Matching

*Robert E. Moore, *Casie Keaton, and Christopher Watts
*Mobile, Alabama, and Harrisonburg, Virginia

Summary: Accurate control of vocal pitch (fundamental frequency) requires

coordination of sensory and motor systems. Previous research has supported
the relationship between perceptual accuracy and vocal pitch matching accuracy. The purpose of this study was to investigate the role of memory for pitch
in pitch matching and pitch discrimination ability. Three experimental tasks
were used. First, a pitch matching task was completed, in which the participants
listened to target tones and vocally matched the pitch of the tones. The second
task was a pitch discrimination task that required the participants to judge the
pitch (same or different) of complex tone pairs. The third task was pitch discrimination with memory interference task that was similar to the pitch discrimination task except interference tones were added. Results of the pitch
matching and pitch discrimination tasks yielded a significant correlation between these values. When there was memory interference, pitch discrimination
ability was poorer, and there was no significant correlation between pitch discrimination and pitch matching. These results support earlier findings of a relationship between pitch discrimination and pitch matching abilities. The results
also suggest a possible role of pitch memory in both tasks. These findings may
have implications for abilities related to accurate pitch control.
Key Words: Pitch memoryPitch matchingPitch discriminationFundamental frequency control.

INTRODUCTION
Research has established that qualities of auditory stimuli and innate abilities influence perceptual
accuracy, and that perceptual accuracy is highly
correlated with the accuracy of vocal pitch or fundamental frequency (F0) control.1,2 One method
used to evaluate the accuracy of F0 control is a pitch
matching task. Those individuals who have accurate pitch discrimination abilities tend to be accurate pitch matchers. This relationship has been
substantiated in numerous populations, including
school-age children, musicians, and accurate
singers.1,3,4 In most populations, a continuum exists

Accepted for publication November 8, 2005.

From the *Department of Speech Pathology & Audiology,
University of South Alabama, Mobile, Alabama; and the
Department of Communication Sciences & Disorders, James
Madison University, Harrisonburg, Virginia.
Address correspondence and reprint requests to Robert E.
Moore, Department of Speech Pathology & Audiology, University of South Alabama, Mobile, AL 36688. E-mail:
rmoore@usouthal.edu
Journal of Voice, Vol. 21, No. 5, pp. 560567
0892-1997/$32.00
2007 The Voice Foundation
doi:10.1016/j.jvoice.2006.04.004

560

PITCH MEMORY & PITCH DISCRIMINATION/MATCHING

as to individual accuracy of F0 control. Related to
vocal pitch matching abilities, research has shown
that the timbre and frequency of an auditory stimulus can influence the accuracy of F0 control.2,5
However, less evidence is available, which has
demonstrated the influence of other variables on
F0 control.
In vocal pitch matching procedures, to accurately
match the pitch of a target stimulus, one must make
use of sensory abilities via the auditory system and
motor abilities via the subsystems of vocal production. In most pitch matching paradigms, a period of
time passes between the presentation of a target
stimulus and the vocal production (the matching attempt). During this interval, it would be necessary
to use memory for pitch, where an individual holds
a trace of the pitch for the target tone in memory
while the system readies for production. This type
of memory is immediate and has been called working or trace memory. Unlike long-term memory
that is used for retrieval of information, it is subject
to interference and capacity limits. While a pitch
matching paradigm would allow direct investigation of this topic, to date this methodology has
not been used to investigate the influence of pitch
memory on either perceptual accuracy or pitch
matching accuracy.
Deutsch6 suggested that memory for pitch is
a function of a specialized system of memory.
Deutsch7 also suggested that pitch memory is the
function of many frequency components or regions
in the cortex. She stated that these regions are activated by specific pitches of tones and are thought to
be organized tonotopically on a log frequency continuum. A functional magnetic resonance imaging
(fMRI) study has suggested that the supramarginal
gyrus and dorsolateral cerebellum play an important role in short-term retention of pitch information.8 In another fMRI study, musicians appeared
to process pitch memory more in the right temporal
and supramarginal gyrus while nonmusicians
showed more activation in the right primary and
left secondary auditory cortex.9
Deutsch7 has stated that there are certain inhibitory interactions that take place with these different
components that are a function of the distance
between interacting regions. When interference
sources, such as speech stimuli, pure tones, and

561

complex tones are presented between two tones in

a pitch discrimination task, they can serve as inhibitors of pitch memory. Of the three, complex tones
are most effective in disturbing pitch memory.
When the interpolated complex tones are within
the same octave as the two tones, they have the
greatest effect on pitch memory. Semal et al10 studied the effect of interpolating complex tones that
were composed of several harmonics of a missing
fundamental frequency. They concluded that memory interference depended on the pitch of the interpolated tones and not on the spectral composition
of the interpolated tones. Interpolated tones that
were close in pitch to the test tones in a pitch discrimination task were associated with poor performance, regardless of the spectra. The conclusion
was that poorer performance was due to interference in pitch memory by the interpolated tones.
These researchers also found that the amplitude
of the interpolated tones had no effect on pitch
memory. The findings of Clement et al11 suggest
that pitch memory and loudness memory are processed differently.
While pitch discrimination and pitch matching
abilities appear to be related, there are varying degrees of ability across individuals. The various factors involved in this relationship are not well
understood. While it is likely that a combination
of factors, such as physiology, auditory, and memory capabilities plays a role in this relationship,
the degree of the role of each individual factor is
less clear. The purpose of this study was to examine
pitch memory as a possible factor related to an individuals ability to match pitch. Pitch discrimination ability was studied with and without memory
interference and related to pitch matching ability.
We speculated that if pitch memory plays a significant role in pitch matching abilities, the relationship between pitch discrimination and pitch
matching ability would significantly decrease in
the presence of pitch memory interference.
METHODS
Participants
Participants were 30 females recruited from the
University of South Alabama student population
and the community. Due to the effect of age on
Journal of Voice, Vol. 21, No. 5, 2007

562

ROBERT E. MOORE ET AL

frequency discrimination, only individuals 2030

years of age participated.12 Because handedness is
a factor in pitch memory tasks, only right-handed
individuals were evaluated.13 All participants
passed an audiometric screening at 25 dBHL at
500, 1000, 2000, and 4000 Hz.14 Each participant
completed a questionnaire to ensure that there
was no history of speech, language, and hearing
disorders. All potential participants who indicated
on the questionnaire that they had received formal
musical training were excluded from the study.
Equipment and stimuli
All preliminary procedures, pitch matching,
pitch discrimination, and pitch discrimination
with memory interference tasks, were completed
in a sound-attenuated room, meeting specifications
for permissible ambient noise levels.15 The audiometric screening was completed using an audiometer, calibrated to ANSI standards.16 The pure tone
stimuli were delivered through TDH-50 earphones.
Pitch matching task
Eight stimulus tones were created using a Roland
EM-15 Creative Keyboard. The eight tones corresponded to the musical notes A#4 (233 Hz), C4
(262 Hz), C#4 (277 Hz), D#4 (311 Hz), F4
(349 Hz), F#4 (370 Hz), G#4 (415 Hz), and A#4
(466 Hz). These notes were selected because they
are within the physiological frequency range of
a normal female, and their corresponding frequencies are perceptually distinctive.17 These notes
were digitized onto an Alesis Masterlink hard
disk recorder, which was also used to present the
stimuli. Stimuli were presented one at a time in random order in sound field via a Fostex 6301B3E amplified speaker at 75 dB SPL with the participant
seated 1.5 m from the speaker.
Pitch discrimination task
Complex tone stimuli were generated using Cool
Edit Pro (version 2.0) sound editing software. Both
the reference and comparison tones had a total duration of 350 ms, and were gated on and off with
10 ms linear amplitude ramps. The reference and
comparison tones were composed of three equal
amplitude harmonics, which were added in phase.
Five reference tones were generated. The reference
Journal of Voice, Vol. 21, No. 5, 2007

tone frequencies were 110, 131, 156, 186, and

220 Hz. The fundamental frequencies for the reference tones were similar to those used by Duetsch.7
Fifteen comparison tones were generated. The fundamental frequency of a given comparison tone
could be equal to the fundamental frequency of
a given reference tone, 75 cents less than the fundamental frequency of the reference tone, or 75 cents
greater than the fundamental frequency of the reference tone. Therefore, for each reference tone, there
were three comparison tones. The comparison tone
frequencies (in Hz) were 105 (75 cents), 110, 115
(75 cents), 125 (75 cents), 131, 137 (75
cents), 149 (75 cents), 156, 163 (75 cents),
178 (75 cents), 186, 194 (75 cents), 211 (75
cents), 220, and 230 (75 cents) (Table 1). The
timing of the reference and comparison tones for
a given pair is shown in Figure 1. A total of 15 experimental conditions were generated in this manner. Each condition containing a reference tone
and a comparison tone was saved as an individual
.wav file. In each condition, the reference tone

TABLE 1. Frequency (Hz) Relationship of Stimulus

Tones, Interfering Tones, and Comparison Tones
Used in the Pitch Discrimination and Pitch
Discrimination with Memory Interference Tasks
Stimulus Tone
(Hz)

Interfering Tones
(Hz)

Comparison Tone
(Hz)

110

107/113/104/117
104/117/113/107
104/117/107/113

105
110
115

131

123/127/139/135
139/135/127/123
139/135/123/127

125
131
137

156

147/165/152/161
165/161/152/147
165/147/161/152

149
156
163

186

175/197/181/191
197/191/181/175
197/191/175/181

178
186
194

220

228/235/209/191
215/235/228/209
235/209/228/215

211
220
230

PITCH MEMORY & PITCH DISCRIMINATION/MATCHING

FIGURE 1. Stimuli timing pattern used for pitch discrimination task (top) and pitch discrimination with memory interference task (bottom).

and comparison tone were separated by a 6 second

(onsetonset) interval.
Pitch discrimination with memory interference task
The stimuli for this portion of the study were
modeled after those of Semal et al.10 The reference
tones and comparison tones for the pitch memory
task were the same as those generated for the pitch
discrimination task. For this task, four interference
tones were generated for each of the 15 conditions
used in the pitch discrimination task. The first interference tone started 1.5 seconds after the onset of
the reference tone. There was also a 1.5 second interval between the onset of the last interference
tone and the onset of the comparison tone. Each interference tone consisted of three harmonics of
equal amplitude and had a duration of 250 ms.
The interference tones were gated on and off with
10 ms linear amplitude ramps. For a given condition, each interference tone was at random, one of
four fundamental frequencies. These frequencies
were 50 cents and 100 cents less than the fundamental frequency of the reference tone, and 50
cents and 100 cents greater than the fundamental
frequency of the reference tone. For example, all
comparisons to a reference tone of 110 Hz involved
interference tones of 104 Hz (100 cents), 107 Hz
(50 cents), 113 Hz (50 cents), and 117 Hz
(100 cents). A total of 15 experimental conditions
were generated in this manner. Each condition containing a reference tone, four interference tones,
and a comparison tone was saved as an individual
.wav file (Table 1). The timing of the reference, interference, and comparison tones is illustrated in
Figure 1.

563

Procedures
Pitch matching task
Each of eight stimulus complex tones was presented randomly at 75 dB SPL, one at a time via
a speaker. A total of 24 stimulus tones were presented in three trials. Subjects were instructed to
listen to the complex tone presented, and then to
vocally match the pitch of each target tone with
the vowel ah. Subjects were asked to maintain
the ah for 5 seconds, which was timed by the investigator. Subjects pitch matching responses were
recorded using a head-mounted microphone and the
Computerized Speech Lab (CSL; Kay Elemetrics
Corporation, Lincoln Park, NJ), with sampling
rate set at 44.1 kHz. Each subjects response was
saved in an individual file to the hard drive of the
computer for further analysis.
Pitch discrimination task
For the pitch discrimination task, participants
were seated at a desk inside the sound-attenuated
room, with a color monitor and computer mouse.
The investigator monitored the participants responses and progress on a networked computer
from outside the room. The investigator was also
able to see the subject using a closed-circuit camera
and monitor system. Presentation of each reference
and comparison tone condition was controlled by
ECos/Win software. The stimuli were presented
binaurally through E-A-R 5A insert earphones at
75 dB SPL.
Participants were instructed to listen for two
tones and to indicate whether the tones were
same or different in pitch. Participants indicated their judgment of the two tones by using
the computer mouse to click on the appropriate
icon displayed on the monitor. A tone pair was
not presented until the participant had made a judgment regarding the preceding pair. One block of 15
stimuli was presented to each participant as practice. Once it was shown that the participant understood and could complete the task, two further
experimental blocks were presented. The order of
presentation of the tone pairs was randomized individually for each block by the ECos/Win software.
All participant responses were saved to the computer hard drive for further analysis. Only the last
two blocks were scored for the experiment.
Journal of Voice, Vol. 21, No. 5, 2007

564

ROBERT E. MOORE ET AL

Pitch discrimination with memory interference task

The basic procedures for the pitch memory task
were almost identical to those for the pitch discrimination task. Because the participants heard six
tones (one reference, four interference, and one
comparison) in this portion of the study, they
were instructed to indicate whether the first tone
was same or different in pitch to the last tone.

Analysis
Three dependent variables were measured in
this study: pitch matching accuracy, pitch discrimination accuracy, and pitch discrimination accuracy with memory interference. Pitch matching
ability was calculated by analyzing the middle
4 seconds of each participants vocal match. This
time frame was chosen to avoid measuring onset
and offset time, which could lead to variability
in the measurement. The fundamental frequency
of each sample was calculated by the CSL using
the Multi Dimensional Voice Profile software. An
average of the three vocal attempts for each target
tone was calculated. The frequency of the target
tone and average frequency of the pitch matching
attempts was used to calculate a difference in
semitones. Finally, the semitone differences for
each of the eight target tones were averaged to arrive at an overall semitone difference for each participant. A smaller score indicated more precise
pitch matching accuracy. Performance accuracy
for the pitch discrimination and pitch discrimination with memory interference tasks was measured
by calculating the percentage correct score for
each participant from a total of 30 stimulus
presentations.
A one-way repeated measures analysis of variance (ANOVA) was performed to compare the
means of the pitch discrimination and pitch discrimination with memory interference task to ensure that the interference tones had a significant
effect on pitch discrimination ability. The presence
and strength of the relationship between pitch
matching and pitch discrimination, as well as between pitch matching and pitch discrimination
with memory interference, was also evaluated.
This was done using Pearsons product-moment
correlations.
Journal of Voice, Vol. 21, No. 5, 2007

RESULTS
The descriptive statistics for the pitch matching,
pitch discrimination, and pitch discrimination with
memory interference tasks are shown in Table 2.
Standard deviations are shown with the pitch
matching results. In general, variance was greatest
with less precise pitch matching accuracy. As expected, the scores for the pitch discrimination
task were better than the scores for the pitch discrimination with memory interference task for
most of the participants. One participant scored
the same for both the pitch discrimination and pitch
discrimination with memory interference tasks,

TABLE 2. Difference Scores for the Pitch Matching

(PM) Task in Semitones With Standard Deviations in
Parentheses and the Percent Correct Scores for the
Pitch Discrimination (PD) and Pitch Discrimination
With Memory Interference (PDI) for Each Participant
Participant
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

PM (Semitones)
0.09
0.12
3.73
3.53
0.09
0.16
0.19
0.78
0.15
0.07
1.37
0.18
0.13
0.15
0.08
0.29
1.78
1.94
0.09
2.31
0.36
1.96
1.01
1.32
1.14
2.41
2.75
4.75
0.53
0.18

(0.08)
(0.09)
(3.89)
(3.37)
(0.07)
(0.09)
(0.10)
(0.90)
(0.09)
(0.06)
(0.41)
(0.07)
(0.65)
(0.13)
(0.05)
(0.11)
(1.16)
(2.01)
(0.09)
(1.37)
(0.08)
(1.23)
(1.72)
(1.91)
(1.55)
(2.44)
(2.06)
(3.74)
(0.89)
(0.03)

PD (%)

PDI (%)

100.0
70.0
37.7
46.7
93.3
93.3
66.7
43.3
96.7
96.7
63.3
83.3
86.6
80.0
100.0
93.3
50.0
53.3
96.7
93.3
100.0
43.3
46.7
53.3
86.6
73.3
56.7
70.0
76.7
96.7

66.7
66.7
50.0
60.0
60.0
93.3
60.0
50.0
43.3
63.3
43.3
43.3
60.0
66.7
46.7
36.7
40.0
36.7
63.3
60.0
70.0
40.0
43.3
40.0
83.3
60.0
53.3
60.0
66.7
66.7

PITCH MEMORY & PITCH DISCRIMINATION/MATCHING

while three participants had higher scores for the
pitch discrimination with memory interference
task. The average pitch discrimination score was
74.92% (SD 5 20.87) and the average pitch discrimination with memory interference score was
56.44% (SD 5 13.70). Pitch matching difference
scores ranged from 0.07 semitones to 3.73 semitones with an average of 1.02 semitones
(SD 5 1.10).
Results from a one-way repeated measures ANOVA indicated a significant difference between
pitch discrimination and pitch discrimination with
memory
interference
scores
(F 5 28.16,
P ! 0.001). This showed that when pitch memory
was disturbed there was a significant change in
pitch discrimination ability. Mauchlys Test of
Sphericity was not significant (W 5 1.0) indicating
that the condition of sphericity existed.
A bivariate correlation analysis using Pearsons
product-moment correlation was performed to investigate the relationship between pitch matching
and pitch discrimination. The correlation coefficient between these two variables was significant
(r 5 0.665, P ! 0.01). This suggested a moderate negative correlation between pitch matching
and pitch discrimination. This was interpreted as
indicating that when pitch discrimination ability
was high in accuracy, pitch matching ability
was also high in accuracy (and vice versa). To
further investigate the relationship between pitch
matching and pitch discrimination, the coefficient
of determination was found. This suggested that
the shared variance accounted for was 44%
(r2 5 0.44).
To evaluate the relationship between pitch
matching and pitch discrimination with memory interference another Pearsons product-moment correlation was performed on these variables. The
correlation between the two variables was not significant (r 5 0.239, P O 0.05). This indicated
that there was no correlation between pitch matching and pitch discrimination when there was interference in pitch memory. The coefficient of
determination between pitch matching and pitch
discrimination with memory interference was also
found. This suggested that the shared variance
was only 6% (r2 5 0.06).

565

DISCUSSION
Perceptual accuracy is needed for accurate F0
control. One method used to study fundamental frequency control is a pitch matching paradigm. Previous research has demonstrated a relationship
between pitch matching ability and pitch discrimination ability. There is variability in the ability of
individuals to match and discriminate pitch. To
complete a vocal pitch matching task, an individual
must hear the stimulus and process the pitch of the
stimulus. In other words, pitch discrimination is
a part of pitch matching. Based on these first two
functions, the individual then programs motor neurons and reproduces the sound as closely as possible. During the time in which the individual is
preparing to produce the sound, the processed pitch
must be held in memory. Previous researchers have
suggested that short-term memory for pitch is
unique in place and pattern of storage.
The purpose of this study was to examine the
role of pitch memory as it relates to ones ability
to match and discriminate pitch. Specifically, we
were interested in investigating the relationship between pitch matching and pitch discrimination and
how that relationship is affected by interfering with
pitch memory. In agreement with previous studies,
the results of the present study demonstrated a relationship between ones ability to match pitch and
ones ability to discriminate pitch. Most individuals
with increased pitch discrimination accuracy were
able to match pitch more accurately. Pitch discrimination ability explained 44% of the variance in
pitch matching ability. Because pitch discrimination accuracy appears to have a significant influence on an individuals pitch matching accuracy,
any change in pitch discrimination accuracy caused
by pitch memory interference should be greater in
those who are the most precise pitch matchers, if
one assumes that those individuals typically make
better use of pitch memory in the pitch discrimination phase of pitch matching. Other factors that
may influence pitch matching accuracy are motor
control and, of course, vocal accuracy.
To test the role of pitch memory, four interference
tones were placed between the two tones used in the
pitch discrimination portion of the study. Previous
research has indicated that such a paradigm results
Journal of Voice, Vol. 21, No. 5, 2007

566

ROBERT E. MOORE ET AL

in decreased pitch discrimination ability. The decline in pitch discrimination has been interpreted
as pitch memory interference.7 In the present study,
a decline in pitch discrimination ability was seen for
most individuals, and overall, there was a significant
decrease in pitch discrimination ability. Under this
condition, correlational analysis indicated that there
was no significant relationship between pitch
matching and pitch discrimination with memory interference. Pitch discrimination ability in the presence of memory interference explained only 6% of
the variance in pitch matching ability. This would
suggest that pitch memory plays a significant role
in both pitch discrimination and pitch matching. It
appears that an important part of fundamental frequency control, as measured by pitch matching, is
pitch discrimination ability with its inherent component, memory for pitch.
No attempt was made in participant identification
to preselect the participants based on vocal musical
ability. However, previous research results would
suggest that those with superior pitch matching
ability could be predicted to have superior vocal accuracy than those with poorer pitch matching accuracy. Accordingly, as shown in Table 3, the
participants were divided into three groups based
on their pitch matching abilities. One group was
made up of those participants with the ability to
match pitch within 0.1 semitones. These are termed
the superior pitch matchers. A second group was
composed of participants who were unable to
match pitch within 1.0 semitone. These were

TABLE 3. Average Pitch Discrimination (PD) and

Pitch Discrimination with Memory Interference
(PDI) Scores and Differences for Three Groups Based
on Pitch Matching Ability

termed the poor pitch matchers. The third group

is made up of those participants who fall between
the other two groups in pitch matching ability.
These were called the intermediate pitch matchers.
As shown in Table 3, the average difference between pitch discrimination ability and pitch discrimination ability with memory interference for
the three groups was 37.4% for the superior group,
21.93% for the intermediate group, and 8.02% for
the poor group. Clearly, for those participants
with superior pitch matching abilities, there was
a greater change in pitch discrimination ability
when interference to pitch memory was presented.
This also suggests a greater influence of pitch memory on discrimination accuracy for those with superior pitch matching ability.
These results have implications concerning the
development of vocal skills. Geringer18 has suggested that both experience and maturation play
a role in the relationship between pitch perception
and pitch production. In children, the relationship
between pitch matching and pitch discrimination
appears to improve with age. One reason for this
may be cognitive maturation resulting in improved
memory processing, including pitch memory. One
aspect of early musical ability, including vocal control, may be the development of pitch memory. At
the other end of the aging spectrum, it is possible
that a contributing factor to a decline in vocal musical ability may be a decline in cognitive processing with advanced age. Declines in cognitive
abilities, including working memory, have been
documented.1921 In summary, pitch memory may
play a role in both vocal musical ability development and decline with aging. Both possible roles
of pitch memory warrant further research.
CONCLUSION

Group
Superior (5)
Intermediate (12)
Poor (13)

PD (%)

PDI (%)

Difference (%)

97.34
80.55
59.55

60.00
60.28
51.53

37.34
20.27
8.02

The numbers in parentheses following the group name are the

numbers of participants included. Superior group 5 difference
in pitch target and vocal pitch less than 0.1 semitones; intermediate group 5 difference in pitch target and vocal pitch between 0.1 and 1.0 semitones; poor group 5 difference in
pitch target and vocal pitch greater than 1.0 semitones.
Journal of Voice, Vol. 21, No. 5, 2007

While pitch memory was not directly measured

in a pitch matching task in this study, inferences
were made based on the relationship between pitch
matching and pitch discrimination abilities. These
inferences indicate that pitch memory plays a role
in both pitch discrimination and pitch matching
abilities. Because pitch matching is a measure of
fundamental frequency control, this also indicates
a role of pitch memory in fundamental frequency

PITCH MEMORY & PITCH DISCRIMINATION/MATCHING

control. In addition, individuals with good pitch
matching abilities and fundamental frequency control appear to rely on pitch memory more than those
with lesser fundamental frequency control. In future research, it may be appropriate to more directly
measure the role of pitch memory in a pitch matching task to further evaluate this relationship.
REFERENCES
1. Watts C, Moore R, McCaghren K. The relationship between vocal pitch-matching skills and pitch discrimination
skills in untrained accurate and inaccurate singers. J Voice.
2005;19:534543.
2. Watts C, Murphy J, Barnes-Burroughs K. Pitch-matching
accuracy of trained singers, untrained participants with talented singing voices, and untrained participants with nontalented singing voices in conditions of varying feedback.
J Voice. 2003;17:187196.
3. Pederson DM, Pederson NO. The relationship between
pitch recognition and vocal pitch production in sixth-grade
students. J Res Music Ed. 1970;18:265272.
4. Amir O, Amir N, Kishon-Rabin L. The effect of superior
auditory skills on vocal accuracy. J Acoust Soc Am.
2003;113:11021108.
5. Larson C, Burnett T, Kiran S, Hain T. Effects of pitch-shift
onset velocity on voice FO responses. J Acoust Soc Am.
2000;107:559654.
6. Deutsch D. The deterioration of pitch information in memory [unpublished doctoral dissertation]. University of California at San Diego; 1970.
7. Deutsch D. Generality of interference by tonal stimuli in recognition memory for pitch. Q J Exp Psychol. 1974;26:229234.
8. Gaab N, Gaser C, Zaehle T, Jancke L, Schlaug G. Functional
anatomy of pitch memory an fMRI study with sparse temporal sampling. NeuroImage. 2003;19:14171426.

567

9. Gaab N, Schlaug G. The effect of musicianship on pitch

memory in performance matched groups. NeuroReport.
2003;14:22912295.
10. Semal C, Demany L, Ueda K. Speech vs. nonspeech in
pitch memory. J Acoust Soc Am. 1996;100:11321140.
11. Clement S, Demany L, Semal C. Memory for pitch versus
memory for loudness. J Acoust Soc Am. 1999;106:2805
2811.
12. He N, Dubno JR, Mills JH. Frequency and intensity discrimination measured in a maximum likelihood procedure
from young and aged normal-hearing subjects. J Acoust
Soc Am. 1998;103:553565.
13. Deutsch D. Pitch memory: an advantage for the lefthanded. Science. 1978;199:559560.
14. American Speech-Language Association. Guidelines for
manual pure-tone audiometry. ASHA. 1978;20:297301.
15. American National Standards Institute. Maximum permissible ambient noise levels for audiometric test rooms.
ANSI S3.1-1991. New York: American National Standards
Institute; 1991.
16. American National Standards Institute. Specifications for
audiometers. ANSI S3.6-1996. New York: American National Standard Institute; 1996.
17. Baken R, Orlikoff R. Clinical Measurement of Voice and
Speech. San Diego: Singular; 2000.
18. Geringer JM. The relationship of pitch-matching and
pitch-discrimination abilities of preschool and fourthgrade students. J Res Music Ed 1983;9399.
19. Engle RW, Cantor J, Carullo JJ. Individual differences in
working memory and comprehension: a test of four hypotheses. J Exp Psychol Learn Mem Cogn. 1992;18:972
992.
20. Just MA, Carpenter PA. A capacity theory of comprehension: individual differences in working memory. Psychol
Rev. 1992;99:122149.
21. Salthouse TA. Influence of working memory on adult age
differences in matrix reasoning. Br J Psychol. 1993;84:
171199.

Journal of Voice, Vol. 21, No. 5, 2007