Pitch of Frequency-modulated signals.

MISSING IMAGE

Material Information

Title:
Pitch of Frequency-modulated signals.
Alternate title:
Frequency-modulated signals
Physical Description:
x, 71 leaves : ill. ; 28 cm.
Language:
English
Creator:
McClelland, Keener Delaney, 1936-
Publication Date:

Subjects

Subjects / Keywords:
Musical pitch   ( lcsh )
Speech thesis Ph. D   ( lcsh )
Dissertations, Academic -- Speech -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Bibliography: leaves 69-71.
Additional Physical Form:
Also available online.
General Note:
Manuscript copy.
General Note:
Vita.
Statement of Responsibility:
by Keener Delaney McClelland

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 13546304
System ID:
AA00024007:00001

Table of Contents
    Title Page
        Page i
        Page i-a
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
    List of Tables
        Page v
    List of Figures
        Page vi
        Page vii
        Page viii
        Page ix
        Page x
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
    Chapter 2. Procedure
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
    Chapter 3. Result
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
    Chapter 4. Discussion
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
    Chapter 5. Summary
        Page 37
        Page 38
        Page 39
    Appendices
        Page 40
    Appendix A. Electrical spectra and earphone frequency response curves
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
    Appendix B. Individual subject pitch match data
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
    Bibliography
        Page 69
        Page 70
        Page 71
    Biographical sketch
        Page 72
        Page 73
        Page 74
Full Text















PITCH OF FREQUENCY-MODULATED SIGNALS












By

KEENER DELANEY McCLELLAND












A DISSERlTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR IRE
DEGREE OF DOCTOR OF PHILOSOPHY










UNIVERSITY OF FLORIDA 1968



















































Ar
































To

Professor C. Albro Newton

















ACKNOWLEDGMENTS



The writer wishes to express his sincere appreciation to Dr. John F. Brandt for his encouragement and guidance in the planning and execution of the present research, but more especially for the benefit of his skill as a researcher and patience as a teacher.

The author is indebted to the members of the dissertation committee for their interest and constructive criticism. Thanks go to the faculty and students of the Communication Sciences Laboratory of the University of Florida for their support and assistance. Special thanks also go to Dr. Donald C. Teas and Dr. Arnold Paige who read and criticized the original manuscript and made many valuable suggestions.

The work of my wife Nancy in the preparation of graphs, charts and the typing of preliminary drafts was invaluable.

Finally, appreciation is extended to the Vocational Rehabilitation Administration for financial support in the form of a VRA traineeship. In addition, the research was supported from funds provided by the Graduate School of the University of Florida and the National Institutes of Health

(Grant NB 06459)















TABLE OF CONTENTS


Page

LIST OF TABLES . . . . . . . . v

LIST OF FIGURES . . . . . . . . . vi

CHAPTER

I INTRODUCTION . . . . . . . 1

II PROCEDURE . . . . . . . . 6
Apparatus . . . . . . . 6
Stimuli . . . . . . . . 9
Subjects . . . . . . . . 9
General Procedure . . . . . 10

III RESULTS . . . . . . . . 11
Effects of Modulation Frequency
(20dB SL) . . . . . . . 18
Effects of Modulation Frequency
(50dB SL) . . . . . . . 20
Subject Variability . . . . . 21

IV DISCUSSION . . . . . . . 24

V SUMMARY . . . . . . . . 37

APPENDIX A ELECTRICAL SPECTRA AND EARPHONE
FREQUENCY RESPONSE CURVES . . . 41

APPENDIX B INDIVIDUAL SUBJECT PITCH MATCH DATA 48

BIBLIOGRAPHY . . . . . . . . . 69

BIOGRAPHICAL SKETCH . . . . . . 72











iv


















LIST OF TABLES



Table Page

1 Experimental conditions. The X's in
the table represent combinations
of modulating and carrier frequencies. 0 9

2 Frequency modulation stimulus
parameters. 26










































v


















LIST OF FIGURES



Figure Page

1 Schematic drawing of equipment. 8

2 Deviation in semi-tones of grouped pitch
judgments in 20-cent intervals from
the carrier frequency for conditions
300/20 (A) and 300/200 (B). Data collected at 20dB SL. The arrows represent spectral information as
explained in the text. 12

3 Deviation in semi-tones of grouped pitch
judgments in 20-cent intervals from the
carrier frequency for conditions 1000/20 (A) and 1000/200 (B). Data collected at
20dB SL. 13

4 Deviation in semi-tones grouped pitch
judgments in 20-cent intervals from
the carrier frequency for conditions
3000/20 (A), 3000/200 (B) and 3000/2000
(C). Data collected at 20dB SL. 14

5 Deviation in semi-tones of grouped pitch
judgments in 20-cent intervals from the carrier frequency for conditions
300/20 (A) and 300/200 (B) Data
collected at 50dB SL. 15

6 Deviation in semi-tones of grouped pitch
judgments in 20-cent intervals from the carrier frequency for conditions 1000/20 (A) and 1000/200 (B). Data
collected at 50dB SL. 16






vi










7 Deviation in semi-tones of grouped pitch
judgments in 20-cent intervals from the carrier frequency for conditions
3000/20 (A) 3000/200 (B) and 3000/2000
(C). Data collected at 50dB SL. 17

8 Representation of between-subject pitch
matches. 22

9 Sensitivity of the ear as a function of
frequency (after Stevens and Davis, 1938). 29

10 Critical bandwidth based on the minimum
frequency difference between the harmonics
of a complex tone necessary for them to
be heard separately, as a function of
frequency (after Plomp and Mimpen, 1968). 31

11 Electrical spectra for 300/20 (A) and
300/200 (B). Analyzing bandwidth was
10 Hz. 42

12 Electrical spectra for 1000*/20 (A) and
1000/200 (B). Analyzing bandwidth was
10 Hz. 43

13 Electrical spectra for 3000/20 (A) and
3000/200 (B). Analyzing bandwidth
was 10 Hz. 44

14 Electrical spectrum for 3000/2000. Analyzing bandwidth was 10 Hz. 45

1.5 Earphone frequency response curve for
Earphone A. 46

16 Earphone frequency response curve for
Earphone B. 47

17 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject DP for conditions 300/20 and 300/200
at 20 and 50dB SL. 49





vii










is Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject DF
for conditions 1000/20 and 1000/200
at 20 and 50dB SL. 50

19 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject DF
for conditions 3000/20 and 3000/200
at 20 and 50dB SL. 51

20 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject DE for
conditions 3000/2000 at 20 and 50dB SL. 52

21 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject NM for
conditions 300/20 and 300/200 at 20
and 50dB SL. 53

22. Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject NM for conditions 1000/20 and 1000/200 at 20
and 50dB SL. 54

23 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject NM for conditions 3000/20 and 3000/200 at 20
and 50dB SL. 55

24 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject NM for
conditions 3000/2000 at 20 and 50dB SL. 56

25 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject TMV for
conditions 300/20 and 300/200 at 20
and 50dB SL. 57





viii










26 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject TM for conditions 1000/20 and 1000/200 at 20
and 50dB SL. 58

27 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject TM for conditions 3000/20 and 3000/200 at 20
and 50dB SL. 59

28 Deviation in semi-tones of pitch judgments in 20-cent intervals from the
carrier frequency from subject TM for
conditions 3000/2000 at 20 and 50dB SL. 60

29 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject CH for conditions
300/20 and 300/200 at 20 and 50dB SL. 61

30 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject CH for conditions
1000/20 and 1000/200 at 20 and 50dB SL. 62

31 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject CH for conditions
3000/20 and 3000/200 at 20 and 50dB SL. 63


32 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject CH for conditions
3000/2000 at 20 and 50dB SL. 64

33 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject JH for conditions
300/20 and 300/200 at 20 and 50dB SL. 65

34 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject JH for conditions
1000/20 and 1000/200 at 20 and 50dB SL. 66



ix










35 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject JH for conditions
3000/20 and 3000/200 at 20 and 50dB SL. 67

36 Deviation in semi-tones of pitch judgments
in 20-cent intervals from the carrier
frequency from subject JHi for conditions
3000/2000 at 20 and 50dB SL. 68




















































x


















CHAPTER I

INTRODUCTION



The psychophysical phenomenon of pitch has long been considered to be related to a process of mechanical frequency analysis approximately a Fourier analysis (von Helmholtz, 1863). Von B~k~sy (1960) demonstrated both in models and in human cochleas that mechanical frequency analysis occurs, and that it takes the form of a distribution of the various spectral components of an auditory signal along the basilar membrane in an orderly sequence according to frequency. A limited resolution Fourier analysis, in conjunction with neural funneling (von B~k~sy, 1960), can account for the pitch of a pure tone and the absence of pitch characteristic of broad band noise. Tonndorf (1962) has shown that the cochlea performs a combined time and frequency analysis of auditory signals. He demonstrated time and frequency analysis in B6k~sy-type cochlear models showing that the models' response to a sinusoid closely approximated a frequency analysis and its response to an impulse closely approximated a time analysis. Complex signals distributed themselves along a continuum between these two extremes.



1






2



Complex signals consisting of sinusoids spaced at inharmonic intervals produce pitch matches to the middle portion of the energy mass (Ekdahl and Boring, 1934). Thus, if a number of sinusoids is placed within a restricted frequency band, listeners produce pitch matches in the center of the band. However, when there are only a few components and the components are widely spaced in frequency, the ear can resolve the complex into its individual components. This phenomenon, the analysis of a complex signal into its individual frequency components, is called Ohm's acoustic law. Some investigators believe that the capacity to perform such an analysis depends upon the ability to discriminate the separate areas of mechanical excitation along the basilar membrane (Stevens and Davis, 1938; Greenwood, 1961). Plomp (1964) found that pitch matches were made individually to the lower five to seven spectral components of both harmonic and inharmonic complex signals when the spacing of the components exceeded critical bandwidth at the frequency location of the signal.

In addition to mechanical separation of components

time analysis not necessarily peripheral in nature allows for the perception of periodicity pitch (Small, 1955), time separation pitch (Small and McClellan, 1963; McClellan and Small, 1965, 1966, 1967) and pitch of the residue (Schouten, Ritsma






3


and Cardozo 1962; Ritsrna, 1962, 1963a, 1963b, 1967). Small (1955) found that amplitude-modulated (AM) signals elicited

pitch matches corresponding to the repetition rate of the modul ating frequency or the carrier frequency or both as stimulus rise-fall time, duty cycle and carrier frequency were varied. He suggested that perception of a periodicity pitch depends on the stimulus bandwidth, location of the spectral components, and envelope wave form fluctuation and not on the individual spectral components alone.

A quasi frequency-modulated (FM) signal, consisting

of only a carrier frequency and two sideband frequencies, has

been used in the study of the pitch of the residue (Ritsma and Engel, 1964; Schouten, Ritsma and Cardozo, 1962) and certain phase effects (Mathes and Miller, 1947; Goldstein, 1967). The pitch of the quasi FM signal was matched to the pitch of an AM signal with the same carrier frequency when the ratio (f/g) of the carrier frequency (f) to the modulating frequency

(g) was 10 or greater for the FM signal. For low values of f/g (i.e., less than nine) pitch matches were made to the modulating frequency, FM signals have also exhibited pitches corresponding to the carrier frequency, as well (Ritsma and Engel, 1964). These investigators were forced to use a quasi FM signal because three-component spectra occur for sinusoidally frequency-modulated signals only when the





4



modulation index (ratio of frequency deviation to modulation frequency) is less than one. When the modulating index exceeds unity, the number of sidebands increases rapidly and the spectrum no longer resembles that of an AM signal.

FM auditory signals have been used in psychoacoustic investigations to obtain frequency difference limens (Shower and Biddulph, 1931; Filling, 1958; Brandt, 1967) and in experiments concerning phase perception (Goldstein, 1967; Mathes and Miller, 1947; Zwicker, 1962), to determine one pitch related to the fine structure of the wave form of an auditory signal (Ritsma and Engel, 1964; Fischler, 1967) and to specify the sensitivity to unidirectional FM (Sergeant and Harris, 1962).

The limited use that has been made of FM auditory

signals has shown that they elicit a variety of pitch perceptions dependent upon the acoustic characteristics of the particular signal employed. Frequency-modulated sinusoids have been shown to elicit pitches related to the carrier frequency and to the modulation frequency. Most experiments involving FM signals have generally required the subject to match the pitch or some other psychophysical attribute of theFM signal to the corresponding attribute of an AM signal. The investigators (Ritsma and Engel, 1964; Schouten, Ritsma and Cardozo, 1962; Mathes and Miller, 1947; Goldstein, 1967)






5


have taken special care to insure that the spectral characteristics of the FM and AM signals were as similar as possible. Such spectral matching of FM signals to their AM counterparts produces signals whose characteristics may

not be representative of the general properties of FM signals.

Consideration of the frequency analyzing capability

of the ear, demonstrated by Plomp (1964) and Plomp and Mimpen (1968) using complex signals with various frequency spacings of the spectral components, suggests that the ear performs a limited resolution Fourier analysis on such signals and that an FM signal should elicit pitches corresponding to its individual spectral components when' the frequency spacing of the spectral components exceeds the critical bandwidth of the ear's analyzing filter at that frequency.

The pitch of FM signals has not been systematically

investigated in the general case. The present investigation is an attempt to determine the pitch or pitches elicited by FM signals as a function of the carrier frequency, modulating frequency and sensation level of the signal in a free response experiment.

















CHAPTER II

PROCEDURE


Apparatus. Frequency-modulated stimuli were generated by the reactance-tube modulator of a beat-frequency oscillator (Bruel and Kjaer, Type 1014). The modulating signal was generated by an external oscillator (General Radio, Type 1304-B). The modulating signal amplitude was controlled by the oscillator attenuator and an external attenuator (Hewlett-Packard, 350D) and was such that the extent of modulation was always +100 Hz. The carrier frequency was selected on the main frequency dial of the Bruel and Kjaer oscillator. The FM signal was split and attenuated by two pairs of attenuators in series (Hewlett-Packard, 350D) and fed through impedance-matching transformers (United Transformer Company, LS-33) to two earphones (Telephonics, TDH 39-10Z) to permit simultaneous listening by two subjects. Each active earphone was mounted together with a dummy earphone in a standard headband. All earphones were fitted with MX-41/AR cushions.

The subjects sat in an IAC sound-treated room (Model

403-A). Two matching oscillators (General Radio, Type 1313A), two attenuators (Hewlett-Packard, 350D) and two two-position



6





7



switches were arranged so that each subject could select either the experimental FMI stimulus or his own unmodulated matching signal. The matching oscillator dials were covered by white cardboard discs so that no dial markings were visible to the subjects. The frequency and intensity of the matching tone were under subject control. The frequency of each matching oscillator was read out on a frequency counter (Hewlett-Packard, 522B) upon indication from each subject that a match had been made. Figure 1 shows the arrangement of the equipment.

Acoustic stimuli were analyzed using an artificial

ear (Bruiel and Kjaer, Type 4152) in' conjunction with a oneinch condenser microphone and cathode follower (Bruel and Kjaer, Type 4132/2163) and microphone amplifier (Briiel and Kjaer audio frequency spectrometer, Type 2112) whose output was fed to a wave analyzer (General Radio, 1900A) and graphic level recorder (General Radio, 1521B). Electrical stimuli were also analyzed using the wave analyzer and graphic level recorder and carefully monitored throughout the experiment. Although second harmonic distortion was present in the acoustic signal, the amplitude of the distortion products was 45dB or more below the level of the unmodulated carrier for all of the stimulus conditions. Earphone, responses and representative electrical spectra are to be found in Appendix A.

















Z Ld
Ld :D
a D
LLJ 0 (D 0
M Q
LL.








Ld :2 1
cr
0
Lij


LLI ui
z z x




Nt











w z
a Lli LLI






cr LLJ
Lli

<
0



m
z
w
0
2 0 0
< m z
in (D EE
Q
!: C)






9


Stimuli. Stimuli consisted of 300-, 1000- and 3000Hz sinusoids, each frequency-modulated by sinusoids of 20, 200 and 2000 11z (the latter only for the 3000-Hz carrier frequency). The frequency deviation in Hz from the carrier frequency was always +100 Hz. Stimulus intensities of 20 and 50dB relative to the subject's threshold were used. All conditions were randomized except that when two subjects listened simultaneously, they were both presented the same FM signal. The experimental conditions are listed in Table 1.




Table 1. Experimental Conditions. The X' s in the table represent combinations of modulating and carrier frequencies.



20dB Sensation Level 50dB Sensation Level
Modulating Carrier Frequency Carrier Frequency
Frequency 300 1000 3000 300 1000 3000


20 X X X X X X

200 X X X X X X

2000 X X






Subjects. Five normal-hearing adults, who demonstrated

the ability to pitch match to the unmodulated carrier frequency with no more than one semitone frequency error, were used as subjects. Normal hearing was defined for purposes of this

experiment as no loss greater than 15dB (re. ISO, 1964 standards)






10


at any octave frequency over the range from 125 to 8000 Hz.

General Procedure. Subjects were seated comfortably in the sound booth with the stimulus switch, matching oscillator and attenuator arranged for the individual subjects' convenience. Instructions were presented to the subject on a typed card and the investigator answered any questions after the subject had read the card. The instructions were:

1. Listen to the signal to become familiar with it.
2. Match to the most obvious pitch or pitches.
3. Search for other pitches in the signal. There
may be one or as many as seven or more. They
may sound close together or far apart.
4. Match to all the pitches.
5. Once the pitches are established, speed should
be increased while maintaining accuracy.
6. If you should become tired, either turn the
switch to the "off" position for awhile, or
leave the booth for a break.

Subjects were permitted to listen to the stimulus as long as they wished, and to switch back and forth between stimulus and matching signal as they wished before they indicated a match. They were allowed to vary the intensity of. the matching stimulus if so desired. Most listeners preferred the two stimuli to be approximately equal in loudness. A fiveminute break was mandatory after each 30-minute pitch matching period if the subject did not choose to rest at more frequent intervals.

















CHAPTER III

RESULTS



Data were collected in the form of pitch matches to sinusoidally frequency-modulated sinusoids with carrier frequencies of 300, 1000 and 3000 Hz. Modulating frequencies of 20, 200 and 2000 Hz were used, the latter for the 3000-Hz carrier frequency only. The frequency deviation about the carrier frequency was held constant at 100 Hz for.all stimuli. Stimuli were presented at 20dB and 50dB above the subject's threshold for the FM signal. The listeners were instructed to match an unmodulated sinusoid to any and all pitches they perceived in theFM stimulus.

Two-hundred pitch matches were obtained from each of

the five listeners resulting in a total of 1000 pitch matches for each of the 14 stimulus conditions. The pitch matches were tallied as deviations from the carrier frequency in 20-cent intervals. Individual listener data are shown graphically in Appendix B.

Figures 2, 3, 4, 5, 6 and 7 are composite figures

showing the combined data from all listeners for each condition in terms of the deviation of pitch matches from the



11






12






0 0 0





>

Q) u
o 41
-0 0
L\j
0

0 41
-0 x
0 4J W
Cd 4J
C) C
0

co 41


-0
9
U)
0 z 4
C-0 w C\j r C) -0 =) ql, G) C Q) Of
Ld
0 cr- 00 C) co
-0 LL 1- 0 -1
re) c)
CC
0 Ld
m C.) 9
C) 4-)
C\j K)
CQ 0 L) Cd
< Z N 0 e-, W z
0 _0 2 :D Tj 0
0 0 0 H
C\j N 0 ro Of
LLI
cq
0 0 LL
0 0 0 U- 0.1-1 4
K) re) -..0- 4 0 4
Cj Cl) C\j bo 0 '0
F
z 4-1
0 LU Co 0
-0 K) U) 0
z C\j
0
F- N 0
-0
o

C)
a) VI ;-4
0 co 0 z
Go 4-4 0
U2
0 co
-0 Lo
0 o r ':4
4-) Z k
0 0 C) (d cr
0 T -0 (-0
C\j
cz
t 0 LO o 1 0 %j-00 C -0 0 If) Q) cd


0 Cd E-4
on-I -- I -- I I I -n
0 oa) 0 0 0 0 0 0 o(o
cli C\j It C\j (D a)

S3HOiVVY HO Id JO 63Gvq nN =






13






rz m
0 LO 0

cli 4-4 Cq
Cl*j M 4J
0 0 -1 co
0 0 co
cli 0 > 10
N

0 CQ 4J 0
-0, co r. (D
0 p- F--f
P-A
4J 0
0 Po oo
"0 m 00 U)
0 -0 4J
cli 0 0 cd
N 0 14- N M
0 -0 ;i:
0 0 CD
C\j (\j
0 0 --P- 0 0 En gQ
0 0 z (D
w C\j
0 0 (D 0
rz
C) W C\,J bD
L -0 w Cj
I C\j LL
0

o
0 0
>- 4-)
c- G
< z
0 L) w
-02 M
c\j 0 OD Cy w w cq
LL cr :j 1-1
3 -0-44-LL 0 0
0 (1) cr)
Q- P- bD 0
z 4-4 14
w 0
ou r
-0 to- U) 0
z
0
0 0
&0 4005: to 0
w
00
-0 F
0 U) 0

0 0 >1
-0 0 u
cli ul 0 z
-H 0)
4- ::1
co cr
-8 10 H a)
>
Q) 4-1

0 p0 a)
C

Q) Cd

0 0 0 __o
C-i 0 V ow n bD (1)

S3HOIVIN HOIld JO H3eyqnN






14





0 0
C) 0 0
N 0 k cd
4-4

(D
-0 do rf) cd
LO >

N 0)
0 (D +-) 0
-0
r.
-H
CIq
41
0 1. 0
0 CJ


0 0
co cq
ff) r. Cd

0 >- OD
CJ z n 41
LLJ K) r.
M a) 0
cy 0 E o





t -00 (0 rn
ui 0 bO N
-0 cr 0 -0
LL K)
It (t) C)
0 Ld 0
-0 11- Cl)
C j M W >.
cr- N 0 4J < z
L) LAj

CQ CY
0 ui Q) 0 cr cr0 LL LjZ 0 0
-0- k 0
(D C\j
UD 0
z co
0 ui 0 4-1
0 L) 0) 0 (n
65 LO 9
z -.0
00 4-)
-0 OD 0
0
m
"a m 0
0 0 LAJ 0 E ci
(\j m 00 C) 4)
0 V C\j LO ril
Cj 0 0
0 C\l 4-4
0 .0 0 (o -4
0 0 -0 K)
C\j 0 n Z 0 .4
C\j ci
1-1 IN 0 9
0 0 0 -H 4)
0 0 0 0 It 4- :
0 0 0 -0 co V-0
ro C'j 4 Q) C)
> k cq
4) 4-1
< 0 a a +-)
0 co
00 0
-4

0 C\l 4J
1 0 0) 4) CO
0 0 0 0 0 0 0 0 0 0 0 0 8 0 57 0 OD
0 0 0 0 0 0 0 LO 0 to to C\j
bjD Q)
-H 4 0
S3HOIVVY HO.Lld JO 83a vi nN 1:4 41













C\j 0 0 0
qt 0 Lo
1 0 LT)
N cd

0 U) .0
==i -O It rH 4) 1
co r- co 41
0-0 > Q

(0 0 0
0 -0 0 0 0
C\j (0 -H Q

.4..) cd
0 LC) r 4J
-0 n Q) m
0 U)


0 cq
-0
OD It

0 >
-oz C\J 41 0
w V"
D
CY E
OW co bD 0
0 at t- lu Cl)
U. K) :j ,
In

m m ow r- z
a a -0 F=
0 cli w 4-j
to 0
LO z
0 0 w
0 0 -om o M
N C\j 0 M CY a) 0
w
0 0 cr- cr
0 0 o LL f,- U- 0 0
K) -0- (D $.4 C
Cj U) C\j U cy)

co z 4OW 00 0 W
00 n r.
It cli cn 0
c z
c 0
0- 0
C-O L to 5; N r
-4 0
w E C-)
00 a) rj) -4
-0 OD
OD 0
4-4
0 OD
-0 (D r.
0 0
c -H
0 4-)
0 0 0 co v
-0 co 0 -0 Q)
(D to cli >
Q)
0 LO 0
0 8 0
ODA 4C:l

0 Q) cd
8 0 OD
0 0@00 o .0 o 0 o(o
10, cli bD Q)
-H 4 C)
S3HOiVIN HOild AO 83evqnN 44 4J 0






'16






0 Ul) 0 0
Itr k LO
c !i N 4-4
Cj
w co
0 0
C- 0 0 (d
N 0 >


o ODi +-)
0 1,,0
C)
co w Cd
-0 L 0 +-)
0 00 co
LO m
0
0 -0
0 rL 0 CD
Cj
0 0 C-) 0
0 0 Z (D 4J
0 0 Ld C\j 0
0
D rz CNI
CY
LU C\i bO
0 cr
C\j LL

ui 0
-0 EE 0 cr 0 4J Cd
0 H
< Z Q4
C) LLJ
0 'o
o 3):3 Q)
N 0 OD Cy cr LU Q, 0
LL X :I cq
o" -t LL 0 ",
-0 cf) $-4 0
NO c)
0
4-1
0 LLI 0
0
-0 0 0
z r- .,A
0 4-)
0 0 -H
F- 41 -0
-0< Pr) OD
H 0
Lli
013
-0 m
0 0
Z 4-4
-H
d-O 0
0 0
Lo 0

0 LO w cr
-0 -rl Q)
0 lq- > ;A
0 Q) 4 1
00 n
CV 0
C 0
0 0 0
0 Q) co
0 to -4 u
0 0 0 0 0 '0 0 0 0 OD :1
Oo (AD It cli U Q)

S3HOIVA HOIld jo 83evqnN ;%4 4J 92






17





0 F;i d
0 4
co
4-1
U)





C)
OD .,1 0
CQ
4-)
C\j 0) C)
C

0 0 c
0 00 N rd
r- r.

>- co
00 (D
-0 z ro 4C\j uj to

cy 0
Ld 0
-0 cr 0
LL ro

0 LLJ
C"j (0 >- 4< z
0 0 LLJ
-02 ro :D Z
lq- 0 N CY Q CD
Ld Q4

0 LL U_ 0 C
U) bD o

C) LLJ 0 0 rn
0 a) r.
co 0
z .,q
C) 0
-0- 0
F- 00 40 < W
> 0
m o Ld o
a I_ co 00 o
0 N Ln W
0
0 Z 4-4
0 In
0 0 0 (0
0 -0 ro
0 0
C\j N cli 0 z

0 0 0 0
0 0 0 -0 C\J cd
0 0 0 .,q
K) >
I Q) 4-1
< z
-0 0
CC) 0 Q

0
C\l cd C)
0
0 0 0 0 o o o 60
U') 0 0 0 0 0 Ln Lo 0 LO 0 LO C\j
ro (D -4
4 0
S2HOIVA HO.Lld JO 838wnN P4





18



carrier frequency in 20-cent intervals. A second scale is included on each of the figures to show the frequency (in Hz) of the matches. The arrows found above the histograms in the figures represent the locations of spectral components in the electrical signal when the modulation frequency is 200 or 2000 Hz. The arrows found above the histograms of data gathered with a 20-Hz modulation frequency only represent the 3dB-down points on the sharp slopes of the very narrow band spectra.

Effects of modulation frequency (20dB SL). Generally, pitch matches to 20dB SL stimuli modulated at a 20-Hz rate lie within the frequency band containing the spectral energy of the stimulus. At the 300-Hz carrier frequency modulated at a 20-Hz rate (300/20) the majority of the matches (Fig. 2A) are distributed between the carrier frequency and the upper extreme of the band of spectral energy with relatively few matches occurring to frequencies below the carrier frequency. The distribution of matches to the 1000/20 (Fi'g. 3A) stimuli has a prominent maximum near the lower frequency extent of the energy. band and a second less prominent maximum near the upper frequency extent of the energy band. The distribution of matches to the 3000/20 (Fig. 4A) stimulus condition is approximately unimodal although skewed from the lower extent of the energy band. An abrupt decrease in number of matches





19



at the upper and lower extents of the energy band is also evident.

The distributions of pitch matches to the 300/200

(Fig. 2B) and 1000/200 (Fig. 3B) conditions ar.e multi-modal with definite maxima occurring at the frequencies of the spectral components present in the stimulus. The single exception occurs in the 300/200 condition (Fig. 2B) where, in addition to pitch matches to the spectral components, pitch matches corresponding to the modulation frequency are found although no spectral component exists at that frequency. This exception will be discussed in detail later. The distribution of matches to the 3000/200 (Fig. 4B) conditions had a large maximum centered about the carrier frequency and some indication of secondary maxima near the upper and lower extents of the energy distribution. The 3000/2000 stimulus conditions elicited pitch matches whose distribution exhibited a large maximum at the carrier frequency (Fig. 3C).

The general character of the distribution of pitch

matches to stimuli modulated at a 20-Hz rate did not change as the carrier frequencywas increased. The pitch matches generally fell at or within the limits of the spectral bandwidth. As the carrier frequency was increasedthe distribution of pitch matches to stimuli modulated at a 200-Hz rate

changed from multimodal with definite maxima at the frequencies





20



of the spectral components of the stimulus to unimodal with a maximum at the carrier frequency in the case of the 3000/200 stimulus. The 2000-Hz modulating frequency was used only in conjunction with the 3000-Hz carrier frequency because it is not possible to modulate a signal with one of higher frequency. The distribution of pitch matches elicited by the 3000/2000 stimulus condition exhibited very definite maxima at the carrier frequency and at 1000 Hz, the frequency of the first spectral component on the low frequency side of the carrier frequency. Pitch matches did not occur to the first upper spectral component of the 3000/2000 stimulus.

Effects of modulation frequenc_y___(50dB SL). An increase in the intensity of the stimulus from 20dB SL to 5OdB SL produced no change in the major configuration of the distribution of responses to the 300/20, 1000/20, 3000/20 and 3000/200 stimulus conditions although the increase in intensity did tend to restrict the frequency extent of the matches. The pitch matches remained at or within the extent of the spectral bandwidth. The increase in stimulus intensity from 20dB to 5OdB SL for the 300/200, 1000/200 and 3000/2000 conditions resulted in pitch matches to additional audible spectral components to which pitch matches did not occur at the low stimulus intensity. The additional pitch matches occurred to a spectral component on the high side of the energy ba-nd in the





21



cases of the 300/200 and 1000/200 conditions and to the spectral component on the low side of the energy band in the case of the 3000/2000 condition.

Subject variability. Examination of the pitch match data of individual subjects reveals a marked difference between the kinds of pitch match distributions obtained from subjects JH and CH and those obtained from subjects DF, NM and TNI. The pitch match distributions produced by JH and CH were consistently unimodal and centered on or near the carrier frequency while the distributions of pitch matches for DF, NM and TM were generally more representative of the spectral energy distributions of the respective stimuli.

Between-subject variability is illustrated in Figure 8. The distribution of pitch matches elicited by the 1000/200 stimulus presented at 50dB SL from each of the five subjects is shown on a common frequency deviation scale with the frequency location of the spectral components of the stimulus shown by inverted arrowheads. Subjects DF, NM and TM exhibited definite groups of pitch matches to 3, 7 and 2 spectral components respectively, while subjects CH and JH did not exhibit well defined groups of pitch matches that could be related to the stimulus spectrum.

The most common pitch judgments which occur consistently enough to produce discernible maxima in the distribution of pitch matches to other than spectral information







22




C\j
co



co
Go

0
A 0


0
-0



cli
C\j

c

-0 LLJ C)


Lli
0 ir F
-0 LL C\j co -+J


-0 CC C))

< 0
0 z uj
0 r- Z)
0 2 0 (D 0 r- Or x LLJ -4-)
LL
0- 0 U0 U) ro 00 F- (D



0 Z
0
U- 20 0 0
0 0 m 0 0
C\J -0 C\j Lo
1 1 0 co LLJ
I m m 0 LO
t M 0 0 0
U) 0
US
0
c


ro

rr)
(D C\j C\j
0 CC)

co
C\j
00


0 0 0 0 0 0 0 0 0 -0 pC\j 0 0
w 04
S3HO.LVIN HOIld A 0 J 3eAnN






23



were judgments representing octave relationships to spectral components. Octave judgments, or more commonly, octave "errors," occurred at 150, 200 and 400 Hz in the 300/200, 20dB condition (Fig. 2B) and at 2000 and 4000 Hz in the 3000/ 2000, 5OdB condition (Fig. 7C). Constant errors, of course, were also found to exist.

















CHAPTER IV

DISCUSSION



Frequency-modulated sinusoids have been shown to elicit a variety of auditory sensations whose character varies as a function of modulation rate. Very slow modulation rates have been used to study differential thresholds for frequency (Shower and Biddulph, 1931; Filling, 1958; Brandt, 1967). Utilizing a low modulation frequency (2-3 Hz), the just noticeable change in pitch of an FM 1000Hz sinusoid occurs when the extent of modulation ranges from

- -5-10 Hz. In these experiments the pitch change is associated only with modulation of the carrier frequency. From modulation frequencies of one Hz to approximately 10*Hz, a listener hears a single tone rising and falling in pitch at the modulation rate. A modulation frequency of 20 Hz elicits perception of a complex warbling or fluttering sound. The warbling changes to a roughness and disappears as the modulating frequency is increased, leaving the perception of a complex auditory signal. The pitch (es) of FM sinusoids with modulation frequencies greater than 20 Hz become dependent

upon the parameters of the signal.



24





25



The spectral distribut ion of the energy of an FM signal is a function of the modulating frequency in conjunction with the modulation index. The frequency spacing of the spectral components (sidebands) is determined by the modulating frequency and is equal to it. The amplitude of the various spectral components is dependent upon a complex function known as a Bessel function. The bandwidth of an FM signal is defined as the width of the frequency spectrum that contains all components having an amplitude of 1 percent or greater relative to the amplitude of the unmodulated wave (Sheingold, 1951). The number of significant spectral components in an FM signal increases when the modulation index is increased.

Pitch matches to the modulating'frequency of a quasiFM signal have been obtained CRitsma and Engel, 1964). The stimulus used to elicit these pitch matches had a ratio, n, of carrier frequency to modulating frequency of five or six and a modulation index of 2.55. The investigators stated that the modulation index should be as large as possible and that n should be small (less than five or six) to produce pitch matches to the modulating frequency. In addition, it might also be stated that the carrier frequency must be 2000 Hz or higher. Carrier frequencies, modulating frequencies, n's and modulation indexes of the stimuli used in the present investigation appear in Table 2. An asterisk beside an entry





26











Table 2. Frequency modulation stimulus parameters.



Cf f
Cf (Hz) Mf (Hz) n Mod. Index Afa
M f Mf


300 20 15 5*

300 200 1.5* .5

1000 20 50 5*

1000 200 5* .5

3000 20 150 5*

3000 200 15 .5

3000 2000 1.5* .05



a f measured one way from the carrier frequency was always 100 Hz in all stimulus conditions.





27



in the table indicates that the associated tabled value meets the criterion of Ritsma and Engel for the production of pitch matches to the modulation frequency. Inspection of the table shows that none of the stimuli used in the present investigation meet both of the Ritsma and Engel criteria necessary to elicit pitch matches to the modulation frequency. However, in the present experiment, one experimental condition produced pitch matches which seemed related to the modulation frequency.

The 300/200, 20dB SL stimulus condition produced a

substantial number of pitch matches centered on 200 Hz where no spectral energy was present in the stimulus (Fig. 2A). Although periodicity is a possible explanation for the matches that occurred at 200 Hz, it is difficult to reconcile such an explanation with the fact that the experimental conditions do not meet the conditions cited above as requisite for such pitch judgments. An alternative explanation follows.

While not as prominent as the pitch matches at 200 Hz, substantial groups of matches also occurred at 100 and 400 Hz. Since 200 and 400 Hz are the second and fourth octaves of 100 Hz, there is the possibility that the groups of matches to 200 and 400 Hz may represent octave errors of the 100-Hz spectral component which is present. In addition, the clusters of pitch matches which occur at 200 and 400 Hz in response





28



to the 300/200-Hz condition at 20dB SL are generally absent in the distribution of matches obtained at 50dB SL. Although the 100-Hz spectral component is audible at the 20dB SL presentation level (Appendix A, Fig. 11B), its increased intensity at the 50dB SL presentation level may permit more accurate pitch judgments and reduce the number of octave errors. It should also be noted that almost all of the pitch judgments at 200 and 400 Hz at 20dB SL are the result of subjects DF and TIM (Appendix B, Figs. 17 and 25 respectively).

Pitch matches to the 200/20 condition lie within the frequency band of the spectral energy of the stimulus, but the majority of the matches occurred at frequencies between the middle and the upper end of the spectral energy distribution. This result can be explained on the basis of the absolute sensitivity of the ear as a function of frequency (Fig. 9) where the sensitivity of the ear decreases rapidly below 300 Hz and continues to increase from 300 Hz up to about two kHz. Since the spectral energy distribution of the 300/20 condition is relatively symmetrical about the carrier frequency, the sensation level of the stimulus will be slightly higher on the high frequency side of the energy band, where the matches predominantly occur.

The 300/20, 1000/20 and 3000/20 stimulus conditions

have spectral energy distributions whose individual components







29


















4-4










4-4 W 0


0
4J






CQ




co w N

W 4J
cli
0 4-4#
0
CL

LLJ 1-4
> ui .,q
cr 4J U)
w -H .14
U- 0 >
AZ Lu co

C*j


co
a)

0 0 0 0 0 0 0 0. Q)
OD to N cli $.4 G)
>

41 (zAO/3N),o i =ep O M ) EIP NI a inm ld w W






30



are spaced at 20-Hz intervals. The frequency extents of the energy bands of these stimuli are marked by quite abrupt decreases in spectral energy (Appendix A, Figs. 11A, 12A and 13A). Pitch match data obtained by von B'ekbsy (1963) in response to band-pass filtered noise, and by Small and Daniloff (1967) in response to low- and high-pass filtered noise show the pitch of these signals to be related to the end or ends of the noise band where the steep gradient in energy vs. frequency occurs. Data from the present experiment also exhibit pitch matches in response to the ends of the energy, band and suggest the possibility that such pitch matches might be the result of a sl -ope detector or peripheral "edge effect" and neural funneling in the sense of von Bekesy (1960).

An alternative explanation which seems to incorporate all the data may be found in the concept of the critial band hypothesis. Consideration of the critical band data of Plomp and Mimpen (1968) (Fig. 10) shows the critical band, or minimum frequency separation necessary for spectral components of a complex signal to be heard individually, to be approximately 60 Hz. The spectral component separation of the 300/ 20, 1000/20 and 3000/20 conditions is only 20 Hz. A critical band hypothesis would predict that the individual spectral components of such signals could not be heard individually





31

















5




2

N-/
0s





z
Li
LJ



LL





5



2




2 5 102 2 5 105 2 5 104
FREQUENCY (HZ)

Figure 10. Critical bandwidth based on the minimum frequency difference between the harmonics of a complex tone necessary for them to be heard separately, as a function of frequency (after Plomp and Mimpen, 1968).





32



and that pitch matches in response to such stimuli should, therefore, not distribute themselves according to the individual spectral component frequency locations of the signals. Indeed, pitch matches in response to stimuli with a 20-Hz modulation frequency used in the present investigation do not distribute themselves according to the individual spectral components present in the stimuli as expected from the critical band hypothesis. Responses to the 1000/20 stimulus condition distribute themselves bimodally with a mode occurring at each end of the frequency band occupied by the spectral energy. The mode at the high frequency end of the energy band was more pronounced for responses to the 5OdB SL presentation level (Fig. 6A) than for responses to the 20dB SL presentation level (Fig. 3A). The critical bandwidth at 1000 Hz is approximately 175 Hz. The bandwidth of the stimulus energy was approximately 500 Hz. The bimodal nature of the response distribution to the 1000/20 stimulus condition could result from activity in the adjacent critical bands above and below the one containing the carrier frequency of the F11 signal. The distribution of pitch matches to the 3000/20 stimulus condition was broad and unimodal with an abrupt decrease in number of matches above and below the frequency extents of the stimulus bandwidth (Figs. 4 and 7). Critical bandwidth at 3000 Hz is approximately 500 Hz, which coincides with the






33


bandwidth of the 3000/20 stimulus and confines the entire stimulus complex to one critical band.

The critical bandwidth is less than 200 Hz at 300 and 1000 Hz. With one exception, previously discussed, pitch matches to the 300/200 and 1000/200 stimuli formed distributions representative of the audible spectral components present in the stimuli. An additional spectral component became audible at 900 Hz when the 300/200 stimulus was increased from 20dB SL to 50dB SL, and at 1600 Hz when the 1000/200 stimulus was increased from 20dB to 5OdB SL (Figs. 2, 3, 5 and 6). Pitch matches occurred at the frequencies of these components when the higher (5OdB SL) stimuli were presented, but not when the 20dB SL stimuli were presented. These responses suggest a long term frequency analysis as a basis for the pitch perceptions elicited. At the 3000-Hz carrier frequency, 200 Hz is less than critical bandwidth while 2000 Hz exceeds critical bandwidth. Pitch match distributions in response to the 3000/200 stimulus at both the 20dB SL and 5OdB SL presentation levels are similar in character and frequency extent to the distribution of matches obtained in response to the 3000/20 stimuli. The 3000/2000 stimuli, however, elicited response distributions similar to those obtained in response to the 300/200 and 1000/ 200 stimulus conditions. At 2OdB SL presentation level, only





34



the spectral component located at the carri er frequency was of sufficient intensity to be audible. The response distribution consisted of a single sharply-frequency-delimited cluster of pitch matches located at the 3000-Hz carrier irequency. When the presentation level was increased to 50dB SL, the 1000-Hz spectral component was sufficiently intense. to be audible, and a second well-defined group of pitch matches occurred at 1000H Hz. The lack of pitch matches to the 5000-Hz spectral component can be accounted for by the rapid decrease in the sensitivity of the ear between 3000 Hz and 5000 Hz (Fig. 9).

Spectral components of the stimuli used in the investigation by Plomp and Mimpen (1968), from which the critical band data of Figure 10 are taken, did not vary in frequency as a function of time. The individual spectral components that made up the various test stimuli used in the present investigation varied in instantaneous frequency as a function of the modulating frequency and the frequency swing of the carrier. Frequency deviation was held constant at

+1-00 Hz so that all of the spectral components of the stimuli were varying sinusoidally over a 200-Hz frequency extent at the rate of the modulating frequency. When the spectral component frequency spacing of the FM signals used in the present investigation exceeded critical bandwidth (Fig. 10)





35




pitch matches were made to the averaged frequency location of the respective frequency components. Such pitch matches suggest that the ear is able to perform a long-term spectral, or Fourier, analysis upon complex ongoing auditory signals provided that the spacing of the components of the complex exceeds the resolving power of the ear's analyzing system for the frequencies involved. A system of filters is the hypothetical mechanism of choice. The critical band is a measure of the limits of the resolving power of the ear's filter system. Pitch match distributions in response to each of the experimental stimuli whose spec-tral component frequency spacing exceeded critical bandwidth were based upon the frequency locations of the spectral components. Pitch match distributions in response to each of the experimental stimuli whose spectral component frequency spacing was smaller than critical bandwidth were based upon the extents of the energy band, the center of the energy band or some other aspect of the energy distribution not related to the individual spectral components.

The amplitude of the various spectral components relative to the sensitivity of the ear at the frequency locations of those components determines the particular 6omponents that will be heard provided that critical bandwidth spacing is exceeded. In every case, when stimuli whose





36



spectral component spacing exceeded critical bandwidth were increased in amplitude by 30dB SL, an additional spectral component became audible and pitch matches occurred in response to it.


















CHAPTER V

SUMMARY



Frequency-modulated (FM) and quasi FM signals have been used as stimuli in psychoacoustic experiments in the study of frequency difference limens, pitch of the residue and phase perception and have been demonstrated to elicit a variety of pitch perceptions dependent upon acoustic characteristics of the signal used. FM sinusoids have been shown to elicit pitches related to the carrier frequency and modulating frequency. This investigation determined the pitch or pitches elicited by FM sinusoids as a function of the carrier frequency and the modulating frequency at two sensation levels of the signal.

Fourteen stimulus conditions were used. Stimuli consisted of 300-, 1000- and 3000-Hz sinusoids, frequency modulated at modulation frequencies of 20, 200 and 2000 Hz (only for the 3000-Hz carrier frequency) The frequency deviation in Hz from the carrier frequency was always +100 Hz. Stimulus intensities of 20 and 50dB relative to the subject's threshold were used. Five normal hearing adults who demonstrated

the ability to pitch match to the unmodulated carrier frequency 37






38




with no more than one semi-tone frequency error were used as subjects. Stimuli were presented monaurally. Subjects were permitted to listen to the stimuli as long as they wished, and to switch back and forth between the FM stimulus and matching signal (an unmodulated sinusoid) as often as they wished before they indicated a match.


In general, the nature of the pitch match distributions obtained in response to FM sinusoids was dependent upon the frequency spacing of the individual spectral components of the stimulus. Pitch matches occurred to individual spectral components when spectral component frequency spacing exceeded critical bandwidth. When spectral component frequency spacing was less than critical bandwidthpitch matches occurred to the center and/or the ends of the band of spectral energy.

While an increase in the intensity of stimuli with

less than critical bandwidth spectral component spacing changed the pitch match distribution to that condition very littlethe same increase in the intensity of stimuli with greater than critical bandwidth spectral component spacing caused an additional component to become audible and pitch matches occurred to the additional component. In only one stimulus condition did the listeners not match to spectral energy. Such matches





39



corresponded to the modulating frequency. With all of the data considered, these matches were explained as octave judgments to a spectral component. Thus, no pitch matches occurred to modulation frequencies but rather to spectral energy present in the acoustic signal.

All of the individual spectral components that made

up the various test stimuli used in the present investigation varied in instantaneous frequency as a function of the modulating frequency and the frequency swing of the carrier. When the spectral component frequency spacing of the FM signals exceeded critical bandwidth, pitch matches were made to the averaged frequency location of the respective frequency components. Such pitch matches suggest that the ear is able to perform a long-term spectral, or Fourier, analysis upon complex ongoing auditory signals provided that the spacing of the components of the complex exceeds the resolving power of the ear's analyzing system for the frequencies involved. A system of filters which increase in bandwidth with frequency in the same fashion as traditional critical band analyzers is the hypothetical mechanism of choice.





































APPENDICES


















APPENDIX A

ELECTRICAL SPECTRA AND
EARPHONE FREQUENCY RESPONSE CURVES



Electrical spectra of the experimental stimuli were obtained using a wave analyzer (General Radio, 1900A) and a graphic level recorder (General Radio, 1521B). The wave analyzer bandwidth was 10 Hz. The recorder chart speed was five inches per minute and the recorder writing speed was 10 inches per second. Zero dB was equivalent to the amplitude of the unmodulated carrier signal (Figures 11, 12, 13, 14).

Earphone frequency response curves for the two telephonics TDH-39 10Z earphones were obtained using a beat frequency oscillator (General Radio, 1304B) as a signal source. An artificial ear (Braiel and Kjaer, 4152), condenser microphone (Brdel and Kjaer, 4132), cathode follower (Briiel and Kjaer, 2163) and microphone amplifier (Brael and Kjaer audio

frequency spectrometer, 2112) supplied the input to a graphic level recorder (General Radio, 1521B) from the acoustic output of the earphone (Figures 15, 16).




41





42




-10


-20


-30 A 300oo/20o



LtJ
-40
O

-50
C.

-60


-70


-80
t J i i 1 I ji Il I l I
0 500 1000 1500 2000
FREQUENCY (HZ)



-10


-20


-30 B 300/200


-40



-50


-60


-70


8 0 1I I I l l l lt i l
0 500 1000 1500 2000
FREQUENCY (HZ). Figure 11. Electrical spectra for 300/20 (A) and 300/200 (B). Analyzing bandwidth was 10 Hz.





43




-10 -20


-30 -A 1000/20


-40


-50
0..

-60


-70


-80
0 500 1000 1500 2000
FREQUENCY (HZ)



-10
IO











-20


-30 B 1000/200


H
U-40 S-50


-60


-70


8 0 1 1 II I 1 1 1 1
0 500 1000 1500 2000
FREQUENCY (HZ)

Figure 12. Electrical spectra for 1000/20 (A) and 1000/200
(B). Analyzing bandwidth was 10 Hz.





44




-I0


-20


-30 A 3ooon000/20
-o
-40
IuJ

M
S-50
a.

-60


-70

-80 -JAI
2000 2500 3000 30 4000
FREQUENCY (HZ)



-10


-20


-30 B 35000ooo/2o00


-40


7-50
a.

-60


-70


-80 I I I F- 1 ... .. r I r lj
2000 2500 3000 3500 4000
FREQUENCY (HZ) Figure 13. Electrical spectra for 3000/20 (A) and 3000/200
(B). Analyzing bandwidth was 10 Hz.






45




0
-0
LO




0
0
LO
0 0

0 H
0 W
0 0 m
C>
0








bD

N


0
-0
00
z Ld C :D C
LU
0 L.L
-0
U-) C\j cn
4
0



0
0 cli





0
-0
U-) C-)


L

0
-0
0

0 0 0 0 0 0 0
7 LO (0 co
I I I

(ep) 3(3nindAv






46









F 0























ro N co
9

0 0 4-1
-7


C3,
Lc) LLJ LL


0 Ul

















Cd
9 U*)






0 0 0 0ro
0 0) co rl- o LO to C\j U

-1 d S G P






47







0







LO










CQ


0


w


0
z Ld CD LLJ LO = u
LL U2

0












4-1









Ul)






0 0 0 0 0
0) co n N

-ldS SP 44

















APPENDIX B

INDIVIDUAL SUBJECT PITCH MATCH DATA



Pitch matches are plotted as deviations in semi-tones of pitch judgments in 20-cent intervals from the carrier frequency for subjects DF (Figures 17, 18, 19, 20), NM (Figures 21, 22, 23, 24), TM (Figures 25, 26, 27, 28), CH (Figures 29, 30, 31, 32), and JH (Figures 33, 34, 35, 36). The arrows .represent spectral information as explained in the text.



































48






49






rZ
a) Cf)
4
4-)
-0
0 LO

m m
0 0 -0 -a -0
LO C\j 0 (\j 0 U)
0 (D
LO C\j co
0 0 LL 0 LO >
0 0 m
0 0 ro
N C\j 0 C)
1-1 LO
0 0 0 0 LM
0 0 0 0 :D
rr) K) ro ro U) 0 0 C)
-0 r- N
CC)

0
z Cl)
-0 C\j
Ld
:D
C-r a
0 Lli co
-o =
q. ro


LIJ
C-C) rr) >- E
C\' n 0 U
< z Lo
0 Lli r
0 n 0
0 0 cy
0 n

4-) 7:5
0, U- H z 0 U) (D 0
C\j F- cj

C:El Lli
00 co 0
-0- ro U2 q-4
It z C\j
0
0 g:
0 C\l
0
0 +1
> U
LLJ F 0
c C: M (a) *n
0 m
1-0 00
Go U2

0
-0 OD r 0
0 0 0 ;-q
-0
0 0
o
10
0 0 0
0 0
(D o
-0
0 c 0 It
-0 -0
Co 2

COL OD
0 0 0 0 0 0 0 0 0 0 0 02
N 114- cli

S3H O.LVV4 HO-Lld jo 83eNnN,






50




0 C\j
0 co



co
co LO

m 4-1
0 m -0
0 0 m U)
C\j "0 -a
0 E 0
C\j >
0 L U- (D
0 0
N CQ N c
C\, CN
I-D N
0 0 M 0 C\j
CIJ >- C)
C) M co 0
z 0
t 'i
0
0 0 0

bi
C
0 UT co

o

u
C:E r- F -i
Z
Li
0 0
I-00 w x

0- 0 U0 CO n (D 0
z
LLI
0(-) 0 -q
0- 0 0
0 z w 4-1

C) 0 0 a
0 0
C\j LO Lli
0
0



c 0
(D to 0 ;
-j 4,
4
0 ca >,
-0 C\j rq Q
w >
C\j C\j C. a)

0 CC)
-0
OD
C\j 00 4-1


-0
0 0 0 0 0 0 0 0 0 0 0 00
IT C\j CQ N ro

S3HOIVIN HOIld JO 63etNnN P-A u






51




0 0
0 (D Cl)
LO
Q) 4

0 0 Lo 41
0 0 K) LO
0
C\l 0 75

0
0 0
C\j
CQ

(D > 4
410 -0 q*- ; cd
0 ro a)

E r. C
0 >- C\j
-00 (D co z r 4-)
Ld C)

00 K)
C) LJ qt
CD X C\j 0
LL Cd
C) (]!: 0 z
0 Ld


o < co r 0
m 0 (D >- 0 0
rr)
N,) C-)
0 2 z
LO C\j 0 0 Ld
= o =) r 0
0 o U- -0 U- cy -4
0 N 0
0 L11 V 4
C\j C\j
"I 0 0 7)
0 0 0 0 m 0. LL
0 0 0 o =) -0 Z (.D
0 --- >- Cl) N LLI C\l

0- 0
OD 0
0 Z ro
C) C\J




LLJ

0 0
-0 0)
co
co U2


0 0
0 0
0 0 0 ;
H
0 +0 0 cd
0 0 0 H
C\j LO >
C) C)
0 0 (D
L- o -0 ro a)
0 K) ;
CQ (, 4r-i
0 4zl1-0 CQ a) C)
0 0 0 0 0 0 0 0 0
ho ;.4
.,I cd
S3HOJ.VN HO Id AO 838vq nN






52





0


4

rr)
LO 0
m 4-4
_0
0
C) 0

>
0 0 0
0 C> LL 14t
0 C) C\j C) LO
(D I;t 4CQ
C) 0 0
0 m C-) 00 co
Z r
0 LLJ ro
to ro V)
0 OD
0 LLJ 0
0 = n CD Cd
C\j LL '0 N 0

0
0
o uj o

ro c
0 < f,- 0
ou CD >- E Cl)
Cj 0 bz
z
LLJ
M =) 0
0 LL ro Cy N LLJ

r
0
C) C\l Q 0
Ld C\j
0 0
C) 0 0
0 0) Tn ,, 1
co CC) 0
0 r
0 *" +
0 < OD 1 +
C) 5 w ri L)
LLJ
0 0 0 0 C\j X)


0 z 0
-0 ro 0 ;,
ro -,1 4-4

co
0 H
0 C\J >
10 C)

0 C)
0
co 0

0 C)

0 0 0 0 0 0 0 0 0 co
It C\j 0 Go 14- C\j C\j U

S3HOJ-VVI HOIld JO 838A nN






53




0
-0 LO
T- Q0
0 Cl) C10


C)
(D
CQ n
0 CT-4 75
0 0
_:;Lo -0 z
C\j
CO
oj
C)
"o C\l
0 LO
r:
-C\j 0
C\j 0
0 4 CO
0 z

co


Z
UJ C\J 0


0 LLJ
0 =
LL

C)
N ro b.0
< z
LLJ 0
-0 0
0 rc) CY 4 4-) or_ Lli
LL -H
C)
0 CO (D 0
C\j C\l u
z 4-1
LLI 0
m co 0
-0 K) ul
0 0 C)
LO N 0 Z C\l
LO 0 +)
0 0 0
0 0 0 0 < I
C\j C\j C\l N w > C\j C-)
1-1 1 1 1-1
0 0 m
0 0 0
ro re) 0 ul Q
-0 co
OD


0 co
0 CD 0.H
4J
0 Ld >
0 0 0 .,.j C)
0 >
0 0 C', (2)
0 0
2 LO
CL 00 -0
0 0ro
0 -0

0:) ;
C)

U
0 0 0 0 0 0 0 0 0 o 1-0


63HOiVW HOIld JO E3ewnN





54




0 C\j
0 co
0

m 0
-r Go
X) 0
co LO 0
0 LO Oj
0 0 02 0
C\J CIA ci Z -0
0 0 0 > 4-)
0 0 0 4 m
0 0 0 0
-0 Q 41
CQ

0 C\d 0
0 C\j z C)
C\J >- a) C)

0
C=< C= LLJ 0
0 Cd
LL)
0
-0 ul
C\J OD 4 CD
Ld
0 cr
0 =
< C-)
L) z LLJ 'n 0 0 o =) H
0 r- CY 9
LL LIJ
0 0 LL 0
-0 CO CD
z
E.Li 0
0 0
0 ca q-4
w
2 z 4:2
0 0
0 0 4--)
0 < 0 4-)
C\j Lo U
>
LLI
0 0 Ul
0
It ;: ul


0
0
c

Cd >
0
0 >
C\j
(D
LN
0 01)
0
OD
Cl

0

0 0 0 0 0 0 0 0 0 00
N It C\j It
S3HOJ-VA HOild AO 83GmN






55




0 0
--o (D C/)
LO


0 Lo 0 n

(D 0
0 co m
0 _0
0 0 0 0 C)
LO cli 0 0 V) 0
C\j 0 0 0 (D
0 0 0 0 2 0 CD > 4
CQ C\l C\j 0 I;t 4 cd
C\j Z
I- _0 rr) (2)
0 0 4 0
0 0 0 m
0 0 0 n 0
K*) re) U) ou
CO Z
Di

0 Ld c\j 0


0 0 CO o
LJ cq
4-) 0
co z C)
< (D (a) 0
-OC-) n E co
C\j >ro C) bL z Z: U)
C) 0 :
0 LLJ r
0 LL 0 D r-C) 0 4-) Ld -H
-C 7;

0
Ld C\j
0 ;4
0
0 :z rr) M 4-4
C\J G)

0 Z
< .4-1
0 C\j fO > C\j
LLI
0

M OD


co
0 0
H
4-)
0 0 0 0 C\j LO >
C)

0 (.0
n
CY)

0
C) C) 0
0 0 0 0
C\l Nt

S3HO-LVV4 HO I Id AO J3eAnN pz,






56




0
0 0 0 0 cli C)

4
0
-0
LO 0
m
0
0 0 0
CC)
>
0 0 0 ;-,
0 0 -0 C\j (D o
0 0 Z (D


CD z
-0 4
Z r
0 D LJ ro r 0
K) n D Q) N
a CO u
0 LLJ (D 4Cli LL. N
0
0 z 0
0
C) LLJ 0
cr-- ro

C) < n ou C) 0
>- E c) u U z U2
0 uJ
or- CO =)
0 LL rr) C)r H C\J Ljj
U

0 C j

(D LLJ CIJ CH u
0 ;
0 0
-0 0) w 14-4
00 z CO
0
0
-4
0 < CC) I
o 5: u
ui a)
0 0 -n
-0 0 U) -0
C\j LO :l



0
-H


0 C\j
>


0
0 0
L
L-L
C\j C) Q
0 m .,i
_0 00
0 0 C) 0 0 0
N 0 co IQ, Cj U
-H cd
S3HOiVA HOIld JO d3eAnN ;.,, u





57




0
0 LO

0
co

0 0
0/ 0 0 LO
0 0
C\j 0
0 0 > cli 0

10
CO 0 LO cd
-0 >
0 n C)
4- CO
(D H
co

C)
0 1*-07, C\j
0 0 LLJ
0 0 M
Cr CN
0 LL, CD 01, F0 0 LL ro
0 0 r"
CL 0 C pr) >N C\j LLJ
C\j C\j 00 C\j = to 0 E
0 < z U
0 Lli -,I
Oj 0 Z) ::s z
0 c3r 0
0 K)
CY-1
LL crm In 0- t- U0 a =E=: -0 o Cl) (D
Lo C\j C) 0 Cj ci 0
LO z u
LLJ
0 0 0c) co 0
0 0 0 0 -0- n 0
CQ CQ 0i Q- Z C"J U 4-4
0 0 0 0 co
0 0 0 0 =) 0 CJ 0 E
rO K) K) (f) -0>
LLI

-0 CD
0 co

0 CD
-0
0 10
--->. 0
0
0 4-j
0 tt
>
0
22 t 0 0 0
0 K)
cli
Lo 4co N
C 0 co
C) C)
0 0 0 0 0 0 0 0 0 0 02
cli %3* CQ q,4- cli Od

CZ
S"MHO-LVA HOild JO ?J:32AnN





58




0 Oj
0 co
C ; 0 C) 'M
_C: 75
4j 0 0 Lo
1-0 co
co LO 0
4-1 W m m 0
13 -a 0 w 0
0
0
0 >
C\J --- >- 0 cd
0 0 -0
0 0 Nd- C\j
N C\j 0
C\j
0 0 o 0 C\l
0 0 0 co -0 C\j C)
C\j >0 0

Li 0
0
0 :D a 0

0
0
N co
ui C)

-0 Ir 71 Ul


QO r- CY Li
U
0- 0 LL H
-0 co n r).
Lo
co 0 u
z
L.Li 0
0 C-) 0
(D 4-4
0
0
0 E
0 0
0 0
U*)
Ld
M
0 LO








C\j


co
0
co
CQ

0
0
0 0 0 0 o 0 0 0 0 0 00
It- N ;a- C\j It C\i N ro
S3 H OiV IN HOild JO 838v n N






59




0 0 0 (D
LO





0 -0
m
_0 0 0
0 0 4-4 CO
Lo N 0 0
n cli C\j 0
ul C)
0
0 0 0 0 (D
C\j C\J 0
LO
C 0 0
0 0 0 0
ro rf) ro n a) C) 0 0
cc) Z rLd

00 n C) Cl)
0 LLJ 4T
0 r,- cli C pc U- I;T N :
0 a 0 z C
0 Lj ro N

0 < Go 4-3 0
-00 ro >- C) 0
to 0 CO

0 0 LLJ m
a 0 0 LL 0
LIJ
,C-, 4P -,53 a U -H
Lj- 41
C'J z
0
LLJ cli

0 L) 4- u
co 0 ;
-OZ ro 0
0

0 < C) C\J (D > cli
LLJ
m
0 0
0 0') ul Go co


-0 GO
0 (D r. 0
0
-H 4-4 rr 4-)
0
0
LO >


0



-0 C\j
0 0 0 0 0 0 (D
C\j C\j

S3HO.LVW HOild O 13eWnN 44






60





0
C)
C\j

C) CD 4-)
0
0
m m
-0 _0 C\j
0 0
LO co

>
0 0
-0
0 0 (D
C\j C\j 4-1
0 0
0 OD
0 C) 07. r0 0 M I;t
U) LLJ ro
OD
0 UJ QO
Cj or
U- ro N
0
0 11 0
0 C)
o J 0 ro ul
4
C) < n r.
0 C) C)
F co
C\j L) b.0 Z 'C LO 0 LLJ r
C,- co 7) o

LLJ
T
0, U- 4-1
L ILij C\J
0 40 0 0
O;z 0) 0
m 4-1
co C) 00 0

C) 0
C) < +)

LLJ -4 Q
E C)
0 a) *r-:)
0 0 C\j n

0
0 K)
0


0
0 C\j (D >
C)

0
C)
co 0

0
0
cc)
0 0. 0 0 0 0 0 N
14" cli C\j 0 OD N U ;-4

S3HOiVA HOild JO 83GAnN






61




0
-------------- I o l 0 LO



71
: L

co 0
N 0
-0 0
0
/C) VC0,
co
0 LO >
0
LO



co

u 0
CQ M
C\l

co r. Cd
H
LL ro
4J
m r
K)
K) C)
LO 0 <
LO Z '-C ul
C\l
0 0 0 0 LU :
0 2 :D .0
0 0 0 0 0
C\j C\j rr)
"I ct LIJ
0 0 0 0 M LL cr U
0 0 0 0 :D LL 4- -Z
U) -H
0 U) (D o
C\j cli

0 L) co 0
0 -0- re) 0
2 st Z W
r 0
I
0 U
L 0 C\j
0 o >
0 ui
0 m
C\j C)
0 C)
-0 OD
0 o 0 00
-0 0 0 Ul
S2 S2 0 0
IN 0 ro 0 CC) 0
0 C-0 (o 0 ;-4
2 0
0 0 0
E 1-0
r 22 sr 0 0
0 1 -0 -r-4
0 0 0 C\j >
o 0 0 -0
0 _0 0 0
C\j
ro
0 0 0
0 -0
\r N 0 0
C\l I f
0 0 0 0 0 0 0 0 0 0
C\j N C*j cli
U
cd
63HO-LVA HOild JO J3E[NnN






62




0
0
0


0
0
co LO

LM m 0

0
Lo C\j


0 0 C-0 (D
0 0 C) M
C\j C\j N 0 c
C\j z 0
1 1 1-1 [ o -H N
0 C\j
co 0 >- C\j 4-) 0
0 0
z 0) c
Lu 0 0 D C)

Lij

0 U
0 0) C)
c C\l x co N
I Li fr
o a,- 't Q < M 53- 0

0 Lli
0 0 D 0
= I"- CY c LL- LLJ 4.)
0 0 LL
-0 NO :
00 Z w 0
u
Sw Li
L <:) C) 0 ;.4
1-0 (D 0
0 z LO
0
0 0
o o

LLJ
0
0
0 LO





ro 0 q


0
(D
C\l C\l 0
;z
0 ry%
0
C\j

0
0 -0
0 0 0 0 0 0 0 0 0 00
C\j 14- C\j ICQ- N ro

S3 H OIVVq HO-Lid JO 63GAnN






63




0 0 C) (D
LO
r

0 LO 0 n v rl- LO
0
0 0 a 0
LO C\j 0 0 -0 0
LO N N 0
Q0
0 0
0 o 0
0 0 N 0
N C\j Od ro Cd
0 0 0 0 LO
0 0 0 CO 4 0
0 0 0 :D 0 >- C\J Z c
00 (D H
re) ro r?) ro 0 co z
It 4 0
uj
D
00 UJ
=
LL

0 0 0
0 LLJ co C)
ro

o 00 0
(o CD
0 rr) >- C) 0 0 = C')
z
C) 0 = 0
0 C) 0
CY
LLJ 4-)

0 U) U -H
U- 40
9 cli z w
LLJ C\j Q 4-4
0 E6 o
0 z ro 0 C\j

c C) < 0 u
0 C\j > C\j
LLJ -H U
C) 0 0
-0
co co

0
0 CO 0 (D

C) 0 E-0 0







0
-0 cli
0 0 0 00 0 0 0 00 0 2
CD (D It N Ka- cli C\l Cl
bio
-( cd S2HOiVA HO.Lld Ao i 3sAnN 44 u






6-1





0
0
C\,

0

(D

m
0
0 0 w
LO Cj co

>
0 0
0 -0 C\j a)
0 L) cli
0 0 CO
0 m- r- 4
0 jj ro C
ro a) C'l
co
LU 0 ro
of
ro

0
0
0 Ld 0 (N
FE r4)

o< ro C) c Fi co
C\j N U

LLJ
0 cr OD --D 0
LL rO 0 H
CIJ LJ -P

Ct) LL- 4- -C, C\j rj Z
0
LLJ C\l u
0 0
C) 0 C) 0) co z CO
0
C)
C) < OD C) 5: ID
LLI
ca
0
C"l LO

(D
n
ro
4-;

-H U C\j >




co 0


C) C\j

u co
0 0 C\j
co C\l co C\j U

S3HOiVA HOild AO 83ev nN






65




0
0
ID




-0 0cli
cli d N

0 0
-0 C; 0 -->- 0 V,)
N _0 0
Od N

0 r--q
0 LO C
-0 0 n >
CC) U
LO



CO co

0
0
C\J
ID 0 LJ
M

C)
co
0 z
Ll
of
w

ro >- C,
N rc) 0 = < z
0 Lij

t 0 C) 0
or_ ro

o- U- C) C)
0 4- co
r C\j oj m
z
Ld
C) C-) co 0
0 r1r) 41
m z N to
-0 0 C)
m
a 0
0 < C\J 0 0
0 4-)
LO cli
0 LLI
0 a
C\j C\j 0
0 0
0 C-1 0 0 0 0 m 0 co
o 0 0 ID 0 :D
n I K) rr) ro U)

nil 0 co
0 C CD
co

0 0 0
-0 0 0 0 LO
_0 0 C\j
C\j 0 >
Qo a)
0 C) 0
rr)
co N 0
cli 0
-00
0\ - 2 0
I C)I\ l 0
CD 0 0 0 0 0 0 0 0
N It

63HOIVW HOIld AO 13eAmN






66





0 C\j
-0 co DQ



0 4 0
0 co
LO
CD
c 0


m 0
0 Q0 C
C\j -0 N
0 0
C\j 0

0 0 10
0 0 0 ;
cli 0 CD
C\j

0 0 0 0 -D 0 C\j
0 0 0 0 m -0 C'J
CQ
0 0 0
z
LJJ 0
-0 0
ci 0 07
LLJ

0 LL

cli ir- co
1.30
0 r 0



Ld
C) o o 1:1 = 0 LL Lij 2: ;-j
C) L.L
C) Cf) rr) C() Q. 0

Lij
0 c) 0
0- 0
z Lo

0 0 0 n
C) 0
u)
Lij
m
LO

sa
LO


-0 z 0
c 0 0
.H 4-1

0
0 C\j
LC)
cli

rn
0
co
CNJ



0 0 0 0 0 0 0 0 00
cli IcT N It C\j C\j

SAHO.LVW HOIld JO 638wnN






67




C)
0
LO
r
a)
0 LO C
0 01)


0 0 0 0
0
Lo cl


0 r ) X)
C\j
0
C)
CC) C)
LLJ
0 Z) a co
Ld N I
= t- C
LL
0 Of 0 C- LIJ co -H 0
N
ul
co 4-j
0 < (D z C
ro U
z
0 0 Lij
C) cl 0 :D L- 0 CY LLJ
X: -4J
u
0 0 rr) LLC)
cli
Lij C\j Q

co 0
0
C) C\j LQ 4-i


0 C\j 0
>
CQ
m ui
m c = C)
0 0 0
0 0 0)
LO LO OD co

0
0 0 o 0
C\j C\J C-0 CC)
r- 0
0 0 0 0
0 0 0 m
0 0 0 M C) 0
rr) rr) ro rr) (n 0
LO



K)


0 co
0 C\j
0 0 0 0 0 0 0
CQ It C\j

cd
S3HOIVIN HOI.Id JO 83Gvq nN











0
0 0 0 0 C\j



-0
0 LO

C\j
-0 0
0 0 E- 0
LO cli

0 o 0 >
0 0 C\j ;-,
0 (D C) L)
CQ C\j

CD
0 m -oz
0 =) Ld
K)
co
0 a 0
-0 Lij cli = re) cq

0



CD

C) 0 C\j i cl,
2. z
0 w L,
0 = CD Z)
0 LL- rs-) c3r 0 CJ Lij

0 40 CJ -H
z C-1 0
(D LLJ C\J C)
0 0 0
0 a) OD z 00 ul 4-1
0 a)
C) t- ;
C) < co 05: (D
LLJ
0
0 0 0 0 C\j U)

0 CD
0 0

c
0
0 C\J 10 >
a) C) r) ::
0
0 0 00

0 C\j 6n
0 0) 1
0 co
0 0 0 0 0 0 0 C\J
co C\j co o N
63HO-LVN HO-Lid JO 8seNnN CZ
zq u
















BIBLIOGRAPHY


von Bek6sy, G., (1960).Experiments in Hearing (McGrawHill Book Company, New York).

von B'k6sy, G., (1963). "Hearing Theories and Complex
Sounds," J. Acoust. Soc. Am. 35, 588-601.

Brandt, J. F., (1967). "Frequency Discrimination Following
Exposure to Noise," J. Acoust. Soc. Am., 41, 448-457.

Ekdahl, A. G., and Boring, E. G., (1934). "The Pitch of
Tonal Masses," Am. J. Psychol., 46, 452-455.

Filling, S., (1958). Difference Limen for Frequency
(Andelsbogtrykkeriet-Copenhagen, I Obense).

Fischler, H., (1967). "Model of the 'Secondary' Residue
Effect in the Perception of Complex Tones," J. Acoust.
Soc. Am., 42, 759-764.

Goldstein, J. L., (1967). "Auditory'Spectral Filtering and
Monaural Phase Perception," J. Acoust, Soc. Am., 41,
458-479.

Greenwood, D. D., (1961). "Auditory Masking and the Critical
Band," J. Acoust, Soc. Am., 33, 413-417.

von Helmholtz, H. L. F., (1863). Die Lehre von den Tonempfindugen als physiologische Grundlage fir die Theorie der Musik (F. Vieweg and Sohn, Brauschweig), in (1954). On the Sensations of Tone, 2nd English translation by A. L.
Ellis (Dover Publications, Inc., New York).

McClellan, M. E., and Small, A. M., Jr. (1967). "Pitch
Perception of Pulse Pairs with Random Repetition Rate,"
J. Acoust. Soc. Am., 41, 690-699.

McClellan, M. E., and Small, A. M., Jr., (1965). "Time
Separation Pitch Associated with Correlated Noise
Bursts," J. Acoust. Soc. Am., 38, 142-143.




69





70



McClellan, M. E., and Small, A. M., Jr., (1966). "Time
Separation Pitch Associated with Noise Pulses," J.
Acoust. Soc. Am., 40, 570-582.

Mathes, R. C., and Miller, R. L. (1947). "Phase Effects
in Monaural Phase Perception," J. Acoust. Soc. Am.,
19, 780-797.

Plomp, R., (1964). "The Ear as a Frequency Analyzer,"
J. Acoust. Soc. Am., 36, 1628-1636.

Plomp, R., and Mimpen, A. M., (1968). "The Ear as a Frequency Analyzer, II.," J. Acoust. Soc. Am., 43, 764767.

Ritsma, R. J. (1962). "Existence Region of the Tonal
Residue. I ," J. Acoust. Soc. Am., 34, 1224-1229.

Ritsma, R. J., (1963a). "Existence Region of the Tonal
Residue II," J. Acoust. Soc. Am., 35, 1241-1245.

Ritsma, R. J., (1967). "Frequencies Dominant in the Perception of the Pitch of Complex Sounds," J. Acoust.
Soc. Am., 42, 191-198.

Ritsma, R. J. (1963b) "On Pitch Discrimination of
Residue Tones," Int. Audiol., 2, 34-37.

Ritsma, R. J. and Engel, F. L., (1964). "Pitch of
Frequency-Modulated Signals," J. Acoust. Soc. Am.,
36, 1637-1644.

Schouten, J. F., Ritsma, R. J., and Cardozo, B. L., (1962).
"Pitch of the Residue," J. Acoust. Soc. Am., 34, 14181424.

Sergeant, R. L., and Harris, J. D., (1962). "Sensitivity
to Unidirectional Frequency Modulation," J. Acoust.
Soc. Am., 34, 1625-1628.

Sheingold, A., (1951). Fundamentals of Radio Communication
(D. Van Nostrand Company, Inc., Princeton).

Shower, E. G., and Biddulph, R., (1931). "Differential
Pitch Sensitivity of the Ear," J. Acoust Soc. Am., 3,
276-287.





71



Small, A. M1. Jr. (1955). "Some Parameters Influencing
the Pitch of Amplitude Modulated Signals," J. Acoust.
Soc. Am., 27, 751.

Small, A. M., Jr. and Daniloff, R. G. (1967). "Pitch
of Noise Bands," J. Acoust. Soc. Am., 41, 506-512.

Small, A. M., Jr., and McClellan, M. E., (1963). "Pitch
Associated with Time Delay between Two Pulse Trains,"
J. Acoust. Soc. Am., 35, 1246-1255.

Stevens, S. S., and Davis, H., (1938). Hearing (John Wiley
and Sons, Inc., New York).

Tonndorf, J., (1962). "Time/Frequency Analysis along the
Partition of Cochlear Models: A Modified Place Concept,"
J. Acoust. Soc. Am., 34, 1337-1350.

Zwicker, E., (1962). "Direct Comparisons between the Sensations Produced by Frequency Modulation and Amplitude
Modulation," J. Acoust, Soc. Am., 34, 1425-1430.
















BIOGRAPHICAL SKETCH



Keener Delaney McClelland was born November 30, 1936, at Bristol, Virginia. In June 1954 he was graduated from Knoxville East High School., He. received the degree of Bachelor of Arts with a major in Geology from the University of Tennessee in March 1961. From 1961 until 1962 he served in the United States Air Force and was stationed in Germany. In 1963 he enrolled into Graduate School of the University of Tennessee, and held a traineeship from the Vocational Rehabilitation Administration until March 1964 when he received the degree of Master of Arts with a major in Audiology. From 1964 until 1965 he served as Clinical Audiologist at The Jewish Hospital of St. Louis in St. Louis, Missouri. In 1965 he enrolled in the Graduate School of the University of Florida, and held a fellowship in the Department of Speech while pursuing his work toward the degree of Doctor of Philosophy.

Mr. McClelland is married to the former Nancy Dianne Newton. He is a member of the American Speech and Hearing Association, The Acoustical Society of America and The American Association for the Advancement of Science.



72















This dissertation was prepared under the direction of the chairman of the candidate's supervisory committee and has been approved by all members of that committee. It was submitted to the Dean of the College of Arts and Sciences and to the Graduate Council, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 27, 1968









Dean, Graduate School











SUPERV (R COMTTE

CL6 an