Journalof Voice

Vol. 12, No. 1, pp. n.n. "!.9

© 1998SingularPublishingGroup,Inc.

Glottal Configuration Associated with

Fundamental Frequency and Vocal Register

* T h o m a s M u r r y , t J i e Jie X u , a n d :]:Gayle E. W o o d s o n

*University of Pittsburgh, The Eye & Ear Institute, Pittsburgh, Pennsylvania;

"~FirstAffiliated Hospital, Nanjing Medical University, Nanjing, China; alut
SUniversity of Tennessee, Memphis, Tennessee

Summary: Simultaneous measurements of mean airflow rate, vocal intensity

and fundamental frequency were made during flexible video endoscopic record-
ing of the vowel/i/sustained in two vocal registers, modal and falsetto. The
glottal closure patterns of four males and four females were evaluated by visu-
ally inspecting the video images. Acoustic signals were recorded and analyzed
to verify the frequency and intensity criteria. Aerodynamic analysis of mean air-
flow rate was done via Rothenberg mask and commercial software. Incomplete
glottic closure was common in both males and females. The degree of closure
was significantly higher for modal samples than for falsetto samples with fre-
quency and intensity held constant. The shape of the glottal closure did not vary
with changes in the mode of phonation. As expected, the mean airflow rate in-
creased with decreased glottal closure. The results suggest that incomplete glot-
tic closure should be considered as a normal glottal configuration in high fre-
quency modal and falsetto phonation. Moreover, hourglass and spindle glottal
configurations may also be found in both the modal and falsetto registers of nor-
mal subjects. These results also confirm the positive relationships between de-
gree of glottal gap and mean airflow rate. Thus, mean airflow rate may be re-
garded as a criterion for judging degree of glottal closure. Key Words: Glottal
configuration---Vocal register--Aerodynamics--Video laryngoscopy

Identification of disordered glottal closure patterns
requires an understanding of glottal configuration
during normal phonafion. Generally, complete glottal
closure is regarded as a characteristic of normal
phonation. In recent years, however, some authors
Accepted for publication June 20, 1997.
Address correspondence to Thomas Murry, Ph.D., Depart-
ment of Otolaryngology, The Eye & Ear Institute Building,
Suite 500, University of Pittsburgh School of Medicine, 200
Lothrop Street, Pittsburgh, PA 15213.
have considered incomplete glottal closure to be a
normal pattern (1--4). Investigations which have fo-
cussed on the relationships between the degree of
glottal closure, the configuration of the glottis and
gender, age, vocal intensity, and fundamental fre-
quency have all shown the presence of incomplete
glottal closure in the presence of normal phonation.
With regard to gender, the degree of incomplete glot-
tal closure was significantly higher for women than
men (4). In studies by Biever and Bless (1) and
Linville (4) that concentrated on female speakers, the
majority of the subjects showed incomplete glottal

44
GLOTfAL CONFIGURATION ASSOCIATED WITH FUNDAMENTAL FREQUENCY
closure during normal phonation. Posterior glottal
chink (PGC) has also been noted to be a common
finding in young (1,4) and middle aged women (5). In
contrast, elderly women more frequently demonstrat-
ed anterior or mid-membranous gaps (1). It has also
been found that the incomplete closure correlates with
decreased loudness (3,6) and elevated fundamental
frequency (2). In the majority of these studies, the
register of phonation was not completely specified
nor were the changes in intensity over the frequency
range of interest maintained constant. Moreover, fe-
males provided the majority of data from the above-
referenced studies.
To our knowledge, no prior investigations of glot-
tal configuration have provided a clear description
of the changes in glottal configuration related to
phonational register when frequency and intensity
are maintained constant. Although clinical experi-
ence suggests that incomplete glottal closure often
corresponds to high frequency modal or falsetto
phonation (6), experimental evidence under controlled
phonatory conditions is lacking.
An understanding of glottal closure would not be
complete without accounting for airflow rate and vo-
cal intensity during phonation. A number of investi-
gators have found that a relationship exists between
airflow rate and glottal closure. Although insufficient
glottal closure is usually associated with increased
mean airflow rate (7,8), previous investigations that
included glottal closure did not specify the shape of
the glottis nor the change in glottal shape with fre-
quency and intensity parameters maintained at con-
stant levels (5,8).
We investigated the characteristics of glottal config-
uration for two vocal registers in normal male and fe-
male adults. We hypothesized that glottal configura-
tion changes with transition from modal to falsetto
vocal registers. A second purpose of this study was to
examine glottic configuration and aerodynamic
changes from modal to falsetto registers using simul-
taneous videoendoscopy and aerodynamic records.
METHODS
Subjects
Eight normal volunteers (4 males and 4 females),
ages 26-48 years, were selected as subjects for this
45
study. They were nonsmokers without laryngeal,
speech, or hearing disorders and had no neurological
or respiratory problems. None were on medication
for respiratory disorders or allergies. All of the sub-
jects had perceptually normal voices. None of them
were professional singers. Prior to final selection, all
subjects demonstrated at least a three semitone over-
lap in their modal and falsetto registers.
PROCEDURES
Determining the phonatory samples
Prior to testing, each subject's modal/falsetto
phonational range and range overlap were obtained.
The subjects were asked to match prerecorded piano
tones at semitone intervals using the/i/vowel until
he or she produced his or her highest and lowest
modal voice and then his or her highest and lowest
falsetto voice. Two experimenters independently
verified the limits of the modal and falsetto ranges
from tape recordings. The overlap range was then
taken to be the range of frequencies that were pro-
duced in both the modal and falsetto registers. For
this study, one voice sample from the modal register
and one sample from the falsetto register were se-
lected as the test frequencies. This sample was at the
50% point of the overlap register as shown in Fig. 1.
Thus, the target fundamental frequency of the two
samples was the same. Subjects practiced these tar-
gets until they were able to produce them at the same
intensity +1 dB.
Data collection and storage
Simultaneous video laryngoscopic and aerody-
namic measures were obtained from the eight sub-
jects. The equipment used for recording and storing
glottal images and aerodynamic signals is shown in
the Fig. 2. Each subject was seated facing the exam-
iner and topical anesthesia was applied to the nasal
cavity and oropharynx. The standard Rothenberg
mask was modified to accept a flexible fiberoptic en-
doscopie (Olympus ENF- P3A) coupled to a color
video camera (Panasonic GP-KS 152). The endo-
scope was passed into the subject's nose and the vo-
cal folds were visualized. Once the vocal folds were
in view, the face mask was placed against the sub-
ject's face to cover the mouth and nose. Each subject
Journal of Voice, Vol. 12, No. I, 1998
 46 THOMAS MURRY, JIE JIE XU, AND GAYLE E. WOODSON
Overlap I
Fadsotto I --
Modal
I I I I I I
C3 E A C4 E A
I I Monitor
Fiberscope
,v I
Transducer~Computer[
FIG. 2. Diagram of equipment arrangement for recording, stor-
ing and glottal images and airflow signals simultaneously.
was instructed to produce samples of sustained vow-
e l / i / a t the preselected modal and falsetto register
fundamental frequencies. The subjects heard the test
frequency from a tape recorder and monitored inten-
sity from the VU meter connected in line with the air-
flow transducer. The video laryngoscopic images and
aerodynamic signals were simultaneously recorded.
The frequency was controlled to _+2 Hz and the am-
plitude was controlled to _+1 dB for the two samples.
The glottal images were recorded on the VCR
(Modal 9132, BVB-2000), and airflow signals were
imported to and stored in the computer. Subjects pro-
duced samples until the examiners were satisfied
with the video images and the examiners and the sub-
jects agreed that they produced the samples in the de-
sired register.
ANALYSIS
Mean airflow rate was determined from the output
of the Rothenberg mask, coupled to a transducer and
converted to a measure of liters per minute (Lpm) us-
ing the SoundScope (GW Instruments, Somerville,
Journal o f Voice, Vol. 12, No. 1, 1998
W FIG. 1. Selection of test fre-
quencies in modal and falsetto
A !
voice. This figure depicts a typi-
cal overlap pattern for a female
subject.
I I I I
C5 E A C6
MA) commercial analysis system. The airflow rates
were obtained over a 2-second segment of phonation.
The videotapes were visually inspected and the
glottal images, and the degrees and shapes of glottal
closure for the vowel/i/produced in two vocal regis-
ters were reviewed independently by two of the in-
vestigators (T.M., J.J.X.). To assess glottal closure,
each investigator reviewed the video tapes to identi-
fy frames with the greatest closure for each subject in
each register sample. Disagreements were jointly re-
solved with a third reviewer. The degree of glottal
closure produced in two modes of phonation for
eight subjects (16 voice samples) was evaluated with
a scale previously described by Gelfer and Bulte-
meyer (2). Type I indicated complete closure all
along the vocal fold at the frame of maximum clo-
sure, Type 2 indicated incomplete closure, and Type
3 indicated opening all along the vocal cord at maxi-
mum closure.
The shape of glottal closure was evaluated accord-
ing to the procedure outlined by Gelfer and Bulte-
meyer (2) and originally proposed by Hirano (9). Six
distinct shapes during the maximal closed frame
were used as the basis for identifying the shape of the
glottis. Each of two investigators (J.J.X., T.M.) inde-
pendently identified the shape from a video print
made at the frame of maximum glottal closure which
was previously agreed upon.
RESULTS
Glottal configuration
Table 1 summarizes the glottal closure for the eight
subjects in the overlap range. Incomplete glottal clo-
sure was the most common finding in both modes of
phonation. Only two subjects exhibited complete
GLOTTAL CONFIGURATION ASSOCIATED WITH FUNDAMENTAL FREQUENCY 47

T A B L E 1. Glottal configuration in two registers: high modal (HM) and low

falsetto (LF) voice samples. The degree of closure (1, 2, 3) is also indicated.
Overlap
Subject High modal Low falsetto

No. Sex Age S D S D

1 M-I 28 Comp. 1 Incomp. 3

2 M-2 26 PGCm. 2 PGCm. 3
3 F-I 34 • Hourg. 2 Hourg. 3
4 M-3 32 AC 2 AC 3
5 F-2 44 PGCm 2 PGCm. 3
6 M-4 50 Spin. 2 Spin. 3
7 F-3 44 PGCm. 2 PGCm. 3
8 F--4 36 Comp. 2 lncomp. 3

Note: S, glottal shape; D, degree of closure; comp., complete closure; Incomp., Incomplete
closure; PGCm., posterior glottal chink-membraneous portion; Hourg., hourglass gap; AC,
anterior chink; Spin., spindle gap.

glottal closure (Type 1) in high modal phonation.
There was a trend toward more glottal closure in
modal than in falsetto phonation. In modal phona-
tion, two subjects closed their glottis completely; for
the other six subjects glottal closure was incomplete
(Type 2). In falsetto, however, all subjects produced
the vowel with an open glottis (Type 3). No differ-
ence in closure was found between males and fe-
males during falsetto trials; three of eight male sam-
ples and three of eight female samples exhibited
partial closure, while four male and four female sam-
ples were judged completely open.
Examples of all six glottal shapes were found for the
vowel/i/during the high modal phonation production.
Only three of these shapes were observed during the
falsetto samples. As seen in Table 1, two of the four
males changed glottal configuration between modal
and falsetto register. M-3, the subject with the anteri-
or chink, and M-4, the subject with the spindle shape
in modal phonation, did not change the shape of the
glottis when transitioning to falsetto. One female, F-
1, also maintained the same glottal configuration
while changing from modal to falsetto. The other
three females produced falsetto with an open glottis.
MEAN AIRFLOW RATE
Table 2 summarizes the findings of mean airflow
rate in the 16 samples. Two of the modal samples
with Type 1 closure had the lowest mean airflow
rates. Six of the modal samples with Type 2 closure,
partially closed, had higher flow rates and the eight
falsetto samples had the highest flow rates of the 16
samples. For any one subject, the flow rate was al-
ways higher in his or her falsetto than in the modal
register sample. Mean airflow rate is plotted against
glottal opening in Fig. 3. There is a linear relation-
ship between the three categories of glottal closure
and mean airflow rate. There was no apparent rela-
tionship between the shape of the glottis and mean
airflow rate for this group of subjects.
DISCUSSION
Incomplete glottal closure was found in videostro-
boscopic images of the majority of subjects. This
finding is consistent with the results reported by pre-
vious investigators (1-5). Therefore, incomplete
glottal closure is not necessarily a pathologic condi-
tion and some degree of incomplete closure should
be regarded as a normal variation during high fre-
quency phonafion of both males and females in the
modal, as well as falsetto registers.
Glottal opening was greater in falsetto than in
modal phonation. This result is consistent with the
high-speed photography by Rubin and Hirt who
found the vocal folds to touch each other lightly or not

Journal of Voice, Vol. 12, No. 1, 1998

 48 THOMAS MURRY, JIE JIE XU, AND GAYLE E. WOODSON

TABLE 2. Airflow rate in liters per minute (l/min) tromyography and laryngoscopy may elucidate on
f o r the eight subjects the actual mechanism.
f o r the two register conditions. In the present study, no subject had complete glot-
Overlap tal closure in falsetto phonation. This conflicts with
Mean airflow Rate (l/min) findings by previous investigators who noted com-
Subject No. Sex High Modal Low Falsetto plete glottal closure in the falsetto register for some
singers (3). A possible explanation is that voice train-
1 M 6.80 12.61 ing may adjust the muscle function, producing a re-
2 M 6.94 8.35 inforced falsetto and thus, more complete glottal clo-
3 F 6.11 8.87 sure in falsetto for singers. It has also been reported
4 M 1.58 1.63 that glottal gap in normal phonation may be related
5 F 7.76 8.59 to age and gender anatomical characteristics (1,4,5).
6 M 3.18 4.96 The present study suggests that the degree of glottal
7 F 9.53 12.29 closure for males and females is nearly the same. In
8 F 2.49 4.53
contrast to previous research, we did not find a
greater degree of closure for males at the high fre-
quency modal or falsetto phonation (3).
In this study, register change affected the degree of
L8
setto glottal closure but not the glottal shape. Simultane-
ous video laryngoscopy and aerodynamic analysis
/6 demonstrated that increased airflow rate accompa-
m nied decreased glottal closure when subjects shifted
4 from modal to falsetto. Thus, as suggested by Sund-
i
berg (15) falsetto tones consume more air than modal
n2 tones. We found a strong relationship among register,
glottal closure, and mean airflow rate. This is not sur-
0 I
prising. With increased distance between the vocal
Closed Part. Open Open folds, laryngeal resistance falls, and increased air-
FIG. 3. Comparison of mean airflow rates from the three de- flow is required to generate sufficient subglottic pres-
grees of glottal closure for 16 voice samples. sure for excitation of vibration. Additionally, the
tense vocal folds in falsetto probably require a high-
er pressure for excitation, further increasing the air-
flow rate required. Previous studies also demonstrat-
at all in falsetto (10). Large et al. (I 1) also reported a ed that the airflow is affected by the size of glottal
definite "chink" throughout the falsetto phonation gap (5,6,8). Thus, mean airflow rate may be consid-
without describing the degree or shape of the chink. ered as an indirect indicator of glottal closure in clin-
Previous investigators have noted less activation ical examinations.
of intrinsic laryngeal muscles during falsetto phona-
tion as compared with the modal register (12-14). CONCLUSION
Thus, the difference between the two registers may
reflect differences in active adduction of the vocal Incomplete glottal closure was found in the major-
folds. In addition, a likely factor is the tension in- ity of the voice samples from eight normal subjects
duced in the vocal fold by cricothyroid muscle con- in their modal-falsetto overlap register. Therefore,
traction for the higher pitches in the falsetto but not some degree of incomplete glottal closure should be
the modal register. Increased tension in combination considered as a normal pattern. Both the amount of
with decreased contraction of the thyroarytenoid glottal closure and airflow rate were affected by the
muscle should result in vocal fold thinning, a char- mode of vocal register. Glottal closure decreased and
acteristic of falsetto phonation. Simultaneous elec- mean airflow rate increased on transition from modal

Journal of Voice, Vol. 12, No. 1, 1998

GLOTTAL CONFIGURATION ASSOCIATED WITH FUNDAMENTAL FREQUENCY
to falsetto phonation under controlled frequency and
intensity conditions. These changes may be attrib-
uted to reduced movement of intrinsic laryngeal
muscles in falsetto. The positive relationship be-
tween size of glottal gap and airflow rate suggests
that airflow rate may be considered to be an indirect
criterion for judging the degree of glottal closure.
Acknowledgment: We would like to thank Clark A.
Rosen, M.D., Valerie Schweizer, M.D., Robert Yarber,
M.D., Manish Wani, M.D., and Arne Hengestet for their
invaluable research assistance.
REFERENCES
1. Biever DM, Bless DM. Vibratory characteristics of the vocal
folds in young adult and geriatric women. J Voice 1989;3:
120-31.
2. Gelfer MP, Bultemeyer DK. Evaluation of vocal fold vibra-
tory patterns in normal voices. J Voice 1990;4:335--45.
3. Sodersten M, Lindestad P-A. Glottal closure and perceived
breathiness during phonation in normally speaking subjects.
J Speech Hear Res 1990;33:601-11.
4. Linville SE. Glottal gap configurations in two age groups of
women. J Speech Hear Res 1992;35:1209-15.
5. Sodersten M, Hertegard S, Hammarberg B. Glottal closure,
transglottal airflow, and voice quality in healthy middle-aged
women. J Voice 1995;9:182-97.
49
6. Rammage LA, Peppard RC, Bless DM. Aerodynamic, laryn-
goscopic, and perceptual-acoustic characteristics in dys-
phonic females with posterior glottal chinks: a retrospective
study. J Voice 1992;6:64-78.
7. Till JA, Crumley RL, Jafari M, Law-Till C. Aerodynamic
and temporal disruptions of speech in laryngeal insufficien-
cy. Arch Otolaryngol Head Neck Surg 1994;120:317-25.
8. Hertegard S, Gauffin J. Glottal area and vibratory patterns
studied with simultaneous stroboscopy, flow glottography,
and electroglottography. J Speech Hear Res 1995;38:
85-100.
9. Hirano M. Clinical examination of the voice. NewYork, NY:
Springer-Verlag 1981:67-91.
10. Rubin HJ, Hirt CC. The falsetto. A high speed cinemato-
graphic study. Laryngoscope 1960;70:1305-24.
11. Large J, Iwata S, von Leden H. The primary female register
transition in singing: aerodynamic study. Folia Phoniatri
22:197:385-96.
12. Welch GF, Sergeant DC, MacCurtain F. Some physical char-
acteristics of the male falsetto voice. J Voice 1988;2:151--63.
13. Klingholz F, Martin F, Jolk A. The 3-dimensional voice
field. Laryngol Rhinol Otol (Stuttg) (Germany, west).
1986;65:588-591.
14. Pemberton C, Russell A, Priestley J, Havas T, Hooper
J.,Clark P. Characteristics of normal larynges under flexible
fiberscopic and stroboscopic examination: an Australian per-
spective. J Voice 1993;7:382-9.
15. Sundberg J. Perception of singing, In: Deutsch D, ed. The
psychology of music. London, Academic Press; 1982:78.
Journal of Voice, Vol. 12, No. 1, 1998