THE EFFECTS OF INTENSITY UPON THE
INTELLIGIBIIJTY 0F TIME-COMPRESSED
0N0 MONOSYLLABLES

THESIS FOR THE DEGREE OF M. A.
MICHIGAN STATE UNIVERSITY

SHELLEY SCHWIMMER
I 9 7 I

 

 

 

THESIS

’0

(Mix

 

 

 

 

 

It.» ”fit I
ZLTMI £1

«(:an

 

 

 

 

 

,l '13

T’{ACT

THE EFFECTS OF INTEHS1T1 UPO"TFINTLLiGIL11TY
OF TI‘W «cOmIXnSSED Q‘.§C

MONOSYLLABLE
By ;
Shelley Schwimmw

A review of the literature surgests pctzntislly veins
Q

\

'heoretical and diagnostic implications for the use cf aime-

CL

A;
C)
‘3’

'. n
. , “1,; w, .
y \AJiJL'. c110. '

compressed speech stimuli with By“ h”u1ug rsaa. s &i

The few reports to da 0, hc.r ever, have failed to cor.trol for
standaHdizstion of stimuli and/or have utiliPed lucstiomahle

compression procedures and eXperiw.c cntle methocsicgv. Such
variaol es as sensation level, right and left ear effects, and
scimulus reliabili‘“, as mes-LL -.u::cd usin gs‘i‘smiardijgcu stimulis
items, have received li*tle attention in th's area.

The purpose of this sfiudy was to ixvesiigute the rcsronse
accuracy of normal listeners to time-ccmcresscd (TC) speech
presenred at various sense tion levels, using the four
Form B of the Northwestern Inivcrsicr Auditory

‘

Lach oi “the four lisi‘s of 50 m'mosyl}a‘cses' vms time compressed
by 30, 4 l wrough 70%, in 10% steps, Also, there was a 0% TC control
condition. The compression was perfor. ad usin»; tr e Zemlinw
modified version of the Fairbanks Time Compressor. This resulted
in 24 eXperimental tapes, all of which were constructed under

highly controlled exPerim €:nial cor di’cions.

 

Int-'31»

 

Shelley Schw‘mmer

Ninety-six right-handed normal hearing young adults, six
groups of 16 members each, participated as listeners in this
.study. The members of each group received lists one through
four of the NU-6 test under a specified TC condition. Each list
was presented to the listeners at one of the following sensation
levels: 8 dB, 16 dB, 24 dB, or 32 dB. The lists were counter-
balanced relative to the sensation levels, for each 16 member
group. In addition, the stimuli were randomly presented to eight
of the sixteen members of each group via the right ear, and eight
via'the left car. All listening procedures were censonant with
strict GXperimental and clinical criteria.

The results revealed that as the TC percentage increased,
listener accuracy decreased. This effect was most pronounced at
the sensation levels of 8 dB and 16 dB. Further, the Spread of
average articulation scores among the first five conditions of
time compression decreased at the higher sensation levels. The
most dramatic decline in listener accuracy occurred at the highest
time compression ratio for all sensation levels. At each time
compression ratio, the greatest amount of list variability
occurred at 8 dB SL, with generally decreasing variability at
the higher sensation levels. Likewise, the greater the TC ratio,
the greater the associated list variability. In addition, there
appeared to be a trading relationship between sensation level and
TC ratio, such that listener inaccuracy due to the distortion of
one factor could be offset by improving the level of the other

factor. Further analysis revealed no effects due to ear

 

Shelley Schwimmer

.3upferenee.' In view of past research, this finding was not
Inmxpccted considering the fact that this task utilized mono-
lndlabic stimuli under monotic listening conditions.

The results of this study provided needed information
relative to TC speech and its theoretical and clinical signifi-
cance. First, the results suggest that the Zemlin-modified
version of the Fairbanks Compressor provides a more distortion-
free signal than its predecessor. Secondly, the results suggest
relationships between TC ratio and sensation level, which to date
rave not been elucidated, especially under controlled experimental
conditions“. These relationships provide the basis for theoretical
speculation. In addition, the results provide normative data as
a basis for future investigations of this measurement procedure
as an aid to audiological evaluation of individuals with neuro~
logically-based disorders of the higher auditory pathways. These

and other theoretical implications are discussed.

' n .. 1 n-Am ‘m_ h‘ -.,.-= «aunt-‘4 ._ x..u—-.-‘_-_‘_‘— .03.. .n.‘ -- u-—.—.m——._-_-____ -MM— ___.___... .A._._____.__

. o_¥‘1 ”a

 

THE EFFECTS OF INTENSITY
UPON THE INTELLIGIBILITY
OF TIME—COMPRESSED CNC
MONOSYLLABLES

By

Shelley Schwimmer

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of '

MASTER OF ARTS
Department of Audiology and Speech Sciences

1971

 

Accepted by the faculty of the Department of Audiology
and Speech Sciences, College of Communication Arts,
Nuchigan State University, in partial fulfillment of the

requirements for the Master of Arts degree.

  

Director of Thesis

 

\

Guidance Committee: , Chairman

an “Zr/a.
1.!

 

 

 

ACKNOWLEDGMENTS

I wish to extend my gratitude to Dr. Daniel S. Beasley,
my thesis adviser, and to Dr. William F. Rintelmann and Dr.
lmrbert J. Dyer, the members of my guidance committee, for
their valuable assistance in the preparation of this thesis.

I would also like to eXpress my appreciation to Dr.
Ffillard R. Zemlin, for his time and effort spent in processing
the'eXperimental tapes, and to Miss Lisa Holstead, for her
hospitality during this time. s

I further wish to thank all of my subjects who were kind
enough to volunteer much of their own valuable time in order ‘ '

'u3participate in this study. Special thanks are extended to

(T)

arbara Kaier and Barbara Mackey, for their constant encourage-
ment throughout this endeavor.

Finally, I would like to eXpress my sincere gratitude to
fix Beasley, whose role in the creation of both this student
and this study was much greater than he could ever hepe to

realize.

mus study was supported in part by the Rehabilitation Services
Mannistration--Traineeship Grant No. 526-T-7l

\ '. II:‘

». .‘AJ—I- All.-.i._w.hw.mw....ln-m . 1",
.

TABLE OF CONTENTS

_ACKROWLEDGMENTS . . . . . . . . . . . . . . .
LIST or TABLES . . . . . . . . . . . . . . .
LIST or FIGURES . . . . . . . . . . . . . . .
CHAPTER

I. INTRODUCTION .‘. . . . . . . . . . .

Distorted Speech Tests . . . . . . .

Controlled Time Compression Procedures

‘ . Controlled Clinical Stimuli . . . .
Summary and Statement of the Problem

II. EXPERIMEVTAL PROCEDURES . . . . . .
Subjects . . . . . . . . . . . . . .
Stimulus Generation . . . . . . . .
Presentation Procedures‘ . . . . . .
Analysis . . . . . . . . . . . . . .

III. RESULTS . . . . . . . . . . . . . .
Time Compression and Sensation Level

Ear Effects . . . . . . . . . . . .

List Effects . . . . . . . . . .'. .

IV. DISCUSSION . . . . . . . . .'. . . .
Clinical Implications . . . .-. . .

Time Compression . . . . . . . .
Sensation Level . . . . . . .‘. .

Test Ear '.' . . . . . . . . . .

List Effects . . . . . . . . . .

iv

iii
vi

vii

 

TMHE OF CONTEN‘S--Continued

Theoretical Implications . . . ... . . . . . #6

Implications for Future Research . . . . . . 49
v. SURRARY ‘. . . . . . . . . . . a
'IJST OF REFERENCES . . . . . . . . . . . . . . . . . 55
APPENDIX '

A. FOUR LISTS OF FORM E OF NU AUDITORY TEST NO. 6. 61

B. INSTRUCTIONS GIVEN TO LISTENERS . . . . . . 66

' C. ANSWER FORM USED BY THE LISTENERS . . . . . 68

 

.TABLE

3.

LIST OF TABLES

Average articulation scores for each condition
of time compression at four sensation levels
COmPUtedbyteStearo00000-000000

Average articulation scores for each condition
of time compression at four sensation levels
COmPUted by teSt liSt o o o o o o o o o o o 0

Average articulation scores for right and left

ears for four test lists at four sensation
levels 0 O I O O O O O O 0 O O O O l O O O O 0

vi

PAGE

22

23

32

LIST OF FIG RES

. FIGURE PAGE

I” Average articulation scores for six conditions
of time compression plotted by sensation level 25

2. Average articulation scores for right and left
ear presentations plotted by time compression
COI‘ldi-‘tion O 0 O I O U C O 0 O O O O O O O O O 27

3. Average articulation scores for right and left
ar presentations plotted by sensation level 29

4. Average articulation scores for each of four
test lists plotted by time compression condi-
'tion"0.0100000000000000... 31-"

5. Average articulation scores for each Of four '
test lists plotted by sensation level . . . . 36

vii

CHAPTER I

INTRODUCTION

Differential audiometric techniques presently used in
clinical diagnosis have been proven useful in the detection of
lesions in the peripheral auditory system and in the eighth
cranial nerve. The standard battery of clinical tests
(including pure tone, speech audiometry, loudness balancing,
$181, and Bekesy audiometry) yield response patterns which
Often.are peculiarly associated with damage to a particular
anatomical site.

Lesions in the brain-stem and the auditory cortex, by
contrast, elude detection by existing audiometric procedures.
Present evidence suggests that individuals with disorders of
the higher auditory pathways usually exhibit response behavior
which appears essentially normal (Willeford, 1969; Katz, 1969).

'\
It has become apparent that conventional auditory tests lack
the structure and sensitivity necessary for the identification
of lesions in the higher auditory centers.

In the diagnosis of central auditory dysfunction, investi-
gators have abandoned traditional audiometric techniques in
favor of specially designed measures which require cortical
integration of complex signals in order for the subject to
respond apprOpriately (Willeford, 1969), 'The function of the

higher auditory centers, according to Bocca and Calearo (1963),

is to organize simultaneous or successive elements of the acoustic

speech signal into a definite pattern. It is recognized that
the central auditory pathways provide a sufficient degree of
intrinsic redundancy, even when damaged, to allow simple psycho-
’acoustic elements, such as pure tones, to satisfactorily reach
the cortical integrative and interpretive centers (Bocca and
Calearo, 1963). Miller and Licklider (1950), using distorted
speech signals, demonstrated that there existed an excess of
auditory cues needed for adequate intelligibility. They used

a switching arrangement whereby the signals were electronically
switched on and off. It was found that as long as the interrup-
tions occurred more than ten times per second, the interrupted
speech signal was easily understbod. The intelligibility of
monosyllables did not drOp below 90% until 50% of the speech
signal was discarded.

Attention has therefore become centered on tests involving
verbal stimuli which have been altered to render comprehension
more difficult. Since the individual with a lesion in the
auditory cortex eXperiences little difficulty with standard
speech audiometry (Katz, 1969), it is probable that he utilise
,the normally redundant information contained in the speech
signal to compensate for his pathological deficit. A common
goal in the formulation of test material to adequately diagnose
central auditory lesions has therefore been to reduce the
extrinsic redundancy contained in the auditory message so that
it can no longer be used to compensate for the deficit in
intrinsic redundancy in the centrally damaged auditory system

(Willeford, 1969; Calearo and Lazzaroni, 1957).

Distorted Speech Tests

In order to increase diagnostic precision of patients
with central auditory disorders, investigators have attempted
'to acquire differential responses to distorted speech signals
(Jerger, 196A; Bocca, 1967; deQuiros, 1964; Bocca and Calearo,
1963; Calearo and Lazzaroni. 1957). It was theorized that when
a message with reduced redundancy, such as distorted speech,
is passed into a damaged pathway, intelligibility will suffer
when compared to normal responses (Bocca and Calearo, 1963).
Several different forms of distorted speech signals have been
used.- Katz (1962) used staggered spondaic word tests to evaluate
central nervous system (CNS) discrders. Jerger (196R), Bocca

l. (1955), Calearo (1957), and Bocca (1955) eXperimented with

gt
filtered speech and reportedly were able to pinpoint temporal

lobe lesions contralateral to the tested ear. Calearo and

 

DiMitri (1958) and Calearo gt al. (1962) used periodically
switched speech and periodically Switched noise and found that
intelligibility increased as the signal-tO-noise ratio decreased.
Another form of distorted speech used to assess auditory
problems of the central nervous system has been time—compressed
speeCh. Fournier (195M) stressed the importance of the time
factor in speech discrimination and related the increased time
required by the cortex for identificatibn of a message to the
difficulties in speech perception eXperienced by the elderly.
Bordley and Haskins (l955) demonstrated that when words are

presented to the elderly at a high average syllabic rate there

typically results an increased difficulty in intelligibility.

Finzi (1955) showed in a series of systematic tests with
accelerated speech that in aged presbycusic subjects, the
speech reception threshold, which is intensity at which
.fifty percent of the words can be accurately discriminated,
may be reached only rarely, and that the threshold shift is
much increased if compared to one of the normal subjects.
Using the above two findings as a basis, Calearo and
Lazzaroni (1957) did a study of precise discrimination scores
of normal and presbycusic subjects, varying the factors Of
intensity and syllabic rate of “short, significant sentences"
(their phrasing) as stimulus material. These sentences were
recorded at the rates of 1&0, 250 and 350 Words per minute (wpm)
to generate articulation curves for both groups of subjects.
The average threshold shift for the normative subject group.
was between 5 and 10 dB at the 250 wpm presentation, and 10
and 15 dB for the 350 me presentation, but the essential shape»
Of the articulation curve remained unchanged. The three curves
Obtained at progressively increased syllabic rates ran parallel
with each other, demonstrating that in normal subjects, an
increase in syllabic rate is almost completely neutralized by
a simultaneous increase in intensity. When the accelerated
sentence material was presented to the group of presbycusic
subjects, however, threshold shift was increased as much as.
30 dB for the intermediate speed, while a threshold of Speech
perception could not be obtained at any intensity level for
the highest speed. It was further reported that when the

accelerated sentences were presented to a group of subjects

a—_...—_.‘ __

with temporal lobe lesions, these subjects yielded poorer
articulation curves when the speeded message was sent to
the ear contralateral to the lesion.

Although Calearo and Lazzaroni demonstrated that cortical
damage may result in poorer discrimination scores, they also
found the lengthening of synaptic time in the central auditory
pathways, originally attributed to presbycusics by Fournier,
to be a significant factor in impairing the ability to respond
to time-accelerated speech.' It should be noted that in trying.
to assess the conclusions drawn from this study, difficulties
ariSe-frOthhe fact that the authors did not precisely describe
the speech stimuli used in their study, nor did they eXplain
the method of time-compression used to process their speech
stimuli. These omissions necessarily preclude replication of
the study which would attest to the validity and reliability
Of its conclusions.

deQuirOs (l96h) used accelerated speech material to test
20 normal subjects, 15 subjects with peripheral hearing losses,
seven presbycusics, and several well—defined groups of adults
and children with central disorders, including six subjects
with retrocochlear lesions and 28 subjects with central lesions.
The speech material used in his study were "abstract" sentences
of approximately ten words for adult subjects, and "concrete"
sentences of similar length for use with children. Each
sentence was presented at the rates of IQO, 250 and 350 wpm,
1ming the same method of speech acceleration as did Calearo

and Lazzaroni (1957), which lacks further description. The

 

_—..~._‘A,. .4 . -4
o~ A

‘I- "-'2Vl‘ "n_ vr‘ch' "Vf‘w‘ ‘rm‘ 4 .

i _--.4--~.— .—

subjects' articulation scores were considered in relation
to the shape of the articulation curve, speech detection
threshold, speech reception threshold, and maximum articula-
_tion score. deQuiros found that for his normal subjects,
there was usually a common curve for all three speeds of
sentence presentation. In the cases where shifts were noted
between speeds, they were within the range of 10 CD, with
the exception of the threshold of detectability, which tended
to remain constant. For his subjects with conductive hearing
losses, the results Obtained were either similar to those of
the'normal.hearing subjects, or the entire family of curves
was shifted to a higher intensity. In cochlear losses, normal
results were sometimes obtained, but more often, the maximum
articulation score for the 350 wpm condition was lower than
those Obtained for normal subjects. In presbycusic subjects,
there was usually a shift in the threshold of detectability
for all rates of sentence presentation.

his discussion of results Obtained with the groups of
subjects with CNS disorders indicated that in differential
diagnosis, accelerated speech testing provided additional
information, which, when correlated with other findings, may
aid in pinpointing sites of brain lesions, especially within
the temporal lobe. Unfortunately, in view of the small number
Of subjects considered in each category of CNS disorders (e.g.,
brain-stem lesions, cerebellar lesions, central vestibular
disturbances, etc.), the consistency of responses from subject

to SUbjGCt within each category and the differential responses

of subjects in each of the categories cannot be considered
definitive on the basis of this study.

Due to l) the inadequate description of the process

.utilizcd for speech compression, and 2) the lack of standard:

ization of speech stimuli employed in the above studies, their
replication using a larger number of subjects under controlled
conditions is impossible. Thus, if studies of this nature are
to assume any clinical significance, it is imperative that they
use controlled time compression procedures as well as clinically
standardized test materials.

‘ I
O

Controlled Time Compression Procedures

A controlled electromechanical procedure for time
compressing speech stimuli was develOped by Fairbanks,
Everitt and Jaeger (1954). Issentially, the process entails
passing a tape over the curved surface of a cylinder and
wrapping it around the cylinder enough to make contact with
one-quarter of its circumference. This is accomplished by
four tape reproducing heads equally spaced around the circum-
ferenee of the cylinder. When this cylinder is stationary,
and the tape is moving at the same speed at which it moved
during recording, it makes contact with one of the reproducing
headsfiand the signal is reproduced as recorded. When an
adjustment is made for a certain amount of compression,
however, the speedwpf the tape increases and the cylinder
begins to rotate in the direction of the tape motion. Under

conditions of time compression, each of the four heads makes,

p....—.....--—-— ..__.-A—.......-. .4- as. v»... A .. a

.J.

amithen loses, contac” vith th ape 100p. Each head
rmncduces, as recordc , the ma al on tha aortion of
“am tape with which it makes cr IdCt. When 1 cylinder

'fisso positioned that one head is just losing >ntact with

'fim tape while the preceeding he' 8 just m' Lg contact
vdth the tape, the segment of t tape that wrapped E
znpund the cylinder between the e two head aver makes 3

cmntact with a reproducing hca , and is t. refore not .
reproduced. This is referred to as the interval of ‘iscard,
and was found to be max'mally effective when it was 15 to
filmsec in length (Fairbanko and Kodman, 1957). The amount
mfspeeeh compression is dependent upon the number of discard
intervals (non-reproduced segments) per unit of time.
The sampling interval (la) is composed of the discard

interval (Id) and the recorded interval (Ir), such that
\ Ia=Id+Ir

The sampling frequency, PS, i.e., the rate at which the input

'sfignal is sampled, is

PS = 1/1a

The compression ratio, RC, is then defined as

RC = Id/Ia = IdFS

Both RC and Id are independently manipulated since, on the
Fairbanks apparatus, the tape and cylinder speed are inde—

Imndently variable. This allows for variation of the temporal

 

 

value of the discarded portions of the message. (This
procedure has been modified by Zemlin (1971), in order to
eliminate low-frequency buzz at high compression ratios
.in the original Fairbanks apparatus, as well as to stream-
line the procedures involved).

Using this electromechanical apparatus, Fairbanks gt 3;.
(1957) presented a‘pair of independent message-test units to
,normal hearing subjects. Each consisted of an extended ex-
position of technical information and a corresponding test of
factual comprehension. These messages were read at approximately
lul‘me, recorded and compressed electromechanically in time
(by various amounts. Independent groups of subjects were
assigned to five experimental conditions which represented a
series of compressions from 0% to 70%, and to a sixth test
condition in which no message was presented. Listener aptitude
was controlled by forming subgroups of approximately "equal"
aptitude for each condition at four different'levcls. The
effect of the message-test difficulty was assessed by sub-
scoring results according to five message effectiveness levels,
based upon differences in reSponses to test items in the 0%
cbmpression condition and the test-only condition. The curve
of comprehension as a function of message time was found to
be sigmoid. Response scores were approximately 50% of the
maximum when the message was compressed by 60%, and slightly
less than 90% of the maximum when the message was compressed \

\

by 50%. It was concluded that the interaction of time compression
and message effectiveness significantly affects comprehension of

factual material.

...--..-o-o_w ___ —. ...—r-

 

10

Shriner, Beasley and Zemlin (l969)-examined the identifica-
tion and reaction time to frequency—divided, frequency-dividcd/
time-restored, and frequency-divided/time-compressed speech of
_50 elementary school children, using stimulus items taken from
a clinically standardized discrimination test for children
(Picture Identification for Children: A Standardized Index
(PICSI), Nosca, 1965). The results of reaction time measured
in the Shriner, £3 31. study demonstrated that increases in
reaction time are accompanied by a corresponding decrease in
the percentage of responses that are correctly identified.
.Although the overall mean reaction times were not found to be
significantly different, significant differences were shown
to occur when each of the distorted conditions were Compared
to the normal run. Thus, they concluded that when small differ-
ences in identification accuracy scores occur, the inclusion of
accompanying reaction time measures does not appear to resolve
these differences. If larger differences occur, as they did
:hen.each of the eXperimental conditions was compared to the
:normal run, then reaction time appears to be highly dependent
xrpon.percentage of words correctly identified.

, The consideration of reaction time as it affects perceptual
jprocessing of auditory stimuli can be related to the temporally-
tmused perceptual processing concepts, particularly as discussed
In/.Aaronson (1967) and Beasley (1970). Aaronson speculated that
true interval of silence (interstimulus interval) between two
stjjnuli may be utilized in the decoding of individual perceptual

Lurits. and further, that the relationship between stimulus

 

ll

duration and the size of the interstimulus interval may be
critical in the development of perceptual strategies. The

use of sentential material by Fairbanks g3 al. (1957) in
.their investigation of time-compressed speech, fails to allow
fbr consideration of the individual effects of interstimulus
interval and stimulus duration on the comprehension of factual
material.

According to Miron and Brown (1968) it is necessary to
take into account the presentation rate of the stimuli in terms
of syllables and words and the associated variability of these
measurements, the distribution of phonation time, and finally,
information measures of method content as stimulus parameters

' \
in speech compression. The interaction of all of these factors
produce experimental variables which, to date, have been
uncontrolled in the studies of the effects of time compression
on the intelligibility of sentential material,

According to Miller (1951), accurate perception of a spoken
word depends to a considerable extent upon the frame of expecta-
tions in which that word occurs. The use of speech, such as
monosyllables, in place of sentences, would serve to reduce the
amount of syntactic and semantic information available to the
listener as cues for intelligibility of the stimulus. The
listener's frame of expectations would necessarily have less
of an influence in his perception of the words than it would if
the words were arranged in a semantic and syntactically meaning—
ful order. It Should further be noted that studies of the

time compression on intelligibility making use of sentential

.---- “ah-n. a..-_ ‘J- v 7

12

material, rather than word lists, necessarily confound the
variables of interstimulus interval and stimulus duration
(Beasley, 1970; Aaronson, 1967).

In essence, the effects of compressing either stimulus
duration or interstimulus interval on the intelligibility of
the speech stimulus cannot be determined with the sentence

as the speech stimulus unit.

Controlled Clinical Stimuli

 

To isolate the effect of reduced stimulus duration on
intelligibility, Fairbanks and Kodman (1957) presented Egan's
phonetically balanced (PB-50) word list to highly trained
listeners. The words were presented at a constant intensity
at varying ratios of time compression. The results of their
study demonstrated that time compression, up to 20% of the
original signal duration, had no significant.influence on
intelligibility when the stimulus words were presented at a
comfortable listening level. 'However, this study cannot be
considered of prognostic value as a clinical tool because of
its psychoacoustic emphasis on the nature of the stimuli.
Also, specially trained listeners were employed as the subject
group rather than a clinical pepulation of "naive" listeners.
Further, it has been demonstrated that the PB-SO word list
reflects low reliability and a wide range of word difficulty
among the test items (Eldert and Davis, 1951). If time-
compressed speech is to be used clinically, it is necessary

to utilize more reliable and theoretically sound stimuli.

 

13

Although the W-22 word lists were devised (hirsh gt a1, 1952)

 

in an attempt to overcome the reliability problem, it was not
until the develOpment of the Northwestern Auditory Test No. 6
.(NU-6) that the latter difficulty was investigated (Tillman
and Carhart, 1966).

The develOpment of this test was based upon studies
performed with an earlier list, the Northwestern Auditory
Test No. 4 (NU-h) developed by Tillman, Carhart and Wilbur
(1963), Which was used extensively in the Auditory Research
Laboratories at Northwestern University for a two year period.
It proved’to be a valuable tool in the measurement of speech
discrimination. The NU-4 consisted of six randomizations of
two 50 word lists composed of phonemically-balanced mono~
syllables of the consonant-hueleus-consonant variety, selected
from a pool of such words compiled by Lehiste and Peterson
(1959). In addition, the 50 words in each list contained a
preportional distribution of phonemes as found in the Thorndike
and Lorge (1952) word list. It was found that even with six
equivalent forms of each list, the investigation of a large
number of listening conditions could not be accomplished
without several repetitions of the various forms and lists.
Because of this limitation, the NU-u was revised and expanded
into the NU-6, which consists of four randomizations of four
pmonemically balanced word lists, each composed of 50 mono-
syllables. From the studies of Tillman and Carhart (1966),
it was found that NU-6 compares favorably with the NU-4 in

interlist equivalence and test-retest reliability. Further,

 

”gm-A
”-

114.

aswith the NU-U, subjects with conductive hearing losses
5delded articulation functions closely duplicating normal
mdflects, whereas subjects with sensorineural impairment
jdclded more gradually rising articulation functions and
lower mean maximum articulation scores. The NU-6, therefore,
satisfies the two basic requirements of a diagnostically
\meful test of auditory discrimination: a) a substantial
segment of the articulation function, depicted graphically
as an articulation curve, is linear, so that the value of

its slope may be precisely measured, and b) the slepe of

the articulation function diagnostically differentiates

among the various auditory pathologies.

§pmmary and §tatement pf the Problem

In summary, a review of the literature suggests
theoretical and practical bases for the diagnostic value of
time compressed speech stimuli in the identification of
central auditory lesions (Bocca and Calearo, 1963; Calearo
and Lazzaroni, 1957; deQuiros, 1964). However, several
factors have served to undermine the clinical value of
past investigations. First, there has been an inadequate
description of the procedures used to compress the speech
stimuli. Further, much of the research carried out with
time compressed speech has made use of sentential material
(Calearo and Lazzaroni, 1957: deQuiros, 1964; Fairbanks,

23 gl. 1957) which, of necessity, involves the uncontrolled

interaction of semantic and syntactic information as these

 

aflkmt the intelligibility of the message. In addition, the
bets of interstimulus interval and stimulus duration of the
Imssage pgp gg could not be assessed independently with respect
matheir contribution to speech intelligibility. The replace—
nmnt of sentences by speech stimuli consisting of isolated
vmrds obviates these variables. The use of a clinically
Nmndardized word list, such as that used in the clinical
znmessment of speech discrimination, would be of particular
benefit in replicating past studies with different groups of
subjects. Although the influence of stimulus duration on
rmrmative.articulation scores has been examined using words
from a standardized list (Fairbanks and Kodman, 1957), the
effects of monosyllables has yet to be examined with respect
to intensity variations in the stimulus presentation.

The purpose of this study, then, is to isolate the effects
of varying degrees of time compression on.monosyllabic word
intelligibility and to examine them with respect to intensity

\ ,
increases in the speech stimuli using normal hearing subjects.
The evaluation of normative responses to words of compressed
duration presented at various sensation levels will provide
the groundwork upon which comparative studies with patho-
logical subjects can be performed.

Specifically, the following questions will be investigated
relative to the response of normal hearing, young adult subjects
to a standardized;‘monosyllabic word list:

1) Will various degrees of time compression (30% to

70% in 10% steps) result in differential intellig~

ibility scores?

—__._._‘m-.-v—-.”~_. .—. . .

 

.flm....—a--..._--_. m—~-_-t’—_% hu—our‘—p——-MM-

16

2) Will the intelligibility scores at the various
amounts of time compression interact with sensation
level? .

3) Will significant laterality effects result from
varying the ear to which the stimuli are presented?

4) Are the results reliable, as measured by using

equivalent versions of a standardized monosyllabic

word list?

» fi»-. .... w -w‘ w'j"-cq<v—V‘mum—-—__

M-—._..r-

- .‘ -~.-"' - -4 —-4—v»-‘v

. w<l~m~‘ --.-q~. mo“.--- -4 ca. o~-.-.>~ ... .v

*w”.-- v-.-.,.".p.o

-‘p'vp- .--g-v-,.—r o';<- ‘a_~‘-‘

 

CHAPTER .1

EXPERIMENTAL PROCEDURES

This study consisted of 96 sui ts assig 1 to one of
shcexperimental conditions. Eac ondition \ characterized
Myone of six levels of time compression, utilizing four
vmtions of a standardized word list presented at four sensa-
ncn levels (SL), and according to a randomized right ear/left

mn‘paradigm.

. l
o

Subiects

The subjects were 96 normal hearing right-handed young
amflim selected from a university population. These subjects
was randomly assigned to six groups of sixteen each. The
mmt ear for each subject was determined by random selection
dfeight right ears and eight left ears for each of the six
groups . 1

Each subject was required to pass a sweep frequency
smreening test presented at a Hearing Level of 22 dB (re:

ISO, 196M Standard) at octave intervals ranging from 125 Hz
in 8,000 hz to insure the normative status of his hearing
Enlaterally. In addition, a live-voice presentation of the
CH3w-l Word List was administered unilaterally to obtain the
Emeech Reception Threshold for the designated test ear, i.e..

'Hm ear to receive the eXperimental stimuli.

‘ . p-l7-

l..fiw .g..-.... ....V..

~uo- --i p.,

18

\

Sfimulus Generation

The experimental stimuli used in this study were the
fmu‘lists of Form B of the NU-6 (Tillman and Carhart, 1966).
'Emflllist is composed of 50 CNC meaningful units, phonemically
aflanced according to the scheme advocated by Lehiste and
Interson (1962). The four word lists were recorded at normal
mchrsational speech and effort level by a trained white
nmle talker who spoke General American English under controlled
recording procedures (Rintelmann and Jetty, 1968).

‘A cepy of each of the four recorded lists was made using
miAmpex Model 601 tape deck (frequency response 50-12,000 hz,
32 dB) and an Ampex Model 600-2 tape deck (frequency response
50-13,000 hz, :2 dB). These c0pies were then temporally
processed using the Fairbanks electromechanical time compression
apparatus (Fairbanks, Everitt and Jaeger, 1954), as modified
kw Zemlin (1971). Each of the lists was time compressed by
30%, now, 50%, 60% and 70%. In addition, each list was passed
through the time compression apparatus under the 0% time 5\
compression condition in order to control for the effects of
possible fidelity distortion in the time compressed tapes.
This resulted in 24 experimental tape recordings: six time
compressed recordings for each of the four lists.

Cepies of each eXperimental tape were made using an Ampex
hodel_60l tape recorder and an Ampex AG 500-2 (frequency response
50-13,000 Hz, :2 dB) monitored by an Ampex AA 620 power ampli-
fier. During this procedure, approximately five seconds of

silent interval response time was allotted between each stimulus

item. .

a...- A“ w
.7...“ “—0-

 

19
Ehescntation Procedures

The 96 subjects were divided into six groups, correspond-
ing to the six different percentages of time compression under
‘studv. Each subject within a single group was presented with
'Hm four lists of Form B of the NU-6 (see Appendix A), each

ist at one of four sensation levels: 8 dB, 16 du, 24 oh and

H

32 dB. Within each group, the order of presentation of each
of the four sensation levels was rotated. In this manner,
each test list was presented at each sensation level a total
of four times for each time compression condition, and the
sensation levels were counterbalanced to avoid possible order
effects of sensation level presentation. .

All subjects were tested individually in a prefabricated
double walled test chamber (IAC 1200 series) with the experi-
menter seated in the adjacent single-walled control room. The
ambient noise in the test room was sufficiently low (45 dB on
the C—scale of a Bruel and Kjar sound level meter) so as not
to interfere with testing at even the lowest sensation level.

Each subject received standard instructions (see Appendix
B) and was given a four page answer form (see Appendix C). The
experimental tapes were then presented to the listener via a
Viking Model 433 tape deck (frequency response 60-12,000 Hz,

:2 dB) mounted in a Maico MA 24 audiometric unit, through TDH

39 earphones mounted in MX 4l/AR cushions.

Analysig

The data were‘hand-scored by the experimenter and converted

to percentage correct scores. These scores were then plotted

 

 

20

as articulation curves, and the slepes of these articulation
functions were calculated for each condition of time compression.

In addition, graphic data was computed for each of the other

factors and respective interactions studied.

7‘», .nL

CHAPTER III

RESULTS

The results of this study generally support the thesis
'Wmt as the ratio of time compression increases, the intell-
igibility of the word lists decreases. Further, as the sensa-
'mcn level increased, intelligibility was shown to increase
IHMGr the several conditions of time compression. As for the
inter-list equivalency of the NU~6, the results show that
Idst I wasnthe most difficult while List IV was the easiest.
lmwever, the effect of word lists interacted with sensation
level and time compre eee ion. Finally. ear differences in this
study were found to be minima

These results are reported in Tables I through 3 and
depicted in Figures 1 through 5.
TLme Con pres sion and Sensation Level

Tables 1 and 2 and Figures 1 through 3 indicate that as
the percentage of time compression increases, intelligibility
decreases. An exception to this is the 0% and 30% conditions,
vmereby the 0% condi cion is slightly inferior at all sensation

levels. In addition, it should be noted that the decrease in
intelligioility is relatively gradual over the several conditions
(n time compression, until 70%, at which point there is a

dramatic breakdown in intelligibility. This suggests that at

Yugh compression ratios , a normal listener may not have enough

-21-

 

 

 

 

 

 

 

 

 

22
{Eablxa 1. Average articulation scores for each condition of
time compression at four sensation levels computed
by test car.
Percentage Time Compression
0 30 40 50 60 70 E Total
8 I. 60.2 67.0 48.8 51.2 50.2 17.0 49.0
11 66.7 62.0 55.8 47.0 45.5 14.5 48.5
L: 63.4 64.5 52.3 49.1 47.8 15.7 48.8
16 1. 81.3 84.8 79.0 71.2 68.8 31.5 69.4
ES 84.0 87.5 79.2 72.2 71.8 33.3 71.3
h: 82.6 ‘ 86.1 79.1 71.7 70.3 32.4 70.3
24 r‘t89.5 91.8 88.2 88.5 81.2 49.5 81.4
g: 94.3 93.0 89.0 82.8 86.0 50.8 82.6
rd 91.9 92.4 88.6 85.6 83.6 50.1 82.0
32 L 95.8 96.8 94.8 91.2 90.8 69.2 89.7
B. 97.0 96.5 93.2 93.5 90.0 61.8 88.6
re 96.4 96.6 94.0 92.3 90.4 65.5 89.2
‘ tam-1 83.5 84.9 78.5 34.6 40.9 7275

 

- -II—m mW-‘pnt‘W-y mv"m... ‘

time and/or information to perceptually process incoming verbal
stimuli (fianiloff, Shriner and Zemlin, 1968; Shriner, Beasley
and Zemlin, 1969; Aaronson, 1967).

Figure l and Table 1 also reveal that the intelligibility
0f time-compressed speech is significantly affected by sensation
level. Specifically, listeners demonstrated increased discrim-
ination ability at each condition of time compression as sensa-
tiqn level increased. This effect was most evident under 70%
'fimm compression, Where the slepe of the articulation curve is
-Tmar1y linear. Under the other five percentages of time

mmmression, the articulation functions are characterized by

 

6.11... .. - f”

u .
nag—.8196: Lita-.1... 64.“, r .. 6...... .

 

 

 

23

 

 

 

6.66 6.06 0.66 6 66 0 66 6.66 66606 6

6.66 6.66 6.06 6.66 6.66 6.66 66606

6.66 0.66 6.66 6. 66 0.66 6.66 >6

0.66 0.66 0.66 6. 66 0.66 6.66 666

6.66 0.66 6.66 6. 66 6.66 0.66 66 .

6.66 0.06 6.66 6.66 0.66 6.66 6 66

0.66 6.06 6.66 6.66 6.66 6.66 66606

m.mm m.mm 0.60 .30 0.60 0.0w >H

6.66 6.66 0.66 6.66 6.66 6.66 666

6.66 0.6: 6.66 0.06 0.66 0.66 66

6.66 6.6: 0.06 0.66 0.06 0.66 6 mm

6.06 6.66 6.06 6.66 6.66 6.66 66606

6.66 0.66 0.66 6. 66 0.66 0.66 >6

6.66 6.66 0.66 6. 06 0.06 0.06 666

6.66 6.66 0.66 0. 66 6.66 0.66 66

6.66 6.66 0.66 6.66 0.66 6.66 6 66

6.66 6.66 6.66 6.66 6.66 6.66 66606

6.66 0.66 0.66 0.66 0.06 0.66 >6

6.6: 6.66 0.66 0. 66 6.66 6.66 666

6.66 0.66 0.6: 6. 66 6.66 6.66 66

6.66 6.66 6.66 6. 66 0.06 0.06 mi 1m
. . 6666 66

66606 6 06 . 06 06 0: 06 0 .

 

 

 

 

 

VII... '1

scenes. MoFoo ofi6e on moseoxmp

 

PM COdflOOHQEOO 08H oPMwHH )MGJ. Rap COPSQIEOO muHUkKOH COHdemw Crow “.50.“
. L P 60 COHPHUCOO come he? mahoow C066.dHSOHP$w Om.cho>< .N OHwa

6.- ,__._,.,, -0..-.- ~_ 60.--.

 

2b

Figure 1.7-Average articulation scores for six conditions
'of time compression plotted by sensation level

 

I‘l- -I!,E .*‘

Hm>og nowwmewm

 

6606". n... In
$6600" XII le
6606”" o! lo
ROS” Dill—U

6606“ .X X
60 “OIIIAu

 

6m 66 66 6

a
\ I

\
u. \ ..
\
\ \ -
\0 \0 ..
\ \ .
ax \\ o -
I\“\\ a! 1
xx.

 

\1

0d

06
0.6

o:

06

ow

06
06
06

OOH

$031100 gueoaad

_ ,-W_-——

26

Rumre 2.--Average articulation scores for right and lef
ear presentations plotted by time compression
condition

 

\fr ;.. .J

 

.5613. I I "I. .6. iii r

sowmmommaou oEwa ommpswoymm

06 06 06 06 06 0

 

. . _ _ _ . 6.

..r:;

.666... 2666 03110
666 0.696 xix

. N whfimwfiunw

 

OH

ON

on

o:

06

06

06

ow

om

OOH

$091103 quaoxad

 

igue 3. --Average articulation scores for right and left
ear presentations plotted by sensation level

 

 

29

 

 

n Hgf.
62, i
6. _....

_. . . «1:15.!—

H @bth 6.....0 Mywwflm m

66 66

6 .6 ._

OD

" (‘1
C

0MV\\\\\\.\
v6 om
86,6 2666 o [1.0

66m 9906 XIIIX . .. 006

 

.fl 063066

4093103 in

rvilinear progressions in which discrimination scores

crease less with the progressive elevation in intensity,
aching an asymptote at 32 dB SL. .The slopes of the articula-
on curves generated for each condition of time compression

re all found to increase between 2 and 3.5 % per dB increase
'intensity, when computed between the sensation levels of

dB and 16 dB. This is consistent with previous data compiled
undistorted NU-6 lists with respect to the slopes of the
ticulation functions computed between these two sensation
vels (Rintelmann and Schumaier, 1971). At the sensation

vel of 32 dB, the mean discrimination scores under each
ndition of time compression, with the exception of 70%,

11 above 90% correct, which may be considered clinically

rmal.

3 Effects

Table l and Figures 2 and 3 depict the results according
right and left test ear. These data suggest that there was
sentially no difference between right and left ear presenta-
3n, at the several conditions of time compression and sensation
rel under study. An interesting point, revealed in rigure 2,
that the 0% time compression, left ear mean score appears to
abnormally low. This could account, in part, for the anomaly
scussed above, where it was noted that the 0% time compression
1dition was slightly inferior to 30% time compression at the

1r sensation levels.

 

 

List Effects

Tables 2 and 3 and Figures 4 and 5 illustrate the
effects of the four lists of Form B of the NU-6 at the six
'conditions of time compression, the four sensation levels.
and left and right test ear. Inspection of Figure u reveals
a general decrease in intelligibility for all lists. except
List IV. as the time compression ratio increases. Specifi-
cally. this effect occurs beyond 0%, whereas at 0% time
compression. there is essentially no list differences (range
from c. 81§-86%). The rather erratic configuration exhibited
by List IV may be explained by simply noting that List IV
will sustain a greater degree of distortion before intellig-
ibility declines. Support for this contention is evident
from Figure 4 and Table 2 where it can be found that List IV
is the easiest. while List I appears to be the most difficult.
Hbtcver because of the small spread of scores between lists
overall, it can be concluded that the interlist differences
are probably clinically negligible.

Reference to Table 2 and Figure 4 reveal that the inter-
list variability is greatest for the time compression ratios of
M0%, 60% and 70%, whereas 30% and 50%‘show similar range of
scores..

Reference to Table 2 and Figure 5 reveal that as sensa-
tion level decreased. the interlist variability increased.
Again. List IV and List I appear to be the easiest and most

difficult, respectively.

 

 

Tmfle 3. Average articulation scores for right and left ears
for four test lists at four sensation levels.

 

 

 

 

 

 

 

 

LIST
I II III IV E Total
SL
'8 L 31.0 56.3 06.8 60.3 #8.6
B 47.0 96.7 69.3 50.2 53.3
M 39.0 51.5 58.0 55.2 50.9
16 L 60.8 60.0 76.2 73.3 67.5
3 62.2 72.6 70.2 79.5 71.1
m 61.5 66.3 73.2 76.4 69.3
2t L 75.0 83.7 78.2 88.3 81.3
. 3 4.81.8 79.1 83.8 85.3 82.5
'm 78.b 81.4 81.0 . 86.8 81.9
32 L 91.2 ‘ 87.3 19.5 91.1 89.7
3 82.7 91.7 86.2 93.5 88.5
M 86.9 89.5 87.8 92.3 89.1
E Total 66.4 72.1 75.0 77.6 72.9

.7} Finally, Table 3 indicates that there was essentially
no consistent interaction between lists and test ear. how»
ever, this conclusion is highly tentative. due to the small
and unequal number of subjects per cell which resulted from
the fact that test ear was randomly assigned within each

subject group.

 

 

33

Figure 4.--Average articulation scores for each of four
test lists plotted by time compression condition

 

 

34

 

CosmmepmSOU mesa ewepseonem

on ow om 0: on o

 

>H swan
HHH pmag
HH pwaa

H PwfiA.

nagrsnu
Asannxu
nusssnv

- . a .i _ —

«ox

 

0H

ON

ow
on.
ow

om

OOH

ioeaaoo aueoaeg

 

36

 

 

 

3 $3 . slim
HHH swag Av111unu
5 $3 oils

H 983 XIIIX

NM cam
_

.m magmas

c1<34 ><'.

 

oa
om
on
0:.
on
00
Oh
om

om

ooa

1031103 iuaoaed

CHAPTER IV

DISCUSSION
(Dinical Implications

The function of the auditory mechanism becomes pro-
gnessively less definitive when consideration is given to the

Imthway central to the cochlear apparatus (Bocca and Calearo,

 

1963). According to Jerger (1960). the higher in the central L j
nervous system the neurological lesion, the more subtle or i ’
complc dare the stimuli required to uncover the lesion. In
an analogy comparing the auditory system to a "bottleneck".
he states that once conventional auditory tests using pure-
tone and speech stimuli pass through the “bottleneck" of the
peripheral auditory system up to and including the eighth
1erve. more difficult material is necessary to determine the
axistence of neurological auditory disorders. Bocca and
falearo (1963) and Bocca (1967) have suggested the value of
iistorted speech tests to identify the more subtle effects
f lesions in the higher auditory pathways.
As one form of distorted speech signal. the potential
linical.utility of time-compressed speech has been demonstrated.
aleaano and Lazzaroni (1957). using time-compressed sentential
iteriajq fbund that with a group of subjects with ascertained
1trfiJisic lesions of the temporal lobe, the discrimination

ailitqr was clearly worse when the accelerated message was

e37-

38

mmnsmitted to the ear contralateral to the lesion. deQuiros
(1969). using similar stimuli. found in various subject groups
vdth central nervous system disorders, that accelerated speech
.1msting provided useful information, which. when correlated
vdth other findings, could aid in differential diagnosis of
brand lesions.

An advantage of using temporally modified speech signals

'over other forms of speech distortion, such as filtering, in

 

differential diagnosis of higher auditory lesions. is that the
several theoretical models of speech perception sugrest that ’ "
perception is essentially temporally-biased (Aaronson. 1967).
In addition. such signals do not eliminate possibly clinically
relevant spectral information, such as formant structure, whereas
this does not hold true for filtered speech.

Although there are several methods of temporally distorting
Speech signals. the most efficient and controlled method to date
is time c mpression using the electromechanical compressor.
This procedure has the additional advantage of having been
utilized in a large number of studies (Fairbanks, g3 g1, 195?;
Fairbanks and Kodman. 1957; Daniloff. Shriner and Zemlin. 1968;
Shriner, Beasley and Zemlin. 1969). If it is to be used
Clinically, however, it is necessary to obtain normative data
USing standard clinical procedures. This study has provided
the necessary normative data. Before further investigations
are Carried out clinically; however. it 19 necessary to consider

several pertinent points relative to the findings of this study.

 

 

KO
\0

Time Compression. The results of this study indicate a gradual

 

decrease in the intelligibility of monosyllables corresponding
to progressively greater percentages of time compression over
.the range of 30% to 60%. with a dramatic reduction of intellig-
ibility occurring at the 70% time compression condition. These
findings are in agreement with those of Daniloff. Shriner and
Zemlin (1968), who found a significant breakdown in intellig-

ibility to occur at 70% time compression using eleven different

 

vowels placed in an /h-d/ context as their eXperimental stimuli. 3
Fairbanks and Kodman (1957), however. found no appreciable iv
breakdown'in intelligibility of phonetically balanced mono-
syllables until a time compression ratio of 80% was reached.
The discrepancy between the results of Fairbanks and Kodman
(1957) and those of Daniloff. Shriner and Zemlin (1968) and
those of the present study can be accounted for in several
ways.

First. Fairbanks and Kodman (1957) used a ten msec
discard interval in their method of time compression, as
Opposed to the 20 msec discard interval employed in the other
'two studies. According to Daniloff. Shriner and Zemlin (1968),
this probably served to enhance intelligibility at the high
compression ratios employed in these studies since smaller
segments of the message and therefore,-smaller bits of informa-
tion, were deleted at a time. Secondly. Fairbanks and Kodman
(1957) used highly trained listeners as subjects for their

study. whereas "naive" listeners served as subjects in the

other two studies. Finally, Fairbanks and Kodman (1957)

 

3h

sesamepmSOQ meB swaggeonem

on om 6m

03 or

 

>H smug
HHH pmaa
HH pmaa

H meq.

n_ss.ns
Asnzzxu

“unilnv

. . .

<1°><

 

OH

ON

om
os.
om

om

ooa

4091109 aueoaag

 

36

 

 

3 6.2; oils
HS 53 4.1.1.0.
E 33 0.1.19

H 9.63 XIIIX

He>eA fiofipmmfimm

Nm a 0M

— .

‘03

.m cheese

CC)

51(34

 

OH
om
on
0:.
om
00
on
ow

om

00H

1091109 guessed

CHAPTER IV

DISCUSSION
'Clinical Implications

The function of the auditory mechanism becomes pro-
gressively less definitive when consideration is given to the
pathway central to the cochlear apparatus (Bocca and Calearo.
1963). According to Jerger (1960). the higher in the central
nervous system the neurological lesion, the more subtle or
complcQ are the stimuli required to uncover the lesion. In
an analogy comparing the auditory system to a "bottleneck".
he states that once cdnventional auditory tests using pure-
tone and speech stimuli pass through the "bottleneck" of the
peripheral auditory system up to and including the eighth
nerve. more difficult material is necessary to determine the
existence of neurological auditory disorders. Bocca and
Calearo (1963) and Bocca (1967) have suggested the value of
distorted speech tests to identify the more subtle effects
of lesions in the higher auditory pathways.

As one form of distorted speech signal, the potential
clinical utility of time-compressed speech has been demonstrated.
Calearo and Lazzaroni (1957). using time-compressed sentential
material. found that with a group of subjects with ascertained
intrinsic lesions of the temporal lobe. the discrimination

ability was clearly worse when the accelerated message was

-37...

 

_. mum-H

38

ransmitted to the ear contralateral to the lesion. deQuiros
(1964). using similar stimuli. found in various subject groups
with central nervous system disorders. that accelerated speech
.testing provided useful information. which. when correlated
with other findings. could aid in differential diagnosis of
brain lesions.

An advantage of using temporally modified speech signals

'over other forms of speech distortion. such as filtering, in
differential diagnosis of higher auditory lesions. is that the
several theoretical models of speech perception sug_est tha
perception is essentially temporally-biased (Aaronson, 1967).
In addition, such signals do not eliminate possibly clinically
relevant spectral information, such as formant structure, whereas

this does not hold true for filtered speech.

Although there are several methods of temporally distorting

speech signals. the most efficient and controlled method to date
is time compression using the electromechanical compressor.

This procedure has the additional advantage of having been
utilized in a large number of studies (Fairbanks, gt gl, 1)»?;
Fairbanks and Kodman. 1957; Daniloff. Shriner and Zemlin. 1968;

Shriner, Beasley and Zemlin, 1969). If it is to be used

9.)

clinically. however. it is necessary to obtain normative dat
using standard clinical procedures. This study has provided
the necessary normative data. Before further investigations
are carried out clinically. however. it is necessary to considpr

several pertinent points relative to the findings of this study.

 

 

\O
\3

Time ngpression. The results of this study indicate a gradual
decrease in the intelligibility of monosyllables corresponding
to progressively greater percentages of time compression over
.the range of 30% to 60%. with a dramatic reduction of intellig-
ibility occurring at the 70% time compression condition. These
findings are in agreement with those of Daniloff. Shriner and
Zemlin (1968). who found a significant breakdown in intellig-
ibility to occur at 70% time compression using eleven different
vowels placed in an /h-d/ context as their experimental stimuli.
Fairbanks and Kodman (1957). however. found no appreciable
breakdown'in intelligibility of phonetically balanced mono-
syllables until a time compression ratio of 80% was reached.
The discrepancy between the results of Fairbanks and Kodman
(1957) and those of Daniloff. Shriner and Zemlin (1968) and
those of the present study can be accounted for in several
ways.

First. Fairbanks and Kodman (1957) used a ten msec
discard interval in their method of time compression, as
Opposed to the 20 msec discard interval employed in the other
'two studies. According to Daniloff. Shriner and Zemlin (1968),
this probably served to enhance intelligibility at the high
compression ratios employed in these studies since smaller
segments of the message and therefore, smaller bits of informa-
tion. were deleted at a time. Secondly. Fairbanks and Kodman
(1957) used highly trained listeners as subjects for their
study. whereas "naive" listeners served as subjects in the

other two studies. Finally, Fairbanks and Kodman (1957)

 

 

 

1+0

1nescnted their stimuli at sufficiently high sensation levels
(&)dB SL) to insure maximum intelligibility. whereas for the
Imrposes of the present study. the maximum intensity level
mployed was at a sensation level of 32 dB. At this level,
PB Max was not reached for the 70% time compression condition.
An exception to the tendency for intelligibility to
decline with gradual increments in the time compression ratio
of monosyllables in this study seems to occur between the 0%

and 30% time compression conditions. Equivoeal as it is, this

 

is not the only instance in which such a finding has occurred.
Aaronson (1966) conducted a study in which the duration
of the auditory test stimuli was varied for a fixed presentation
rate. Spoken digits were compressed by 35% and compared to the
original seven digit sequence with respect to ease of short
term recall. It was found that when subjects memorized each
sequence and tried to recall it a few seconds after presentation,
they did better with the sequence Of digits that were time»
compressed by 35% than they did under conditions of 0% time
compression. Zemlin.flbaniloff and Shriner (1968). in their
psychological scaling of time compressed speech. found that
normal listeners prefer speech at about 30% time compression.
although.it was scaled as being more difficult to understand.
These findings suggest the possibility that the amount
of redundancy (or lack of it) present in the undistorted speech
signal may inhibit. if only to a slight degree. the discrimina-
tion ability and/or the comprehension ability of normal listen-

ers. and that an Optimal listening duration of signals compressed

 

1a

between 30 and 35% may facilitate information processing.

Future research is needed to clarify this issue.

Sensation Level. The interaction of intensity and the effects

 

'Of time compression suggest a relationship between the two
factors. The data obtained from the present study indicate

that for every increase in intensity. there is a corresponding
increase in intelligibility of the test stimuli for all conditions

of time compression employed. The effects of intensity incre—

 

ments lessen. as would be expected. as an Optimal listening
intensity is approached. This effect is readily observable
from the curvilinear configuration of the articulation function;
of the first five levels of conditions Of compression. (Figure
1). I

The effects of intensity on the comprehension ability
of normal subjects used in the studies of Calearo and Lazzaroni
(1957) and of deQuirOs (1964) cannot readily be compared to the
present findings due to the difference in speech stimuli and
time compression methods employed by each. It should be noted,
however, that a tendency for intensity to neutralize the effects
of speech acceleration was reported by Calearo and Lazzaroni
(1957) for their ndrmal subjects. This holds true to a consid-
,erabde extent in the present study as well. Although the mean
zrrticulation scores for each of the first five conditions of
tflxne compression are not exactly equal even at the highest
sunisation level. they all fall within the clinical range Of

ntnflnal. This is not the case with the 70% time comwression
}

cxnadition. At this percentage of time compression, the

L2

‘articulation function is still linear at the highest sensation
level, indicative of the fact that sufficient intensity levels
have not been reached for maximum discrimination. It is

_doubtful. in fact. that such a level would be reached for thin

degree of time compression.

.iIPI I'V‘II

Since tests of discrimination are clinically administered

2!”!-

at sufficiently high intensity levels for the patient to perform
Optimally (Hirsh. 1952). it can be reasoned. on the basis Of

'this study. that a clinical test of discrimination using time-

 

compressed monosyllables. cannot employ stimuli compressed
beyOnd 60% at the highest sensation level examined. To do so
would allow for the interference of accurate determination of
speech discrimination ability. Further studies using patholog~.
ical subjects should yield valuable information relative to the
precise combinations of time compression and intensity levels
which would be of maximum diagnostic utility in the formulation

of such a discrimination test.

iggt 3gp. In order to utilize validly the same test for both
right and left ears. performance of normal subjects would warrant
that test results between ears be approximately equal. If this
were not. in fact, the underlying assumption of all audiometric
testing, then separate tests would have been devised for each.
ear. taking into account laterality effects on the respective
scores Of each ear. The question of differences between ears

is therefore worthy of consideration in light Of the potential
utility of time-distorted speech as a diagnostic tool for central

auditory disorders. The absence Of ear differences would allow

for the utilization of a single test.

 

 

 

43

There is considerable evidence. however. that under
dichotic and monotic listening conditions. certain kinds of
material are recalled or recognized better if they are
.received by one ear rather than the other. It has further
been found that the ear of preference is dependent upon the
stimulus used.

Under dichotic listening conditions. digits (Kimura.
1961; Broadbent and Gregory. 1964), phonetically matched

words (Borkowski. Spreen and Stutz. 1965) and plosive conson-

 

ants (Shankweiler and Studdert-Kennedy. 1967) are better
received in the right ear, whereas orchestrated melodies
(Kimura. 1964), sonar signals (Chaney and Webster. 1966),
environmental sounds (Curry. 1967). 2-Click thresholds
(Murphy and Venables. 1969) and rapid pitch changes (Darwin.
1969) are better received in the left ear.

If differential ear preference is a function of cerebral
dominance. as Kimura (1961) contends, and it has been found to
be independent of order in which the signals to the two ears
are recalled (Bryden. 1963). then it is reasonable to assume
that there are differences in the ability of the two hemi—
spheres to process different types of sounds. Because it is
aSsumed that each ear has greater neural representation in the
OPposite cerebral hemisphere. the predominance of the left
hemisphere for speech. for example, is reflected in superior
reCognition for words arriving at the right ear (Kimura. 1967).
In effect. the above stxdies (Kimura, 1961; Broadbent and

Gregory, 1969; Borkowski, Spreen and Stutz. 1965; Shankweiler

 

an

and Studdert—Kcnnedy. 1967) with dichotically presented pairs
of speech stimuli support this contention.

Studies using monotically presented speech stimuli
.have investigated the question of whether the hemispheres,
when stimulated unilaterally. will function comparably with
each other. or whether. in fact. there is a dominance Of one
hemisphere for particular stimuli, as has been evidenced in
dichotic listening tasks. If the latter holds true, then

superior performance of the ear contralateral to the dominant

 

hemisphere for that stimulus could be expected. Dirks (1964)
presented'filtered phonetically balanced words both diehotically
and monotically to normal listeners. He Obtained a significant
right ear superiority under the dichotic listening condition.
but only a very small, nonsignificant right ear superiority
under the monotic condition. Kimura. using both words and
nonsense syllables, found significant right ear advantages
when subjects reported only one ear at a time in both dichotic
listening tasks. but no significant difference in the monotic
tasks. Corsi (1967) found no significant ear advantages in
three different tests of speech perception when they were
presented monotically. I

It should be noted that in all of the above monotic
.studies; only Kimura's word study failed to find at least
some small difference in favor of the right ear. It could
thus be argued that there may, in fact, be a right ear
superiority under monotic presentation conditions, but one

which was too small to be detected in a statistical analysis

 

.1: v“

.04

I
«I-vv“

L... .‘a.

v‘ "Q

Inuit

H vb

a... .

'ii-

 

Fi“

‘\‘.
r‘)

(l)

C}
D

W _
l (“N (to
d l I) I o)

“”5

Of the above studies. This contention is substantiated by
the results of a survey of 3.465 subjects (Glorig. g: g_.
1958). A slightly lower right ear threshold for both speech
.and pure tones was found with monotic presentation of the
stimuli, but this was of a lower magnitude than is typical
under dichotic listening conditions. Thus, it is necessary
to consider ear effects in clinical audiometry. although to

date. such effects appear to be minimal in monotic listening

tasks. Since this study also revealed no ear effects. it , 1

appears that it is clinically useful monotically and will not \ '~ 5

 

be confounded by ear laterality effects. This conclusion is
tentative. however. due to the small and unequal number of
subjects per right or left ear condition presented with each

Of the four test lists at a given sensation level in this study.

Further research is thus indicated to establish its validity.

List Effects. All lists are essentially equal at 0% time
compressiOn whereas higher ratios of time compression as well
as lower sensation levels serve to increase the interlist
variability. Further. List IV overall is the easiest and
List I is the hardest. It should be recognized, however,
that the order of list presentation remained constant for all
subjects under each condition of time cOmpression. Despite
the fact that intensity order was counterbalanced within each
of these conditions. it is possible (see Figure 4) that order
effects of presentation come into play under the conditions
of greater degrees of time compression. establishing Lists 1.

II. III. and IV as hardest to easiest. respectively. Therefore,

 

 

86

in the formulation of a clinical discrimination test using
time—compressed speech stimuli. the effects of adapting to
the discrimination‘task. resulting in improved scores is
.worthy of further investigation in order to avoid confusing

“the effects of practice with discrimination ability.

Theoretical Implications

 

Temporal considerations with regard to the processing

of speech stimuli have received a considerable amount of 1

 

'attention in the literature. much of which has been focused f

on the interpretation of shortfterm memory functions.
Theoretical models typically describe two stages of percep-
tion involved in the performance of immediate recall tasks
(Pollack. 1959; Mackworth. 1959; Sperling, 1960. 1963;
Sternberg. 1964; Broadbent, 1957a, 1957b, 1958).

In the first stage. the stimulus representation are
thought to be unidentified sensations which are unstable and
subject to change over the passage of time or as a result of
intervening events. These representations are held in a large
capacity perceptual storage system which can receive more the.
one item simultaneously. In the second stage. processing Or
encoding of the individual stimulus items taken from the \
storage system occurs. Each element is encoded one at a time
and these representations are more permanent than those in
stage one. .

In most models of short-term memory. according to
Aaronson (1967), the extent to which errors occur during

Perception rather than during subsequent retention or retrieval

1 .

‘2
AL—— ".-

 

 

 

L17 .
of the information cannot be determined directly. The roles
of stimulus duration and the rate of stimulus presentation
have been speculated upon in relation to the aforementioned
,theoretical framework. Aaronson (1967) for example. suggests

that stimulus duration might determine the amount of stimulus

‘7

information that goes into a buffer storage (stage one), while
duration of the interstimulus interval might affect time avail—

able for encoding the representation. She further contends

 

that the amount of time between stimulus items, during which
the stimulus is available to the subject for perceptual pro-
cessing. may be a more impertant factor than physical stimulus
duration in determining recall accuracy. In support of this
contention. she cites the study by Fairoanks, 23 gl. (1957)

in which subjects were presented time-compressed messages and
messages of normal rate. The comprehension ability was judged
to be poorer for the condition of time compression than that
for the normal rate. To isolate the effects Of stimulus

duration on the intelligibility of the message, Fairbanks and

1+
'V

03

Kodman (1957) used time-Compressed monosyllables presented
uniform interstimulus intervals to normal listeners. They
found that 80% of the original duration of the words could be
dis carded before intelligibility would drOp below 95% correct.
6 The inference draVIn by Aaronson from the two preceding
studies concerni gthe relative importance of interstimulus
interval over stimulus duration in the processing of auditory

information is contraindicated to an extent by the results of

the present study. The present findings reveal that once a

3.-—_-
l

"'1

 

 

-.-.—.,. o... H -. ._~. ~_

#8

aficical duration is reached. at which point between 60% and
Ci of the signal is deleted. the intelligibility of the
onosyllabic unit is sigpificantly reduced. Further. the
iberal amount of time provided after each word presentation,
orresponding to what would be considered the encoding stage,
id not appear to compensate for the reduced informational
edundancy provided by the acoustic signal which was thus
educed in duration.

Beasley (1970) provided further indication of the
aportance of word duration as Opposed to interstimulus
xterval duration in the perception of various orders of
:ntential approximations. He states, ". . . increasing
i0 interstimulus interval does not necessarily offset the
gradating effects of a decrease in word duration in
ntential stimuli“ (p. 36). His findings indicate. in
ct. that word duration was a more crucial variable in
call accuracy.

Thus. in tasks not involving comprehension, the rela-

ve importance of the interstimulus interval in neutralizing

('3

3 effects of reduced stimulus duration in minimized. In th-
waessing of complex material, the use of the interstimulus
:erval.to process semantic as well as non-verbal cues may be

cxnisiderable importance. In the intelligibility of isolated

‘J

 

 

‘J—fb-A. I
i

I.
tug-'91.

Tr: or synthetic sentences. however._where semantic constraints
iiot as great. interstimulus interval duration cannot compen-
<3.for the reduced word duration in improving performance once

chiration of the word has been reduced beyond a critical

ation.

 

”9

Although the study bnyeasley is not exactly comparable
to the current study due to stimulus differences, it does
indicate that the use of complex strings of verbal stimuli,
.such as synthetic sentences (Speaks and Jerger, 1965; Beasley
(1970), as a stimulus for testing higher auditory centers,

may be unnecessary. That is. the monosyllabic lists, which :

..... ""f

provide less variability than sentential material, may be used
for this purpose if they are apprOpriately distorted. Further
work on pathological pepulations should provide valuable

information relative to this contention.

 

I
a

implications for Further Research

Perhaps the most obvious area requiring further research
is that concerned with the responses of a clinical pepulation
to the stimuli used in this study. Specifically. the effects
of time-compressed monosyllables on the discrimination ability
of subjects with conductive. cochlear. retrocochlear, and
central nervous system disorders must be examined to assess
its actual utility in the differential diagnosis of central
auditory lesions. Further. such an investigation would serve
to derive precise combinations of time compression and sensa-
tion level which wouldmbe considered most diagnostically
Valuable.

Another potential area of investigation lies in the
comparison of time-compressed monosyllables with other forms
0f distorted speech signals developed for the assessment of

central auditory lesions. A discussion of specific techniques

 

is of little value at the present time since there are
variations to each type of test. Further. standardization
of stimulus material. procedures and instrumentation for
these tests is presently lacking. On the basis of current
research. however, each distinct method of speech distortion
provides diagnostic information relative to particular sites
.
'in the central auditory system. To date, frequency distortion
‘has been found to be most useful in the identification of
lesions above the third order neuronal level, whereas inter-
rupted, switched and time-distorted speech signals are more
indicative of lesions at the second order neuron level
(Bocca and Calearo, 1963). Additional diagnostic utility
has been indicated for time-distorted speech, however, in
the identification of diffuse damage of the higher auditory
pathways and of lesions occurring at or above the level of

est scores 1n

the third order neuron. resulting in adverse t
the ear contralateral to the site of lesion (Calearo and
Lazzaroni. 1957; deQuiros. 1964). The roles of each form of
speech distortion can be viewed in the context of a test
battery. for potential use in site of lesion testing. The
attempt of this study to use standardized clinical methodology
and stimulus items in the presentation of time—compressed
speech‘stimuli can be seen as a springboard towards the
development of such a battery.

The findings of this study suggest the need for investi-
gating the effects of prolonged exposure to time—compressed '

speech stimuli on the discrimination ability of the listener.

 

III —lL-‘ L—h;
. r .

 

 

.
,..‘.A

..->%'-

'0
‘p—-.'

villi.

Int—mm

I!|V~.
-

.'fi"’
.,

1
‘vV. v

'rr- g

-..‘; k

I
“I‘m
..

'uv.\

A)

q.

'a.‘

!"n'.

“mu”

7";-

L.
Gluy

IAN»,
‘th

.
m...
\

u‘v‘

’1,“-
”VF

51

Present results using four clinically standardized word lists
fbund to be essentially equivalent in difficulty under normal
conditions, suggest that under high percentages of time
.compression. the relative difficulty of the listening task
decreased with each successive word list presentation. Since
the order of list presentation was not rotated in this study,
there remains a need to determine whether this effect is due
to inequalities in the difficulty of the word lists under the
examined conditions of time compression, or whether, in fact,
it is due to order effects of presentation. It would appear.
if the latter should be the case, that lengthy tests employing
time-compressed stimuli should be avoided so as not to con-
found learning effects with an accurate appraisal of the
subject's discrimination ability.

A further area of investigation indicated by the
results 01 this study is related to the finding that. for
the left ear, maximum articulation scores were obtained at
the 30% time compression condition, rather than at the
condition of 0% time compression, as would be expected. This
would imply that perhaps the acoustic redundancy contained in
the speech signal serves to impede the Optimal functioning of
the auditory system relative to its processing of speech. In
-support of this contention, prior research has indicated that
normal hearing subjects preferred listening to speech compressed
by 30% (Zemlin. Daniloff and Shriner, 1968). Research directed
toward the question of whether an optimum signal duration for

the auditory processing of speech actually exists must

I a.- “u .

 

uninsur-

A. p- .. —- - _ E l- n. . - -a—p—--¢—a~m-». —.—o—--. __,, _

m--.~-rn- ..-. -p‘ .- -

 

52

necessarily take into account the context of the speech signal.'
iwwever, definitive findings concerning this issue may hold
significant implications for the training of individuals with
.auditorily based disorders of language processing.

Finally. the matter of ear differences relative to the
reception of time-compressed speech stimuli has only been
incidentally investigated in the present study. Due to the
random selection of test ear for each condition of time
compression. the number of subjects administered each of the
four lists at the same sensation level differed for each ear.

In additidn, the small number of subjects used to examine ear
effects for each condition of time compression does not allow
for a generalization concerning ear effects to be made at this
time. More rigorous investigation of possible laterality
effects. using larger and equal numbers of subjects for each
test ear, is warranted in order for a definitive statement ‘\

concerning laterality effects in the auditory processing of

time-compressed speech stimuli to be made.

CHAPTER V

SUMMARY

The results of this study, within the limits of the
experimental procedures employed, provide significant informa-
tion relative to the intelligibility of time-compressed speech
stimuli. It was found that increases in the compression ratio
up to the level of 60% generally resulted in gradual decreases
in the intelligibility of the word lists. An exception to this
was found with the 0% and 30% conditions of time compression,
in which case the mean left ear articulation scor-s for the
30% time compression condition exceeded the mean articulation
scores for the 0% time compression condition at all sensation

levels. Normative articulation functions, graphically depicting

the effects of intensity upon the intelligibility of the standard-

ized clinical word lists under varying degrees of time
compression, provide a basis for comparison with future studies
involving hearing-impaired listeners. The articulation scores
generated by the normal listeners examined in this study were
found to increase at the rate of 2-3.5%/dB for all conditions
of time compression. In addition, for all except the 70%
time compression condition, the mean articulation curves were
found to be curvilinear, reaching an asymptote at the sensation
level of 32 dB.

No differences between right and left ears were found.
Implications of this finding may have bearing upon current

theories of cerebral dominance for speech with respect to

monotic yersus dichotic listening tasks.

1 -sp— swan-um- .. . 1 '

- 21112: i;-

 

 

 

35

Figure 5.--Average articulation scores for each of four
test lists plotted by sensation level

 

 

 

54

Finally, the interlist comparisons of the Northwestern
Auditory Test No. 6 revealed essentially no list differences
at the 0% time compression condition, but for the other condi-
.tions of time compression, List I was generally found to be

most difficult whereas List IV was found to be easiest.

 

I 14""D‘: ' A;

.hw-wdc-o

 

 

LIST OF REFERENCES

Aaronson, D., Immediate recall of compressed speech. Unpub.
manuscript. Harvard Univ.. Center for Cognitive Studies:

Boston (1966).
Aaronson, D., Temporal factors in perception and short term
memory. Psychological Bulletin. 67, 130-144 (1967). 1

Beasley. D. S., Auditory analysis of time-varied sentential
approximations. Unpublished doctoral dissertation,
University of Illinois (1970).

Bocca, E.. Binaural hearing: another approach. Laryngoscone,
65, 1164-1171 (1955). \«

 

0
if)” .u a
‘

Bocca, E.. Distorted speech tests. Sensorineural Hearing
Processes and Disorders, A. B. Graham, ed., Boston:
Little, Brown, and Co. (1967).

Bocca. E.. Calearo, C., Cassinari, V.. and Migliavacca, F..
Testing cortical hearing in temporal lobe tumors.
Acta Oto-Laryngologica, 45, 289-303 (1955).

Bocca, E.. and Calearo. 0., Central hearing processes. Epdcrn
ngelopments in Audiology. J. Jerger. ed.. New York:
. . .r. /w
Academic Press (1963). 1
z \
Bordley, J. E.. and Haskins, H. L., The role of the cerebrum
in hearing. Annals of Otologyi,Bhinologv, and Larypu

gology, LXIV, 370-382 (1955).

Borkowski. J. G., Spreen. 0., and Stutz. J. 2., Ear preference
and abstractness in dichotic listening. Psychonomica;

Science, 3, 547-548 (1965).

.Breadbent, D. E.. Immediate memory and simultaneous stimuli.
Quarterly Journal of Experimental Psychology, 9, 1-11

(1957).

Broadtmnt. D. E.. A mechanical model for human attention and
immediate memory. Psychological Review, 64, 205-215

Brtmnlbent, D. E.. and Gregory. M.. Accuracy of recognition for‘
speech presented to the right and left ears. Quartgrly
Journal of Experimental Psychology, 16, 359-360 (1964).

 

 

 

J...

(1
‘V‘l

("j

I")

56

Bryden, M. P., Ear preference in auditory perception. Journal
of Egperimental Psychology, 65, 103-105 (1963).

Calearo, C., Binaural summation in lesions of the temporal
lobe. Acta Oto-Laryngologica, 47, 392-397 (1957).

.Calearo, C., and diMitri, T., Sulla intelligibilita della voce
periodicemente alternata da un orchio all altro.
Congress di Audiolggica. Padova, Italy (1958).

Calearo. C., Teatini. G. P..and Pestalozza, 0., Speech
intelligibility in the presence of interrupted noise.
Journal of Auditory Research, 2, 179-187 (1962).

Calearo, C., and Lazzaroni, A., Speech intelligibility in
relation to the speed of the message. Laryngoscope,
67. Lao-419 (1957) . \

Chaney, R. B.. and Webster, J. C., Information in certain
multidimensional sounds. Journal of the Acoustical
Society of America, 40, 447-455 (1966).

 

Corsi, P., The effects of contralateral noise upon perception
and immediate recall of monaurally-presented verbal
materials. Unpublished Master's Thesis (1967).

Curry, F. H., A comparison of left-handed and right-handed
subjects on verbal and non-verbal dichotic listening
' tasks. Cortex, 3, 343-352 (1967).

Daniloff. R. C., Shriner, T. H., and Zemlin, W. R., Intelli-
gibility of vowels altered in duration and frequency.
Journal of the Acoustical Society of America. 44,

700-707 (1968).

Darwin, C. J., Unpublished doctoral dissertation, University
of Cambridge (1969).

deQuiros, J., Accelerated speech audiometry, and examination
of test results (Trans. by J. Tonndorf). Translations
Egltone Institute of Hearing_Research, No. 17, Beltone
Institute of Hearing Research: Chicago (1964).

Dirks,~D., Perception of dichotic and monotic verbal material
and cerebral dominance for speech. Acta 0to-Laryngologica,

589 73’80 (1.964).

Eldert, E.. and Davis, H., The articulation function of patients
with conductive deafness. Lgryngoscgpe, 61, 891-909

(1951).

 

Fairbanks, 0., Everitt, W., and Jaeger, R., Methods for time
or frequency compression-eXpansion of speech.

Fairbanks, C., Guttman, N., and Miron. H., Effects of time
compression upon the comprehension of connected Speech.
Journal of Speech and Hearing_Disorders, 22, 10-19

(1957).

Fairbanks. G.. and Kodman,.F., Word intelligibility as a
function of time compression. Journal of thg
Acoustical Society of Amerigg, 29, 636:641 (1957).

Finzi, A., 11 comportamento della soglia di intellezione del
giovani con poacusia percetteiva e dei presbiaeusici
verso tests audiometrici vocali sensibilizzati.

Arch. It. di 0tol., 67 (1956).

Fournier. J. E.. L'analyse et l'identification du message
sopore., JQur. Franc. 0to-Rhino-Laryngol.. 67 (1956).

Glorig, A., A report of two normal hearing studies. Apnals
9f Otology. Rhinology and Laryngology, 67, 93-112

(1958).

Hirsh, I. J., The 'Measurement of Hearing. New York: McGraw-
Hill Book Co. (1952) p. 1M7.

Hirsh, I. J., Davis, H., Silve11an, S. R., Reynolds, E. G.,
Eldert. E.. and Benson, R. H., Development of materials
for speech audiometry. Journal of Spgech and Hearing
Disorderg, XVI, 321-328 (1952).

Jerger, J., Auditory tests for disorders of the central auditory
mechanism. Neurological Aspects of Auditory and Vesti-
bular Disorders. W. S. Fields and B. R. Alford, eds.
Springfield: C. C. Thomas (1964).

 

Jerger, J., Audiological manifestations of lesions in the
audétpry nervous system. Laryngoscope. 70, 417-425
19 0 .

Katz, J., The use of staggered spondaic words for assessing
the integrity of the central auditory nervous system.
Journal of Auditory Research, 2, 327-337 (1962).

Katz, J., Differential diagnosis of auditory impairments.
Audiometry for the Retarded. Robert T. Fulton and
Lyle L. Lloyd, eds. Baltimore: Williams and Wilkins

Company (1969).

 

 

 

 

Kimura, D., Some effects of temporal-lobe damage on auditory
perception. Canadian Journal of Psychology, 15, #3,

156-165 (1961).

Kimura, D., Functional assymetry of the brain in dichotic
listening. Cortex, 163-178 (1967).

' Lehiste, I., and Peterson, 0.. Linguistic considerations in
the study of speech intelligibility. Journal of the
ggcustical Society of America, 31, 280-286 (1959).

"53

Nackworth, J. F.. Paced memorizing in a continuous task.
Journal of EXperimental Psychology, 58, 206-211 (1959).

Miller, G. A., Language and Communication. New York: McGraw-
Hill Book Co. (1951) p. 77.

 

l$fl-fl-35§--_ he)! I '3" n ' ."s '
I

Miller. C., and Licklider, J., The intelligibility of inter-
rupted speech. Journal of the Acoustical Society of

America, 22, 167-173 (I950).

Miron, M.. and Brown, E.. Stimulus parameters in speech
compression. Journal of Communications, 18, 219-235

(1968).

Murphy, E. H., and Venables, P. H., Ear asymmetry in the thres-
hold of fusion of two clicks: A signal detection analysis.
Quarterly Journal of EXperimental Psychology, 288-300

(1970).

JPollack, 1., Message uncertainty and message reception. Journal
of the Acoustical Society of America, 31, 1500-1508

(1959).

Rintelmann, W. F. and Jetty, A. J., Reliability of speech \
discrimination testing using CNC monosyllabic words.
Unpublished study, Michigan State University (1968).

Ffixrtelmann, W. F. and Schumaier, D., Personal communication

(1971).

Sharflmweiler, D. P., and Studdert-Kennedy, M., Identification of
consonants and vowels presented to left and right ears.
Quarteply Journal of Exocrimental Psychology, 19,

59:5(1937) .

Shrdxuar, T. H., Beasley, D. S., and Zemlin,-W. R., The effects
of frequency division on speech identification in
children. Journal of Speech and Hearing_Research, 12,
No. 2, 413-4229(1969).

 

 

59

Speaks. C., and Jerger, J., Method for measurement Of speech
identification. Journal of Speech and HearinggResearch,

8, 185-194 (1965).

Sperling, C., The information available in brief visual
presentations. Psyghological Monographs, 74 (11, Whole
No. 498) (1960).

Sperling, C., A model for visual memory tasks. Human Factors. --. fl
5. 19—31 (1963). .4

Sternberg, 8., Two Operations in character recognition: Some ' .
evidence from reaction time measures. In, Models for
the perception of speech and visual form. AFRCL
Symposium, Boston (1964).

 

 

Thorndike, E. L., and Lorge, 1., Tgagper's Word Bock Of430,000
Words. New York: Bureau of Publications, Teachers
College, Columbia University. (1950).

Tillman, T. w.. and Carhart, R., An expanded test for speech
discrimination using CNC monosyllabic words: North-
western University Auditory Test No. 6. (1966).

Tillman, T. H., Carhart, R., and Wilbur, L. A., United States
Air Force School of Aerospace Medicine, SAM-TDR-62-135,
AD NO. 403275, Brooks Air Force Base, January, 1963.

willeford, J, A., Audiological evaluation of central auditory
1...")

disorders. Maico Audiological Library_Series, VI, 1 ,

(1969).

Zemlin. w, R., Daniloff, R. C., and Shriner. T. H., The
Difficulty of listening to time-compressed speech.
Journal of Speech and Hearinngesearch, 11, 875-881
{1968). ~m

 

- v“ A. - —_ .-
-- “mu—-7_‘—. ‘¢-. _. u-,..--- —% s b .n A.

 

APPENDIX A

FOUR LISTS OF FORM B
OF NU AUDITORY TEST NO. 6

 

 

19.
2o.
21.
22.
23.
2a.

25,

burn
lot
sub
home
dime
which
keen

yes

.boat

sure
hurl
door
kite

sell

death
love
tough
gap

moon

choice

king

61

\

N. U. AUDITORY TEST #6
N. U. RESEAICH P.B. LIST I FORM B

(or witch) 31.

37.
38.
39.
no.
41.
42.
43.

an;

-. ,}+5.
46.
47.
48.
1+9.
50.

size
pool
vine
chalk
laud

puff
jar
reach
rag

mode

' limb

third
jail
knock
whip

met

2‘ .A v’.' L
i
c
v. -.p-M

 

N.U. A

62

'1':

UDITORY TEST #6

N.U. RESEARCH P.B. LIST II FORM B

live
voice
ton
learn
match
chair
deep

pike

room '

read (to read)
calm
book
dab
loaf
goal
shack
far
witch
rot
pick
fail '
said
wag
haze

white

26.
27.
28.
29.
3o.

31.

32.

33.

3a.
35.
36.
37.
38.
39.
no.
41.
42.

m.
wu.‘
"#5.

H6.

. 47.

48.
“9.
50.

hush

dead

 

hate

.turn

.rain

shawl
bought
thought
bite
lore

south

23.
24.

I ‘1
A).

N.U.

sheep
cause
rat
bar
mouse

talk

' hire

search

'luck“

pain
base
mop

mess

 

N.U. AUDITORY TEST #6
RESEARCH P.B. LIST III FORM B

26.

27.

28.

29.

30.

31.

32.

33.

, 31+.

~ '- 35.
36.

37.

38.

39.

4o.

41.

42.

43. .
. 44.-

" '45,
46.
1+7.
48.
49.
so.

germ
thin
name
ditch
tell
cool
seize
dodge‘
youth
hit
late
ins
wire
walk
date
when
ring

check

.note

gun
beg
void
shall
lid

good

 

 

 

N.U.

wheat
thumb
near.
lease
yearn

kick

6M

N.U. AUDITORY TEST #6

RESEARCH P.B.

LIST IV FORM B

26.
27.
28.
29.
3o.
31.
32.
33.
34.

49 .
50 0

back
hall
bath
tire
peg
perch
chain
make
long
wash
food'
mood
neat
tape
ripe
hole
gas

came

' vote

lean
red
doll
shirt
sour

wife

a;

 

" ”fi—flfiuuV‘ .

I
c

. h- >4~.A- H¥V“HHI *~--uw-~—_.In..

APPENDIX B

INSTRUCTIONS GIVEN TO LISTENERS-

, .r g... —.. ohm—.4-

 

. “-2;

INSTRU‘TIOKS GIVEN TO LISTEPERS

You will now hear a tape recording of four lists, each
composed of fifty monosyllabic words. Each word is preceded by
.a carrier phrase, "You will say". Your task will be to write
down the word immediately following the carrier phrase in the
apprOpriate space provided on the answer sheet. For example,
if you hear, "You will say dog", you would be eXpected to write

the word "dog". There will be an ample amount of time provided

 

 

immediately after each word presentation for you to write down
your.response.

Each of the four lists will be presented to you at a
different intensity level. although all of the fifty words on
the same list will be equally loud. Some of the lists may
sound extremely soft, so it is of extreme importance that you
pay careful attention to the listening task. In addition, it
may seem that the words are spoken on this tape in an unusually
rapid manner, so again, pay close attention to what you hear
and respond to the best of your ability. If you are uncertain
of a response item, you are encouraged to guess. When you have
completed an entire word list, there will be approximately
twelve seconds before the items from the next list will be

presented. Are there any questions?

APPENDIX C

ANSWER FORM USED BY THE LISTENERS

fil’

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.kI.—" . I. . ‘A a’.
A

 

 

-—.-_.- h" ‘11".

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No. _ __ Form _
Name 1 Sex
Age yrs. mos. Date
1 26
2 27
3 28
4' 29
5 30
. 6 ‘ 31
7 a 32
8 33
9 p 34
10 35
11 . 36
12 # 37
13 f 38
14 f 39
15 , A 40
16 . _ f 41
' 17 ,_ _. 42
18 4.3,.
19 ~ . . 44
20 45
21 '46
22 ' 47
23 _‘ 48. _g
1 491

 

25 1 . _ I so

’rm-flw'rfi"' ''''''' A~—.-o,'—rrv-‘—..-1-o--v)--..ooe_pv ..... v o. J _-U ,' » - . ‘ - . ' ._ __ . on - - .. I ‘ , . 7 “

'74'1'3AN STATF l'J“fRS'T'v L'B'v‘-'

\“ “ l‘ i‘:|| | |W3|I""||
1

1293 03174 53 20