J Am Acad Audiol 22:491–500 (2011)

Performance of Normal Adults and Children on

Central Auditory Diagnostic Tests and Their
Corresponding Visual Analogs
DOI: 10.3766/jaaa.22.8.2

Teri James Bellis*†

Jody Ross*‡

Abstract

Background: It has been suggested that, in order to validate a diagnosis of (C)APD (central auditory
processing disorder), testing using direct cross-modal analogs should be performed to demonstrate that
deficits exist solely or primarily in the auditory modality (McFarland and Cacace, 1995; Cacace and
McFarland, 2005). This modality-specific viewpoint is controversial and not universally accepted (Amer-
ican Speech-Language-Hearing Association [ASHA], 2005; Musiek et al, 2005). Further, no such analogs
have been developed to date, and neither the feasibility of such testing in normally functioning individuals
nor the concurrent validity of cross-modal analogs has been established.

Purpose: The purpose of this study was to investigate the feasibility of cross-modal testing by examining
the performance of normal adults and children on four tests of central auditory function and their corre-
sponding visual analogs. In addition, this study investigated the degree to which concurrent validity of
auditory and visual versions of these tests could be demonstrated.

Research Design: An experimental repeated measures design was employed.

Study Sample: Participants consisted of two groups (adults, n 5 10; children, n 5 10) with normal and
symmetrical hearing sensitivity, normal or corrected-to-normal visual acuity, and no family or personal
history of auditory/otologic, language, learning, neurologic, or related disorders.

Data Collection and Analysis: Visual analogs of four tests in common clinical use for the diagnosis of
(C)APD were developed (Dichotic Digits [Musiek, 1983]; Frequency Patterns [Pinheiro and Ptacek,
1971]; Duration Patterns [Pinheiro and Musiek, 1985]; and the Random Gap Detection Test [RGDT;
Keith, 2000]). Participants underwent two 1 hr test sessions separated by at least 1 wk. Order of sessions
(auditory, visual) and tests within each session were counterbalanced across participants. ANOVAs (anal-
yses of variance) were used to examine effects of group, modality, and laterality (for the Dichotic/Dichoptic
Digits tests) or response condition (for the auditory and visual Frequency Patterns and Duration Patterns
tests). Pearson product-moment correlations were used to investigate relationships between auditory and
visual performance.

Results: Adults performed significantly better than children on the Dichotic/Dichoptic Digits tests. Results also
revealed a significant effect of modality, with auditory better than visual, and a significant modality 3 laterality
interaction, with a right-ear advantage seen for the auditory task and a left-visual-field advantage seen for the
visual task. For the Frequency Patterns test and its visual analog, results revealed a significant modality 3
response condition interaction, with humming better than labeling for the auditory version but the reversed
effect for the visual version. For Duration Patterns testing, visual performance was significantly poorer than
auditory performance. Due to poor test-retest reliability and ceiling effects for the auditory and visual gap-
detection tasks, analyses could not be performed. No cross-modal correlations were observed for any test.

Conclusions: Results demonstrated that cross-modal testing is at least feasible using easily accessible
computer hardware and software. The lack of any cross-modal correlations suggests independent pro-
cessing mechanisms for auditory and visual versions of each task. Examination of performance in indi-

*Department of Communication Sciences and Disorders, University of South Dakota, Vermillion; †Division of Basic Biomedical Sciences, Sanford
School of Medicine, Vermillion, SD; ‡Winnipeg Hearing Centres, Winnipeg, Manitoba, Canada
Teri James Bellis, PhD, Professor and Chair, Department of Communication Sciences and Disorders, University of South Dakota, 414 E Clark St,
Vermillion, SD 57069; Phone: 605-677-6201; Fax: 605-677-5767; E-mail: tbellis@usd.edu

491
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
Journal of the American Academy of Audiology/Volume 22, Number 8, 2011

viduals with central auditory and pan-sensory disorders is needed to determine the utility of cross-modal
analogs in the differential diagnosis of (C)APD.

Key Words: Central auditory processing, dichoptic, dichotic, gap detection, modality specificity, temporal
patterning
Abbreviations: (C)APD 5 central auditory processing disorder; LVFA 5 left-visual-field advantage;
REA 5 right-ear advantage; RGDT 5 Random Gap Detection Test; SRT 5 Speech Recognition
Threshold

M
odality specificity of central auditory process-
ing disorder ([C])APD) has been a key topic of
controversy in recent years. Cacace and
McFarland (2005; McFarland and Cacace, 1995) have
recommended that, in addition to central auditory diag-
nostic tests, performance on analogous tests in other
sensory modalities should be examined for possible
multimodal, pan-sensory, and/or task-related (rather
than auditory-specific) difficulties in children suspected
of (C)APD (central auditory processing disorder). These
authors have advocated for the development and use of
direct visual or other sensory analogs of behavioral
tests of central auditory function for purposes of differ-
ential diagnosis of (C)APD. Specifically, they state that,
because behavioral tests of central auditory function
can be affected by several nonauditory factors, includ-
ing memory, attention, language, and others, it is nec-
essary to vary systematically the nature of the stimulus
while holding other task-related factors constant to
demonstrate that performance deficits observed are
due, in fact, to the nature of the stimulus (e.g., auditory
versus visual), rather than to more global confounds.
Musiek, Bellis, and Chermak (2005) expressed concern
regarding the use of direct analogs to validate diagnosis of
(C)APD for several reasons. These include the fact that vis-
ual and related testing is not in the scope of practice of the
audiologist and the concern that there are no data demon-
strating the neurophysiologic substrates or sensitivity/
specificity of such tasks. As such, it is unclear whether
these analogs recruit similar processing mechanisms.
Indeed, in an investigation of the concurrent validity of
a dichotic listening task and its corresponding visual ana-
log, Voyer and Boudreau (2003) found no significant cross-
modal correlations between the auditory and visual tasks,
suggesting that each task measured independent brain
processes. These findings were consistent with those of
Wexler and King (1990) and Bellis et al (2008), who
also found no cross-modal correlations between audi-
tory and visual tasks in normal participants. However,
Bellis et al (2008) did find significant cross-modal corre-
lations between an auditory dichotic task and its corre-
sponding visual analog in a group of children diagnosed
with (C)APD presumed to involve the interhemispheric
pathways, which may suggest some degree of concur-
rent validity of these tasks in some individuals.
A more important concern revolves around the diffi-
culty inherent in creating true cross-modal analogs of
central auditory tests. That is, even when all efforts
are made to hold task-related factors constant to the
greatest extent possible, can the resulting measures ever
be truly analogous and equivalent? Not only is it difficult
to render such tasks equivalent in terms of difficulty lev-
els, it is also highly unlikely that such tasks could be
equivalent in terms of processing mechanisms required,
given the significant differences that occur in auditory
processing versus processing in other sensory modal-
ities. For example, as discussed in depth by King and
Nelken (2009), a substantial amount of auditory process-
ing occurs prior to processing by primary auditory cortex
(A1) via multiple subcortical relays. As a result, process-
ing in A1 is far more complex and represents a higher
level of processing than occurs in primary visual cortex
(V1), which receives projections directly from retinal
cells by way of visual thalamus only. These neurophysio-
logic and psychophysical differences between auditory
and visual processing represent a significant barrier
to constructing truly analogous auditory-visual cross-
modal tasks. Therefore, for purposes of this paper, the
terms analog and analogous are used for convenience;
however, the caveat that these measures may not be
truly analogous should be kept in mind throughout.
In short, the use of visual analogs of central auditory
tests in the diagnosis of (C)APD remains a contentious
topic. To compound the problem further, no direct vis-
ual or other analogs of behavioral central auditory tests
in current clinical use exist for this purpose at the
present time. Before one can assess the utility of such
analogs in the differential diagnosis of (C)APD from
other pan-sensory, multimodal, or task-related con-
founds, the feasibility of conducting such testing must
be demonstrated, and performance of typically func-
tioning adults and children must be elucidated.
The purpose of this study was to investigate the feasi-
bility of using visual analogs of commonly used central
auditory tests in the assessment of normal adults and chil-
dren and also to provide a baseline against which the per-
formance of individuals with disorders can be compared.
Specifically, this study sought to elucidate the per-
formance of normal adults and children on four central
auditory tests in common clinical use for the diagno-
sis of (C)APD and four corresponding visual analogs
developed by the investigative team to alter the mode
of stimulation (auditory to visual) while attempting
to hold all other task-related factors constant. Further,

492
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta

in an effort to determine whether similar processing
mechanisms are employed for each auditory/visual com-
bination, this study investigated the degree to which con-
current validity on these tests could be demonstrated via
cross-modal correlations.
METHOD
Participants
Participants for this study consisted of 10 adults
(mean age 5 23.2 yr, SD 5 2.04) and 10 children (mean
age 5 12.6 yr, SD 5 2.17) recruited from the general
community as part of a larger study investigating a
variety of auditory, visual, and electrophysiological
tasks in normal and disordered auditory processing
and attention. All participants evidenced normal and
symmetrical hearing sensitivity as measured by pure-
tone air- and bone-conduction thresholds ,15 dB HL
and #5 dB HL interaural difference for octave frequen-
cies 250–8000 Hz, normal findings on tympanometry
and acoustic reflex testing, and normal or corrected-
to-normal visual acuity. In addition, Speech Recog-
nition Thresholds (SRTs) were obtained to provide
for determination of intensity levels for the auditory
experimental tasks. Further, participants had no fam-
ily or personal history of diagnosed or suspected au-
ditory, language, learning, or related disorders, and
no history of otologic or neurologic disorder or insult.
Finally, all participants were native speakers of Eng-
lish and were consistently right-handed for nine of the
most reliable items on the Edinburgh Handedness
Inventory (Oldfield, 1971; Raczkowski et al, 1974).
All participants were naı̈ve to the experimental tasks
and were paid for their participation. Informed con-
sent was obtained from each participant or his/her
legal guardian; child assent was obtained from each
child. Ethical approval to conduct the study was
obtained through the Institutional Review Board of
the University of South Dakota.
Procedures
Testing was conducted over two 1 hr test sessions,
separated by at least 1 wk. Order of sessions (i.e., vis-
ual, auditory) and tests administered within each ses-
sion were counterbalanced to control for possible order
or learning effects. All auditory tests were conducted
in a sound-treated audiological test booth. Stimuli
were routed from a compact disc player through a two-
channel audiometer meeting ANSI (2004) standards
and were delivered via ER3a insert earphones. All vis-
ual tests were conducted in a private office with the
door closed and outside noise minimized, with black
stimuli presented on a white background using a com-
puter and 17 in LCD monitor.
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
Visual Analogs of (C)APD Tests/Bellis and Ross
Materials
Four central auditory tests were selected for in-
clusion in this study: Dichotic Digits (Musiek, 1983),
Frequency Patterns (Pinheiro and Ptacek, 1971), Dura-
tion Patterns (Pinheiro and Musiek, 1985), and the
Random Gap Detection Test (RGDT; Keith, 2000).
These tests were selected because they are in common
clinical use for the diagnosis of (C)APD, they represent
categories of central auditory testing recommended by
current consensus documents for central auditory assess-
ment (American Speech-Language-Hearing Association
[ASHA], 2005; American Academy of Audiology [AAA],
2010), and their test constructs rendered them particu-
larly suitable for translation into the visual modality.
Further, three of the tests—Dichotic Digits, Frequency
Patterns, and Duration Patterns—have been shown to
be sensitive, specific, and efficient for the identification
of dysfunction of cortical and interhemispheric central
auditory pathways (e.g., Musiek et al, 1980; Musiek,
1983; Musiek, Gollegly, et al, 1984; Musiek, Kibbe,
et al, 1984; Musiek and Pinheiro, 1987; Musiek et al,
1990; Musiek, 1994). The fourth test, the RGDT, was
selected because of its commercial availability and com-
mon clinical use at the time of the study and because gap-
detection/auditory-fusion paradigms have been shown
to be sensitive to cortical, particularly left-hemisphere,
dysfunction (Lackner and Teuber, 1973).
Visual analogs of each central auditory test were
developed using SuperLab Pro Experimental Labora-
tory Software version 2.0 (Cedrus Corporation, 1999).
All attempts were made to mirror precisely the task
demands; response requirements; content, number,
and order of stimulus items; instructions to the partic-
ipant; and all other characteristics of the analogous
auditory task. However, some minor stimulus-related
modifications were necessary to render the tests feasi-
ble in the visual modality. Further, for all of the visual
tasks, the interstimulus intervals varied as the signal to
advance to the next stimulus was triggered following
the participant’s response. This allowed for maximiza-
tion of performance without the confounding effect of a
time-based task. For the same reason, and consistent
with clinical administration of these tests, it was deter-
mined in advance that the auditory tasks would be
paused between stimuli if additional response time
was required; this situation did not occur during the
current study for any participant. A description of each
auditory test and its corresponding visual analog,
including modifications, follows.
Dichotic/Dichoptic Digits
The Dichotic Digits test (Musiek, 1983) consists of
25 stimulus items with each composed of two digit
pairs separated by an interstimulus interval of 3 sec
493
Journal of the American Academy of Audiology/Volume 22, Number 8, 2011
presented to each ear simultaneously in a dichotic man-
ner at 50 dB SL re: SRT, for a total of four digits per
stimulus item. Stimuli are numbers from 1 to 10
(excluding 7) delivered by a male speaker of Standard
American English. Participants were instructed to
report all four numbers heard, or as many as possible,
in a free-recall paradigm, and were encouraged to guess
if unsure. Performance was scored as a function of per-
cent correct per ear, and a subtraction of left- from
right-ear performance yielded a measure of right-ear
advantage (REA). Practice items were provided before
testing to ensure participants’ understanding of the
task.
For the visual analog (Dichoptic Digits), stimuli con-
sisted of the same digit pairs in the same presentation
order as in the auditory paradigm (two digits on each
side of a central fixation marker), except that the num-
ber 1 was substituted for the number 10, which could be
construed by the participant as two separate digits
(1 and 0). The substitution of 1 was chosen because the
alternative (0) could have been viewed as the capital let-
ter O by the participants and also because verbal report
of the number 0 requires a bisyllabic utterance. Black
stimuli were presented tachistoscopically on a white
background using a computer and 17 in LCD monitor.
A black cross (1 cm width and height) served as a cen-
tral fixation marker. One pair of digits (one digit to the
right and left of the central fixation marker) was pre-
sented for a duration of 200 msec to allow for recogni-
tion of both digits while at the same time ensuring
visual half-field presentation by eliminating the oppor-
tunity for optical hemisaccades that would allow for the
acquisition of the targets via central (foveal) vision.
This was followed by an interstimulus interval of 100
msec, after which the second pair of digits was pre-
sented for a duration of 200 msec. Therefore, the overall
stimulus duration was, by necessity, longer for the vis-
ual dichoptic digits task as compared to the auditory
dichotic task. Digit stimuli were 1.5 cm in height and
1 cm in width, with the distance between the central
fixation marker and the inner edge of the leading stim-
ulus subtending 3˚ of visual angle. A chin rest was used
to reduce head movement. Instructions to the partici-
pants, provision of practice items, and methods of scor-
ing were the same as those for the auditory dichotic
listening task, with performance scored as percent cor-
rect per visual field. In addition, a subtraction of the
left-visual-field score from the right-visual-field score
yielded an intratest comparison measure of Visual Field
Advantage (VFA).
Frequency Patterns/Visual High-Low
Stimuli for the Frequency Patterns test (Pinheiro and
Ptacek, 1971) consist of triads of tone bursts that differ
in frequency (high 5 1112 Hz, low 5 880 Hz) delivered
494
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
binaurally at 50 dB SL re: 1000 Hz pure-tone threshold.
Stimuli were 200 msec in duration with an interstimu-
lus interval of 150 msec. Consistent with clinical recom-
mendations for administration of this test, 25 stimulus
items were presented for each of two response condi-
tions: linguistic labeling, in which the participant
was instructed to label the three-tone pattern presented
(e.g., high-high-low, low-high-low), and humming, in
which the participant was instructed to mimic nonver-
bally the pattern. Percent correct per response condi-
tion was calculated. As with all tests administered,
practice items were provided prior to testing to ensure
participants’ comprehension of the task.
A Visual High-Low paradigm was created as a visual
analog to Frequency Patterns. Stimuli consisted of a
black rectangle measuring 6 cm in width by 4 cm in
height presented at horizontal midline on a white back-
ground. Triads of stimulus presentations were con-
structed with the visual stimuli appearing at “high”
(bottom edge at 4 cm above center) or “low” (top edge
at 4 cm below center) locations on the screen. Twenty-five
stimulus triads were presented for each response condi-
tion (linguistic labeling, humming) in the same order as
in the auditory Frequency Patterns task. Each individual
stimulus item was presented for a duration of 500 msec,
with a within-triad interstimulus interval of 50 msec. The
duration and interstimulus interval were chosen based
on pilot studies during the test development stage that
showed that these parameters were more than adequate
for normal adults to perform the task while, at the same
time, not being so long as to tax working memory. Fur-
ther, pilot studies clearly demonstrated that the 50 msec
interstimulus interval was entirely sufficient for the per-
ception of a clear demarcation between successive stim-
uli, even when they occupied the same spatial position on
the screen. Spatial position (high versus low) was chosen
rather than visual frequency (e.g., color) so as to hold the
task and responses constant across auditory and visual
modalities (i.e., labeling and humming of “high” and
“low” stimuli). In addition, there is evidence to suggest
that brain regions involved in processing demands for
this task are similar across modalities in that processing
of both visual-spatial position and tonal patterns are, at
least initially, subserved primarily by right-hemisphere
brain regions (e.g., Blumstein and Cooper, 1974; Joseph,
1988). Instructions to participants, response conditions
(i.e., labeling, humming), provision of practice items,
and methods of scoring were the same as for the analo-
gous auditory task.
Auditory and Visual Duration Patterns
The auditory Duration Patterns task (Pinheiro and
Musiek, 1985) is similar to the auditory Frequency Pat-
terns task in that stimuli consist of triads of tone bursts.
For this task, 25 triads of 1000 Hz tone bursts differing

in duration (long 5 500 msec, short 5 250 msec) were
presented at 50 dB SL re: 1000 Hz pure-tone threshold
in each of the two response conditions, linguistic label-
ing and humming. Scoring of responses was identical to
that of the auditory Frequency Patterns task.
The visual Duration Patterns task utilized the same
black rectangles described previously for the Visual
High-Low task. For this task, stimuli (long 5 800 msec,
short 5 250 msec) were presented at horizontal and ver-
tical midline, with a within-triad stimulus interval of
100 msec. Twenty-five stimulus items were presented
in each report condition and in the same order as in
the auditory paradigm. It should be noted that the “long”
visual stimulus (800 msec) was longer in duration than
the “long” auditory stimulus (500 msec) due to the in-
creased difference limen for duration discrimination in
the visual modality (e.g., Silva et al, 2007). A pilot study
conducted during the task-development stage for the vis-
ual paradigms confirmed that a larger difference was
necessary for typically developing individuals to perceive
a difference consistently. Scoring per response condition
was identical to that described previously for the other
temporal patterning tasks.
Auditory and Visual Gap Detection
To assess auditory gap-detection performance, the
RGDT (Keith, 2000) was employed. This test consists
of 500, 1000, 2000, and 4000 Hz 500 msec pure tones
and clicks in which a silent gap is inserted. The duration
of the gap is varied randomly from 0 to 40 msec, with
gap intervals of 0, 5, 10, 15, 20, 25, 30, 35, and 40 msec.
The listener’s task is to indicate whether one or two
stimuli were perceived for each stimulus presentation.
The entire test consists of nine stimulus presentations
per frequency, and the lowest gap identified at each
stimulus frequency is ascertained. Practice items con-
sisting of nine nonrandom stimuli with progressively
larger gap intervals are provided. The composite gap
detection threshold is calculated by the simple arith-
metic mean of the results obtained at each stimulus fre-
quency. In addition, the click gap detection threshold is
calculated by identifying the lowest gap correctly iden-
tified for the click stimuli.
The visual gap-detection paradigm was constructed
to be identical to the auditory RGDT (click stimuli) in
the areas of stimulus presentation order, gap durations,
and response requirements. The same black rectangles
used in the previous two paradigms were utilized and
were presented on a white background at horizontal
and vertical midline for a duration of 500 msec.
Blank-screen gaps of 0 to 40 msec were inserted into
stimulus presentations in the same order as in the audi-
tory RGDT (click stimuli) for both practice and test
items. As with the auditory task, participants were
instructed to indicate whether one or two stimuli were
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
Visual Analogs of (C)APD Tests/Bellis and Ross
perceived. The gap detection threshold was calculated
in the same manner as in the auditory task.
RESULTS
F or the digits and temporal patterning tasks, two
levels of data analysis were performed, with the
first examining group differences in test performance
and the second examining cross-modal correlations.
In addition, Cohen’s d was calculated as a measure of
effect size for each pairwise comparison. As will be dis-
cussed below, statistical analyses were not possible for
the gap detection tasks due to inconsistent test perform-
ance across groups for the auditory version and ceiling
effects for the visual version.
Dichotic/Dichoptic Digits
Results of Dichotic/Dichoptic Digits were subjected to
a 2 3 2 3 2 analysis of variance (ANOVA) with Group
(adults, children), Modality (auditory, visual), and Lat-
erality (right, left) as the between-group, or independ-
ent variables, and percent correct performance as the
within-group, or dependent, variable. ANOVA revealed
a significant effect of Group (F(1,72) 5 7.093; p , .01;
d5 .55), with adults exhibiting better overall perform-
ance than children. ANOVA also revealed a significant
effect of Modality (F(1,72) 5 11.156; p , .01; d 5 .71),
with overall performance in the auditory modality bet-
ter than that in the visual modality. These results are
displayed in Figure 1. Finally, ANOVA revealed a sig-
nificant Modality 3 Laterality interaction (F(1, 72) 5
5.254; p , .05), with an REA observed in the auditory
modality but the reversed effect (a left-visual-field
advantage [LVFA]) observed in the visual modality.
This is illustrated in Figure 2. No significant effect of
Laterality (F(1,72) 5 .0, p 5 1.0; d 5 .0) and no other sig-
nificant interactions were found. In fact, when collapsed
across group and modality, mean scores for Laterality
were identical (right595.525, left595.525), although var-
iability was somewhat larger for right-ear/visual-field
scores (SD 5 6.75) than for left (SD 5 4.536). This find-
ing was due to an equal and opposite effect of the REA
for Dichotic Digits and the LVFA for Dichoptic Digits in
this group of normal children and adults.
Frequency Patterns/Visual High-Low
Results of Frequency Patterns/Visual High-Low were
subjected to a 2 3 2 3 2 ANOVA with Group (adults,
children), Modality (auditory, visual), and Response Con-
dition (humming, labeling) as the between-group, or inde-
pendent, variables, and percent correct performance as
the within-group, or dependent, variable. ANOVA revealed
no significant main effects of Group (F(1,72)5 1.178; ns;
d5.297), Modality (F(1,72)5.550, ns; d5.164), or Response
495
Journal of the American Academy of Audiology/Volume 22, Number 8, 2011

Figure 1. Performance of Group 1 (Adults) and Group 2 (Children) on Dichotic/Dichoptic Digits by ear/visual field (right, left) and modal-
ity (auditory, visual). Error bars represent standard error of the mean.

Condition (F(1,72) 5 .198, ns; d 5 .098). A significant
Modality 3 Response Condition interaction was found
(F(1,72) 5 4.951, p , .05), with performance in the hum-
ming condition better than in the labeling condition for
the auditory task, but the reverse finding for the visual
task. These results are shown in Figure 3. There were no
other significant interactions.
Auditory and Visual Duration Patterns
As with the previous test, results of auditory and visual
Duration Patterns were subjected to a 2 3 2 3 2 ANOVA
with Group, Modality, and Response Condition as
between-group factors and percent correct performance
as the within-group factor. ANOVA revealed a significant
main effect of Modality (F(1,72) 5 45.674; p , .01; d 5
1.53), with performance for the visual task significantly
poorer than that for the auditory task. There were no sig-
nificant main effects of Group (F(1,72) 5 .594, ns; d 5 .14)
or Response Condition (F(1,72) 5 1.037, ns; d 5 .18) and
no significant interactions. These results are displayed
in Figure 4.
ipants exhibited visual gap detection thresholds of
5 msec. Therefore, statistical analyses could not be per-
formed on results of the gap-detection testing.
Cross-Modal Correlations
Pearson product-moment correlations revealed no sig-
nificant correlations between overall test performance in
the auditory and visual modalities for Dichotic/Dichoptic
Digits, Frequency Patterns/Visual High-Low, or Auditory/
Visual Duration Patterns. As previously stated, correla-
tion analysis could not be performed on gap-detection
results. These results indicate that individual overall test
accuracy in the auditory modality is not related to overall
task accuracy in the visual modality for any of these tasks
in normal individuals.
Similarly, no significant cross-modal correlations were
found for any ear/visual-field combination for the Dichotic/
Dichoptic Digits task. To compute response asymmetries,
left-ear or visual-field responses were subtracted from

Auditory and Visual Gap Detection

Performance of all participants on the auditory RGDT

was highly variable with extremely poor test-retest reli-
ability, which rendered it impossible to obtain reliable
gap detection thresholds for all participants. Specifically,
gap detection thresholds of 5 msec were obtained from
two children and three adults. Thresholds could not be
calculated for three children and four adults due to
inconsistent responses (e.g., reporting two stimuli during
0 or 2 msec gap durations yet reporting one stimulus dur-
ing significantly longer gap durations). The remaining
participants reported hearing two stimuli for all gap Figure 2. Ear and visual-field advantages obtained from Group
1 (Adults) and Group 2 (Children) on Dichotic/Dichoptic Digits by
durations, including the 0 ms condition. These problems degree of advantage (positive values indicate right advantage,
persisted despite reinstruction to ensure understanding negative values indicate left advantage) and modality (auditory,
of the task. For the visual gap-detection task, all partic- visual). Error bars represent standard error of the mean.

496
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
 Visual Analogs of (C)APD Tests/Bellis and Ross

Figure 3. Performance of Group 1 (Adults) and Group 2 (Children) on auditory Frequency Patterns and Visual High-Low tasks by
response condition (humming, labeling) and modality (auditory, visual). Error bars represent standard error of the mean.

right-ear or visual-field responses and divided by the sum
of the two values [(R 2 L)/(R 1 L)] (Repp, 1977). This pro-
vides an index of laterality that is independent of overall
performance. Analysis revealed no significant relationship
between auditory and visual laterality indices for this
task. These results are consistent with those of Wexler
and King (1990), Voyer and Boudreau (2003), and Bellis
et al (2008), who also found no significant cross-modal cor-
relations between auditory and visual laterality indices.
Finally, no significant cross-modal correlations were found
between auditory and visual modalities for the other two
tasks in either the humming or labeling response condi-
tions. The lack of any cross-modal correlations suggests
that the auditory tasks and their corresponding visual
analogs invoke independent processing mechanisms in
normal individuals.
DISCUSSION
oping children on four tests of central auditory function
and their corresponding visual analogs. A secondary
purpose was to explore the concurrent validity of these
auditory and visual homologous tasks through the use
of cross-modal correlation analyses.
With respect to the first purpose, results of this study
indicate that, for the Dichotic/Dichoptic Digits task,
adults’ overall performance was significantly better
and less variable as compared to the children. Although
the effects of maturation on dichotic listening have been
well-documented with results for the Dichotic Digits
task reaching adult values by 11 to 12 yr of age, there
are no studies to date that have examined maturational
effects on dichoptic performance. Therefore, these re-
sults suggest that, if such visual analogs are to be used
clinically for differential diagnostic purposes as sug-
gested by Cacace and McFarland (2005; McFarland
and Cacace, 1995), age-specific normative data must
be obtained for the visual measures.

T he primary purpose of this study was to investigate

performance of normal adults and typically devel-
For the auditory Frequency Patterns/Visual High-
Low and the Auditory/Visual Duration Patterns, no

Figure 4. Performance of Group 1 (Adults) and Group 2 (Children) on Duration Patterns by response condition (humming, labeling) and
modality (auditory, visual). Error bars represent standard error of the mean.

497
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
Journal of the American Academy of Audiology/Volume 22, Number 8, 2011
effects of subject group were observed. Although this
finding may suggest that maturation has little effect
on these tasks, this is inconsistent with previous find-
ings indicating clear maturational effects on auditory
temporal patterning tasks (e.g., Musiek, Gollegly,
et al, 1984; Bellis, 2003). A more likely explanation is
that participants in the current study (mean age 5
12.6 yr) were performing at adult levels as predicted
by previous studies showing that performance on these
tasks reaches adult levels by approximately age 11 or
12. As such, it is likely that, had Group 2 consisted of
younger children, significant group effects on these
tasks would have been observed.
For two of the measures (Dichotic/Dichoptic Digits
and Auditory/Visual Duration Patterns), auditory per-
formance was significantly better and less variable than
visual performance. An effect of Modality was not
observed for the Frequency Patterns/Visual High-Low
task. It is important to note that ceiling effects were
observed for right-ear Dichotic Digits performance for
the majority of the participants (7/10 adults and 7/10
children). However, ceiling effects were observed for
left-ear performance on the auditory task in only 2/10
adults with none of the children exhibiting 100% per-
formance for the left ear. For the visual dichoptic digits
task, ceiling effects were observed in 3/10 adults and
2/10 children for right-visual-field performance and in
4/10 adults and 3/10 children for left-visual-field per-
formance. Ceiling effects also were observed for the
auditory Frequency Patterns task in both the humming
(6/10 adults, 7/10 children) and the labeling (6/10
adults, 4/10 children) response conditions; similar ceil-
ing effects were observed for the Visual High-Low task
(7/10 adults and 3/10 children for humming; 9/10 adults
and 8/10 children for labeling). Finally, although ceiling
effects were observed in auditory Duration Patterns in
both the humming (7/10 adults, 2/6 children) and label-
ing (4/10 adults, 1/10 child) response conditions, such
effects were not observed in the visual task in either
response condition for either group.
The presence of ceiling effects on the auditory tasks
used in this study in normal individuals has been shown
previously (e.g., Bellis, 2003). As such, the utility of these
tasks for investigation of the magnitude of hemispheric
laterality effects in normal subjects likely is limited.
However, the sensitivity and specificity of these tasks
for the detection of dysfunction in the central auditory
pathways have been well-documented (e.g., Musiek et al,
1980; Musiek, 1983; Musiek, Kibbe, et al, 1984; Musiek
and Pinheiro, 1987; Musiek et al, 1990; Musiek, 1994).
Therefore, despite the presence of ceiling effects in nor-
mal subjects, these auditory tasks have been shown to
have clinical utility for the diagnosis of central auditory
dysfunction in disordered populations. Further, for the
Dichotic/Dichoptic Digits task, a significant Modality 3
Laterality interaction was observed in that an REA was
498
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
seen in the auditory task; however, an LVFA was seen
in the visual task. This finding was most pronounced in
the children. These results are consistent with previous
reports of reversed cerebral asymmetry for auditory
dichotic versus visual dichoptic performance, particu-
larly in children (Voyer and Boudreau, 2003; Bellis
et al, 2008). It is hypothesized that this effect likely is
due to the preeminence of the right hemisphere for the
processing of visual-spatial features of orthographically
presented stimuli, in contrast to the well-recognized
left-hemisphere dominance for the processing of verbally
presented stimuli (Voyer and Boudreau, 2003; Bellis
et al, 2008). As such, a left-visual-field (right hemisphere)
advantage might be predicted for dichoptic stimuli such
as these.
An interaction between Modality and Response Con-
dition also was observed in the auditory Frequency
Patterns/Visual High-Low task, with performance in
the labeling condition poorer than in the humming
condition for the auditory modality, but the reversed find-
ing in the visual modality. The presence of better hum-
ming than labeling performance for the auditory version
of this task has been well-established in the literature,
with this difference being attributed to the need for inter-
hemispheric transfer when verbal report is required (e.g.,
Musiek, Gollegly, et al, 1984). A likely explanation for the
reversed pattern in the visual homolog may simply
be that this finding reflects difficulty in transferring a
visual-spatial code into a pitch-based humming code as
is required by the humming response condition. In con-
trast, the verbal report of “high” or “low” relative to spa-
tial position on the screen may simply be an easier and
more intuitive task. Whatever the reason, this finding
clearly indicates that different processing mechanisms
come into play for these auditory-visual analogs, even
when attempts have been made to maintain the same
overall task requirements across modalities.
Finally, the presence of significant variability and poor
test-retest performance for the auditory gap-detection
task used in this study (the RGDT) precluded the ability
to evaluate group differences or cross-modal correlations
for this task. These findings also may raise concern re-
garding the clinical utility of this task. It should be noted
that shortcomings to the RGDT as a clinical tool have been
reported previously (e.g., Chermak and Lee, 2005).
As to the second purpose of the study—the degree to
which the auditory and visual analogs demonstrate con-
current validity as measured by cross-modal correlation
analyses—results revealed no relationship between the
auditory tasks and their corresponding visual analogs.
These findings are consistent with previous reports of
no correlation between auditory and visual dichotic/
dichoptic tasks and have been interpreted as indicating
that the auditory and visual tasks clearly represent
separate and distinct cognitive processing mechanisms
(e.g., Wexler and King, 1990; Voyer and Boudreau,

2003; Bellis et al, 2008). However, an alternative con-
clusion is equally likely. That is, due to the ceiling
effects and reduced range observed, performance on
these tasks may have precluded the ability to evaluate
cross-modal relationships in these healthy individuals.
That this is a possibility is supported by the findings of
Bellis et al (2008), who did find significant cross-modal
correlations in dichotic/dichoptic digits testing in children
with central auditory disorders. Therefore, it may be dif-
ficult to determine the degree to which these tasks recruit
independent or separate processing mechanisms in nor-
mal adults or typically developing children. Further, the
mechanisms underlying the auditory paradigms used in
this study are relatively well-understood and have been
supported by both behavioral data from listeners with
known lesions of the central auditory nervous system
as well as through physiologic means (e.g., Kimura,
1961; Milner et al, 1968; Sparks and Geschwind, 1968;
Musiek et al, 1980; Musiek, 1983; Musiek, Gollegly,
et al, 1984; Musiek and Reeves, 1986; Musiek and
Pinheiro, 1987; Musiek et al, 1990; Musiek, 1994; Aiello
et al, 1994; Hugdahl et al, 1999; Wioland et al, 1999;
Westerhausen et al, 2006). In contrast, and as empha-
sized by Musiek et al (2005), we do not know what neuro-
physiologic and processing mechanisms underlie visual
analogs of central auditory tests. Certainly, the fact that
auditory and visual processing differ substantially from
one another from both a neurophysiologic and psycho-
physical perspective (King and Nelken, 2009) suggests
that such “analogs” quite likely are not truly analogous.
Therefore, the utility of visual analogs of auditory tests
in the differential diagnosis of central auditory dysfunc-
tion remains unclear at the present time.
Finally, results of this study indicate that, for two
of the measures, the visual tasks were more difficult
than the auditory tasks. This may be due, at least in
part, to the need to transfer from a visual input mode
to an auditory/verbal output mode which could be ex-
pected to recruit additional resources not necessary
for a purely auditory/verbal task. Whatever the rea-
son, this finding underscores the difficulty inherent
in creating cross-modal tasks that are equal in diffi-
culty while, at the same time, maintaining similarity
of task demands to the greatest extent possible. As
such, differences in difficulty and presence or absence
of ceiling effects must be taken into account in any
attempt to use cross-modal comparisons for clinical
purposes.
CONCLUSIONS
I t has been suggested by some that examination of anal-
ogous tasks in other sensory modalities is needed to
validate the diagnostic construct of central auditory pro-
cessing disorders and to differentiate such disorders from
pan-sensory or global disorders that may likewise influ-
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
Visual Analogs of (C)APD Tests/Bellis and Ross
ence performance on tests of central auditory function
(McFarland and Cacace, 1995; Cacace and McFarland,
2005). Inherent in this suggestion is the concept of
modality specificity, with the primary argument being
that, for (C)APD to be a valid diagnostic construct, it
must be shown that deficits are either confined to or
are more pronounced in the auditory modality as com-
pared to the visual (or other sensory) modality.
Results of the current study do indicate that although
such testing is at least feasible using readily available
computer hardware and software, constructing truly
analogous tasks across auditory and visual modalities
is difficult, particularly regarding differences in diffi-
culty levels and uncertainty concerning the degree to
which the underlying mechanisms might truly be anal-
ogous. This, combined with findings in the literature
that the expectation of modality-specificity of (C)APD
is inconsistent with the current evidence regarding audi-
tory and cognitive processing (e.g., ASHA, 2005; Musiek
et al, 2005; King and Nelken, 2009), may limit the applic-
ability of these paradigms in the clinical setting. Further,
even if one accepts the theoretical argument that truly
analogous tasks can be constructed and validated and
that the addition of these tasks will improve the diagnos-
tic specificity of (C)APD, the differential diagnostic utility
of cross-modal testing in disordered populations cannot
be inferred simply from performance of normal popula-
tions on these tasks. Therefore, cross-modal testing of
participants with auditory and pan-sensory disorders
is needed to determine whether the inclusion of such ana-
logs aids in the differential diagnosis of (C)APD.
Acknowledgments. This work was supported by a subpro-
ject of National Institutes of Research Resources Grant P20
RR-015567 to T.J.B. that is designated as a Center of Biomed-
ical Research Excellence (COBRE).
REFERENCES
Aiello S, Sotgiu S, Sau GF, Manca S, Conti M, Rosati G. (1994)
Long latency evoked potentials in a case of corpus callosum agen-
esis. Ital J Neurol Sci 15:497–505.
American Academy of Audiology (AAA). (2010) Guidelines for the
Diagnosis, Treatment and Management of Children and Adults
with Central Auditory Processing Disorder. http://www.audiology.
org/resources/documentlibrary/Documents/CAPD%20Guidelines%
208-2010.pdf.
American National Standards Institute (ANSI). (2004) Specifica-
tions for Audiometers (ANSI S3.6-2004). New York: American
National Standards Institute.
American Speech-Language-Hearing Association (ASHA). (2005)
(Central) Auditory Processing Disorders. http://www.asha.org/
members/deskref-journals/deskref/default.
Bellis TJ. (2003) Assessment and Management of Central Auditory
Processing Disorders in the Educational Setting: From Science to
Practice. 2nd ed. Clifton Park, NY: Delmar Thompson Learning.
499
Journal of the American Academy of Audiology/Volume 22, Number 8, 2011
Bellis TJ, Billiet C, Ross J. (2008) Hemispheric lateralization of
bilaterally presented homologous visual and auditory stimuli in
normal adults, normal children, and children with central audi-
tory dysfunction. Brain Cogn 66:280–289.
Blumstein S, Cooper W. (1974) Hemisphere processing of intona-
tion contours. Cortex 10:146–158.
Cacace AT, McFarland DJ. (2005) The importance of modality spe-
cificity in diagnosing central auditory processing disorder. Am J
Audiol 14:112–123.
Chermak GD, Lee K. (2005) Comparison of children’s performance
on four tests of temporal resolution. J Am Acad Audiol 16:
554–563.
Hugdahl K, Bronnick K, Kyllingsbaek S, Law I, Gade A,
Paulson OB. (1999) Brain activation during dichotic presentations
of consonant-vowel and musical instrument stimuli: a 150-PET
study. Neuropsychologia 37:431–440.
Joseph R. (1988) The right cerebral hemisphere: emotion, music,
visual-spatial skills, body-image, dreams, and awareness. J Clin
Psychol 44:630–673.
Keith RW. (2000) Random Gap Detection Test. St. Louis, MO:
Auditec.
Kimura D. (1961) Cerebral dominance and the perception of verbal
stimuli. Can J Psychol 15:166–171.
King AJ, Nelken I. (2009) Unraveling the principles of auditory
cortical processing: can we learn from the visual system? Nat Neu-
rosci 12:698–701.
Lackner JR, Teuber HL. (1973) Alterations in auditory fusion thresh-
olds after cerebral injury in man. Neuropsychologia 11:409–415.
McFarland DJ, Cacace AT. (1995) Modality specificity as a crite-
rion for diagnosing central auditory processing disorders. Am J
Audiol 4:32–44.
Milner B, Taylor S, Sperry R. (1968) Lateralized suppression of
dichotically presented digits after commissural section in man.
Science 161:184–185.
Musiek FE. (1983) Assessment of central auditory dysfunction: the
Dichotic Digits Test revisited. Ear Hear 4:79–83.
Musiek FE. (1994) Frequency (pitch) and duration pattern tests.
J Am Acad Audiol 5:265–268.
Musiek FE, Baran JA, Pinheiro ML. (1990) Duration pattern rec-
ognition in normal subjects and patients with cerebral and coch-
lear lesions. Audiology 29:304–313.
Musiek FE, Bellis TJ, Chermak GD. (2005) Nonmodularity of the
central auditory nervous system (CANS): implications for (central)
auditory processing disorder. Am J Audiol 14:128–138.
500
IP : 191.180.158.77 On: Fri, 14 Feb 2020 16:30:43
Delivered by Ingenta
Musiek FE, Gollegly K, Baran JA. (1984) Myelination of the corpus
callosum in learning disabled children: theoretical and clinical cor-
relates. Semin Hear 5:231–242.
Musiek FE, Kibbe K, Baran J. (1984) Neuroaudiological results
from split-brain patients. Semin Hear 5:219–229.
Musiek FE, Pinheiro M. (1987) Frequency patterns in cochlear,
brainstem, and cerebral lesions. Audiology 26:79–88.
Musiek FE, Pinheiro ML, Wilson DH. (1980) Auditory pattern
perception in “split-brain” patients. Arch Otolaryngol 106:610–612.
Musiek FE, Reeves A. (1986) Effects of partial and complete corpus
callosotomy on central auditory function. In: Lepore F, Ptito M,
Jasper HH, eds. Two Hemispheres—One Brain: Functions of the
Corpus Callosum. New York: Alan R. Liss, 432–433.
Oldfield RC. (1971) The assessment and analysis of handedness:
the Edinburgh Inventory. Neuropsychologia 9:97–113.
Pinheiro ML, Musiek FE. (1985) Sequencing and temporal order-
ing in the auditory system. In: Pinheiro ML, Musiek FE, eds.
Assessment of Central Auditory Dysfunction: Foundations and
Clinical Correlates. Baltimore: Williams and Wilkins, 219–238.
Pinheiro ML, Ptacek PH. (1971) Reversals in the perception of
noise and tone patterns. J Acoust Soc Am 49:1778–1782.
Raczkowski D, Kalat JW, Nebes R. (1974) Reliability and validity
of some handedness questionnaire items. Neuropsychologia 12:
43–47.
Repp BH. (1977) Measuring laterality effects in dichotic listening.
J Acoust Soc Am 62:720–737.
Silva O, Morrone MC, Banks M, Burr D. (2007) Duration discrim-
ination thresholds for brief bimodal events follow a dipper func-
tion. Perception 36 ECVP Abstract Supplement.
Sparks R, Geschwind N. (1968) Dichotic listening in man after sec-
tion of neocortical commissures. Cortex 4:3–16.
Voyer D, Boudreau VG. (2003) Cross-modal correlation of auditory
and visual language laterality tasks: a serendipitous finding.
Brain Cogn 53:393–397.
Westerhausen R, Kreuder F, Dos Santos Sequeira S, et al. (2006)
The association of macro- and microstructure of the corpus cal-
losum and language lateralization. Brain Lang 97:80–90.
Wexler BE, King GP. (1990) Within-modal and cross-modal con-
sistency in the direction and magnitude of perceptual asymmetry.
Neuropsychologia 28:71–80.
Wioland N, Rudolf G, Metz-Lutz MN, Mutschler V, Marescaux C.
(1999) Cerebral correlates of hemispheric lateralization during a
pitch discrimination task: an ERP study in dichotic situation. Clin
Neurophysiol 110:516–523.