Q 1990 Gordon and Breach Science Publishers S.A.
Printed in the United Kingdom

Intern. J . Neuroscience, 1990, Vol. 54, pp. 139-146
Reprints available directly from the publisher
Photocopying permitted by license only

Case Study
DICHOTIC LISTENING IN AN APHASIC MALE PATIENT AFTER A
SUBCORTICAL HEMORRHAGE IN THE LEFT FRONTO-PARIETAL
REGION

’

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

‘KENNETH HUGDAHL ’KNUT WESTER and ARVE ASBJORNSEN
’Department of Somatic Psychology, University of Bergen, Haukeland Hospital
’Department of Neurosurgery, University of Bergen
(Received Junuury 18, 1990)

Dichotic listening (DL) performance to consonant-vowel (CV) syllables is reported in a n aphasic righthanded male patient. The patient suffered initially from a complete expressive aphasia after an intracerebral
hemorrhage caused by the rupture of an arteriovenous malformation (AVM) underlying Broca’s area. The
patient was tested three times with DL; one week after the hemorrhage, five weeks after, and after at about
6 months. At the first test he could not speak, but understood well. At the second test he uttered one-syllable
words, and could answer “yes” and “no” to questions. At the third test he could speak whole sentences,
although slow and “stutter-like.” Dichotic listening performance showed an almost perfect match with
speech recovery. At the first test he showed a left ear advantage which changed to no ear advantage at the
second test, and to a right ear advantage a t the third test. The overall performance was markedly reduced
at the first test, but improved at the following tests. I t is concluded that dichotic listening may be a valid
complement to traditional language assessment procedures after unilateral brain lesions.
Keywordr: dichotic listening: uphusiu; subcortical hemorrhuge

An often used technique for noninvasive assessment of language laterality is dichotic
listening (DL) (see Hugdahl, 1988 for a review). DL was introduced into neuropsychology in the early sixties by Kimura (e.g., 1961) who reported that subjects with left
hemisphere language were more accurate in reporting items presented to the right ear.
This right-ear advantage (REA) is a surprisingly robust empirical phenomenon in
most individuals (see Bryden, 1988; Geffen & Quinn, 1984). Conversely, individuals
with right hemisphere dominance show a left ear advantage (LEA) on verbal tasks
(e.g., Strauss, Gaddes and Wada, 1987).
The most favored explanation for the REA phenomenon in a free-recall DL task
is the structural model suggested by Kimura (1967). According to Kimura (1967) (see
also Bradshaw & Nettleton, 1988) the REA effect is the result of the interaction
between four factors: 1) that one hemisphere (usually the left) is dominant, or
specialized for speech and language; 2 ) that the contralateral auditory pathways are
more preponderant than the ipsilateral ones (Rosenzweig, 1951; Maximilian, 1982);
3) that the ampler contralateral projection from the right ear to the left hemisphere
will suppress or inhibit the ipsilateral left ear input to the left hemisphere; 4) that
contralateral information from the left ear to the right hemisphere must be transferAddress all communications to: Kenneth Hugdahl, Department of Somatic Psychology, University of
Bergen, Aarstadveien 2 I , N-5009 Bergen, Norway.
The present research was financially supported by a grant to Kenneth Hugdahl and Knut Wester from
Nansen-fondet, and to Knut Wester and Kenneth Hugdahl from the Norwegian Council for Research in
the Medical Science.
139

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

I40

K. HUGDAHL, K. WESTER AND A. ASBJPJRNSEN

contralateral information from the left ear to the right hemisphere must be transferred
across the corpus callosum in order to be processed.
As a consequence, ear differences in DL performance may reflect either the relative
processing superiority of one hemisphere, or information loss due to interhemispheric
transmission, or to an interaction between the two.
Support for the structural model has come from the classic studies by Sparks and
Geschwind ( !968), and Milner, Taylor and Sperry (1968) who found marked reduction in left ear performance in commissurotomy patients under dichotic presentations.
This was explained by Sparks and Geschwind (1968) as due to the fact that auditory
signals from the left ear, reaching the right auditory cortex, could not be transferred
to the left hemisphere for oral report because of the disruption of the callosal
commissures (see also Eslinger & Damasio, 1988).
If this explanation is correct, then also damage to areas in the left hemisphere
responsible for language functions should produce an “ear extinction” phenomenon
(cf. Damasio & Damasio, 1979; Eslinger & Damasio, 1988). However, in addition to
producing left ear extinction, a lesion in the left hemisphere should also produce right
ear extinction, i.e.. both left and right ear correct recalls should be reduced compared

FIGURE I Serial, horizontal CT scans rrom a right-handed male with an intracerebral hemorrhage
caused by an A V M in the left rrontoparietal region Note that the left hemisphere is shown to the right and
vice versa according to radiological conventions. Top row: At the time of the first DL-test Middle row: At
the time of the second DL-test Borron7 r m , : At the time of the third DL-test

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

DICHOTIC LISTENING A N D APHASIA

141

to the normal healthy individual. This effect should further be more dramatic the
more damaged the frontoparietal areas are.
It could finally be argued that a reduced ear advantage should be most marked in
the acute phase of an infarct or hemorrhage, with an increase in overall performance
as well as in right ear advantage (REA) along with recovery of speech and language
functions.
In the present study we present DL data from a patient who suffered from complete
expressive aphasia acutely after an intracerebral hemorrhage caused by an arteriovenous malformation (AVM) covering the left parietal operculum and overlapping the frontal speech area (Broca).
METHOD

The patient
The patient was a 27 year old right-handed male. One week before the first test
session, he woke up in the middle of the night, unable to speak. He was acutely
admitted to the hospital, where a complete expressive aphasia and a slight right-sided
hemiparesis was found. His consciousness was left undisturbed by the event. Although completely unable to speak, he was able to write simple messages, and seemed
to understand all instructions. The deep reflexes were augmented on the right side, but
the plantar reflexes were normal. The motor deficits improved considerably during the
next 24 hours. A CT scan revealed an intracerebral, subcortical hemorrhage in the left
frontoparietal region, corresponding to the area of Broca (Figure 1).

FIGURE 2 Left carotid angiogram showing a small arteriovenous malformation (AVM) situated medially to the hemorrhage in the left hemisphere. Arrows indicate the extent of the AVM. Oblique projection:
anterior to the left.

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

I42

K . HUGDAHL, K . WESTER AND A. ASBJ0RNSEN

An acute cerebral angiography revealed an AVM in the same region, with an
arterial blood supply from the middle cerebral artery (Figure 2).
The patient’s handedness was determined by using a modified version of the
handedness-questionnaire developed by Raczkowski, Kalat and Nebes (1974). The
patient was asked to indicate by nodding or shaking his head whether he used his
right, left or both hands to a total of 15 questions related to manual items. The patient
was 100% right-handed.
The patient was tested with free recall D L on three different occasions: 1) six days
after the acute event, 2) after another four weeks, and finally, 3) six and a half months
after the bleeding.
His neurological condition was as follows on the three DL-test occasions:
First test Complete expressive aphasia, with understanding well preserved. Able to
follow simple instructions (like “raise your arms”), although with slight difficulties.
Moderate right facial paresis, and slight hemiparesis. He could walk, but had some
impairment of toe movements. His right arm/hand was weak.
Second test Speech improved, was now able to say brief words like “yes” and “no,”
but was completely unable to form sentences. Complete recovery of hemiparesis, but
still a slight facial asymmetry.
Third test No signs of hemiparesis. Remarkable improvement of speech. He could
now form sentences and interact in verbal communication. However, his speech was
slow, hesitating and “stutter-like,” and he had to search for certain words.
Dichotic listening procedure

The dichotic stimulus-materials consisted of the six stop-consonants b, d, g, p , t. k
which were paired with the vowel a to form six basic consonant-vowel (CV) syllables
(ha, ga, p a , etc.). The syllables were paired with each other for all possible combinations, thus yielding 36 dichotic pairs including the homonymic pairs (the homonyms
were excluded from statistical analyses). Each pair was randomly recorded three times
on the tape. Thus, the total number of trials on the tape was 108. The intertrial
interval (ITI) between stimulus presentations varied between 4 and 5 s.
The dichotic tape was prepared on a PDP 11/45 computer with 12 bits A/D and
D/A converters, and with D/A multiplexer. Each CV-syllable had a duration of
320 ms, and temporal alignment between channels was set at the first visually identifiable energy-release in the consonant-segment of each syllable. Maximum onset
difference between the channels was 0.5 ms due to D/A multiplexer resolution and the
sampling frequency ( 1 0 KHz).
Each syllable was originally read by a male voice and fed to the computer. After
computer analyses, the syllables were recorded onto a NAGRA IV tape-recorder. In
order to test the patient more easily in the hospital environment the NAGRA tape was
copied onto a chrome dioxide cassette and played from a SONY WM D D minicassette player. The output of the minicassette player was calibrated and the mean
intensity was 84 dB A SPL measured with a Bruel and Kjaer 2204 sound level meter.
Tvst -procedure

The patient was tested at the University Hospital the first time and at the dichotic
listening laboratory in the Department of Somatic Psychology the second and third

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

DICHOTIC LISTENING A N D APHASIA

143

listening laboratory in the Department of Somatic Psychology the second and third
times. Because the patient could not speak at all during the first test, a modified
“monitoring procedure” (Geffen & Sexton, 1978) was used. The patient was asked to
point with his left index finger to the syllable he heard on each trial from a list
containing all six CV-syllables used in the test. Since the patient was quite slow in
monitoring trial-by-trial, a I0 s interrupt procedure was used; i.e., the experimenter
waited 10 s after each syllable-pair presentation for a manual response. If no response
had occurred during the 10s interrupt interval, the trial was scored as an omission
trial. The monitoring procedure was also used at the two other test occasions, in order
to keep the test situations as identical as possible.

RESULTS
Percentage correct recall from right and left ears, respectively, for the three tests are
seen in Figure 3.
As can be seen in Figure 3 , there are two main trends in the data. First of all, there
is an overall increase in correctly reported items from both ears across time (and
particularly from the right ear). This was statistically significant with the Chi-square
test (Chi’(2) = 2 9 . 0 4 , ~< .OOl). The second trend in the data was a “flip-over’’ from
a left ear advantage (LEA) at Test I to a right ear advantage (REA) at Test 3, with
no ear advantage (NEA) at Test 2 (see Figure 3). This was statistically evaluated by
calculating a laterality-index for each test-occasion and then applying the Chi-square
test on the index scores. The laterality-index was calculated according to the formula:

R, - L,
x 100
R,
L,

+

with R, = Correct right ear reports, L, = Correct left ear reports. Thus, a negative

-

Rlmt Ear

Leftsided A W-Aphasia
Left Ear

t

X

20

10

n
Test 3 16.5 r m l

FIGURE 3 Percentage correct recall in dichotic listening for the three test-occasions. I wk = 1 week
after the hemorrhage, 5 wk = after 5 weeks, 6.5 mo = after 6.5 months.

K. HUGDAHL. K. WESTER A N D A. ASBJ0RNSEN

I44

index indicates a LEA, a positive index a REA, and a zero index a NEA. Index scores
for the three test-occasions were -42.8 (Test I ) ; - 2.54 (Test 2) and 7.18 (Test 3).
A chi-square test showed a highly significant difference between the three indexscores, (Chi’ (2) = 110.41, p -= .OOl).
Evaluating finally the difference between correct right and left ear scores for each
test-occasion showed a significant LEA for Test 1 (chi’ ( I ) = 5.94, p < .05).

+

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

DISCUSSION
To sum up the findings, Figure 3 reveals an almost perfectly linear increase in right
ear correct reports across test occasions (together with an increase in overall correct
performance). This development of dichotic listening performance across time was
nicely corroborated by the development of speech functioning. At the first test the
patient was completely mute, and did not utter a single word. At the second test he
was able to utter single-syllable words like “yes” and “no.” At the third test occasion
(about 6 months after the hemorrhage) he did speak, although with reduced pace and
with some errors.
Thus, there was an interesting parallel in speech-recovery, and the increase in
overall DL-performance and particularly the switch from a LEA at test 1 to a REA
at test 3. Note that the patient was tested with the same procedure (pointing to the
list of CV-syllables) at all three test-occasions.
The present data thus support Kimura’s (1967) structural model for explaining
dichotic listening, with the REA-phenomenon conditional o n the progress of left
hemisphere functioning.
A comparison of the patient’s D L performance with normative data for righthanded healthy males shows a striking difference at the first test with a clear tendency
towards “normalization” across time. Normative data for righthanded adults are
shown in Figure 4.

=

Sample &ta
&ta
REar

healthy a&lts

frcm 3 Ekpwimsntd

EBB Ear

90 I

75

t
E!

1

1 -

6o

+
45i

9x

30
15

0

FUi Males

RH Females

FIGURE 4 Normative data for dichotic listening performance in healthy righthanded adults.
R H = righthanded.

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

DICHOTIC LISTENING A N D APHASIA

I45

As can be seen in Figure 4,percentage correct right ear reports normally varies
between 55-75% with percentage correct left ear scores varying between 40 and 54%.
Thus, the REA lies within a 10-20% difference for the right and left ear reports (see
Hugdahl & Anderson, 1986; Bryden, Munhall, & Allard, 1983 for examples of
normative data). The patient switched from a 15% LEA to a 8% REA accross the
three test occasions.
The LEA at Test 1 is probably best explained as an effect of failure of the left
hemisphere to process the contralateral right ear signal at the acute stage of the
hemorrhage (cf. Eslinger & Damasio, 1988).
Thus, the LEA effect is primarily not caused by an increase in left ear correct
reports, but to a drastic decrease in correct right ear reports. This, however, recovers
to a “normal” REA as a function of speech rehabilitation.
It is interesting to speculate as to the nature of the almost perfect match between
the development in DL performance and speech recovery. It may be argued that DL
basically is a test of verbal recognition, or perception (Bradshaw & Nettleton, 1983)
with no obvious connection to speech production. More than 95% of all righthanders have speech localized to the left hemisphere (Wada & Rasmussen, 1960).
However, DL usually correctly identifies language dominance in only about 85-90%
of individuals tested (Strauss, Gaddes, & Wada, 1987). Thus, there is a discrepancy
of about 10% in reliability between invasive techniques like the sodium-amytal test,
and a noninvasive technique like DL. This has been explained that the invasive
technique basically monitors expressive language function while DL mainly monitors
impressive function.
However, the patient in the present study had an intracerebral hemorrhage mainly
overlapping frontoparietal speech areas, seemingly leaving temporal areas intact.
Since the perisylvian fissure and surrounding temporal areas are considered important
in language perception (Brodal, 1981), one could argue that dichotic listening performance should not have been affected.
As is obvious from Figure 3 this is, however, not the case. On the contrary, DL
performance closely follows speech performance over time. This leaves the conclusion
that dichotic listening is not uniquely tied to language perception and speech reception, but that in addition it also reflects changes in the functional status of frontoparietal areas, and consequently in speech production per se.
The impairment of speech and DL performance was probably caused by a permanent destruction of some neural elements as well as a temporary dysfunction of
other neural elements due to local pressure from the hematoma and the oedema
surrounding it. Consequently, the parallel improvement in both speech and DL
performance was caused by the gradual resolution of the hematoma and normalization of tissue pressure in the affected area.
If the present findings are valid and replicable, then dichotic listening opens up an
objective way of complementing traditional speech-tests in assessing overall language
functioning after unilateral cortical lesions.
REFERENCES
Bradshaw, J. L., & Nettleton, N. C . (1983). Humun ccrebrul usymmefry. Englewood CliKs, N. J.: PrenticeHall.
Brodal, A. (I98 I).Neurological unarumy (3rd Edition). New York: Oxford University Press.
Bryden, M. P. (1988). An overview of the dichotic listening procedure and its relation to cerebral
organization. In K. Hugdahl (Ed.), Hutidhook of dichotic listening: Theory, methods und research.
Chichester, UK.: Wiley.

Int J Neurosci Downloaded from informahealthcare.com by University of Newcastle on 12/31/14
For personal use only.

146

K. HUGDAHL, K. WESTER A N D A. ASBJBRNSEN

Bryden, M. P.. Munhall, K . , & Allard, F. (1983). Attentional biases and the right-ear effect in dichotic
listening. Brain and Language. 18. 236-248.
Damasio, H., & Damasio, A. R. (1979). “Paradoxic” ear extinction in dichotic listening: Possible anatomic
significance. Neurolog!., IY, 644-659.
Eslinger, P. J., & Damasio, H. (1988). Anatomical correlates of paradoxic ear extinction. In K. Hugdahl
(Ed.). Hundhook of ciichoric lisrening: Theory, merliud,s and research. Chichester. U.K.: Wiley.
GeKen, G., & Sexton. M. A. (1978). The development of auditory strategies of attention. Develupmenrcil
P.yj’chology, 14, 11-17.
Geffen, G., & Quinn. K . (1984). Hemispheric specialization and ear advantages in processing speech.
Psyekolagicul Bullrtin, 96, 273-291.
Hugdahl, K., & Anderson, L. (1986). The “forced-attention” paradigm in dichotic listening to CV-syllables: A comparison between adults and children. Corres. 22. 417-434.
Kimura, D. (1961). Cerebral dominance and the perception
of verbal stimuli. Canadian Journal of Psychol.
.
~ ~ g15,
y , 166-171.
Kimura. D. (1967). Functional asvmmetrv of the brain in dichotic listenina. Corres. 3, 163-178
Maximilian. V. A. (1982). Corticai blood kow asymmetry during monaurayverbal stimulation. Brain and
Lunguage, 15, 1-1 I .
Milner. B., Taylor. L., & Sperry. R. W. (1968). Lateralized suppression of dichotically presented digits after
commissural section in man. Science, 161. 184-186.
Raczkowski, D., Kalat. J. W., Nebes, R. D. (1974). Reliability and validity of some handedness questionnaire items. Ncwupsyrholugiu, I2. 43-47.
Rosenzweig, M. R. (1951). Representations of the two ears at the auditory cortex. American Journal uf
P I ~ ~ s ~ o167,
I o ~147-214.
~,
Sparks, R., & Geschwind, N. (1968). Dichotic listening in man after section of neocortical commissures.
Cortez. 4, 3-16.
Strauss, E., Gaddes, W. H.. & Wada, J. (1987). Performance on a free-recall verbal dichotic listening task
and cerebral dominance determined by the carotid amytal test. ~‘eu,op.yycholiigicJ,25, 747-753.
Wada, J.. & Rasmussen, T. (1960). Intracarotid injection of Sodium amytal for the lateralization ofcerebral
speech dominance. Jourriul uf Neurosurgery, 17, 266-282.
~~

~I