How the Brain Integrates Affective
and Propositional Language Into a
Unified Behavioral Function

Hypothesis Based on Clinicoanatomic Evidence
Elliott D. Ross, MD; John H. Harney, MD; Christine deLacoste-Utamsing, MPhil; Phillip D. Purdy, MD

\s=b\ Recent publications suggest that the
right hemisphere dominates in modulating the affective components of language.
Disorders of language from right-sided

focal brain lesions have been called
"aprosodias" and can be classified in a
manner similar to the aphasias. We
describe a patient with motor aprosodia
who subsequently died and underwent
neuropathologic examination. From the
neuropathologic findings and recent observations concerning the neurology of
depression, we hypothesize that the
motor integration of propositional and
affective language takes place in the
brainstem, whereas their higher-order
integration takes place via the callosal
connections between Wernicke's area on
the left and its homologue on the right.
Direct application of these functional and
anatomic relations can help clinicians to
properly interpret the often incongruous
and disparate behavioral and language
responses encountered in brain-damaged

patients.
(Arch Neurol 1981;38:745-748)
Accepted for publication April 6, 1981.
From the Departments of Neurology (Drs
Ross, Harney, and Purdy, and Ms deLacoste\x=req-\
Utamsing) and Psychiatry (Dr Ross), University
of Texas Health Science Center at Dallas; and the
Department of Anthropology (Ms deLacoste\x=req-\
Utamsing), Columbia University, New York.
Reprint requests to Department of Neurology,
University of Texas Health Science Center at
Dallas, 5323 Harry Hines Blvd, Dallas, TX 75235
(Dr Ross).

IX/Totor aprosodia is a disorder of
affective language consequent to
right frontoparietal opercular dam¬
age.' It is characterized by monotone

speech, poor ability to repeat sen¬
tences with prosodie-affective varia¬
tion, and excellent comprehension of
the affective components of speech
and gesturing; propositional lan¬

guage, a dominant function of the left
hemisphere,- is intact. The syndrome

is one of six recently described disor¬
ders of language called the "aproso¬
dias,"1 secondary to focal lesions of
the right hemisphere, which provides
further evidence to support the hypo¬
thesis that the right hemisphere is
dominant for modulating the affec¬
tive components of language.ln This
dominant language contribution by
the right hemisphere seems to be ana¬
tomically and functionally organized
along the lines of propositional lan¬
guage in the left hemisphere.1 De¬
pending on the specific combination
of affective language disturbances,
the disorders of affective language
appear to be classifiable in a manner
similar to the aphasias. Six varieties
of aprosodia, motor, sensory, global,
transcortical motor, transcortical sen¬
sory, and mixed, have been described
and correlated to lesions observed on

computed tomography (CT).'

In this report, we describe a patient
with motor aprosodia who subse-

quently died and underwent neuropa¬
thologic examination. Although his
lesion proved to be in an unusual
location when compared with those of
previously described patients with
motor aprosodia,' the functional-ana¬
tomic correlations in this patient, in
conjunction with recent observations
concerning the neurology of depres¬
sion," provide insights into how the
brain anatomically integrates the propositional and affective components
of language into a unified behavioral
function.

REPORT OF A CASE
Clinical Examination

A 47-year-old, right-handed man was
transferred to the neurology service at the
Dallas Veterans Administration Hospital
on Aug 30, 1980, after suffering a sudden
left hemiplegia with dysarthria ten days
prior to transfer. His medical history was
significant only for non-insulin-dependent
diabetes mellitus and hypertension, con¬
trolled by appropriate medication. There
was no history consistent with a previous
stroke, transient ischemie attack, or focal
neurologic deficit.
Examination on admission disclosed a
moderately obese man who was alert,
attentive, and fully cooperative. Results of
his physical examination were normal.
Immediate recall, recent and remote mem¬
ory, similarities, proverb interpretation,
and judgment were intact. He had a flat¬
tened affect but denied depression. There

Downloaded From: http://archneur.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/20/2015

Fig 1 .—Computed tomographic scan sections of patient. Arrows point to infarctions.
was

implicit denial of his left hemiplegia in

that he

was

aware

of his deficit but

thought unrealistically that he would have
full recovery in a few days. He was mildly
dysarthric but not aphasie. Propositional
speech, comprehension, repetition, object
naming, writing, and reading were com¬
pletely normal. He had no difficulty with
finger recognition, right-left orientation,
or calculations. Constructions were good
with no evidence of neglect.
Cranial nerve examination demon¬
strated full visual fields and normal fundi.
Bilateral simultaneous visual or auditory
stimuli were perceived normally. There
was moderate left-sided facial weakness
but no other deficits in cranial nerve func¬
tions. Specifically, there was no evidence
for pseudobulbar palsy or dysphagia on
examination; he could protrude and wiggle
his tongue and elevate his palate on
request without difficulty. He had a dense
left hemiplegia with hyperreflexia and a
left-sided Babinski's sign. Results of somatosensory examination were entirely nor¬

mal, including graphesthesia, two-point
discrimination, stereognosis, and bilateral
simultaneous stimulation.

Evaluation of Affective

Components of Language
Spontaneous Prosody and Gesturing.—His
voice was flat and monotone, lacking the
alterations in intonation, pitch, and
rhythm that normally color speech with

emotions. There was no discernible differ¬
ence in his voice when discussing his sad¬
ness over his illness, his prolonged absence
from his family, or his excitement and
anticipation about his upcoming weekend
pass. There was a marked paucity of spon¬
taneous gesturing and facial animation,
which caused the patient to have a flat
affect.

Repetition.—This was
markedly impaired. For instance, when
Prosodie-Affective

asked to mimic a statement void of emo¬
tional words in an angry, tearful, indiffer¬
ent, or happy voice, the patient was totally
unable to repeat the sentence with affec¬
tive intonation. The only prosodie variation
he displayed was minimal lowering or rais¬
ing of the loudness of his voice at the end of
a sentence.

Prosodie-Affective

Comprehension.—His

Fig 2—Drawings of coronal brain slices involved by infarctions. Black areas indicate
complete necrosis; hatched areas, demyelination without necrosis. Rostral-caudal
sequence is indicated by numbers under each drawing.

prosodie comprehension, however, was
completely normal. He was easily able to
identify various affective tones, including
surprise, anger, disinterest, sadness, or ela¬
tion, when incorporated into declarative
statements that were void of emotional

words.

Comprehension of Emotional Gesturing.—

This was also intact. The patient could
identify all the aforementioned emotional
tones by name when they were expressed
as a pantomime involving face and limbs.
Results of routine laboratory studies,
including a lumbar puncture and EEG,
were normal. Computed tomography done
on Sept 10, 1980, showed an area of low
absorption in the deep portion of the coro¬
na radiata and rostral internal capsule on
the right and a small round lucency in the

left mid-internal capsule that were consis¬
tent with vascular infarctions (Fig 1).

During hospitalization, repeated testing

of affective language revealed a persistent
motor aprosodia with no notable change
from the initial examination. On Sept 23,
1980, the patient sustained a massive
pulmonary embolism that precipitated a
cardiopulmonary arrest; he could not be
resucita ted.
METHODS

The brain was removed at autopsy and

placed in a 4% formaldehyde solution. Four
weeks later the leptomeninges were
removed and the brain embedded in

a

gelatin mixture under negative followed
by positive pressure (deLacoste-Utamsing

Downloaded From: http://archneur.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/20/2015

"silent" since there were no rightsided or pseudobulbar findings by
examination or symptoms consistent
with a previous stroke by history.
Therefore, for discussion purposes the
left-sided lesion will be ignored
although it is conceivable that it could
have contributed to the patient's syn¬
drome even though the patient's fami¬
ly agreed that his change in voice ("no
life in it") was a direct consequence of
the left hemiplegia.
COMMENT

Fig 3.—Horizontal reconstruction display¬
ing the extent of involvement in right and
left internal capsules by infarctions.

et al, unpublished technique). After the
gelatin was hardened in a 4% formalde¬
hyde solution for two days, the brain was
coronally sectioned at 5-mm intervals on a
rotary sheer. The gelatin surrounding each
slice was gently removed. The slices were
then stained with a trichrome staining
procedure (deLacoste-Utamsing et al, un¬
published technique). Appropriate sections
were also taken for routine histopathologic
examination, using hemotoxylin-eosin and
Loyez (myelin) stains, which confirmed

that the lesions were ischemie infarctions
and that their distributions were accurate¬
ly reflected by the trichrome staining pro¬
cedure. The stained brain slices were then
photographed and the slides back projected
onto a glass table. Each brain slice involved
by the infarctions was outlined on polyes¬
ter film tracing paper (Fig 2) so that the
extent of the lesions could be accurately
reconstructed (Fig 3).
RESULTS

The right-sided infarction involved
the posterior two thirds of the anteri¬
or limb, the entire genu, and the ante¬
rior third of the posterior limb of the
rostral internal capsule (Fig 3). The
left-sided lesion was much smaller,
partially cystic, and at least two
months older than the right-sided
lesion by pathologic examination. It
involved only the midcentral portion
of the rostral genu of the internal
capsule (Fig 3). Both lesions were con¬
fined to the white matter and located
within the internal capsule below the
corticocortical transcallosal connec¬
tions. In terms of clinicopathologic
correlations, the right-sided lesion
effectively destroyed almost the en¬
tire motor output of the right hemi¬
sphere with sparing of the sensory
radiations,7 which correlated well with
the patient's severe and persistent
pure motor (left) hemiplegia. The leftsided lesion was small and clinically

If we accept the hypothesis that the
left hemisphere is dominant for modu¬
lating the propositional components
of language and the right hemisphere
is dominant for modulating the affec¬
tive components of language,1 ·* then
an important issue to resolve is how
the brain anatomically integrates
these two parallel and distinct lan¬
guage functions.
Our case would indicate that the
actual motor integration of the lan¬
guage outputs from each hemisphere
takes place in a subcortical site, below
the level of the deep telencephalic
nuclei. Were the motor integration of
affective and propositional language
transcallosal, our patient should not
have manifested a motor aprosodia as
his right-sided lesion was entirely con¬
fined to the internal capsule and did
not involve the transcallosal connec¬
tions. In keeping with current knowl¬
edge concerning the organization and
termination of the pyramidal and
nonpyramidal motor systems,8" the
patient's right-sided lesion essentially
prevented the entire upper motor neu¬
ron output of the right hemisphere
from innervating the right and left
medial and lateral reticular formation
and the right and left motor nuclear
groups in the brainstem. Thus, the
motor integration of the propositional
and affective components of language
appears to take place in the brainstem, in the aforementioned areas.
There is evidence, however, that a
"higher" order integration takes place
between the hemispheres, via the cor¬
pus callosum, so that the propositional
and affective components are correct¬
ly matched prior to their motor inte¬
gration in the brainstem. Gazzaniga
and Sperry'" published a study con¬
cerning language capacities in pa¬
tients undergoing transection of the
cerebral commissures. No significant
changes in personality or language
behavior were observed. However,
they did report that one patient would
occasionally display mismatching of
emotional behavior to the proposition¬
al content of her conversations during

the first two postoperative weeks,
thus suggesting that commissural
mechanisms do, in fact, exist for uni¬
fying propositional and affective lan¬
guage at cortical levels. For instance,
when conversing in a normal manner,
the patient would suddenly become
sad and be on the verge of tears for
"no apparent reason." This dissocia¬
tive type of behavior, where the pros¬
odie-affective components of lan¬
guage are mismatched to proposition¬
al content, has also been observed in
patients with right-sided brain dam¬
age who have sensory and transcortical sensory forms of aprosodia.1
In a publication dealing with the
neurology of depression, Ross and
Rush6 provide evidence that the syn¬
thesis of an appropriate verbal-cogni¬
tive set—which depends on the propo¬
sitional language areas in the left
hemisphere to accurately reflect a
patient's inner affective state (mood),
which in turn is apparently generated
by the right hemisphere—requires the

posterior-superior temporoposteriorinferior parietal lobes on the right,

Wernicke's area on the left, and the
callosal connections between both ar¬
eas to be intact. (The previously men¬
tioned area in the right hemisphere
seems to have homologous functions
to Wernicke's area as lesions involv¬
ing this region produce comprehen¬
sion deficits for the affective compo¬
nents of language.') Thus, whereas
proper matching of the affective and
propositional components of language
appears to be the result of transcallo¬
sal interactions between the "senso¬
ry" language areas in each hemi¬
sphere, the actual motor integration
of these components of language
appears to be accomplished subcortically, in the brainstem.
Knowledge of the different levels
and proposed anatomic pathways for
the integration of affective and propositional language is useful to both
neurologists and psychiatrists so that
when they deal with brain-damaged

patients they can properly interpret
the patient's affective and proposi¬
tional language responses since brain

lesions may dissociate these functions,
causing disparate, confusing, and of¬
ten incongruous behaviors.1 A direct
clinical application of the issues dis¬
cussed here can be found in an article
by Ross and Rush" in which the vexing
problems of diagnosing depression in
brain-damaged patients are ad¬
dressed. For instance, a patient with
motor aprosodia may be improperly
diagnosed as being depressed because
of the flat character of his speech and
behavior; conversely, if he becomes

Downloaded From: http://archneur.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/20/2015

depressed, the clinician may overlook
his propositional language statements
concerning dysphoria and suicide
since they will be delivered without
affective intonation. Some patients
with large right-hemisphere lesions
may verbally deny all signs and symp¬
toms of depression, including vegeta¬
tive symptoms, and yet be severely
depressed as evidenced by the pre¬
sence of vegetative signs, irritability,
psychomotor retardation, and a posi¬
tive dexamethasone suppression test,
all of which respond to treatment with
tricyclic antidepressants. The recent
book by Gazzaniga and LeDoux"

should also be consulted for further
elucidation of these particular issues
in split-brain patients.
A final note concerns the anatomic
localization of motor aprosodia. As

indicated in the introduction, pre¬
viously described patients with motor
aprosodia have had right frontoparietal opercular damage,' whereas our
patient's infarction was confined to
the internal capsule, destroying al¬
most the entire motor output of the
right hemisphere. Although the in¬
farction was subcortical, the cell bod¬
ies of the destroyed motor axons
resided in cortex; thus, functionally,
the lesion can be considered as having
"cortical" component. Whether
a
there are differences between motor
aprosodias resulting from frontoparietal opercular damage, which also
entails disruption of callosal and
intrahemispheric connections, and
motor aprosodias resulting from deep
subcortical lesions that spare callosal
and intrahemispheric connections

cannot be answered at this time.

Recent evidence in patients who have
aphasia with preserved comprehen¬
sion from left capsular-putaminal
damage, however, has shown that
their speech output is grammatically
correct although produced with a
severe and painfully slow dysarthria.1This is in contrast to patients with
Broca's aphasia secondary to left frontoparietal opercular damage, whose
speech output is nonfluent and agrammatical. Whether similar differences
will hold true for motor aprosodias
from opercular vs diencephalic lesions
remains to be determined.
Loralu Raburn, MD, helped to obtain autopsy
permission; Howard Feit, MD, PhD, and Ste¬
phanie Ross, MS, provided valuable criticisms of
the manuscript; and Denise Wernet provided
secretarial assistance.

References
1. Ross ED: The aprosodias: Functional-ana\x=req-\
tomic organization of the affective components
of language in the right hemisphere. Arch Neu-

rol, 1981;38:561-569.

2. Benson DF: Aphasia, Alexia, and Agraphia.
New York, Churchill Livingstone, 1979.
3. Heilman KM, Scholes R, Watson RT: Auditory affective agnosia: Disturbed comprehension
of affective speech. J Neurol Neurosurg Psychia-

try 1975;38:69-72.
4. Tucker DM, Watson RT, Heilman KM: Discrimination and evocation of affectively intoned
speech in patients with right parietal disease.
Neurology
1977;27:947-950.
5. Ross ED, Mesulam M-M: Dominant Ian-

guage functions of the right hemisphere? Proso-

dy

and

emotional

gesturing. Arch

Neurol

1979;36:144-148.
6. Ross ED, Rush AJ: Diagnostic issues and
neuroanatomical correlates of

depression in

brain-damaged patients: Implications for a neurology of depression. Arch Gen Psychiatry, to be
published.
7. Carpenter MD: Human Neuroanatomy.

New York, Williams & Wilkins Co, 1976, pp
462-465.

8. Kuypers HGJM: Corticobulbar connections

to the pons and lower brainstem in man. Brain

1958;81:364-388.
9. Kuypers HGJM: The anatomical organiza-

tion of the descending pathways and their contributions to motor control especially in primates, in
Desmedt JE (ed): New Developments in EMG and
Clinical Neurophysiology. New York, S Karger
AG, 1973, vol 3, pp 38-68.
10. Gazzaniga MS, Sperry RW: Language
after section of the cerebral commissures. Brain

1967;90:131-148.

11. Gazzaniga MS, LeDoux JE: The Integrated
Mind. New York, Plenum Press, 1978.
12. Naeser MA, Alexander MP, Helm-Esta\x=req-\
brooks N, et al: Aphasia with predominantly
subcortical lesion sites: Description of three capsular/ putaminal aphasia syndromes. Arch Neurol, to be published.

Increase in Price for Color Reproduction and Printing.—After 18 years of asking the
author to contribute $275 per page toward the cost of color reproduction and printing, the
AMA finds it necessary to increase this sum to $400. This covers the author's share of the
cost for up to six square-finished illustrations that can be arranged on a one-page layout.
Any additional illustrations or special effects will be billed to the author at cost. This
increase in the author's share for color reproduction and printing will be effective on
manuscripts accepted after Dec 1,1981. Previous commitments based on the old rate will
still be honored.

Downloaded From: http://archneur.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 06/20/2015