J Neurol (1994) 241:470-474
© Springer-Verlag 1994

ORIGINAL

Hideo Tohgi
Katsumi Watanabe
Hiroaki Takahashi
Hisashi Yonezawa
Kentaro Hatano
Toshiaki Sasaki

Prosopagnosia without topographagnosia
and object agnosia associated
with a lesion confined
to the right occipitotemporal region

Received: 30 June 1993
Received in revised form:
15 November 1993
Accepted: 14 December 1993
H. Tohgi ([~) • K. Watanabe
H. Takahashi - H. Yonezawa
Department of Neurology,
Iwate Medical University, 19-1 Uchimaru,
Morioka, 020 Japan
K. Hatano • T. Sasaki
Nishina Memorial Cyclotron Center
(Japan Radioistope Association)
and Cyclotron Research Center,
Iwate Medical University, Morioka, Japan

COMMUNICATION

Abstract A patient is presented
who developed prosopagnosia with
a recent occipitotemporal infarct in
the distribution of the right posterior
cerebral artery. He did not have topographical agnosia or object agnosia.
He regained the ability to recognize
faces of familiar persons, whereas
he remained unable to identify faces
of persons w h o m he met after the
disease onset. This case demonstrates
that prosopagnosia may occur as a
deficit of matching a perceived face

Introduction
The inability to recognize familiar faces is a specific agnosia (prosopagnosia; Bodamer in 1947 [7]; Hdcaen and
Angelergues in 1962 [18]). At least two major questions
remain unsettled concerning the pathogenetic mechanisms of prosopagnosia. First, it is uncertain whether
prosopagnosia is (1) a partial symptom of a general impairment of visuospatial perception, (2) a defect of a presumed specific perceptual process for faces, or (3) a
deficit of matching a correctly perceived face to a memory store of the face. This is largely bcause the previously
reported patients almost always presented with other visuospatial disturbances such as topographical agnosia or object agnosia, in addition to both retrograde and anterograde prosopagnosia. Second, there is a discrepancy between autopsy and clinical studies which have not determined whether bilateral lesions or a right occipitotemporal lesion alone are required for prosopagnosia to occur.
The pathology reports available have suggested that symmetrical lesions in the medial occipital region are generally necessary for prosopagnosia to occur [1, 3, 11, 16, 18,
19, 25, 32, 33, 45], because all the autopsy cases except

to a memory store of the face, and
that the failure to recognize unfamiliar faces may be due to the inability
to form memory stores of new faces.
These deficits can occur in association with a lesion confined to the
right occipitotemporal region.

Key words Prosopagnosia
Visual memory • Right
occipitotemporal lesion - Magnetic
resonance imaging • Positron
emission tomography

three [18, 23, 44] displayed bilateral occipitotemporal lesions [12]. In most of these reports, autopsy studies were
performed long after the onset of prosopagnosia. However, there have also been reports of patients with
prosopagnosia having a right occipitotemporal lesion
alone. These have been on the basis of autopsy findings
shortly after the development of the symptoms [23], or of
clinical and radiological evidence [10, 13, 17, 21, 24, 28,
30, 40, 44].
We report the case of a patient with an acute onset of
prosopagnosia who did not develop topographagnosia or
object agnosia. Three months after onset, his ability to
recognize faces of familiar persons recovered, but he remained unable to recognize those whom he first met after
the onset. Computed tomography, magnetic resonance
imaging (MRI) and positron emission tomography (PET)
displayed anatomical and functional abnormalities confined to the right occipitotemporal region.

Case report
A 62-year-old, right-handed man was admitted for evaluation of
his inability to recognize faces. He was educated for 6 years in a

471

Fig. 1 T2-weighted MRI
(TPUTE 3,000/100 ms, 1.5 T, 5
m m thickness) of (a) axial, (b)
coronal (3 cm anterior to the
occipital pole), and (c) sagittal
(1.5 cm lateral to the midline)
sections, demonstrating a hyperintensity area in the cortical
and subcortical areas in the distribution of the right posterior
cerebral artery. In a and b, the
right side of the head is on the
right

primary school, worked as a local government employee until aged
48 years, and as a chief executive of a real estate business thereafter. He had been well except for hypertension and a history of
acute myocardial infarction. One morning a month before admission, he was watching television when he suddenly noticed that
faces on the TV screen appeared blurred and he could not recognize them. The faces appeared reddish. Subsequently, he could no
longer recognize the faces of family members, relatives, and photographs of faces of famous people. However, he could find his
way to familiar places. As his complaint, which caused him great
concern, persisted, he visited our department.
On admission, he was alert, and well oriented for both time and
space. He had a Mini-Mental-State examination score of 30 (full
score). The Wechsler Adult Intelligence Score (WAIS) was 83 for
the verbal tests, and 70 for the performance tests. The low score
for the verbal tests may be due to his short education (6 years); in
addition, the low performance test score was mainly due to his impaired ability in the picture arrangement tests. However, his score
for block design tests was excellent, suggesting unaltered synthesizing and abstract ability in visuospatial functions. He had a left
upper quadranopsia in the right eye, but did not neglect the left half
of his visual space. Visual acuity was 0.5 (1.5 with glasses) in the
right eye, 0.6 (1.5 with glasses) in the left. His colour perception
was normal: he named and matched colours correctly, and had no
problem with Ishihara's test for colour blindness. However, he
could no longer recognize relatives or other familiar persons by
their faces, but could recognize them immediately by voice. He remembered the hospital staff only by voice, but not by face. He
could identify a face as a face in real life or on photographs, and
describe characteristics of faces. However, he saw faces as partly
distorted (metamorphopsia), stating that the right eye of persons
was twinkling, and that the right palpebra appeared "long and narrow as a fox's eye". This metamorphopsia was restricted to h u m a n
faces, and was not seen for faces of animals. He could identify
faces of animals (cats, dogs, horses, etc.). W h e n he was presented
with a single front-view photograph of an unfamiliar face, he could
identify it in several front-view photographs from other unfamiliar
faces; his responses were correct in three such tests. However, he
always failed to identify it in several oblique-view photographs
from other unknown faces.
He did not suffer from topographical agnosia: he could go to
clinics of other departments for examinations, find the way back to
his room on the ward, and draw a plan of the ward accurately; he
could draw a map of Japan, and place the names of cities correctly.
He did not have object agnosia: he could identify and name more
than 50 objects or pictures of objects (bed, chair, glasses, cars. etc)
presented in both canonical (conventional) and non-canonical (unconventional) views, and state the size and shape of presented pictures. He responded correctly to 45 out of 60 pictures of the Boston
Naming Test (40-55 for the average Japanese). His language ability was normal as regards speaking, auditory comprehension, reading and writing. He neither displayed apraxias nor agnosias for the
auditory and somatosensory modalities.

Fig.2 PET images of cerebral blood flow (CBF), oxygen extraction fraction (OEF), and cerebral metabolic rate of oxygen (CMR02)
at the horizontal planes 2 0 - 4 0 m m above and parallel to the orbitomeatal line (Headtome IV, Shimadzu, Japan, with the full width at
half-maximum of 4.5 mm). The CBF and CMRO: were determined
using C1502 and 1502, respectively. The CBF and C M R O , are decreased in the right occipitotemporal areas (arrowheads). The right
side of the head is on the left

When he was discharged 2 months after admission, his symptoms had partially recovered, and he could recognize faces of familiar persons whom he met before his illness. He could also recognize all photographs of families, relatives and old friends, and
90% of the photographs presented of famous people such as politicians, actors and professional wrestlers. He could discriminate
genders and races by face. He could label emotional expressions on
people and photographs as laughing, crying, or prim and proper.
However, he remained unable to recognize faces of persons whom
he met after the onset of illness. He identified these people only by
their voice, height and hair. He still had difficulty in evaluating age
by face. The cranial nerves, muscular strength, motor coordination,
deep tendon reflexes and sensation were all normal throughout the
course.
T2-weighted MRI (Fig. 1) demonstrated an area of hyperintensity in the cortical and subcortical areas in the right occipitotemporal areas, indicating an infarction in the territory of the posterior
cerebral artery. The high-intensity area included the fusiform
gyms, lingual gyrus, tapetum, inferior longitudinal fascicle and a
part of the optic radiation. In other areas, only a few lacunes were
demonstrated. PET studies, using C15COa and 1502 inhalation
methods, demonstrated a reduction in regional cerebral blood flow
and cerebral metabolic rate of oxygen in the infarct areas indicated

472

on MRI (Fig. 2). There were no other areas of the brain with impaired blood flow or metabolism.

disturbed perception of colours [28, 35]. The frequent association of prosopagnosia in patients with achromatopsias (12 out of 14 patients reviewed by Meadows in 1974
[29]) may suggest a close proximity of their pathogenetic
loci. His low scores for picture arrangement in the WA1S
Discussion
tests suggest an impaired ability to discrimate essential
The main points to be addressed concerning the prosopag- parts from non-essential details. However, this cannot acnosia of our patient are: (1) that it was relatively isolated; count for his prosopagnosia, because he could describe
(2) that recognition of faces known to him before the ill- the characteristics of faces presented to him.
Experimental studies in monkeys indicated that visual
ness recovered, whereas recognition of faces encountered
after the illness remained impaired; and (3) that neu- information is relayed successively from the striate to the
roimaging studies ifidicated morphological and functional prestriate cortex, OB and OA (according to the terminolchanges in the occipitotemporal areas on the right side ogy of Bonin and Bailey), and the posterior temporal areas, TEO and TE, and finally to the medial temporal lobe
only.
The difficulty of our patient in recognizing faces could memory system, hippocampus, and the adjacent related
not be attributable to a less specific neuropsychiatric dis- cortex [31, 39]. This system has widespread reciprocal
turbance, because he had neither topographical agnosia connections with the association cortex, where memories
nor object agnosia. Most patients with object agnosia are are recognized, consolidated and stored; as time passes,
without prosopagnosia, and in some patients like ours, the memories stored in association cortex become indeprosopagnosia is not associated with object agnosia [6, pendent of the medial temporal lobe [31, 39]. The visual
12, 20], suggesting that the process of face recognition is field is re-represented in the striate and prestriate cortex
distinct from that of object recognition. In addition to the and the posterior temporal areas, which in turn perform
preserved object recognition, our patient was able to iden- submodality processing. For face recognition, the facetify familiar faces after his illness had partially recovered, sensitive cells have been demonstrated in the monkey
and to interpret facial expressions from the beginning of [46]. Provided that facial information in the human is
his illness. The dissociations between deficits of familiar processed in a similar way, the isolated face recognition
and unfamiliar face recognition [4, 9, 26], and between difficulty in our patient may be attributable to selective
impairments in face identification and expression analysis damage to one of the modalities in the prestriate area. The
[9, 37] have been described in previous reports. Unlike retrograde face memories may be executed independently
our patient, some patients with prosopagnosia were able from the medial temporal lobe memory system, if the strito recognize unfamiliar faces [2, 5, 35, 42]. Such dissoci- ate and prestriate cortex and their connections are intact;
ations may be well explained by the functional model put the anterograde face memories may be selectively imforward by Bruce and Young [8], which assumes that paired if connections from the prestriate cortex to the merecognition of familiar and unfamiliar faces uses different dial temporal lobe are damaged. Thus, the dissociation beroutes, and that face identification and expression analysis tween anterograde and retrograde face memories in our
are processed independently. The inability of our patient patient can be explained by the subcortical white matter
to recognize familiar faces in the early phase of the illness lesion in the occipitotemporal areas, which may be lomay have occurred as a deficit of matching a perceived cated more anteriorly than in patients with both anteroface to a memory store of the face that had been con- grade and retrograde prosopagnosia [41 ].
While most of the previous autopsy cases with
structed; the residual deficit for unfamiliar faces appeared
to be due to his inability to form memory stores of new prosopagnosia had bilateral cerebral lesions, some cases
faces. His failure to match unfamiliar faces photographed did not have left occipitotemporal lesions [18, 23, 33],
at different angles suggests that his inability to integrate suggesting that bilateral involvement of the visual system
abstract and structural visual codes from view-centred may not be a prerequisite for prosopagnosia [23]. This
(pictorial) information of unfamiliar faces prevented fur- contention was further supported by many other reports
based upon clinical and radiological findings [10, 13, 17,
ther processing in the visual cognitive system.
Although his metamorphopsia for the eye region may 21, 24, 28, 30, 40, 44]. In our patient, we could clearly
have played some role in his prosopagnosia in the early rule out lesions and functional impairments in the left ocperiod [5], it does not seem to have contributed to the cipitotemporal areas, and in areas other than the right ocprosopagnosia in the later period. This is because after cipitotemporal areas. Although the mechanisms for propartial recovery he could identify familiar faces, when he sopagnosia with a right occipitotemporal lesion remain to
still complained of distortion of eye regions, although it be clarified, previous researchers suggested a frequent cowas less remarkable than in the earlier period. Our patient occurrence of prosopagnosia and other agnosias for space
also complained that the faces appeared red-tinted, and which are associated with right posterior hemisphere lethat the right eye appeared excessively bright. Similar sions [22, 28]. The left-hemisphere template could be
dyschromatopsias have been described in patients with based on separate elements of the face, and have a

473

favoured position for verbal evocation, whereas the righthemisphere template might be more global [12]. It has
b e e n suggested that while association or semantic categorization were impaired with left hemisphere damage, the
visual identification and discrimination of c o m p l e x patterns [14] and perceptual categorization [43] were deficient with right hemisphere lesions. Studies on n o r m a l
subjects [15, 27] and split-brain patients [38] have
demonstrated different strategies of facial recognition between the left and right brain. Thus, there m a y be various
pathogenetic m e c h a n i s m s of prosopagnosia, d e p e n d i n g
u p o n the location and size of the occipitotemporal lesions.
In addition, there m a y be i n d i v i d u a l variation in the dep e n d e n c y of facial recognition on the right and left occip-

itotemporal areas complicated by different cognitive
strategies. It is, therefore, possible that a small proportion
of individuals whose left occipitotemporal area cannot assume the function of the corresponding areas on the right
side m a y suffer from prosopagnosia. The findings in our
patient indicate that relatively pure prosopagnosia and anterograde prosopagnosia with preserved retrograde facial
recognition can occur with a lesion confined to the right
occipitotemporal area.
Acknowledgements We thank Prof. Toru Yanagisawa and members of the Nishina Memorial Cyclotron Center for their cooperation, Dr. Paul Langman for reviewing the manuscript, and Miss
Kikuko Ymazaki for secretarial assistance.

References
1. Arseni C, Botez MI (1958) Consideraciones sobre un caso de agnosia de
las fisonomias. Rev Neuropsiquiatr
Lima 21 : 583-593
2. Assal G (1969) R6gression des troubles
de la reconnaissance des physionomies
et de la mdmoire topographique chez
un malade op6rd d'un h~matome intrac6rdbral pari6to-temporal droite. Rev
Neurol (Paris) 121 : 184-185
3. Benson DF, Segarra J, Albert ML
(1974) Visual agnosia-prosopagnosia:
a clinicopathologic correlation. Arch
Neurol 30:307-310
4. Benton AL (1980) The neuropsychology of facial recognition. Am Psychol
35 : 176-186
5. Benton AL, Van Allen MW (1972)
Prosopagnosia and facial discrimination. J Neurol Sci 15 : 167-172
6. Blanc-Garin J (1984) Perception des
visages et reconnaissance de la physionomie dans l'agnosie des visages.
Annee Psychol 84 : 573-598
7. Bodamer J (1947) Die Prosop-Agnosie
(die Agnosie des Physiognomieerkennens). Arch Psychiatr Nervenkr 179 :
6-53
8. Bruce V, Young A (1986) Understanding face recognition. Br J Psychol 77 :
305-327
9. Bruyer R, Laterre C, Seron X, Feyereisen P, Strypstein E, Pierrard E,
Rectem D (1983) A case of prosopagnosia with some preserved covert remembrance of familiar faces. Brain
Cogn 2 : 257-284
10. Christen L, Landis T, Regard M (1985)
Left hemispheric functional compensation in prosopagnosia? A tachistoscopic study with unilaterally lesioned
patients. Hum Neurobiol 4 : 9-14
11. Cohn R, Neumann MA, Wood DH
(1977) Prosopagnosia: a clinicopathological study. Ann Neurol 1 : 177-182

12. Damasio AR, Damasio H, Ban Hoesen
GW (1982) Prosopagnosia: anatomic
basis and behavioral mechanisms. Neurology 32 : 331-341
13. De Renzi E (1986) Prosopagnosia in
two patients with CT evidence of damage confined to the right hemisphere.
Neuropsychologia 24:385-389
14. De Renzi E, Scotti G, Spinnler H
(1969) Perceptual and associative disorders of visual recognition: relationship to the side of cerebral lesion. Neurology 19 : 634-642
15. Ellis HD, Shepherd JW, Davies GM
(1979) Identification of familiar and
unfamiliar faces from internal and external features: some implications for
theories of face recognition. Perception
8 : 431-439
16. Gloning I, Gloning K, Jellinger K,
Quatember R (1970) A case of "prosopagnosia" with necropsy findings.
Neuropsychologia 8 : 199-204
17. Habib M, Sirigu A (1987) Pure topographical disorientation: a definition
and anatomical basis. Cortex 23 : 73-85
18. H6caen H, Angelergues R (1962) Agnosia for faces (prosopagnosia). Arch
Neurol 7 : 92-100
19. Heidenhain A (1927) Beitrag zur
Kenntnis der Seelenblindheit. Monatsschr Psychiatr Neurol 66:61-116
20. Jeeves M A(1984) The historical roots
and recurring issues of neurobiological
studies of face perception. Hum Neurobiol 3 : 191-196
21. Landis T, Cummings JL, Christen L,
Bogen JE, Imhof HG (1986) Are unilateral right posterior cerebral lesions
sufficient to cause prosopagnosia?
Clinical and radiological finding in six
additional patients. Cortex 22 : 243-252
22. Landis T, Cummings JL, Benson DF,
Palmer EP (1986) Loss of topographic
familiarity: an environmental agnosia.
Arch Neurol 43 : 132-136

23. Landis T, Regard M, Bliestle A, Kleihues P (1988) Prosopagnosia and agnosia for noncanonical views. An autopsied case. Brain 111:1287-1297
24.Lhermitte F, Pillon B (1975) La
prosopagnosie: R61e de l'hdmisph~re
droit dans la perception visuelle (~t propos d'un cas consdcutif ~tune lobectomie occipitale droite). Rev Neurol
(Paris) 131:791-812
25. Lhermitte F, Chain F, Escourolle R,
Ducarne B, Pillon B (1972) I~tude
anatomoctinique d'un cas de prosopagnosie. Rev Neurol (Paris) 126:329346
26. Malone DR, Morris HH, Kay MC,
Levin HS (1982) Prosopagnosia: a
double dissociation between the recognition of familiar and unfamiliar faces.
J Neurol Neurosurg Psychiatry 45 :
820-822
27. Marzi CA, Berlucchi G (1977) Right
field superiority for recognition of famous faces in normals. Neuropsychologia 15:751-756
28. Meadows JC (1974) The anatomical
basis of prosopagnosia. J Neurol Neurosurg Psychiatry 37 : 489-501
29. Meadows JC (1974) Disturbed perception of colonrs associated with localized cerebral lesions. Brain 97 :
615-632
30. Michel F, Poncet M, Signoret JL
(1989) Les ldsions responsables de la
prosopagnosie sont-elles toujours bilatdrales? Rev Neurol (Paris) 146:
764-770
31. Mishkin M (1982) A memory system
in the monkey. Philos Trans R Soc
Lond [Bid] 298 : 83-95
32. Nardelli E, Buonanno F, Coccia G, Fiaschi A, Terzian H, Rizzuto N (1982)
Prosopagnosia. report of four cases.
Eur Neurol 21 : 289-297

474

33. Pevzner S, Bornstein B, Loewenthal M
(1962) Prosopagnosia. J Neurol Neurosurg Psychiatry 25 : 336-338
34. Pillon B, Bakchine S, Lhermitte F
(1987) Alexia without agraphia in the
left-handed patient with a right occipital lesion. Arch Neurol 44:1257-1262
35. Rondot P, Tzavaras A, Garcin R
(1967) Sur un cas de prosopagnosie
persistant depuis quinze ans. Rev
Neurol (Paris) 117 : 424-428
36. Schwartz M, Smith ML (1980) Visual
asymmeties with chimeric faces. Neuropsychologia 18 : 103-106
37. Shuttleworth EC Jr, Syring V, Allen N
(1982) Further observations on the nature of prosopagnosia. Brain Cogn 1 :
307-322

38. Sperry RW (1982) Some effects of disconnecting the cerebral hemispheres.
Science 217 : 1223-1226
39. Squire LR (1987) Memory: neural organization and behavior. In: Mountcastle VB (ed) MD Handbook of physiology, section 1, vol 5, part 1. American
Physiological Society, Bethesda, pp
295-371
40.Torii H, Tamai A (1985) The problem
of prosopagnosia: report of three cases
with occlusion of the right posterior
cerebral artery. J Neurol 232 [Suppl
140]
41. Tranel D, Damasio AR (1985) Knowledge without awareness: an autonomic
index of facial recognition by prosopagnosis. Science 228 : 1453-1454
42. Tzavaras A, Merienne L, Masure MC
(1973) Prosopagnosie, amndsie et troubles de langage par 16sion temporale
gauche chez un sujet gaucher.
Encdphale 62 : 383-394

43. Warrington EK, Taylor AM (1978)
Two categorical stages of object recognition. Perception 7 : 695-705
44. Whiteley AM, Warrington EK (1977)
Prosopagnosia: a clinical, psychological, and anatomical study of three patients. J Neurol Neurosurg Psychiatry
40 : 395-403
45. Wilbrand H (1892) Ein Fall yon Seelenblindheit und Hemianopsie mit Sectionsbefund. Dtsch Z Nervenheilkd 2 :
361-387
46. Young MP, Yamane S (1992) Sparse
population coding of faces in the inferotemporal cortex. Science 256 :
1327-1331