Journal of Clinical and Experimental Neuropsychology
1989, Vol. I I , No. 2, pp. 206-218

0168-8634/89/1102-0206 $3.00
@ Swets & Zeitlinger

Memory Functioning Following
a Left Medial Thalamic Hematoma*
Gregory G. Brown, Stuart Kieran, and Suresh Patel
Departments of Psychiatry, Neurology, and Diagnostic Radiology,
Henry Ford Hospital, Detroit.
ABSTRACT
A patient with a discrete, left medial thalamic hematoma developed a severe
anterograde amnesia. The memory deficit was most consistent for verbal materials, but was also observed on several tests of figural memory. The patient
appeared to have a normal short-term memory (STM) capacity, performed normally on a STM distractor task and was not abnormally sensitive to proactive
inhibition. This case is evidence that unilateral damage to the medial thalamus
alone can produce a severe memory deficit.

Clinicopathological studies of patients with alcoholic Korsakoff syndrome
suggest that damage to the medial thalamus, particularly the dorsomedial
nucleus (DM) of the thalamus, might cause the amnesia seen in that disease
(Martin, 1974; Victor, Adams, & Collins, 1971). Reports of memory impairments involving patients with a variety of focal lesions support the belief that
damage to the medial thalamus can cause severe memory dysfunction (Ignelzi &
Squire, 1976; Michel, Laurent, Foyatier, Blanc, & Portafaix, 1982). However,
in reviewing cases of amnesia associated with thalamic damage, Markowitsch
(1982) found that the lesions in most cases also involved cortical structures,
other subcortical nuclei, or white-matter tracts surrounding the thalamus. For
example, the well-known case of a fencing foil injury to the left DM probably
also involved the orbital-frontal cortex, neostriatum, the rostrum of the corpus
callosum and white matter near the thalamus (Squire & Moore, 1979). Patients
with alcoholic Korsakoff syndrome also have lesions outside of the medial
thalamus. Post-mortem studies have nearly always found extensive bilateral
nerve cell loss and gliosis of the mammillary bodies, in addition to involvement
of DM (Gamper, 1928; Victor, et al., 1971). In the two cases of alcoholic
Korsakoff syndrome reported by Mair, Warrington, and Weiskrantz (1979), the
mammillary bodies were involved to a greater extent than were the dorsomedial

* This study was in part supported by funds from NINCDS grant NS23393-01 to the
Stroke Center, Henry Ford Hospital. We wish to thank Steven Levine for his helpful
comments on a n early draft and Lawrence M. Binder for scoring the Rey-Osterrieth
Complex Figure.
Address correspondence to: Gregory G . Brown, Ph.D., Neuropsychology Division,
Henry Ford Hospital (K-1 I), 2799 W. Grand Blvd., Detroit, MI 48202, USA.
Accepted for publication: October 26, 1987

GREGORY G. BROWN ET AL

207

nuclei. Since Markowitch’s review, several cases of infarction of the medial
thalamus have been reported (Graff-Radford, Damasio, Yamada, Eslinger, Lk
Damasio., 1985; Michel et al., 1982; Rousseaux et al. 1986). Some of these
reports are limited by the presence of lesions existing outside of the thalamus.
Others are limited by a minimal behavioral analysis of the memory deficits
observed. Two cases of material-specific memory deficits following unilateral
infarction restricted to the medial thalamus were thoroughly studied by Speedie
and Heilman (1982, 1983). But each patient had a history of psychopathology.
The case with the left thalamic lesion was a paranoid schizophrenic, while the
64-year-old male with the right thalamic infarction had a long-standing depression. Since schizophrenia and depression can both impair memory functioning
(Koh, 1978; Stromgren, 1977) the effects of neuropathology and psychopathology are confounded in each case. The material-specific nature of the memory
dysfunction might appear to discount explanations based on psychopathology.
However, some investigators have reported that schizophrenia might affect
verbal functions more than visuospatial ones, while affective disorder might
have the reverse effect (Flor-Henry, 1976).
Extrathalamic damage or psychopathology has been reported among virtually
all previous cases of amnesia associated with lesions of the medial thalamus.
Consequently, the relevance that damage to the medial thalamus has for amnesia remains uncertain. This paper is a report of memory functioning after a
discrete, left-medial thalamic hematoma, occurring in a patient with no history
of psychopathology. Memory for daily events was severely impaired and performance on verbal memory tasks that involved long-term memory processing
was poor. To facilitate comparison with alcoholic Korsakoff syndrome, the
patient was examined by methods used to study alcoholic Korsakoff patients.
We also administered tests useful in investigating material-specific memory
deficits.

CASE REPORT
History and Neurological Examination
J.E. was a 69-year-old, right-handed Black male with 12 years of education who
was a retired postal clerk when he experienced a sudden onset of “confusion” on
1/3/86. The patient’s wife noticed the confusion as she drove him home from an
outpatient biopsy of the prostate. He was slightly drowsy and had memory
problems throughout the day; his wife took him to the emergency room the
following day. There was no previous history of neurologic disease. His medical
history was significant only for glaucoma and urolithiasis. In particular he had
no history of hypertension or heart disease; his prostate biopsy revealed normal
tissue.
Initial examination showed him to be oriented to person and time, but not to
place. He was slightly lethargic, but easily arousable. He was unable to

208

THALAMIC HEMATOMA

remember any of three objects in 5 minutes; did not remember the prostate
biopsy from the previous day; and h ad a patchy memory for remote events in his
life. Naming, repetition, fluency, comprehension, reading, and writing were all
normal. Motor functioning was intact; he could distinguish left from right on his
body. Although pain, touch, proprioception, graphesthesia, two-point discrimination, and stereognosis were intact, J.E. tended to ignore the right side of the
body on double-simultaneous stimulation. The following day, his alertness and
ability to recognize stimulation on the right side of the body improved. His
memory deficits persisted.
An EEG performed on the day of admission was significant only for some
depression of sleep spindle activity in the left hemisphere. This finding was
interpreted as an underlying disturbance within the thalamocortical radiation
system, possibly involving the thalamus. Other features of the EEG record were
compatible with J.E.’s age.

Neuropsychological Assessment
General NeuropsychologicaZ Functioning. J.E.’s initial neuropsychological assessment was performed on 1/28/86, three and one-half weeks after the ictus.
Form I1 of the WMS was administered on 3/5/86, the day of his nuclear magnetic
resonance imaging study. On the WAIS-R this patient’s FSIQ was 95, while his
VIQ was 92 and his PIQ was 101. The patient performed consistently within the
normal range across the various IQ subtests. All but two of the patient’s
age-corrected scores varied between 8 and 11. His Picture Arrangement score
was 12, while his Comprehension score was a 7. On this latter subtest the
patient’s answers were often insufficiently specific to merit credit. J.E. performed well on the Wisconsin Card Sorting Test obtaining all 6 concepts and
making only 14 errors. He also scored well on the Facial Recognition Test
(Benton, Hamsher, Varney, & Spreen, 1983), where he obtained a projected,
long-form score of 49. Using Taylor’s criteria, J.E. obtained a score of 32 when
copying the Rey-Osterrieth design (See Figure la). His performance placed him
at the 40th to 50th centile (Lezak, 1983, p. 401), and was more typical of the
performance of normal controls than of cases with right or left hemispheric
stroke (Binder, 1982).
Language Functioning. J.E.’s speech was fluent with normal phrase length,
prosody, and grammar. There were no paraphasias in his conversational
speech. His auditory comprehension was normal; he scored at the 90th to 100th
centile on three of the four components of the auditory comprehension subtest
of the Boston Diagnostic Aphasia Examination (Goodglass & Kaplan, 1983).
He scored only at the 65th centile on the complex ideational component, but his
poor performance on that test occurred only for the comprehension of story
paragraphs, where his memory deficit may have limited his performance. J.E.3
speech comprehension also fell within the normal range on the 36-item version
of the Tokens Test (DeRenzi & Faglioni, 1978), where he obtained a score of 32

209

GREGORY G. BROWN ET AL

A

B

Fig. 1 . JL’spcrformance on the Rcy-Osterrieth Complex Figure Test (a) copy, (b) recall,
45-minute delay.

(adjusted for education). The patient’s repetition was intact; he performed perfectly on both the low and high probability items of the repetition subtest of the
Boston Diagnostic Aphasia Examination. J.E. earned a score of 32 (adjusted for
age, sex, and education) on the FAS version of the Controlled Oral Word
Association Test (Lezak, 1976). This performance placed him in the bottom
quarter to one-third of the normal population. J.E. obtained a score of 55 on the
Boston Naming Test, indicating intact confrontational naming (Kaplan, Goodglass, & Weintraub, 1983). Overall, the patient’s language functioning was
normal.

Memory Functioning. The patient’s Wechsler Memory Quotient was 86. He was
oriented to person, place, and time, but had some trouble remembering the
President of the United States prior to Reagan and difficulty remembering the
current Governor of Michigan. J.E. recalled 6 digits forward and 4 reversed. He
recalled seven pieces of information from the Visual Reproduction subtest, an
above average performance when his score is compared with individuals of
similar age (Hulicka, 1966). His scores of 2.25 on the Logical Memory paragraphs and 4.5 on the Associative Learning items indicate a severe level of
dysfunction, when compared with age-peers. After a 20-minute delay, the
patient recalled only 33% of the original information that he remembered from
the logical paragraphs, but 57% of the information from visual reproduction.
J.E.’s scores on Form I1 of the WMS, performed on 3/5/86, were very similar to
his Form 1 scores, from 1/28/86. He was severely impaired on the Immediate
(3.5) and Delayed Recall of Prose (0%). He again recalled only easy word pairs,
earning an impaired score of 4.0. His memory for geometric designs was again
much better than his memory for verbal material with an Immediate Recall score

210

THALAMIC HEMATOMA

of 13 and a percentage retained of 54%.
His performance on Form 1 of Hannay and Levin’s version of the Selective
Reminding Task (Hannay & Levin, 1985) was also dramatically impaired. We
analyzed these results both from Buschke’s framework (Buschke, 1973) and
from the framework of the Search of Associative Memory theory (Raaijmakers
& Shiffrin, 1981). After correcting for age and sex, J.E. earned a score of 73 on
the long-term store parameter and a score of 25 for consistent long-term
memory retrieval. When compared with individuals age 60-64, these scores fell at
the 4th and 2nd centile, respectively (Larabee & Levin, 1984). On later trials,
J.E. typically retrieved three or four items from short-term memory, an abnormally large number of short-term memory retrievals. Together these results
suggest that J.E. had a pathological tendency to rely on short-term memory
mechanisms to retain items.
The results of the selective reminding task were also analyzed by fitting the
parameters of a model based on the Search of Associative Memory theory to
J.E.’s scores (Raaijmakers & Shiffrin, 1981). A simulation of performance on
the selective reminding task was developed. We estimated the values of three
parameters: r - the capacity of the short-term memory buffer, a - the increment
in the strength of association between item and context during each rehearsal
period, and kmux - the maximum number of consecutive rehearsal failures a
subject will tolerate before terminating memory retrieval. The parameter values
were determined by a search through a large segment of three-parameter space.
Performance on the Selective Reminding Task was simulated 100 times for
each triplet of parameter values. Each simulation produced a matrix representing correct (scored 1) and incorrect (scored 0) recall for each of the 12 items on
each of the 12 trials. The concordance between each cell of the matrix produced
by a simulation and each cell from J.E.’s matrix was determined by performing
an exclusive-or on the two values. When homologous cells from the two matrices
agreed (0,Oor 1, l), the exclusive-or equaled 0, otherwise it equaled 1. The extent
to which the two matrices agreed was determined by summing the results of the
exclusive-or performed on each of the 144 pairs of cells. The values that
minimized the average summed exclusive-or, computed for the 100simulations,
were chosen as the estimates of the three parameters. The average estimates for
each parameter were: r = 3.85, u=0.04, kmax= 7.14. While the result for r, the
capacity of the STM buffer, was similar to 4, the value typically reported in the
SAM literature for young, healthy subjects (Raaijmakers & Shiffrin, 1981), the
values of the other two parameters are low. Low values on a and kmax would
suggest both a deficit in encoding contextual associations and a premature
closure of memory search. The results from the model suggested that J.E.
probably had a normal STM capacity, but was impaired on aspects of long-term
memory functioning. These results converge with those from the analysis based
on Buschke’s framework.
The patient’s short-term memory functioning was further studied on two
short-term memory (STM) distractor tasks. The first required J.E. to recall

GREGORY G. BROWN ET A L

21 1

consonant trigrams after counting backwards by one for zero, three, nine, or
eighteen seconds (Cermak, Butters, & Goodglass, 1971). The second fixed the
counting interval at 9 seconds and required J.E. to recall three words from the
same semantic category on four consecutive trials. The semantic category was
switched on the fifth trial, permitting an assessment of the release from proactive inhibition (Cermak, Butters, & Moreines, 1974). Table 2 contains J.E.’s
scores as well as those of eight aged-matched controls (mean age = 65.6 years,
mean education = 11.6 years). The controls displayed a buildup of proactive
inhibition (Pi) on trials 1 to 4, with a release from Pi on trial 5. J.E. performed
normally for the memory of trigrams and showed no appreciable buildup of
proactive inhibition. (See Table 1).
Table 1
Short-Term Memory Distractor Tasks
Consonant Trigrams
Interval
0s

3s
9s
18s

Replications Recalled”
JE
Controls
10
7.9
6
5.8
3.3
7
7

2.4

Consonants Recalledb
JE
Controls
30
25.5
26
24.8
26
17.2
25
14.6

Pi Paradigm‘
Trial
1
2

3
4

5 (release trial)

Words Recalled”
JE
Controls
15
13.2
14
10.6
13
8.4
14
8.0
12
12.5

a - 10 trigrams at each distractor interval
b - 30 consonants total
c -the release from proactive inhibition (Pi) paradigm
d - 15 words total, 5 triads per trial

J.E. was given Rausch’s version of Kimura’s Recurring Figures Test, which
omits the geometric designs from Kimura’s stimuli (Personal Communication to
K. M. Adams, June 28, 1984). J.E. obtained a score of 7, indicating a severely
impaired performance; his errors were generally misses. His performance on the
Rey-Osterrieth Designs was both sparse and inaccurate after a 45-minute delay
(see Figure 1b).
J.E.’s remote memory was assessed on two subtests of the Boston Remote
Memory Battery (Albert, Butters, & Levin, 1979). We compared J.E.’s scores
with the healthy elderly controls from Wilson, Kaszniak, and Fox’s (198 1) study.
These subjects were elderly, community volunteers with a mean age of 69.4

212

THALAMIC HEMATOMA

Table 2
Two Subtests of the Boston Remote Memory Battery
Decade
Difficulty
Level

Percent
Correcta

Easy
Hard

88 (60)
58 (35)

Easy
Hard

42 (50)
33 (35)

'

1920

1930

1940

1950

1960

1970

3
3

2
2

4
1

4
1

1
0

4
1

1
1

1
2

Famous Faces
4
4

4

3

Verbal Recall
2
2

1
2

Note. Maximum Score for each entry by decade = 4
a-the percentages in parentheses are values for age-matched controls (R. S . Wilson,
personal communication, June, 1986).

years and an average educational attainment of 13.2 years. As reported in Table
2, J.E.'s scores were better than controls on both the easy and hard items from
Famous Faces and similar to controls on Verbal Recall. J.E. appeared to have a
gradient on the hard items of the Famous Faces Test, where his recall was
progressively worse for more recent information.
In summary, J.E. demonstrated an anterograde memory deficit for verbal
and geometric material, when the amount of material recalled was large relative
to his short-term memory capacity or when he had to recall the material after the
elapse of several minutes. In distinction, his verbal short-term memory functioning was intact. J.E. appeared to have no consistent remote memory deficit;
despite the appearance of a gradient on the hard items of the Famous Faces
Test, his overall scores on tests of remote memory were normal.

Neuroradiologic Findings
Computerized tomography (CT) of the head was performed on 1/6/86. It
showed a left, medial thalamic hematoma, surrounded by minimal edema (see
Figure 2). There was no contrast enhancement. Some blood was present in the
body of the left lateral ventricle. The third ventrical and septum pellucidum fell
along the midline; there was no obstructive hydrocephalus. A cerebral angiogram performed three days after admission showed no evidence of an arteriovenous malformation, aneurysm, or major vessel disease.
J.E. had a magnetic resonance imaging scan on 3/5/86, two months after his
admission. The study was performed on a 1.5 tesla magnet. Serial T, weighted
images (TR600, TE25) were obtained in the sagittal plane using a partial
saturation technique. Subsequently, spin echo techniques were used to obtain
T, weighted images (TR2500, TE25,70,75, 100) in coronal, axial, and sagittal
planes. The T, weighted image showed a small, low-intensity lesion in the

GREGORY G. BROWN ET AL

213

A

B

c
Fig. 2. ( a ) CT bcan of medial thalamic hemorrhage, axial cut (b) N M K imaging scar1
showing involvement of anterior a n d dorsomedial thalamic nuclei, axial cut (c)
N M R image of a coro11al view of the left thalamus.

214

THALAMIC HEMATOMA

medial, superior thalamic region involving mainly the dorsomedial thalamic
nucleus. On the T,weighted,axial images, a low intensity signal was seen in the
same region of the medial left thalamic region as the TI weighted lesion. (See
Figure 2.) Most of the lesion spans the dorsomedial thalamic nucleus, but some
of the anterior nucleus is also involved. Much of the pulvinar was spared, as
were more lateral nuclei. The findings were compatible with the resolved or
resolving hematoma, with hemosiderin deposition, seen on the CT scan. Views
of the hippocampus, mammillary bodies, and the region ventral to the anterior
comrnissure (including the substantia innominata) appeared normal. There was
no evidence of edema, mass effects, or blood in the ventricles.

DISCUSSION
This case supports the thesis that damage to the thalamus alone can produce a
severe memory deficit. Although some edema of surrounding white matter was
noted on the CT performed acutely, the magnetic resonance imaging studies
performed two months later showed no chronic changes in the area of the edema.
In reviewing cases of amnesia associated with thalamic damage, Markowitsch
(1982) found that the lesions in most cases also involved cortical structures,
other subcortical nuclei, or white matter tracts surrounding the thalamus. In
our case, NMR imaging studies showed no abnormalities in the cortex, hippocampus, mammillary bodies, or the region ventral to the anterior comniissure.
Except during the acute period, the white matter surrounding the thalamus,
including nearby segments of the fornix, also appeared normal. The area of
damage in J.E. seemed well restricted to the thalamus.
The hematoma in this case primarily involved the dorsomedial thalamic
nucleus. However, the anterior thalamic nucleus and part of the anterior
portion of the internal medullary lamina were also involved. Whether damage
occurred to adjacent pulvinar 01 adjacent intralaminar nuclei could not be
determined and remains a possibility. The involvement of the anterior nucleus
in this case is compatible with Markowitsch’s view that memory impairment
accompanies damage to dorsomedial thalamic nucleus only if some other limbic
nucleus is involved. Involvement of both nuclei coincides with the results from
animal studies, which show that combined lesions of anterior and posterior
medial thalamic regions produce a greater impairment of visual object recognition than damage to either region alone (Appleton & Mishkin, 1983a, 1983b).
Markowitsch’s hypothesis, based on anatomical data, may have a broader
metabolic corollary. Some patients with subcortical lesions have a metabolic
depression of ipsilateral caudate and cortex, as measured by the method of
l*F-fluorodeoxyglucose, positron emission computed tomography (Metter et
a]., 1983). In particular, case N.A. with a fencing foil injury to the left dorsomedial thalarnic nucleus showed a depression of glucose metabolism throughout
the left hemisphere (Metter et al., 1983). Although our patient did not have

GREGORY G. BROWN ET A L

215

metabolic studies, sleep spindles occurring only in the left hemisphere were
observed in his scalp EEG. These observations suggested a cortical effect of his
thalamic lesion. Speedie and Heilman’s (1983) case of right thalamic infarction
also had asymmetrical slowing of his EEG on the side ipsalateral to his infarction, If this depression of cortical metabolism is associated with depression of
cortical functioning, then extrathalamic metabolic effects could contribute to
the memory deficit associated with thalamic damage.
Despite finding that J.E. was as impaired as patients with alcoholic Korsakoff
syndrome on the logical memory and paired associates subtests of the WMS, his
short-term memory functioning was normal. In particular, he rehearsed three to
four words on a verbal, multi-trial, list-learning task; performed normally
following 9 and 18 s of distracting activity on an STM distractor task; and was
resistant to the buildup of proactive inhibition. Each of these three findings
distinguishes him from many cases of alcoholic Korsakoff syndrome (Brown,
Rosenbaum, Lewis, & Rourke, 1980; Cermak et al., 1971; Cermak et al., 1974;
Cermak, Naus, & Reale, 1976). The differences between J.E. and alcoholic
Korsakoff patients in short-term memory functioning cannot be explained by
method variance. Both the STM trigram task and the release from proactive
inhibition task followed the procedure used by Cermak and Butters in their
studies of alcoholic Korsakoff patients (Cermak et al., 1971; Cermak et al.,
1974). Rather, bilateral thalamic damage or lesions outside the medial thalamus
might very well cause the poorer STM functioning of patients with alcoholic
Korsakoff syndrome. In particular, Moscovitch (1982) has argued that an
abnormal sensitivity to P1 can be found among patients with frontal-lobe
dysfunction.
We found only weak evidence of a material-specific, anterograde, memory
deficit. J.E. performed better on the visual reproduction subtest of the WMS,
than on the recall of prose or word pairs. However, his delayed recall of
geometric designs was impaired. Further, the patient was severely impaired on
the Kimura Recurring Figures Task, a test performed well by Speedie and
Heilman’s case of left DM infarction. Although J.E. performed poorly, his error
pattern was uncharacteristic of cases of right temporal lobectomy (Kimura,
1967). While lobectomy cases make a large number of false alarms, J.E.’s errors
were misses. His poor performance on this task might be attributable to a
conservative response bias, rather than a figural memory deficit. Consequently
the basis of J.E.’s poor performance on some tests of figural memory was
unclear.
J.E. had a gradient on the hard items from the Famous Faces Test. This
finding is compatible with the view that damage to the medial thalamus might
be responsible for the remote memory gradient observed in alcoholic Korsakoff
syndrome. At least two factors vitiate this hypothesis. While patients with
alcoholic Korsakoff syndrome show a gradient on both the easy and hard items
of the Famous Faces and Verbal Recall tests, J.E. had a gradient on only one of
these four conditions. Further, remote memory deficits have not been routinely

216

THALAMIC HEMATOMA

observed in other cases of left medial thalamic damage whose remote memory
was studied (Cohen & Squire, 1981; Michel et al., 1982; Speedie & Heilman,
1982).The generality of the one gradient that we found is probably limited. J.E.
did perform relatively worse on the Verbal Recall subtest than the Famous
Faces subtest. This finding raises the possibility of a material-specific remote
memory deficit.
Future studies of the effects of thalamic damage on memory should include
measures of functional neuronal activity, in addition to an anatomical description of the lesion. Such studies are needed to understand the role that abnormalities in neuronal activity outside of the thalamus make to the amnesia associated with thalamic damage.
REFERENCES
Appleton, J. P., & Mishkin, M. (1983a). Visual recognition impairment following medial
thalamic lesions in monkeys. Neuropsychologia, 21, 189-197.
Appleton, J. P., & Mishkin, M. (1983b). Memory impairments following restricted
medial thalamic lesions in monkeys. Experimental Brain Research, 52, 199-209.
Albert, M. S., Butters, N., & Levin, J. (1979). Temporal gradients in the retrograde
amnesia of patients with alcoholic Korsakoff s disease. Archives of Neurology, 36,
21 1-216.
Benton, A. L., Hamsher, K. de S., Varney, N. R., & Spreen, 0. (1983). Contributions fo
neuropsychological assessment: A clinical manual (pp. 30-43). New York: Oxford,
Binder, L. M. (1982). Constructional strategies on complex figure drawings after unilateral brain damage. Journal of Clinical Neuropsychology, 4, 51-58.
Brown, G. G., Rosenbaum, G., Lewis, R., & Rourke, D. (1980). The effects of rehearsal
rate on serial recall in Korsakoff amnesia. Neuropsychologia, 18, 185-191.
Buschke, H. (1973). Selective reminding for analysis of memory and 1eaming.Journal of
Verbal Learning and Verbal Behavior, 12, 543-550.
Cermak, L. S., Butters, N., & Goodglass, H. (1971). The extent of memory loss in
Korsakoff patients. Neuropsychologia, 9, 307-315.
Cermak, L. S., Butters, N., & Moreines, J. (1974). Some analyses of the verbal encoding
deficit of alcoholic Korsakoff patients. Brain and Language, I, 141-150.
Cermak, L. S., Naus, M. J., & Reale, L. (1976). Rehearsal strategies of alcoholic Korsakoff patients. Brain andlanguage, 3, 375-385.
Cohen, N. J., & Squire, L. R. (1981). Retrograde amnesia and remote memory impairment. Neuropsychologia, 19, 337-356.
DeRenzi, E., & Faglioni, P. (1978). Normative data and screening power of a shortened
version of the Token Test. Cortex, 4, 41-49.
Flor-Henry, P. (1976). Lateralized temporal-limbic dysfunction and psychopathology.
Annals of the New York Academy of Science, 280, 777-795.
Gamper, E. (1928). Zur Frage der Polioencephalitis haemorrhagica der chronischen
Alkoholiker. Anatomische Befunde beim chronischen Korsakow und ihre Beziehungen zum klinischen Bild. Deutsche Zeitschrvt fur Nervenheilkunde, 102, 122- 129.
Goodglass, H., & Kaplan, E. (1983). Assessment of aphasia and related disorders. Philadelphia: Lea & Febiger.

GREGORY G . BROWN ET AL

217

Graff-Radford, N. R., Damasio, H., Yamada, T., Eslinger, P. J., & Damasio, A . R.
(1985). Non-haemorrhagic thalamic infarction: Clinical neuropsychological and
electrophysiological findings in four anatomical groups defined by computerized
tomography. Brain. 108, 485-5 16.
Hannay, H. J., & Levin, H. S. (1985). Selective Reminding Test: An examination of the
equivalence of four forms. Journal of Clinical and Experimental Neuropsychology, 7,
25 1-263.
Hulicka, I. M. (1966). Age differences i n Wechsler Memory Scales scores. .Iournal of
General Psychology. 109, 135-145.
Ignelzi, R. J., & Squire, L. R. (1976). Recovery from anterograde and retrograde amnesia
after precutaneous drainage of a cystic craniopharyngioma. Journal of Neurology,
Neurosurgery and Psychiatry,39, 1231-1235.
Kaplan, E., Goodglass, H., & Weintraub, S. (1983). Boston Naming Test. Philadelphia:
Lea & Febiger.
Kimura, D. (1967). Right temporal lobe damage. Archives of Neurology, 3, 163-178.
Koh, S. D. (1978). Remembering of verbal materials by schizophrenic youngadults. In S.
Schwartz (Ed.), Language and cognition in schizophrenia (pp. 55-89). New York:
Lawrence Erlbaum Associates.
Larrabee, G. J., & Levin, H . S . (1984, February). Verbal, visualandremote memory test
performance in a normalelderly sample. Paper presented at the meeting of the International Neuropsychology Society, Houston, TX.
Lezak, M. (1976). Neuropsychologicalassessment. New York: Oxford.
Lezak, M. D. (1983). Neuropsychological assessment (2nd ed.) New York: Oxford.
Mair, W. G . P., Warrington, E. K., & Weiskrantz, L. (1979). Memory disorder in
Korsakoff s psychosis: A neuropathological and neuropsychological investigation of
two cases. Brain, 102, 749-783.
Markowitsch, H. J. (1982). Thalamic mediodorsal nucleus and memory: A critical
evaluation of studies in animals and man. Neurosciences andBiobehavioralReviews, 6,
351-380.
Martin, J. J. (1974). Thalamic system atrophies in Gayet-Wernicke’s polioencephalopathy. Acta Neurologica Belgica, 74, 25-35.
Metter, E. J., Riege, W. H., Hanson, W. R., Kuhl, D. E., Phclps, M. E., Squire, L. R.,
Wasterbain, C. G., & Benson, D. F. (1983). Comparison of metabolic rates, language,
and memory in subcortical aphasias. Brain andLanguage 19, 33-47.
Michel, D., Laurent, B., Foyatier, N., Blanc, A,, & Portafaix, M. (1982). Infarctus
thalamique paramedian gauche. Revue Neurologigue. 138, 533-550.
Moscovitch, M. (1982). Multiple dissociations of functions in amnesia. In L. S. Cermak
(Ed.), Human memory and amnesia (pp- 337-370). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Raaijrnakers, J. G. W., & Shiffrin, R. M. (1981). Search of associative memory. Psychological Review, 88, 93- 134.
Rousseaux, M., Cabaret, M., Lesoin, F., Devos, P., Dubois, F., &Petit, H. (1986). Bilan
de I’amnesie des infarctus thalamiques restreints - 6 cas. Cortex, 22, 213-228.
Speedie,L., & Heilman, K. (1982). Amnesic disturbance following infarction of the left
dorsomedial nucleus of the thalamus. Neuropsychologia, 20, 597-604.
Speedie, L., & Heilman, K. (1983). Anterograde memory deficits for visuospatial material after infarction ofthe right thalamus. Archives of Neurology, 49, 183-186.
Squire, L. R., & Moore, R. Y . (1979). Dorsal thalamic lesions in a noted case of chronic

218

THALAMIC HEMATOMA

memory dysfunction. Annals of Neurology, 6, 503-506.
Stromgren, L. S. (1977). The influence of depression on memory. Acta Psychiatrica
Scandinavica, 56, 109-128.
Victor, M., Adams, R. D., & Collins, G . H.(1971). The Wernicke-KorsakoffSyndrorne.
Philadelphia: F. A. Davis.
Wilson, R. S., Kaszniak, A.W., & Fox, J. H. (1981). Remote memory in senile dementia.
Cortex, 17, 41-48.