89

SHORT REPORT

Paradoxical recovery in a bilingual patient with aphasia after
right capsuloputaminal infarction
A Garcı́a-Caballero, I Garcı́a-Lado, J González-Hermida, R Area, M J Recimil, O Juncos Rabadán,
S Lamas, G Ozaita, F J Jorge
...................................................................................................................................
J Neurol Neurosurg Psychiatry 2007;78:89–91. doi: 10.1136/jnnp.2006.095406

We report the case of a bilingual dextral patient, who
presented with an uncommon pattern of aphasic deficit
following a right capsulo-putaminal infarction. In this patient,
the linguistic deficit concerned the use of her mother tongue
(Galician, L1) much more than the lesser practised second
language (Spanish, L2). Our patient presented spontaneous
fluent speech in L2 but not in L1, automatic translation into L2,
and impaired repetition in L1, whereas comprehension was
spared in both L1 and L2. Reading and writing were less
valuable due to educational interference (reduced schooling).
Spontaneous speech 16 months after the stroke showed the
stability of the impairment.
This is the first reporting of a crossed subcortical aphasia in a
bilingual patient.

A

lthough subcortical aphasia cases in monolinguals have
been extensively studied during the past decades,1–5 only
two cases of bilingual aphasia in adults have been
described after subcortical stroke. In the first case the lesion
was located in the left capsuloputaminal region, and the patient
showed a paradoxical recovery of her less-practised second
language (Italian, L2), being no longer able to speak her mother
tongue (Venetan, L1).6 7 In the second case, impairment of the
mother tongue (Croatian, L1) appeared after infarction of the left
caudate nucleus.8 During follow-up, improvement of the mother
tongue was accompanied by worsening of the second language
(Italian, L2) after extension of the ischaemic lesion to the cortex.
Crossed aphasia is a term coined by Bramwell8a in 1899 to
refer to language dysfunction appearing after right hemispheric
brain lesions in dextrals.9 The precise incidence of this type of
anomalous language lateralisation is unknown, ranging from
1% to 13% of all patients with aphasia and being present in up
to 38.5% of neurologically intact dextrals.10 11 Two types of
crossed dysphasia have been described, with either dissociated
or simultaneous language and visuospatial deficit due to
reversed representation of hemisphere-specific functions or
transfer of most cognitive functions to the right hemisphere.12
Subcortical crossed aphasia, although rare, have been
described previously in monolinguals after infarction of the
posterior limb of the internal capsule and head of the caudate
nucleus, and anterior part of the internal capsule, lentiform
nucleus and head of the caudate.13 14 Given these facts, it was
thus intriguing to think how subcortical right hemisphere
lesions could affect language performance in bilingual patients.

CASE REPORT
The patient is a 91-year-old, right-handed housewife who, in August
2004, suddenly stopped speaking in her most fluent and practised
Galician (L1), and started to speak in her less fluent Spanish (L2).
She showed no further neurological features. She had no history of
hypertension, ischaemic heart disease or diabetes mellitus.

Galician is the patient’s mother tongue and the language she
had used all her life. She grew up and lived in a rural hamlet
where only Galician was spoken. She attended 4 years of school
in Spanish, although she managed to be a passionate reader,
especially of religious literature in L2. She watched television 2–
3 h/day (1/3 of the time in Galician and the rest in Spanish).
Before the stroke she hardly ever spoke in L2; its use was
restricted to speaking with some of her grandsons born in
Castile when they came to visit her once or twice a year.
Initially, her family did not pay much attention to the fact that
she stopped speaking in Galician. They thought this fact was related
to other mild behavioural disturbances they had been noticing
during the previous years. These symptoms consisted of mild
disinhibition (ie she asked their guests for a cup of coffee, a
cigarette or even a glass of liquor!) and fluctuating memory deficits,
which were considered by her family to be a normal consequence of
aging. There was no impairment of her daily living abilities, but
they were just restricted to bathing, dressing and reading religious
literature (in L2). Four months after the episode of language switch,
an in-law of the patient who was a young trainee, reported the case
to us and asked for a cognitive evaluation to discount dementia.
Neurological examination
The general physical examination was unremarkable. She
showed no hemiparesis or hypoesthesia. Tendon reflexes were
symmetrical and Babinski sign was absent. Strikingly, she
continued speaking Spanish during the evaluation, although
her daughter and the doctor spoke to her in L1.
Neuroimaging
A magnetic resonance imaging study performed in January 2005
showed a right capsuloputaminal infarction (figs 1 and 2).
Severe corticosubcortical atrophy (figs 1 and 2) and small lesions
of the subcortical white matter were also present bilaterally.
Neuropsychology
Neuropsychological examination was performed in L2. She was
right handed when assessed by the Edinburgh Handedness
Inventory (laterality quotient +100).15 She was evaluated using
the Spanish version of the Addenbrooke’s Cognitive Examination (ACE).16 ACE score was 65/100. She scored 26/30 in the
Spanish version17 of the Mini-Mental Status Examination.18 She
was oriented in time and space. Memory was apparently
unaffected, because she remembered three out of three words
after 5 min and correctly named the current and former President
and the Pope’s name. Learning curve of a list of 10 unrelated words
was considered normal for her age and educational level.19 Five
trials were allowed (3–3–4–6–7) and a delayed recall 5 min later
(six correct words).19 Afterwards she was shown a recognition list
where she correctly identified the words of the first list and
Abbreviations: ACE, Addenbrooke’s Cognitive Examination; L1, mother
tongue; L2, second language

www.jnnp.com

90

Subcortical crossed aphasia

was shown a recognition list where she correctly refused the
confusion items, missing two items of the first list.
Finally, to differentiate genuine memory deficits from
apparent memory deficits due to inefficient strategies and
impaired attention, an enhanced cued recall test was used.22 In
this case, we used the Spanish version of the free and cued
learning test,23 included in the 7-min neurocognitive screening
battery.24 Free recall was 6/16, cued recall was 8/16 and total
score was 14/16, a performance considered normal for her age
and educational level (between the 50th and 60th centiles).23 It
has to be noted that her cued recall 10 min later was 15/16.

Figure 1 Magnetic resonance image executed 6 months after the onset,
showing a right capsuloputaminal infarct.

correctly refused the confusion items. Verbal fluency was impaired
(animals: 6; letter p: 4) and mixed items in L1 and L2. She showed
a constructional apraxia on both copy (pentagons) and free-hand
drawing. The clock drawing test following ACE’s instructions was
severely impaired.20 Her writing to command (L2) was impaired,
showing agrammatisms and spelling errors (letter omission).
These features were less valuable because of educational interference. She was not able to perform serial seven subtractions or its
easier Spanish counterpart, serial three subtractions. Clinical
Dementia Rating Scale score was 0.5 (questionable impairment).21
Remarkably, 1 year later her cognitive status remained
stable, her Mini-Mental Status Examination scored 27, and
she kept performing the same daily activities.
One and a half years after the first neuropsychological
evaluation, the learning curve of a list of 10 unrelated words
was repeated.19 Five trials were allowed (3–5–5–6–7) and a
delayed recall 5 min later (seven correct words).19 Later, she

Language tasks
In December 2004, our patient did not produce spontaneous
speech in her mother tongue and only on request would she
translate simple single words but not phrases. She showed
spontaneous translation into L2 when repeating words and
phrases in L1. This was also true for repetition of well-known
refrains in L1. Reading was impaired in L1 but not in L2. On the
other hand, comprehension was spared in both L1 and L2 (table 1).
One year later she was given the Galician25 and the Spanish26
version of the Bilingual Aphasia Test. Table 1 summarises the
results. There was a high percentage of words (97%) in the nonrequested language (L2) during sessions in L1; by contrast, words
in L1 were absent during sessions in L2. There were no differences
in reading, repetition, naming and comprehension tasks.

DISCUSSION
Studies on bilingual aphasia have suggested that dominant basal
ganglia lesions impair access to the mother tongue while sparing
L2.6 7 This was coherent with a hypothesis posed by Paradis,27 who
suggested that, given that L1 uses unconscious learning strategies
typical of implicit memory, it should be anatomically organised in
cortical and subcortical structures that differ from those implicated
in L2, generally learnt by means of conscious strategies, typical of
explicit memory. Given that basal ganglia are involved in implicit
(unconscious-repetitive) memory but not in explicit (conscioussingle exposition) memory and also in language organisation, a
lesion of the basal ganglia should impair L1 but not L2.6 Moreover,
it has been suggested that grammatical operations in L1 might be
sensitive to basal-ganglia prefrontal cortex network dysfunction
present, for instance, in patients with Parkinson’s disease.28
Galician is a romance language different from Spanish, which
belongs to the Iberian romance family of languages. It has
common aspects with Portuguese (morpho and some vocabulary).
From the linguistic point of view, the boundaries of the Galician
Table 1 Patient’s performance during the initial evaluation
and 1 year later using the Bilingual Aphasia Test
Number of correct items
(total trials)

Figure 2 Coronal section magnetic resonance image showing a right
putaminal lacunar infarct (arrows) and corticosubcortical atrophy. Head of
the caudate nucleus is not affected.

www.jnnp.com

Initial evaluation
Spontaneous speech
(total number of words in 5 min)
Repetition
Word reading
One year later (BAT results)
Spontaneous speech (dialogue in L1;
total number of words in 5 min)
Spontaneous speech (dialogue in L2;
total number of words in 5 min)
Repetition
Word reading
Naming
Auditory comprehension

L1

L2

5

209

16 (21)
1 (6)
L1
11

23 (23)
6 (6)
L2
356

0

387

22 (25)
7 (10)
9 (10)
7 (9)

25 (25)
10 (10)
10 (10)
9 (9)

BAT, Bilingual Aphasia Test; L1, mother tongue; L2, second language.

Subcortical crossed aphasia

language are: Portuguese in the South and Astur-Leonese in the
East. Galicia has a population of 2.7 million inhabitants, 94% of
the population understand Galician and 88% speak it.29
Traditionally, Galician was spoken in everyday life, whereas
Spanish was limited to formal contexts. During Franco’s régime,
the Galician language was banned and therefore formal education
in Galician was prohibited until the 1980s. Owing to this
sociolinguistic situation, the patient never learnt to read or write
in the Galician language. In the patient’s case, Galician was the
only language she was mainly exposed to until she went to school
at the age of 8 years. The reason is that she was born and lived in a
very small hamlet made up of just a few farms. This relative
isolation resulted in a restricted exposure to the Spanish language.
These questions are relevant, because native language grammar is
acquired implicitly with involvement of basal ganglia structures
that are mature at birth, whereas syntactic skills in L2 are learnt by
using declarative memory, which mainly relies on temporal
systems that complete their maturation in later infancy.28 It is also
relevant because reading and writing in Galician is not valuable or
useful for comparisons. The same is true for other tasks such as
discrimination between words and pseudowords,30 or phonetic
oral fluency,31 tasks that require phonemic awareness.
From the clinical point of view, it can be argued that the
behavioural disturbances the patient showed were prodromal
features of a degenerative disorder. However, serial neuropsychological testing dealing with memory deficits showed no
deterioration over a follow-up period of 1K years. Moreover,
enhanced cued recall, a well-known measure to avoid educational and attentional bias,22 was normal given her age and
educational level.23 Visuoconstructional tasks are less valuable in
a patient with only 4 years of formal education.32 These facts, in
our opinion, rule out an Alzheimer’s-type dementia. Accordingly,
a frontotemporal dementia seems improbable given her age and
clinical features; it is more reasonable to attribute her mild
disinhibition to vascular disruption of frontosubcortical circuits.
Vascular changes have been demonstrated by magnetic resonance imaging, and a vascular aetiology of the language switch is
not only probable, given a compatible lesion in the right
capsuloputaminal region, but likely, given the sudden onset of
the language impairment and its evolution during the next year.
The case we report was clinically similar to those reported in
the literature,6–8 but the patient’s lesion was the mirror image of
what should be expected.
Cases of subcortical crossed aphasia, although rare, have
been described previously in monolinguals,13 14 but as far as we
know this is the first report of the occurrence of a subcortical
crossed aphasia in a bilingual. As we have seen, our findings
support the hypothesis that basal ganglia are essential nodes in
the neural network subserving the mother tongue.

ACKNOWLEDGEMENTS
We thank the patient and her daughter for their collaboration.
.......................

Authors’ affiliations

A Garcı́a-Caballero, I Garcı́a-Lado, J González-Hermida, M J Recimil,
S Lamas, Servicio de Psiquiatrı́a, Complexo Hospitalario de Ourense,
Ourense, Spain
R Area, Servicio de Psiquiatrı́a, Complexo Hospitalario Xeral-Calde, Lugo, Spain
O Juncos Rabadán, Departamento de Psicoloxı́a Evolutiva e da Educación,
Universidade de Santiago de Compostela, Santiago de Compostela, Spain
G Ozaita, Servicio de Neurologı́a, Complexo Hospitalario de Ourense, Ourense,
Spain
F J Jorge, Departamento de Ciencias Morfológicas, Universidade de Santiago de
Compostela, Santiago de Compostela, Spain
Competing interests: None declared.
Informed consent has been obtained for publication of the patients’ details
in this paper.

91
Correspondence to: Dr A Garcı́a-Caballero, Servicio de Psiquiatrı́a,
Complexo Hospitalario de Ourense, R/ Ramon Puga 54, 32001, Ourense,
Spain; Alejandro.Garcia.Caballero@sergas.es
Received 10 April 2006
Revised 9 July 2006
Accepted 17 July 2006

REFERENCES
1 Damasio AR, Damasio H, Rizzo M, et al. Aphasia with nonhemorrahgic lesions
in the basal ganglia and internal capsule. Arch Neurol 1982;39:15–20.
2 Naeser MA, Alexander MP, Helm-Estabrooks N, et al. Aphasia with
predominantly subcortical lesion sites: description of three capsular/putaminal
aphasia syndromes. Arch Neurol 1982;39:2–14.
3 Weiller C, Wilmes K, Reiche W, et al. The case of aphasia or neglect after
striatocapsular infarction. Brain 1993;116:1509–25.
4 Mega MS, Alexander MP. Subcortical aphasia: the core profile of capsulostriatal
infaction. Neurology 1994;44:1824–9.
5 D9Esposito M, Alexander MP. Subcortical aphasia: distinct profiles following left
putaminal hemorrhage. Neurology 1995;45:38–41.
6 Aglioti S, Fabbro F. Paradoxical selective recovery in a bilingual aphasic
following subcortical lesions. Neuroreport 1993;4:1359–62.
7 Aglioti S, Beltramello A, Girardi F, et al. Neurolinguistic and follow-up study of
an unusual pattern of recovery from bilingual subcortical aphasia. Brain
1996;119:1551–64.
8 Moretti R, Bava A, Torre P, et al. Bilingual aphasia and subcortical-cortical
lesions. Percept Mot Skills 2001;92:803–14.
8a Bramwell B. On crossed aphasia and the factors which go to determine whether
the ‘leading’ or ‘driving’ speech-centres shall be located in the left or in the right
hemisphere of the brain, with notes of a case of ‘crossed aphasia’ (aphasia with
right-side hemiplegia) in a left-handed man. Lancet 1899;i:1473–9.
9 Kirsher HS. Handbook of neurological speech and language disorders. New
York: Marcel Dekker, 1995.
10 Alexander MP, Annet M. Crossed aphasia and related anomalies of cerebral
organization: case reports and a genetic hypothesis. Brain Lang 1996;55:213–39.
11 Pujol J, Deus J, Losilla JM, et al. Cerebral lateralization of language in normal
left-handed people studied by functional MRI. Neurology 1999;52:1038–43.
12 Kapur N, Dunkley B. Neuropsychological analysis of a case of crossed dysphasia
verified at postmortem. Brain Lang 1984;23:134–47.
13 Sapir S, Kokmen E, Rogers PJ. Subcortical crossed aphasia: a case report.
J Speech Hear Disord 1986;51:172–6.
14 Dobrzynska J. Crossed aphasia in right handed patients. Neurol Neurochir Pol
1991;25:31–9.
15 Oldfield RC. The assessment and analysis of handedness: The Edinburgh
inventory. Neuropsychologia 1971;9:97–113.
16 Garcı́a-Caballero A, Garcı́a-Lado I, González Hermida J, et al. Validation of the
Spanish version of the Addenbrooke’s cognitive examination in a rural
community in Spain. Int J Geriatr Psychiatry 2006;21:1–7.
17 Lobo A, Saz P, Marcos G, et al. Revalidación y normalización del Mini-Examen
Cognoscitivo (primera versión en castellano del mini-mental status examination)
en la población general geriátrica. Med Clin (Barc) 1999;112:767–74.
18 Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical method for grading
the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–98.
19 Christensen AL. El diagnóstico neuropsicológico de Luria, 2. ”, eds. Madrid:
Visor, 1987.
20 Garcı́a-Caballero A, Recimil MJ, Garcı́a-Lado I, et al. The ACE clock scoring. A
comparison with eight standard correction methods in a population of loweducational level. J Geriatr Psychiatr Neurol 2006;19:216–19.
21 Hughes CP, Berg L, Danzinger WL, et al. A new clinical scale for staging of
dementia. Br J Psychiatry 1982;140:566–72.
22 Grober E, Buschke H, Crystal H, et al. Screening for dementia by memory testing.
Neurology 1988;38:900–03.
23 Del Ser T, Sánchez Sánchez F, Garcı́a de Yébenes MJ, et al. Versión española
del Test de los 7 Minutos. Datos normativos de una muestra poblacional de
ancianos de más de 70 años. Neurologı́a 2004;19:344–58.
24 Solomon PR, Hirschoff A, Kelly B, et al. A 7 minute neurocognitive battery highly
sensitive to Alzheimers disease. Arch Neurol 1998;55:349–55.
25 Paradis M, González M, Juncos O, et al. Test da afasia nos bilingües, Galician
Version. Hillsdale, NJ: Lawrence Erlbaum Associates, 1989.
26 Paradis M, Elias J. Test de afasia para bilingües, European Spanish Version.
Hillsdale, NJ: Lawrence Erlbaum Associates, 1989.
27 Paradis M. Neurolinguistic aspects of implicit and explicit memory: implications
for bilingualism and second language acquisition. In: Ellis NC, eds. Implicit and
explicit learning of languages. London: Academic Press, 1994:393–419.
28 Zanini S, Tavano A, Vorano L, et al. Greater syntactic impairments in native
language in bilingual Parkinsonian patients. J Neurol Neurosurg Psychiatry
2004;75:1678–81.
29 Sousa X. Galician. Introductory page. http://www.usc.es/,ilgas/galicianhtml,
1995.
30 Castro-Caldas A, Petersson KM, Reis A, et al. The illiterate brain. Learning to
read and write during childhood influences the functional organization of the
adult brain. Brain 1998;121:1053–63.
31 Ratcliff G, Ganguli M, Chandra V, et al. Effects of literacy and education on
measures of word fluency. Brain Lang 1998;61:115–22.
32 Ostrosky F, Ardila A, Rosselli M, et al. Neuropsychological test performance in
illiterates. Arch Clin Neuropsychol 1998;13:645–60.

www.jnnp.com