Available online at www.sciencedirect.com

Brain and Language 105 (2008) 50–58
www.elsevier.com/locate/b&l

Speech and language functions that require a functioning Broca’s area
Cameron Davis a, Jonathan T. Kleinman a, Melissa Newhart a, Leila Gingis a,
Mikolaj Pawlak d, Argye E. Hillis a,b,c,*
a

Johns Hopkins University School of Medicine, Department of Neurology, 600 North Wolfe Street, Baltimore, MD 21287, USA
b
Johns Hopkins University School of Medicine, Department of Physical Medicine and Rehabilitation, MD, USA
c
Johns Hopkins University, Department of Cognitive Science, MD, USA
d
University of Pennsylvania, Department of Radiology, USA
Accepted 19 January 2008
Available online 5 March 2008

Abstract
A number of previous studies have indicated that Broca’s area has an important role in understanding and producing syntactically
complex sentences and other language functions. If Broca’s area is critical for these functions, then either infarction of Broca’s area or
temporary hypoperfusion within this region should cause impairment of these functions, at least while the neural tissue is dysfunctional.
The opportunity to identify the language functions that depend on Broca’s area in a particular individual was provided by a patient with
hyperacute stroke who showed selective hypoperfusion, with minimal infarct, in Broca’s area, and acutely impaired production of grammatical sentences, comprehension of semantically reversible (but not non-reversible) sentences, spelling, and motor planning of speech
articulation. When blood flow was restored to Broca’s area, as demonstrated by repeat perfusion weighted imaging, he showed immediate recovery of these language functions. The identification of language functions that were impaired when Broca’s area was dysfunctional (due to low blood flow) and recovered when Broca’s area was functional again, provides evidence for the critical role of Broca’s
area in these language functions, at least in this individual.
Ó 2008 Published by Elsevier Inc.
Keywords: Aphasia; Broca’s area; Magnetic resonance perfusion imaging; Acute ischemic stroke

1. Introduction
Studies of chronic aphasia have indicated an essential
role of Broca’s area in various speech and language functions. The definition of ‘‘Broca’s area” is controversial,
and the precise neuroanatomical boundaries vary across
individuals, whether the boundaries are defined by the
cytoarchitecture or borders of sulci and gyri (Amunts
et al., 1999; Rademacher, Caviness, Steinmetz, & Galaburda, 1993; Van Essen, Drury, Joshi, & Miller, 1998; Whitaker & Selnes, 1976). Nevertheless, Broca’s area is
sometimes defined as Brodmann’s area 44 and 45—areas
*
Corresponding author. Address: Johns Hopkins University School of
Medicine, Johns Hopkins Hospital, Department of Neurology, 600 North
Wolfe Street, Baltimore, MD 21287, USA. Fax: +1 410 614 9807.
E-mail address: argye@JHMI.edu (A.E. Hillis).

0093-934X/$ - see front matter Ó 2008 Published by Elsevier Inc.
doi:10.1016/j.bandl.2008.01.012

of cytoarchitecture generally found in the left posterior
inferior frontal cortex. Some investigators claim that Broca’s area is essential for creating and maintaining traces
or chains of movement in movement-derived sentences
(e.g., passive and object relative clause sentences; Grodzinsky, 2000). Others argue that this area is more generally
required for thematic role assignment or for verbal (semantic or phonological) working memory. Evidence for these
hypotheses primarily comes from patients with chronic
Broca’s aphasia, who are assumed to have lesions involving
Broca’s area. For example, many studies have reported
that individuals clinically diagnosed with Broca’s aphasia
understand simple active sentences, but fail to understand
semantically reversible passive sentences (Friedmann &
Shapiro, 2003; Grodzinsky, 2000; Schwartz, Saffran, &
Marin, 1980). Some of these studies have found that they
also understand subject relative clause sentences and sub-

C. Davis et al. / Brain and Language 105 (2008) 50–58

ject clefts, but fail to understand object relative clause sentences and object clefts (Caramazza & Zurif, 1976; Friedmann & Shapiro, 2003; Grodzinsky, 1989). The latter
observation has led Grodzinksky and colleagues to claim
that Broca’s area is specifically involved in creating and
maintaining traces of movement. However, the ‘‘trace deletion hypothesis” has also been criticized, leading to a lively
discussion on the role of Broca’s area in sentence comprehension (see Beretta, 2000; Berndt, Mitchum, & Haendiges,
1996; Caplan, Alpert, Waters, & Oliveri, 2000; Caplan, Hildebrandt, & Makris, 1996; Dick & Bates, 2000; Dickey,
Choy, & Thompson, 2007; Drai & Grodzinsky, 2006; Drai,
Grodzinsky, & Zurif, 2001; Hickok, 2000; Hickok & Avrutin, 1995; Kolk & Hartsuiker, 2000; Muller & Knight,
2006).
Other lesion/deficit association studies have implicated
Broca’s area in agrammatic speech (Caramazza & Hillis,
1990; Kolk & Friedericim, 1985; Miceli, Silveri, Romani,
& Caramazza, 1983), verb production (Hillis, Tuffiash, &
Caramazza, 2002; Hillis, Wityk, Baker, Ulatowski, &
Jacobs, 2003; Zingeser & Berndt, 1990), complex verb
argument structure (Shapiro, Gordon, Hack, & Killackey,
2002; Thompson, Lange, Schneider, & Shapiro, 1997), or
spelling (Hillis, Chang, & Breese, 2004; Hillis, Kane, Tuffiash et al., 2002). However, the evidence for some of these
associations comes from patients selected for having ‘‘Broca’s aphasia” (a controversial classification in itself, e.g.,
Caramazza, Capitani, Rey, & Berndt, 2001; Miceli, Silveri,
Romani, & Caramazza, 1989), rather than patients selected
for lesions involving Broca’s area. There is not a one to one
correspondence between Broca’s aphasia and lesions in
Broca’s area. Some patients with damage to Broca’s area
do not have Broca’s aphasia, and some patients with Broca’s aphasia do not have lesions in Broca’s area (Bramwell,
1898; Dronkers, 2000; Dronkers, Shapiro, Redfern, &
Knight, 1992; Marie, 1906; Mohr, 1976; Willmes & Poeck,
1993). Furthermore, the majority of patients with chronic
Broca’s aphasia have large lesions involving additional
areas, rather than damage specifically to Broca’s area.
Therefore, it is difficult to determine whether deficits are
due to Broca’s area or surrounding areas.
If one were interested in identifying the language functions that critically depend on Broca’s area, it would seem
more straightforward to test language in patients with
damage to Broca’s area. However, many patients with
damage to Broca’s area (or any other perisylvian area)
have language deficits initially, but recover over time, often
after rehabilitation. Recovery may be due to reorganization; that is, other parts of the brain may assume functions
of the damaged part. In other cases, rehabilitation may
have provided alternative strategies or cognitive mechanisms for completing the task. In either case, some patients
with damage to Broca’s area might show normal performance, even on tasks that initially depended on Broca’s
area. To avoid these complications, one could test language
functions immediately after damage to Broca’s area, before
reorganization or recovery.

51

However, neuroimaging of patients with acute stroke
have demonstrated that surrounding acute infarction, there
is an area of dysfunctional but potentially viable ischemic
tissue that may be contributing to the observed deficits.
For example, early studies using intracarotid xenon-133
cerebral blood flow indicated that aphasia in patients with
subcortical lesions is generally due to hypoperfusion and
functional loss in cortical language zones (Olsen, Bruhn,
& Oberg, 1986). More recent studies using perfusion
weighted imaging (PWI) have confirmed that aphasia only
occurs after subcortical infarcts when there is concomitant
hypoperfusion (Hillis, Wityk et al., 2002) and that language
deficits in acute stroke are attributable to the entire region
of dysfunctional brain tissue – the infarct plus the hypoperfused tissue (Croquelois, Wintermark, Reichhart, Meuli, &
Bogousslavsky, 2003; Hillis, Barker et al., 2004; Hillis,
Wityk et al., 2001; Hillis, Work et al., 2004; Reineck, Agarwal, & Hillis, 2005).
It has been demonstrated that a variety of interventions
can restore blood flow to the hypoperfused tissue surrounding the infarct, with concomitant improvement in
behavioral functions (Croquelois et al., 2003; Hillis, Barker
et al., 2001; Hillis, Kane, Tuffiash, Ulatowski et al., 2002;
Hillis, Ulatowski et al., 2003). For example, in a study of
44 patients with acute subcortical stroke, the patients
who had aphasia all showed concurrent cortical hypoperfusion (Hillis, Wityk et al., 2002). Among the patients in this
study who received intervention that successfully restored
cortical perfusion, six out of six showed immediate resolution of aphasia. Interventions to restore perfusion have
included intravenous and intra-arterial thrombolysis, stenting, urgent endarterectomy and other vascular surgical
procedures (e.g., Wityk, Hillis, Beauchamp, Barker, & Rigamonti, 2002), and induced blood pressure elevation (Hillis, Ulatowski et al., 2003; Rordorf, Koroshetz,
Ezzeddine, Segal, & Buonanno, 2001). The last intervention is based on animal studies showing that in hypoperfused tissue, there is a loss of autoregulation, such that
increased blood pressure results in a linear increase in
regional cerebral blood flow (Astrup, Symon, Branston,
& Lassen, 1977).
Functional imaging studies have also identified activation in Broca’s area (or parts of Broca’s area) associated
with a wide variety of language tasks and nonlinguistic
or cognitive processes, not all of which can be reviewed
in a case report. Some of these cognitive processes may
be critical to the language functions that are impaired
in Broca’s aphasia. For example, a reliable finding from
fMRI is activation in Broca’s area and surrounding areas
during tasks that demand phonological working memory
(Fiez et al., 1996; Paulesu, Frith, & Frakowiak, 1993). A
number of authors have proposed that phonological
working memory deficits may underlie agrammatic
speech and/or asyntactic comprehension that are frequently observed in Broca’s aphasia (Stowe, 2000; Vallar
& Baddeley, 1984). Other studies have shown activation
in Broca’s area associated with a variety of semantic

52

C. Davis et al. / Brain and Language 105 (2008) 50–58

tasks (Bookheimer, 2002) or selection of a response from
semantic memory (Thompson-Schill, Aguirre, D’Esposito, & Farah, 1999; Thompson-Schill & Gabrieli, 1999;
Thompson-Schill, Kurtz, & Gabrieli, 1998). Some studies
also show activation of Broca’s area during verb production (Tranel, Grabowski, Lyon, & Damasio, 2005) or
spelling (Beeson et al., 2003; Rapp & Hseih, 2001). Other
functional imaging studies show that different subregions
of Broca’s area are responsible for different deficits. For
example, a number of studies converge on the hypothesis
that anterior Broca’s area is critical for semantic tasks
(Gitelman, Nobre, Sonty, Parrish, & Mesulam, 2005),
while posterior Broca’s area is critical for word recall
and repetition (Buckner et al., 1995) or processing of letters or syllables (Booth et al., 2002; Poldrack et al.,
1999).
Many other functions have been found to be associated
with activation in Broca’s area, with or without activation
in its right hemisphere homologue. However, functional
imaging studies can only show that Broca’s area is engaged
in a process, not that it is essential for the process (Chatterjee, 2005; Fellows et al., 2005). Ideally, functional imaging
studies and lesion studies should be used together to identify areas that are necessary and sufficient for a particular
cognitive function (Chatterjee, 2005; Muller & Knight,
2006; Price, Mummery, Moore, Frakowiak, & Friston,
1999). However, this recommendation has rarely been
practiced.
One type of supporting evidence that an area is essential to a particular function in an individual is provided
when a particular language function is impaired when
the area is dysfunctional and recovers when that area
becomes functional again. This sort of evidence can be
obtained with temporary ‘‘lesions” (e.g., transcranial
magnetic stimulation, direct cortical stimulation, or transient ischemia). To illustrate the last method, we previously reported that accessing orthographic word forms
of verbs (but not nouns) for written naming was
impaired in two patients with low blood flow in Broca’s
area, and recovered when blood flow was increased in
Broca’s area (in one case by carotid stenting, and in
one case by blood pressure elevation) (Hillis, Wityk
et al., 2003).
We hypothesized that other language functions, such as
understanding and producing syntactically or semantically
complex sentences are dependent on Broca’s area (perhaps as one cog in a complex neural system) and will
therefore be acutely compromised when the neural tissue
within Broca’s area is selectively hypoperfused. The following case report illustrates support for this hypothesis.
We studied a variety of language functions in a patient
with hyperacute stroke, who showed selective hypoperfusion, with minimal infarct, in Broca’s area. Language testing was repeated after Broca’s area was reperfused to
confirm that language functions dependent on Broca’s
area in this individual recovered with reperfusion of Broca’s area.

2. Case report
MJE, a 67 year old, right-handed man with atrial fibrillation developed sudden muteness at 4 AM, the day after a
surgical procedure. Due to recent surgery he was not an
appropriate candidate for intravenous thrombolysis.
Neurological examination took place 34 min after acute
onset. MJE was found to be severely aphasic. Initially, he
was mute, but within an hour he began to articulate words
and phrases with numerous and varied off-target attempts,
particularly with polysyllabic words (e.g., hospital-> [taspIdEl]. . .[hastIbEl]) despite normal strength, rate and range
of movements of the muscles of articulation (consistent
with apraxia of speech). He was alert and interactive, but
his speech production was limited to short phrases or single
words, and he made frequent paraphasic errors, mostly
semantic in nature.
Although MJE could accurately follow simple commands, he was unable to repeat sentences and had difficulty
naming objects. He described the ‘‘cookie theft” picture as
‘‘kis-kitchen. . .in kitchen. . .picture. The laundry running
over.. . .kid on the thing.” Although the patient’s sensation
was intact and motor strength was normal in both upper
and lower extremities, he could not form letters to write.
Immediate head CT was normal. On the same day, MRI
scans were obtained at four hours post-onset of symptoms.
MRI scans included sequences of diffusion weighted imaging (DWI), which is highly sensitive to infarct in the first
hours of stroke, and perfusion weighted imaging (PWI),
which shows areas of poor perfusion (low blood flow).
Areas of hypoperfusion were defined by a time-to-peak
(TTP) delay in arrival of contrast of >2.5–4.5 s compared
to the homologous voxels on the right. MRI showed a very
small (<3 mm) infarct in the left posterior frontal lobe, surrounded by a much larger perfusion deficit in a region that
approximates Broca’s area (an area in which we would predict the cytoarchitecture of Brodmann area (BA) 44 and 45
based on studies by Amunts et al., 1999 and others) (Figs. 1
and 2). The correspondence between his area of hypoperfusion and Broca’s area defined in this way was confirmed by
computational analysis of the percentage of voxel overlap
with the probabilistic map of Amunts et al. (1999). To
allow us to compute voxel overlap we registered the
patient’s diffusion and perfusion sequences to the MNI
atlas, using a mutual-information maximization algorithm
(Jenkinson & Smith, 2001) with the patient’s high resolution spoiled gradient echo (SPGR) sequence obtained at
the same time. Once we had registered the perfusion
(TTP) map to the atlas, we could automatically locate
the midsagittal plane, and thus could readily identify voxels
that showed diminished perfusion relative to their counterparts in the right hemisphere. We placed these registered
diffusion and perfusion maps and the Amunts et al.
(1999) probabilistic map (already registered to the MNI
atlas) into a brain-image database (BRAID; Herskovits,
2000) to calculate the percent overlap. Our patient’s area
of hypoperfusion overlapped with 85% of the voxels where

C. Davis et al. / Brain and Language 105 (2008) 50–58

53

Fig. 1. DWI and PWI before intervention (Day 1) and after intervention (Day 2). (A) DWI (upper panel) and PWI (lower panel) obtained at 4 h postonset of symptoms (pre-treatment). We show all slices that showed any lesion or hypoperfusion. PWI slices show substantial hypoperfusion (green areas)
within Broca’s area. (B) DWI and PWI obtained again at 36.5 h post-onset of symptoms (post-treatment). PWI scans were superimposed on the axial T1
weighted images, to show the sulci and gyri. These scans are in anatomical convention, with the left hemisphere on the left.

10/10 patients in the Amunts et al. study had cytoarchitecture of BA 45, and covered 47% of voxels where any patient
in that study had cytoarchitecture of BA 45 (Fig. 2b). The
area of cytoarchitecture of BA 44 in the Amunts et al.
study was more variable; there was no area where all 10
patients in their study had BA 44. Our patient’s area of
hypoperfusion overlapped with 67% of the voxels where
7/10 patients had cytoarchitecture of BA 44, and covered
51% of the voxels where any patient had BA 44 (see
Fig. 2a).
2.1. Pre-intervention language performance
Language tests were administrated at 5 h post-onset of
symptoms (pre-treatment), and again 32 h post-onset of
symptoms (post-treatment). The language battery
included: oral and written naming, oral reading, spelling

to dictation, and repetition, with 17 or 34 word stimuli
for each task, matched in word length and frequency across
tasks and versions given on separate days. Additionally, 25
pseudowords matched in length in syllables and phonemes
across sessions and across tasks were administered for oral
reading and oral spelling to dictation. Norms for these lexical and sublexical processing tests were obtained from 50
neurologically normal subjects with a mean age of
64.7 ± 10.7 years. The median score for normal controls
on all of the tests administered was 100% correct; no normal subject scored below 90%. Two film clip/sentence verification tests were administered; one with auditory and
one with written sentences. Each of 13 film clips depicted
a semantically reversible subject–verb–object relationship
(e.g., the man pushed the girl). Each film clip was presented
four times in counterbalanced order with: correct active
sentence; correct passive sentence; incorrect active sen-

54

C. Davis et al. / Brain and Language 105 (2008) 50–58

Fig. 2. Three dimensional reconstruction of the area of MJE’s hypoperfusion at Day 1. (A) Overlap between MJE’s area of hypoperfusion (shown in
purple) and the area where any patient had cytoarchitecture of BA 44 in the probabilistic map of Amunts et al. (1999; shown in yellow). (B) Overlap
between MJE’s area of hypoperfusion (shown in purple) and the area where any patient had cytoarchitecture of BA 45 in the probabilistic map of Amunts
et al. (1999; shown in orange).

tence; and incorrect passive sentence (semantically reversed
forms), intermixed with other sentence/video pairs. Both
auditory and written verification tasks required working
memory, but sentence production was not required. Subjects were not required to produce a sentence; they were

asked to simply answer yes/no as to whether or not the sentence described the videoclip.
MJE scored normally on oral reading of words and
pseudowords (57/59 or 96.6% successful trials; normal
mean = 96.9 ± 5.3) and repetition of words and pseudo-

C. Davis et al. / Brain and Language 105 (2008) 50–58

55

intervention (due to the risk of systemic bleeding). Instead,
he was treated with intravenous saline to increase his blood
pressure, to improve blood flow to the ischemic region. In
addition, because he was in atrial fibrillation (a basis for
cardioembolic source of stroke), he was anticoagulated
with heparin to prevent recurrent stroke. Although there
is some risk of hemorrhagic conversion of infarct with heparin, particularly after cardioembolic stroke, this risk was
considered minimal in view of the tiny size of the infarct.

words (57/59 or 96.6% correct; normal mean = 99.0 ± 2.6),
when his self-corrected articulatory (apractic) errors were
not penalized. He was 100% correct in answering five simple yes/no questions (e.g., Do dogs fly?) that were semantically non-reversible. However, he performed poorly on
other language tests (Fig. 3). Excluding self-corrected articulatory errors, he had mild impairment of oral naming (15/
17; 88.2% correct; normal mean 99.2 ± 2.1). His errors
were semantic paraphasias (e.g., bee named as ‘‘mosquito”). He was unable to form letters for written spelling
to dictation, and had a moderate impairment in oral spelling to dictation (44/58 or 75.9% correct; normal
mean = 98.7% ± 3.3). His spelling errors were phonologically implausible nonwords, with deletions, substitutions,
and transpositions of letters. On the remaining tasks his
performance was not significantly different from chance.
He scored 20% correct (1/5) in answering semantically
reversible yes/no questions (e.g., Is a dog larger than an elephant?). On auditory sentence/film clip verification with
semantically reversible sentences, he had 16/26 successful
trials (61.5% correct) for passive sentences and 17/26
(65.4% correct) for active sentences. On the written form
of this verification test, he was also at chance level: 11/26
or 42.3% correct for passive and 14/26 or 53.8% correct
for active sentences.

Repeat MRI scans, including DWI and PWI sequences,
were obtained 36.5 h post-onset of symptoms. DWI
sequences showed a few punctuate bright areas, consistent
with acute infarct, within Broca’s area. PWI showed complete reperfusion of Broca’s area (Fig. 1). The cause of
reperfusion of the area is unknown; reperfusion may have
been due to either spontaneous recanalization of the vessel
supplying this territory or intervention resulting in reperfusion through collateral flow. The cause of the reperfusion is
irrelevant to the current study, the purpose of which was to
determine what language functions were impaired when
Broca’s area was hypoperfused and recovered when the
area reperfused.

2.2. Intervention

2.4. Post-intervention language testing

Because his cortical infarct was so tiny, it was our clinical impression that it was very unlikely to be the cause of
his deficits. Rather, it seemed clear that MJE’s significant
language deficits were due to hypoperfusion of Broca’s
area, and that the only way to restore function was to
restore blood flow to this area of cortex. Although he presented within the 3 h window for intravenous thrombolysis,
his very recent surgery made him a poor candidate for this

Language tests were re-administrated at 32 h post-onset,
after he had undergone treatment designed to improve
blood flow to the hypoperfused region of the brain. He
scored at ceiling (or at normal or above average levels)
on all language tasks (Fig. 3). MJE’s spontaneous speech
also returned to normal. He produced fluent, grammatical
sentences in describing the ‘‘cookie theft picture”. He
scored 100% (17/17 successful trials) on oral and written

2.3. Evidence of reperfusion

Fig. 3. MJE’s performance on language tests administrated on Day 1 (5 h post-onset of symptoms; pre-treatment), and again on Day 2 (32 h post-onset of
symptoms; post-treatment).

56

C. Davis et al. / Brain and Language 105 (2008) 50–58

naming, and at normal levels on repetition (58/59 or 98.3%
correct; normal mean = 99.0 ± 2.6), oral and written spelling to dictation (both 56/59 or 94.9% correct; normal
mean = 98.7% ± 3.3). He scored 100% correct on auditory
and written sentence/film clip verification for passive and
active sentences, and 10 yes/no questions (semantically
reversible and non-reversible).
3. Discussion
This case confirms the hypotheses that adequate function of Broca’s area was necessary for a number of language comprehension and production tasks for this
individual. Results suggest that Broca’s area played an
essential role in his comprehension and production of
semantically reversible sentences, spelling, and motor planning/programming of speech articulation. Acute hypoperfusion of Broca’s area was associated with severe
impairment in comprehension of active and passive semantically reversible sentences, and moderate impairment of
motor speech, grammatical sentence production, and spelling, without impairment of comprehension of semantically
non-reversible sentences. Restoration of normal blood flow
to Broca’s area resulted in immediate recovery of these
functions. These results provided evidence that deficits in
these language functions can be due to hypoperfusion of
Broca’s area in at least some patients and that reperfusion
can restore these functions in the same individuals. These
functions might have relied on Broca’s area or on areas
functionally connected to Broca’s area that were compromised when Broca’s area was dysfunctional.
When Broca’s area was hypoperfused, MJE’s performance was no better than chance in understanding semantically reversible questions and sentences, but was above
chance in understanding non-reversible sentences. Because
reperfusion was achieved very quickly, we did not administer enough tasks to determine if his sentence comprehension impairment was due to a general problem with
thematic role assignment, working memory, semantic comparison, or more generally due to impaired selection of a
response from semantic memory.
MJE was also impaired in motor planning and/or programming of speech articulation, manifest as numerous
and varied off-target attempts at articulating polysyllabic
words. This finding is consistent with studies indicating
that acute lesions in Broca’s area are associated with
apraxia of speech (Hillis, Work et al., 2004; Mohr, 1976).
Studies of chronic stroke have shown variable areas of
overlap (including the anterior insula; Dronkers et al.,
1992) in patients with chronic apraxia of speech. However,
this result may reflect the fact that patients with chronic
speech deficits have large infarcts, which often include the
insula.
Finally, MJE was profoundly impaired in writing, and
moderately impaired in oral spelling when Broca’s area
was hypoperfused. His deficit in oral spelling was not due
to his motor speech impairment, since his rare errors in

articulating names of letters were not penalized. Rather,
he seemed impaired in accessing orthographic representations for output, or in maintaining the representations in
the graphemic buffer while motor plans for writing or oral
spelling were articulated. Again, there was not sufficient
time to tease apart these deficits. Of note, though, his errors
in spelling were not phonologically plausible, indicating
that he was not able to rely on sublexical phonology-toorthography conversion to spell. His impairment in forming letters may have been due to impaired retrieval of letter-shape specific motor plans or converting from
graphemes (abstract letter identities) to letter shapes, with
a particular case and font (see Rapcsak & Beeson, 2004;
Rapp & Hseih, 2001). A similar deficit was previously
reported in patients who all had lesions or hypoperfusion
involving parts of Broca’s area and Brodmann’s area 6
(Hillis, Chang et al., 2004; Hillis, Rapp, & Caramazza,
1999). It became clear the day following treatment that
his deficits in spelling and writing were not premorbid,
since he recovered these functions with reperfusion.
In summary, this case provided the unique opportunity
to identify the speech and language functions that were
compromised when neural tissue in Broca’s area was dysfunctional in this individual, before substantial reorganization or recovery. It is possible that MJE’s transient
impairments had a single underlying cause. Each of the
impaired tasks (comprehending semantically reversible sentences, grammatical sentence production, spelling, and
motor planning/programming of speech articulation)
requires maintaining the sequential order of elements for
output or for further processing. One can speculate that
Broca’s area was essential for maintenance of the serial
order of phonemes, graphemes, and other linguistic and
motor elements (see Knutson, Wood, & Grafman, 2004
for evidence consistent with this hypothesis). Alternatively,
Broca’s area may include functionally distinct cortical
areas with different connections as well as specializations
(see Grodzinsky & Amunts, 2006 for extensive discussion;
see also Hillis, 2006).
MJE’s transient deficits occurred when there was temporary hypoperfusion of an area of left posterior inferior
frontal cortex, often described as Broca’s area. Although
we do not know the cytoarchitecture of the hypoperfused
region, it generally corresponds to an area within which
10 individuals studied at autopsy had cytoarchitecture of
Brodmann’s areas 44 and 45 (Amunts et al., 1999). A single
case cannot provide evidence for a hypothesis about structure–function relationships, since there is almost certainly
some individual variability in such relationships, at least
within heteromodal association cortex (or all areas other
than ‘‘primary” or unimodal cortex, such as striate cortex
and Heschl’s gyrus). Nevertheless, a series of studies of
additional patients with focal areas of tissue dysfunction
caused by ischemia or transcranial magnetic stimulation
of parts of Broca’s area would complement functional
imaging studies of normal subjects in sorting out the various roles of Broca’s area in language comprehension and

C. Davis et al. / Brain and Language 105 (2008) 50–58

production across individuals. Moreover, this case provides clear evidence of rapid recovery of language when
blood flow is restored to areas critical for language.
Acknowledgments
The research reported in this paper was supported by
NIH (NIDCD), through RO1 DC 05375. We gratefully
acknowledge this support and the cheerful participation
of MJE.
References
Amunts, K., Schleicher, A., Burgel, U., Mohlberg, H., Uylings, H. B. M.,
& Zilles, K. (1999). Broca’s region revisited: Cytoarchitecture and
intersubject variability. Journal of Comparative Neurology, 412,
319–941.
Astrup, J., Symon, L., Branston, N. M., & Lassen, N. A. (1977). Cortical
evoked potential and extracellular K+ and H+ at critical levels of
brain ischemia. Stroke, 8, 51–57.
Beeson, P. M., Rapcsak, S. Z., Plante, E., Chargualaf, J., Chung, A.,
Johnson, S. C., & Trouard, T. P. (2003). The neural substrates of
writing: A functional magnetic resonance imaging study. Aphasiology,
17, 647–665.
Beretta, A. (2000). Why the TDH fails to contribute to a neurology of
syntax. Behavioral and Brain Sciences, 1, 23.
Berndt, R. S., Mitchum, C. C., & Haendiges, A. N. (1996). Comprehension of reversible sentences in ‘‘agrammatism”: A meta-analysis.
Cognition, 58, 289–308.
Bookheimer, S. (2002). Functional MRI of language: New approaches to
understanding the cortical organization of semantic processing. Annual
Review of Neuroscience, 25, 151–188.
Booth, J. R., Burman, D. D., Meyer, J. R., Gitelman, D. R., Parrish, T.
B., & Mesulam, M. M. (2002). Modality independence of word
comprehension. Human Brain Mapping, 16, 251–261.
Bramwell, B. (1898). A remarkable case of aphasia. Brain, 21, 343–373.
Buckner, R. L., Petersen, S. E., Ojemann, J. G., Miezin, F. M., Squire, L.
R., & Raichle, M. E. (1995). Functional anatomical studies of explicit
and implicit memory retrieval tasks. Journal of Neuroscience, 15,
12–29.
Caplan, D., Hildebrandt, N., & Makris, N. (1996). Location of lesions in
stroke patients with deficits in syntactic processing in sentence
comprehension. Brain, 119, 933–949.
Caplan, D., Alpert, N., Waters, G. S., & Oliveri, A. (2000). Activation of
Broca’s area by syntactic processing under conditions of concurrent
articulation. Human Brain Mapping, 9, 65–71.
Caramazza, A., & Zurif, E. B. (1976). Dissociation of algorithmic and
heuristic processes in language comprehension: Evidence from aphasia. Brain and Language, 3, 572–582.
Caramazza, A., & Hillis, A. E. (1990). Where do semantic errors come
from?. Cortex, 26, 95–122.
Caramazza, A., Capitani, E., Rey, A., & Berndt, R. S. (2001). Agrammatic
Broca’s aphasia is not associated with a single pattern of comprehension performance. Brain and Language, 76, 158–184.
Chatterjee, A. (2005). A madness to the methods in cognitive neuroscience? Journal of Cognitive Neuroscience, 17, 847–849.
Croquelois, A., Wintermark, M., Reichhart, M., Meuli, R., & Bogousslavsky, J. (2003). Aphasia in hyperacute stroke: Language follows
brain penumbra dynamics. Annals of Neurology, 54, 321–329.
Dick, F., & Bates, E. (2000). Grodzinsky’s latest stance – or just how
specific are ‘‘lesion – specific” deficits. The Behavioral and Brain
Sciences, 23–29.
Dickey, M. W., Choy, J. J., & Thompson, C. K. (2007). Real-time
comprehension of wh-movement in aphasia: Evidence from eyetracking while listening. Brain and Language, 100, 1–22.

57

Drai, D., & Grodzinsky, Y. (2006). A new empirical angle on the
variability debate: Quantitative neurosyntactic analyses of a large data
set from Broca’s aphasia. Brain and Language, 96, 117–128.
Drai, D., Grodzinsky, Y., & Zurif, E. (2001). Broca’s aphasia is associated
with a single pattern of comprehension performance: A reply. Brain
and Language, 76, 185–192.
Dronkers, N. F. (2000). The gratuitous relationship between Broca’s
aphasia and Broca’s area. The Behavioral and Brain Sciences, 30–31.
Dronkers, N. F., Shapiro, J. K., Redfern, B., & Knight, R. T. (1992). The
role of Broca’s area in Broca’s aphasia. Journal of Clinical and
Experimental Neuropsychology, 14, 52–53.
Fellows, L. K., Heberlein, A. S., Morales, D. A., Shivde, G., Waller, S., &
Wu, D. H. (2005). Method matters: An empirical study of impact in
cognitive neuroscience. Journal of Cognitive Neuroscience, 17, 850–858.
Fiez, J. A., Raife, E. A., Balota, D. A., Schwartz, J. P., Raichle, M. E., &
Petersen, S. E. (1996). A positron emission tomography study of the
short-term maintenance of verbal information. Journal of Neuroscience, 16, 808–822.
Friedmann, N., & Shapiro, L. (2003). Agrammatic comprehension of
simple active sentences with moved constituents: Hebrew osv and ovs
structures. Journal of speech, Language, and Hearing Research, 45,
288–297.
Gitelman, D. R., Nobre, A. C., Sonty, S., Parrish, T. B., & Mesulam, M.
M. (2005). Language network specializations: An analysis with parallel
task design and functional magnetic resonance imaging. Neuroimage,
26, 975–985.
Grodzinsky, Y. (1989). Agrammatic comprehension of relative clauses.
Brain and Language, 37, 480–499.
Grodzinsky, Y. (2000). The neurology of syntax: Language use without
Broca’s area. The Behavioral and Brain Sciences, 23, 1–71.
Grodzinsky, Y., & Amunts, K. (2006). Broca’s region. NY, New York:
Oxford University Press.
Jenkinson, M., & Smith, S. M. (2001). A global optimisation method for
robust affine registration of brain images. Medical Image Analysis, 5,
143–156.
Herskovits, E. (2000). An architecture for a brain-image database.
Methods of Information in Medicine, 39, 291–297.
Hickok, G., & Avrutin, S. (1995). Representation, referentiality, and
processing in agrammatic comprehension: Two case studies. Brian and
Language, 50, 10–26.
Hickok, G. (2000). Functional anatomy of speech perception and speech
production: Psycholinguistic implications. Journal of Psycholinguistic
Research, 30, 225–235.
Hillis, A. E. (2006). Language and frontal cortex. In B. L. Miller & J.
Cummings (Eds.), The Human frontal lobes (2nd ed., pp. 306–316).
New York: The Guilford Press.
Hillis, A. E., Barker, P. B., Beauchamp, N. J., Winters, B. D., Mirski, M.,
& Wityk, R. J. (2001). Restoring blood pressure reperfused Wernike’s
area and improved language. Neurology, 56, 670–672.
Hillis, A. E., Barker, P. B., Wityk, R. J., Aldrich, E., Restrepo, L., Breese,
E. L., et al. (2004). Variability in subcortical aphasia is due to variable
sites of cortical hypoperfusion. Brain and Language, 89, 524–530.
Hillis, A. E., Chang, S., & Breese, E. (2004). The crucial role of posterior
frontal regions in modality specific components of the spelling process.
Neurocase, 10, 157–187.
Hillis, A. E., Kane, A., Tuffiash, E., Beauchamp, N., Barker, P. B., Jacobs,
M. A., et al. (2002). Neural substrates of the cognitive processes
underlying spelling: Evidence from MR diffusion and perfusion
imaging. Aphasiology, 16, 425–438.
Hillis, A. E., Kane, A., Tuffiash, E., Ulatowski, J. A., Barker, P.,
Beauchamp, N., et al. (2002). Reperfusion of specific brain regions by
raising blood pressure restores selective language functions in subacute
stroke. Brain and Language, 79, 495–510.
Hillis, A. E., Rapp, B. C., & Caramazza, A. (1999). When a rose is a rose
in speaking, but a tulip in writing. Cortex, 35, 337–356.
Hillis, A. E., Tuffiash, E., & Caramazza, A. (2002). Modality-specific
deterioration in naming verbs in nonfluent primary progressive
aphasia. Journal of Cognitive Neuroscience, 14, 1099–1108.

58

C. Davis et al. / Brain and Language 105 (2008) 50–58

Hillis, A. E., Wityk, R. J., Tuffiash, E., Beauchamp, N. J., Jacobs, M. A.,
Barker, P. B., et al. (2001). Hypoperfusion of Wernicke’s area predicts
severity of semantic deficit in acute stroke. Annals of Neurology, 50,
561–566.
Hillis, A. E., Wityk, R. J., Barker, P. B., Beauchamp, N. J., Gailloud, P.,
Murphy, K., et al. (2002). Subcortical aphasia and neglect in acute
stroke: The role of cortical hypoperfusion. Brain, 125, 1094–1104.
Hillis, A. E., Wityk, R. J., Baker, P. B., Ulatowski, J. A., & Jacobs, M. A.
(2003). Change in perfusion in acute nondominant hemisphere stroke
may better estimated by tests of hemispatial neglect than by the
NIHSS. Stroke, 34, 2392–2398.
Hillis, A. E., Ulatowski, J. A., Barker, P. B., Torbey, M., Ziai, W.,
Beauchamp, N., et al. (2003). A pilot randomized trial of induced
blood pressure elevation: Effects on function and focal perfusion in
acute and subacute stroke. Cerebrovascular Diseases, 16, 236–246.
Hillis, A. E., Work, M., Breese, E. L., Barker, P. B., Jacobs, M. A., &
Maurer, K. (2004). Re-examining the brain regions crucial for
orchestrating speech articulation. Brain, 127, 1479–1487.
Knutson, K. M., Wood, J. N., & Grafman, J. (2004). Brain activation in
processing temporal sequence: An fMRI study. Neuroimage, 23,
1299–1307.
Kolk, H. H., & Friedericim, A. D. (1985). Strategy and impairment in
sentence understanding by Broca’s and Wernicke’s aphasics. Cortex,
21, 47–67.
Kolk, H. H., & Hartsuiker, R. J. (2000). Agrammatic sentence processing:
Severity,complexity,andpriming.BehavioralandBrainSciences,23,39–40.
Marie, P. (1906). Revision de la question de l’aphasie: la troisième
circonvolution frontale gauche ne joue aucun rôle spécial dans la
fonction du langage. Semaine Médicale, 26, 241–247.
Miceli, G., Silveri, M. C., Romani, C., & Caramazza, A. (1983).
Contrasting cases of italian agrammatic aphasia without comprehension disorder. Brain and Language, 19, 65–97.
Miceli, G., Silveri, M. C., Romani, C., & Caramazza, A. (1989). Variation
in the pattern of omissions and substitutions of grammatical morphemes in the spontaneous speech of so-called agrammatic patients.
Brain and Language, 36, 447–492.
Mohr, J. P. (1976). Broca’s area and Broca’s aphasia. In H. Whitaker & H.
A. Whitaker (Eds.), Studies in neurolinguistics (pp. 201–235). New
York: Academic Press.
Muller, N. G., & Knight, R. T. (2006). The functional neuroanatomy of
working memory: Contributions of human brain lesion studies.
Neuroscience, 139, 51–58.
Olsen, T. S., Bruhn, P., & Oberg, R. G. (1986). Cortical hypoperfusion as
a possible cause of subcortical aphasia. Brain and Language, 393–410.
Paulesu, E., Frith, D. C., & Frakowiak, R. S. (1993). The neural correlates
of the verbal components of working memory. Nature, 362, 342–345.
Poldrack, R. A., Wagner, A. D., Prull, M. W., Desmond, J. E., Glover, G.
H., & Gabrieli, J. D. (1999). Functional specialization for sematic and
phonological processing in the left inferior prefrontal cortex. Neuroimage, 10, 15–35.
Price, C. J., Mummery, C. J., Moore, C. J., Frakowiak, R. S., & Friston,
K. J. (1999). Delineating necessary and sufficient neural systems with
functional imaging studies of neuropsychological patients. Journal of
Cognitive Neuroscience, 11, 371–382.
Rademacher, J., Caviness, V. S., Steinmetz, H., & Galaburda, A. M.
(1993). Topographical variation in the human primary cortices:

Implications for neuroimaging, brain mapping, and neurobiology.
Cerebral Cortex, 3, 313–329.
Rapp, B., & Hseih, L. (2001). Functional magnetic resonance imaging of
the cognitive components of the spelling process. Cognitive Neuroscience Society Meeting; San Francisco, Calif.
Rapcsak, S. Z., & Beeson, P. M. (2004). The role of left posterior inferior
temporal cortex in spelling. Neurology, 62, 2221–2229.
Reineck, L., Agarwal, S., & Hillis, A. E. (2005). The ‘‘diffusion-clinical
mismatch” predicts early language recovery in acute stroke. Neurology,
64, 828–833.
Rordorf, G., Koroshetz, W., Ezzeddine, M. A., Segal, A. Z., & Buonanno,
F. S. (2001). A pilot study of drug induced hypertension for treatment
of acute stroke. Neurology, 56, 1210–1213.
Schwartz, M., Saffran, E., & Marin, O. (1980). The word-order problem in
agrammatism: I. Comprehension. Brain and Language, 10, 249–262.
Shapiro, L. P., Gordon, B., Hack, N., & Killackey, J. (2002). Verbargument structure processing in complex sentences in Broca’s and
Wernicke’s aphasia. Brain and Language, 45, 423–447.
Stowe, L. A. (2000). Sentence comprehension and the left inferior frontal
gyrus: Storage, not computation. Behavioral and Brain Sciences, 23, 51.
Thompson, C. K., Lange, K., Schneider, S. L., & Shapiro, L. (1997).
Agrammatic and non-brain-damaged subjects’ verb and verb argument structure production. Aphasiology, 11, 473–790.
Thompson-Schill, S. L., Kurtz, K. J., & Gabrieli, J. D. E. (1998). Effects of
semantic and associative relatedness on automatic priming. Journal of
Memory and Language, 38, 440–458.
Thompson-Schill, S. L., & Gabrieli, J. D. E. (1999). Priming of visual and
functional knowledge on a semantic classification task. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 25,
41–53.
Thompson-Schill, S. L., Aguirre, G. K., D’Esposito, M., & Farah, M. J.
(1999). A neural basis for category and modality specificity of semantic
knowledge. Neuropsychologia, 37, 671–676.
Tranel, D., Grabowski, T. J., Lyon, J., & Damasio, H. (2005). Naming the
same entities from visual or from auditory stimulation engages similar
regions of left inferotemporal cortices. Journal of Cognitive Neuroscience, 17, 1293–1305.
Vallar, G., & Baddeley, A. D. (1984). Fractionation of working memory:
Neuropsychological evidence for a phonological short-term store.
Journal of Verbal Learning and Verbal Behavior, 23, 151–162.
Van Essen, D. C., Drury, H. A., Joshi, S., & Miller, M. (1998). Functional
and structural mapping of the human cerebral cortex: Solutions are in
the surfaces. Proceedings of the National Academy of Sciences of the
United States of America, 95, 788–795.
Whitaker, H. A., & Selnes, O. (1976). Anatomic variations in the
cortex: Individual differences and the problem of the localization of
language functions. Annals of the New York Academy of Sciences,
280, 844–848.
Willmes, K., & Poeck, K. (1993). To what extent can aphasic syndromes
be localized? Brain, 116, 1527–1540.
Wityk, R., Hillis, A. E., Beauchamp, N., Barker, P. B., & Rigamonti, D.
(2002). Perfusion weighted MRI in adult moyamoya syndrome:
Characteristic patterns and change after surgical intervention: Case
report. Neurosurgery, 51, 1499–1506.
Zingeser, L., & Berndt, R. S. (1990). Retrieval of nouns and verbs in
agrammatism and anomia. Brain and Language, 39, 14–32.