RESEARCH ARTICLE - IA de Investigación especializada en el MIT

ARTÍCULO DE INVESTIGACIÓN

Canonical Sentence Processing and the Inferior
Frontal Cortex: Is There a Connection?

un acceso abierto

diario

Nicholas Riccardi1

, Chris Rorden1,2

, Julius Fridriksson2,3

, and Rutvik H. Desai1,2

1Department of Psychology, University of South Carolina, Columbia, CAROLINA DEL SUR
2Institute for Mind and Brain, University of South Carolina, Columbia, CAROLINA DEL SUR
3Department of Communication Sciences and Disorders, University of South Carolina, Columbia, CAROLINA DEL SUR

Palabras clave: stroke, Broca’s area, sentence comprehension, anterior temporal lobe, lesion-symptom
mapping, conectividad

ABSTRACTO

The role of left inferior frontal cortex (LIFC) in canonical sentence comprehension is
controversial. Many studies have found involvement of LIFC in sentence production or complex
sentence comprehension, but negative or mixed results are often found in comprehension
of simple or canonical sentences. We used voxel-, region-, and connectivity-based lesion
symptom mapping ( VLSM, RLSM, CLSM) in left-hemisphere chronic stroke survivors to
investigate canonical sentence comprehension while controlling for lexical-semantic,
executive, and phonological processes. We investigated how damage and disrupted white
matter connectivity of LIFC and two other language-related regions, the left anterior temporal
lobe (LATL) and posterior temporal-inferior parietal area (LpT-iP), affected sentence
comprensión. VLSM and RLSM revealed that LIFC damage was not associated with
canonical sentence comprehension measured by a sensibility judgment task. LIFC damage was
associated instead with impairments in a lexical semantic similarity judgment task with high
semantic/executive demands. Damage to the LpT-iP, specifically posterior middle temporal
gyrus (pMTG), predicted worse sentence comprehension after controlling for visual lexical
access, semantic knowledge, and auditory-verbal short-term memory (STM), but not auditory
single-word comprehension, suggesting pMTG is vital for auditory language comprehension.
CLSM revealed that disruption of left-lateralized white-matter connections from LIFC to LATL
and LpT-iP was associated with worse sentence comprehension, controlling for performance
in tasks related to lexical access, auditory word comprehension, and auditory-verbal STM.
Sin embargo, the LIFC connections were accounted for by the lexical semantic similarity judgment
tarea, which had high semantic/executive demands. This suggests that LIFC connectivity is
relevant to canonical sentence comprehension when task-related semantic/executive demands
are high.

INTRODUCCIÓN

Comprehending a spoken sentence is a complex process that requires coordination of multiple
cognitive resources, such as phonological, executive, lexical, syntactic, and semantic opera-
ciones. Reflecting this complexity, studies demonstrate that numerous distributed brain areas
functionally contribute to sentence comprehension (Dronkers et al., 2004; Friederici, 2012;
Hagoort & Indefrey, 2014; Walenski et al., 2019). Of these areas, the contribution of the left
inferior frontal cortex (LIFC) in sentence comprehension remains particularly controversial.

Citación: Riccardi, NORTE., Rorden, C.,
Fridriksson, J., & Desai, R. h. (2022).
Canonical sentence processing and
the inferior frontal cortex: Is there
a connection? Neurobiology of
Idioma, 3(2), 318–344. https://doi.org
/10.1162/nol_a_00067

DOI:
https://doi.org/10.1162/nol_a_00067

Recibió: 1 Abril 2021
Aceptado: 21 Enero 2022

Conflicto de intereses: Los autores tienen
declaró que no hay intereses en competencia
existir.

Autor correspondiente:
Nicholas Riccardi
riccardn@email.sc.edu

Editor de manejo:
Liina Pylkkanen

Derechos de autor: © 2022
Instituto de Tecnología de Massachusetts
Publicado bajo Creative Commons
Atribución 4.0 Internacional
(CC POR 4.0) licencia

La prensa del MIT

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

While many studies have found involvement of LIFC in sentence production and in complex
sentence comprehension, negative or mixed results are often found in the comprehension of
simple or canonical sentences. A more complete understanding of the role of LIFC in canonical
sentence comprehension would help inform neuroanatomical models of language processing.

The LIFC has traditionally been associated with language production, but a growing num-
ber of studies report its involvement in comprehension (Desai & Riccardi, 2021; Fadiga et al.,
2009; Rogalsky & Hickok, 2011). In regards to sentence processing specifically, results of neu-
roimaging studies of the LIFC, here defined as Brodmann Areas (BA) 44 y 45, ha sido
inconsistent (see Kemmerer, 2021). A meta-analysis of 53 neuroimaging studies comparing
sentence listening or reading to control conditions found that the inferior frontal gyrus pars
opercularis (IFGoper) and pars triangularis (IFGtri) were only activated in 13 y 23 estudios,
respectivamente (Hagoort & Indefrey, 2014). En efecto, multiple neuroimaging studies have found
that reading or listening to sentences passively does not activate the LIFC compared to word
liza (Humphries et al., 2006; Mazoyer et al., 1993; Rogalsky & Hickok, 2009). Sin embargo, un
additional subanalysis within the Hagoort and Indefrey (2014) investigation revealed that
semantic or syntactic demands (such as violation detection) during sentential processing reli-
ably activated areas in the LIFC. Multiple neuroimaging studies have found LIFC activation for
simple phrases and canonical word orders during tasks such as semantic/syntactic violation
detection or meaningfulness judgment (Graessner et al., 2021; Schell et al., 2017; Zaccarella
& Friederici, 2015).

These findings suggest that LIFC may be involved in the comprehension of simple phrases
and canonical sentences, especially when there are task-related demands requiring attention
to meaning or form. This predicts that damage to the LIFC should be associated with
canonical/simple sentence comprehension impairments when measured by tasks that orient
attention to semantic/syntactic error detection or meaningfulness. Sin embargo, neuropsychological
evidence supporting this prediction is relatively scarce, as briefly reviewed below. While consid-
ering these studies it is important to note that, even if noncanonical sentence comprehension is
significantly more associated with LIFC disfunction than canonical comprehension, it does not
follow that the LIFC does not also functionally contribute to canonical comprehension.

Two studies of patients with gliomas in the LIFC (Kinno et al., 2009, 2014) found that, com-
pared to healthy controls, patients were impaired at two-argument active canonical sentence
comprehension as measured by a sentence-picture matching task. They used voxel-based
lesion-symptom mapping ( VLSM) to demonstrate that these sentence impairments were signif-
icantly associated with damage to portions of the LIFC. Wilson et al. (2016) found that, com-
pared to healthy controls, patients with primary progressive aphasia (PPA) were less accurate
and responded more slowly to canonical sentences, especially for longer sentences, as mea-
sured by a sentence-picture matching task. Voxel-based morphometry ( VBM) implicated LIFC
and surrounding frontal cortices in the accuracy impairments. Sin embargo, the VBM analysis of
comprehension accuracy in that study grouped canonical and noncanonical accuracies
together, and LIFC atrophy was significantly associated with slower response times for nonca-
nonical sentences specifically, controlling for canonical performance. In a VLSM study of
patients with left-hemisphere stroke, Magnusdottir et al. (2013) found that damage to white
matter underlying LIFC was associated with worse sentence-picture matching performance
for canonical sentences, in addition to damage to posterior temporal regions. Caramazza
et al. (2005) demonstrated that, in a group of patients with Broca’s aphasia and confirmed
damage to LIFC, 17 out of 38 patients were unable to perform significantly better than chance
accuracy on a sentence-picture matching task of active canonical reversible sentences. Cómo-
alguna vez, patient performance was highly variable, and the lack of an explicit anatomical lesion-

Neurobiology of Language

319

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

deficit association analysis makes it impossible to ascribe these impairments to LIFC damage
specifically instead of other areas commonly damaged in patients diagnosed with Broca’s apha-
sia (p.ej., insula, superior temporal gyrus (STG); Dronkers et al., 2007; Fridriksson et al., 2015).

Por otro lado, many studies have found no relationship between LIFC damage and
canonical sentence comprehension. In a seminal VLSM study, Dronkers et al. (2004) found
no relationship between damage to IFGoper or IFGtri and sentence-picture matching for
canonical sentences. Newhart et al. (2012) also found no association between LIFC damage
and canonical sentences using sentence-picture matching and sentence reenactment, con
LIFC damage being associated instead with auditory-verbal and working memory deficits as
measured by forward and backward digit span. Also using sentence-picture matching,
Rogalsky et al. (2018) found no association between LIFC damage and canonical sentence
comprensión, even in a subset of patients (norte = 11) with relatively focal lesions to LIFC.
En cambio, these studies find evidence either that damage to posterior temporal areas is related
to worse sentence comprehension (p.ej., Kristinsson et al., 2020; Thothathiri et al., 2012), o
that LIFC damage is only associated with impairments for complex syntactic structures. Taken
together, these studies suggest that the LIFC does not functionally contribute to canonical sen-
tence comprehension, with its role instead being dependent on either production demands or
syntactic complexity.

These mixed findings raise questions about the functional involvement of LIFC during com-
prehension of simple phrases and sentences. One consideration is the relative lack of task
variety used in patient studies for measuring canonical comprehension. A large majority of
those studies use sentence-picture matching. In that task, an incompatible picture trial for
canonical sentences is typically created either by switching the places of the subject and
object (for reversible sentences), or by replacing the subject or object with a different
entity/item than what is described in the sentence (for nonreversible). Sentence-picture match-
ing for simple sentences therefore focuses mainly on aspects of comprehension related to
thematic role assignment (p.ej., who is the agent and who is the patient) or single word under-
de pie, especially related to nouns (p.ej., whether the item/entity in the picture matches the
word in the sentence). Other aspects of sentence comprehension, such as those related to the
comprehension of verbs and their compatibility with the nouns in the sentence, are not nec-
essarily measured. Using sentence comprehension tasks with demands that differ from
sentence-picture matching may provide novel insights about the functional contribution (o
lack thereof ) of LIFC to canonical sentence comprehension.

Another consideration is that most patient studies of canonical sentence comprehension
have used traditional lesion overlap methods, such as VLSM, to examine the relationship
between behavior and brain damage. A limitation of these methods is that they can only detect
areas of overlapping necrosis/gliosis. Alternativamente, white matter damage/disconnection has
been demonstrated to have wide-ranging consequences outside of the necrotic area and
can lead to functional disruption of cortical regions that are spared by the lesion (Bonilha
et al., 2014; Bonilha & Fridriksson, 2009; Bonilha, Rorden, & Fridriksson, 2014; Catani
et al., 2012; Catani & ffytche, 2005; Fridriksson et al., 2007). VLSM also requires a number
of patients to have overlapping damage within a given area of interest in order to detect behav-
ioral associations, a limitation that may lead to false negatives or inconsistent results between
studies depending on sample size and etiology of the brain damage (p.ej., glioma, PPA, middle
cerebral artery stroke). Connectivity-based lesion-symptom mapping (CLSM; Gleichgerrcht
et al., 2017) can complement traditional VLSM methods by detecting effects of disrupted white
matter connectivity resulting from damage anywhere along white matter tracts that connect
two grey matter regions. This provides two advantages. Primero, it can detect effects associated

Neurobiology of Language

320

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

with disconnection between two grey matter areas even if those areas are spared by the lesion.
Segundo, because it is senstive to damage located anywhere along a given white matter tract
(even if the specific areas of damage do not overlap between patients), it does not necessarily
require the same amount of lesion overlap as VLSM. Using complementary connectivity-based
measures in addition to traditional VLSM may reveal effects associated with LIFC disconnec-
tion from the larger language comprehension network that cannot be detected using VLSM in
isolation.

En efecto, prior studies have raised the possibility that, despite negative VLSM findings
regarding the association between LIFC damage and sentence comprehension, structural or
functional disconnection of LIFC from the larger language network may have behavioral
consequences (den Ouden et al., 2019; Fridriksson et al., 2018; Lukic et al., 2021; Turken
& Dronkers, 2011). Por ejemplo, although Dronkers et al. (2004) did not find a relationship
between LIFC damage and sentence comprehension, a follow-up study in healthy participants
showed that portions of the LIFC display functional and structural connectivity to the posterior
temporal area that was associated with sentence comprehension in the original lesion study
(Turken & Dronkers, 2011). The current investigation seeks to expand upon that finding by
using CLSM to explicitly examine the behavioral associates of structural disconnection of LIFC.
Además, Fridriksson et al. (2018) found that, while damage to LIFC was not associated
with sentence comprehension impairments (canonical and noncanonical included together),
white matter disconnection between the IFGoper and IFGtri did predict sentence comprehen-
sion impairments. Den Ouden et al. (2019) found similar results, but specifically for more
complex syntactic structures. The current study uses similar methods, but focuses specifically
on canonical sentences, to provide additional information about the role of LIFC in sentence
comprensión.

Aquí, in a retrospective study using data collected as part of a previously existing language
and cognition task battery, we used VLSM, CLSM, and region-based lesion symptom mapping
(RLSM) in a group of unilateral left-hemisphere chronic stroke survivors to investigate compre-
hension of canonical sentences while controlling for related cognitive abilities (p.ej., lexical
semantics, auditory single-word comprehension, auditory-verbal short-term memory (STM)).
Our focus was to investigate how damage and disrupted white matter connectivity of the LIFC
and areas within two other language-related regions, the left anterior temporal lobe (LATL) y
left posterior temporal-inferior parietal area (LpT-iP), affected canonical sentence comprehen-
sion and related processes. We used an auditory sentence sensibility task on declarative sen-
tences which required participants to determine if a sentence made sense, as opposed to the
commonly used sentence-picture matching task. We included four control tasks with varying
lexical-semantic, executive, and phonological demands to interrogate the LIFC’s involvement
in these processes (for a full description of the demands of each task, see Materials and Pro-
cedure). A visual lexical decision task with low semantic, executive, and phonological
demands was used to control for lexical access. Auditory word comprehension, which has
relatively higher semantic, executive, and phonological demands, was used to control for
auditory single-word comprehension. A visual semantic similarity judgment task, which had
high semantic and executive demands, but low phonological demands, was used to control for
semantic retrieval. The forward digit span task was used to control for auditory-verbal STM.

Several hypotheses regarding the role of LIFC can be tested given this retrospective lan-
guage and cognition battery. Primero, the merge hypothesis states that the LIFC is involved in
the binding of two or more elements into a hierarchical structure (Zaccarella & Friederici,
2015; Zaccarella et al., 2017), starting with smaller units (p.ej., the boy) and expanding into
larger structures as the phrase or sentence continues (p.ej., the boy kicks). This process would

Neurobiology of Language

321

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

be important for successful completion of the sentence sensibility task but less so for the other
tasks included here. Merge would be important for the sentence task because to judge the
sensibility of a sentence, a hierarchical structure must be built such that the meaningfulness
of the semantic and grammatical relations among the subject, verb, and object can be
accepted or rejected. Two of the other tasks (visual lexical decision, visual semantic similarity
judgment) have the merge operation at the most basic level (p.ej., nouns and verbs are pre-
ceded by the and to, respectivamente), but they only contain this basic two-word structure as
opposed to phrase-level combinations in the sentence sensibility task. If LIFC disruption is
associated with worse sentence comprehension, but not with the other tasks, it could be taken
as evidence that the LIFC functionally contributes to merge or related structure-building pro-
cesses at the phrase or sentence level.

A second hypothesis is that the LIFC contributes to sentence comprehension via executive
task-related demands, especially pertaining to detection of syntactic or semantic violations.
Evidence for this account comes from previously discussed neuroimaging findings suggesting
that LIFC is especially active in sentence comprehension tasks that require special attention to
syntactic or semantic information (Hagoort & Indefrey, 2014; Hasson et al., 2006; Love et al.,
2006; Rogalsky & Hickok, 2009). Because the current sentence task explicitly orients attention
to the semantic meaningfulness of the sentences, support for this hypothesis would come from
LIFC being a shared neural substrate for the sentence comprehension task and the semantic
similarity judgment, as that task also requires explicit semantic analysis.

Tercero, the LIFC may contribute to sentence comprehension through auditory-verbal STM.
Evidence for this hypothesis comes from neuroimaging studies showing that some areas in the
LIFC are activated by both syntactic demands and STM (Matchin et al., 2017; Rogalsky &
Hickok, 2011; Rogalsky et al., 2008). Neuropsychological studies also find that LIFC damage
is correlated with reduced digit span and impaired comprehension of sentences (Pettigrew &
Hillis, 2014). To successfully complete the auditory sentence sensibility task, Participantes
must hold the subject, verb, and object in their STM long enough to make the meaningfulness
judgment. En el estudio actual, finding that forward digit span and sentence comprehension
share the LIFC as a common neural substrate would support the auditory-verbal STM
hypothesis.

Finalmente, the LIFC may be involved in a variety of operations that subserve general lexical-
semantic processing (Fiebach et al., 2002; Heim et al., 2009; Kotz et al., 2002, 2010; Fallar
et al., 2008), which may contribute to sentence comprehension. The current sentence sensi-
bility task requires participants to understand individual words, as well as conceptual relation-
ships between words, to judge meaningfulness. Finding that LIFC is a common neural substrate
for sentence comprehension and the tasks requiring lexical-semantic search and access (es decir.,
auditory word comprehension, visual lexical decision, and visual semantic similarity), would
support this hypothesis.

MATERIALES Y MÉTODOS

Participantes

Seventy-five (24 femenino) native English-speaking participants with unilateral left-hemisphere
stroke were recruited. Fourteen participants did not complete the sentence sensibility task
due to time constraints on the day of testing or technological malfunction, leaving 61 (17 femenino)
participants for the neuroanatomical analyses (ver tabla 1 for demographic information). Past
power analysis has demonstrated that samples of ∼50 or greater provide adequate power to

Neurobiology of Language

322

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Mesa 1.

Participant demographic information

Participant
M2025

Age
56

Gender
METRO

Educación
(años)
16

Dominant hand
prior to stroke
bien

Aphasia type
Broca’s

WAB: AQ
64.6

M2071

M2005

M2036

M2006

M2002

M2069

M2007

M2020

M2141

M2142

M2046

M2014

M2074

M2061

M2078

M2143

M2072

M2144

M2145

M2146

M2075

M2059

M2040

M2076

M2147

M2082

M2079

M2149

M2151

M2152

METRO

bien

izquierda

bien

izquierda

bien

izquierda

bien

Broca’s

Anomic

Broca’s

Anomic

Broca’s

Anomic

None

Conduction

Anomic

Broca’s

Wernicke’s

Conduction

None

Global

None

Anomic

None

Anomic

Broca’s

Conduction

None

Anomic

Global

None

Anomic

63.6

55.2

76.2

83.2

80.4

90.5

26.1

86.2

99.1

99.2

51.5

59.4

52.7

88.8

99.2

23.6

99.6

93.2

98.5

94.2

58.2

57.5

72.1

97.3

91.1

25.3

98.6

96.9

93.1

Neurobiology of Language

323

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Participant
M2153

Age
48

Gender
METRO

Mesa 1.

(continued )

Educación
(años)
14

Dominant hand
prior to stroke
bien

M2086

M4138

M2088

M2044

M2087

M2155

M2030

M2156

M2158

M2159

M2160

M2094

M2162

M2031

M2103

M4180

M2164

M2106

M2110

M2109

M2114

M2111

M2029

M4209

M2119

M2121

M2117

M2127

METRO

bien

izquierda

bien

izquierda

bien

Aphasia type
Anomic

Conduction

Anomic

Broca’s

None

Broca’s

None

Broca’s

None

Wernicke’s

Conduction

None

Anomic

Wernicke’s

Global

Anomic

Broca’s

Conduction

None

Broca’s

Transcortical
sensorial

WAB: AQ
87.6

73.5

77.8

87.5

74.8

48.9

98.9

57.2

96.7

99.6

99.1

96.6

64.6

98.4

31.2

82.9

45.2

90.1

91.3

47.8

15.2

93.4

74.6

97.5

98.1

49.1

57.8

Neurobiology of Language

324

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

detect medium to strong effects in the majority of brain areas (Kimberg et al., 2007). Participantes
were at least 6 months post-stroke (4.02 years ± 4.4), and a mean age at time of testing of 58.75
years ± 9.68. The Western Aphasia Battery ( WAB; Kertesz, 2007) Aphasia Quotient mean score
era 75.7 ± 23.6. All participants signed informed consent, and the University of South Caro-
lina Institutional Review Board approved the research.

Materials and Procedure

Auditory sentence sensibility

One hundred canonical declarative and 50 low-meaningfulness or nonsense sentences were
presented auditorily to the participants. Sentences were sound files recorded in a noise atten-
uated room by a male speaker who was instructed to read as clearly and naturally as possible,
as if they were reading aloud to someone else. Sentences were denoised using Audacity audio
editing software (https://www.audacityteam.org/) and ranged from 2–3.5 s in length. To pre-
serve a more naturalistic sound for the stimuli, speech rate was not artificially modified
post-recording. In a sound attenuated room, sentences were played aloud to participants on
a laptop PC running E-prime software (versión 1.2, Psychology Software Tools, Cª; https://
pstnet.com/products/e-prime/). Ten practice sentences were presented before the real trials
to ensure that the participant could hear the sentences and understood the task. The partici-
pant was instructed to determine as quickly and accurately as possible whether the sentence
made sense or not by pressing one of two response buttons. Participants had 10 s from the
onset of the sentence to respond.

Fifty of the sensible sentences were literal. Half of the literal sentences described physical
hand/arm actions (The repairman bent the cable for her), while the other half were more
abstract or cognitive in nature (The bank ignored the pleas from her). El otro 50 sensible
sentences were figurative. Half of the figurative sentences were idioms (The defense picked
holes in the argument), and half were metaphors related to physical action (The discovery lifted
this nation out of poverty). The task was originally designed to examine literal/figurative sen-
tence processing and action-relatedness (Fernandino et al., 2013). In the current group of
patients, t tests revealed that response times (RT) and accuracies (CAC) did not significantly
differ (RT p = 0.35; ACC p = 0.24) between the literal (M RT = 4,802 EM, SD RT = 963 EM;
M ACC = 77.7%, SD ACC = 15.3%) and figurative conditions (M RT = 4,638 EM, DE = 954 EM;
M ACC = 80.9%, SD ACC = 14.5%), so the two conditions were combined. There is evidence
eso, in sentences, high action-relatedness and figurative language may involve distributed
brain areas in addition to classic language regions (p.ej., Johari et al., 2021), but core
language-related regions (the focus of the current study) are still expected to be involved in
the majority of language processing (Binder & Desai, 2011; Binder et al., 2009), especially
after collapsing all sentence types together.

Sentences followed subject-verb-object order except for four sentences that lacked an
object. All verbs were transitive or ambitransitive and were used transitively except for the four
aforementioned sentences. For sensible sentences, 25% of the subject nouns and 10% del
object nouns were grammatically animate. For low-sensibility sentences, 28% of the subject
nouns and 4% of the object nouns were grammatically animate. Low-sensibility sentences
were well formed grammatically but constructed such that the verb was incompatible with
one or both of its arguments (The taxpayer seized the planets with his small arm; The company
twisted the shot in the dark; The tape rejected the air in the sky). Detecting these sentences as
nonsensible involved understanding the subject/object nouns, verb, and their semantic as well
as grammatical relationships. Sensible and nonsense sentences were matched in length

Neurobiology of Language

325

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Mesa 2. Means (and standard deviations) for characteristics of sensible and nonsense sentences

Letras

Syllables

Words

Sensible
43.8 (7.3)

11.1 (2.2)

7.9 (1.2)

Nonsense
44.5 (8.9)

10.6 (2.4)

7.8 (1.7)

Accuracy in healthy adults

0.91 (0.08)

t test p
0.61

0.25

0.84

0.54

(number of letters, number of syllables, and number of words) and difficulty, as measured by
response accuracy from a pilot study using healthy adults (Mesa 2; Fernandino et al., 2013).
Control tasks were as follows.

Visual lexical decision

Visual lexical decision consisted of 80 verbos, 80 nouns, y 160 phonologically plausible
pseudowords. Pseudowords were chosen from the English Lexicon Project (ELP; Votar
et al., 2007). Words and pseudowords were matched in number of letters, bigram frequency,
orthographic neighborhood size, and visual lexical decision accuracy (Mesa 3). Visual lexical
decision consisted of the presentation of a fixation cross (500 EM), a mask (‘########’,
100 EM), a prime (50 EM), mask (100 EM), followed by the target. The prime was the same
as the target word/pseudoword in capital letters for half of the stimuli, and a consonant string
also in capital letters for the other half. For the purposes of the present investigation, we do not
investigate priming effects and collapse primed and unprimed trials together. Los participantes fueron
instructed to indicate as quickly and as accurately as possible whether the target was a real
word or not by pressing one of two buttons. Participants had 5 s to respond.

Half of nouns were manipulable objects (the phone, the pen), while the other half were
concrete but comparatively nonmanipulable (the ocean, the stadium), as determined by
body-object interaction ratings (Pexman et al., 2019). All were inanimate. Half of the verbs
referred to voluntary hand/arm actions (to pour, to pinch) while the other half referred to sen-
sory or cognitive concepts (to observe, to notice). All verbs were transitive or ambitransitive
except for four (5% of verb stimuli) that were intransitive. This task was originally designed to
examine priming effects, as well as effects of action-relatedness and manipulability (Desai
et al., 2015). The current study focuses on core language-related regions that are expected
to be involved in lexical-semantic processing for many word types (Binder & Desai, 2011;
Binder et al., 2009), especially when collapsing across word categories. Había 160 ensayos,

Mesa 3.
visual lexical decision task

Psycholinguistic variable means (and standard deviations) for words and nonwords in the

Length

Orthographic neighborhood

Words
5.53 (1.48)

3.75 (4.85)

Nonwords
5.36 (1.74)

3.59 (3.92)

Bigram frequency

1,607.78 (713.06)

1,580.48 (738.27)

LD ACC

0.96 (0.05)

0.97 (0.03)

Nota. LD = Lexical decision measures from the ELP database.

t test p
0.51

0.81

0.33

326

Neurobiology of Language

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Mesa 4. Means, desviaciones estandar (in parenthesis), and ranges for response times, exactitud, and d 0 for all tasks

CAC

d 0

Auditory sentence
sensibility
4,690 (920)
2,814–6,737

0.79 (0.14)
0.31–0.98

1.88 (1.09)
−0.86–4.15

Visual semantic
semejanza
2,944 (616)
1,849–4,030

0.78 (0.16)
0.45–0.98

–

Visual lexical
decisión
1,015 (327)
543–2,029

0.96 (0.07)
0.7–1

3.5 (1.1)
0.2–5

Nota. Bolded boxes indicate the measures used in the neuroanatomical analyses.

Auditory word
comprensión
–

0.88 (0.19)
0.13–1

–

Forward
digit span
–

6.6 (1.95)
0–12

–

divided equally between words and pseudowords, in both the verb and noun versions of the
visual lexical decision task. This task was the least demanding task in the battery with respect
to executive and semantic processing, as it merely required participants to recognize a word as
real or not, and there were no distractor items (reflected by high accuracies and low response
veces; Mesa 4). Semantic demands for lexical decision are relatively low compared to the
other tasks included here, as participants do not need to explicitly access word meaning in
order to successfully complete the task.

Visual semantic similarity judgment

The visual semantic similarity judgment task consisted of sets of 240 verbs and 240 nouns (ver
Mesa 5 for lexical characteristics). Each set was organized into 80 verb and 80 noun triplets
such that, for each triplet, the target word was more similar in meaning to one of the two
choices (p.ej., to thrill, to excite, to harm; bold indicates the target word). Similar to visual lex-
ical decision, the nouns were equally divided into manipulable and concrete nonmanipulable,
and the verbs into hand/arm actions and cognitive/sensory concepts. All nouns were inani-
compañero. All verbs were transitive or ambitransitive except for two (2.5% of verbs). This task
was originally designed to detect deficits related to manipulability and action-relatedness in
patients (Riccardi et al., 2019, 2020). Collapsing noun and verb performance together across
all categories provides a measure of general lexical-semantic processing and is expected to

Mesa 5.
judgment tasks

Lexical characteristics of words in the visual lexical decision and semantic similarity

Letras

Phonemes

Syllables

Log F

LD RT

LD ACC

Lexical
decisión
5.32

3.4

1.26

1.18

640.75

0.965

Semantic similarity
judgment
5.40

4.32

1.47

1.09

655.5

0.945

Nota. Log F = logarithmic frequency, LD = Lexical decision measures from the ELP database.

Neurobiology of Language

327

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

involve core lexical-semantic language regions. This task is also relatively executively
demandante, as participants must choose the correct response in the presence of a distractor
item that is from a similar semantic category, matching one of two words to the target instead
of making a binary judgment. These increased lexical-semantic and executive demands are
reflected by low accuracy scores and high response times compared to other control tasks
(Mesa 4).

The sentence sensibility, visual lexical decision, and visual semantic similarity judgment
tasks were presented on a laptop PC running E-prime software (versión 1.2, Psychology Soft-
ware Tools, Cª). Participants indicated their response by pressing one of two buttons. El
position of the bottom words was counterbalanced across participants. Participants could
use whichever hand they preferred and were asked to respond as quickly and accurately as
posible. The words remained on the screen for 5 s, after which the next triplet was presented.
Había 80 trials in both the verb and noun versions.

Auditory word comprehension

Auditory word comprehension was administered by a licensed speech-language pathologist as
part of the WAB (Kertesz, 2007). It consists of 60 real objects and pictures coming from 10
categories: real objects, drawn objects, formas, letters, numbers, colores, furniture, body parts,
fingers, and right-left body parts. There are six stimuli per category. The speech-language
pathologist speaks aloud the name of one of the pictures/objects, and the participant must
point to the correct item. Participants are given a point for each item that they correctly point
a, for a maximum of 60 puntos. Auditory word comprehension requires participants to com-
prehend a spoken word, and then to correctly recognize its corresponding visual form. A diferencia de
the semantic similarity judgment task, it does not require explicit access to deeper semantic
features in order to successfully complete the task.

Forward digit span

A series of digits (1 a través de 9) were read aloud to the participant. The participant was
instructed to repeat as many of the digits as they could, en orden, with series length increasing
after successful repetition.

MRI Data Acquisition

MRI data were obtained with a Siemens 3T Trio System with a 12-channel head coil and a
Siemens 3T Prisma System with a 20-channel coil. Participants underwent two anatomical
MRI sequences: (i) T1-weighted imaging sequence with a magnetization-prepared rapid-
gradient echo (MPRAGE) turbo field echo (TFE) sequence with voxel size = 1 mm3, campo
of view (FOV) = 256 × 256 mm, 192 sagittal slices, 9° flip angle, repetition time (TR) =
2,250 EM, inversion time (TI) = 925 EM, tiempo de eco (EL) = 4.15 EM, generalized autocalibrat-
ing partial parallel acquisition (GRAPPA) = 2, y 80 reference lines; y (ii) T2-weighted
MRI with a 3D sampling perfection with application optimized contrasts by using different
flip angle evolutions protocol with the following parameters: voxel size = 1 mm3, FOV =
256 × 256 mm, 160 sagittal slices, variable flip angle, TR = 3,200 EM, TE = 212 EM, y
no slice acceleration. The same slice center and angulation were used as in the T1 sequence.
Diffusion tensor imaging (DTI) scans consisted of two scans with a 180° flip, TR = 4,987 EM,
TE = 79.2 EM, matrix = 90 × 90, FOV = 207 × 207 mm2, slice thickness = 2.3 mm, y
50 transversal slices.

Neurobiology of Language

328

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Preprocessing of Structural Images

Lesions were defined in native space by a neurologist (l. Bonilha) in MRIcron (Rorden et al.,
2012) on individual T2-weighted images. Preprocessing started with coregistration of the
T2-weighted images to match the T-weighted images, allowing the lesions to be aligned to
native T1 space. Images were warped to standard space using enantiomorphic (Nachev
et al., 2008) segmentation-normalization (Ashburner & Friston, 2005) custom Matlab script
(https://github.com/rordenlab/spmScripts/blob/master/nii_enat_norm.m) to warp images to an
age-appropriate template image found in the Clinical Toolbox for SPM (https://www.nitrc.org
/scm/?group_id=881). The normalization parameters were used to reslice the lesion into stan-
dard space using linear interpolation, with subsequent lesion maps stored at 1 × 1 × 1-mm
resolution and binarized using a 50% límite. (Because interpolation can lead to fractional
probabilities, this step confirms that each voxel is categorically either lesioned or unlesioned
without biasing overall lesion volume.) Normalized images were visually inspected to verify
quality.

Preprocessing of DTI Data

Diffusion data were processed in the method described in Bonilha et al. (2015). MRTrix tools
were used to perform Gibbs artifacts removal (Kellner et al., 2016) and de-noising (Veraart
et al., 2016). FMRIB Software Library’s (FSL) TOPUP (Andersson et al., 2003) and eddy
(andersson & Sotiropoulos, 2015) were used to attenuate spatial distortion. FSL’s dtifit was
used to compute tensors, fractional anisotropy (FA), and mean diffusivity maps, and bedpost
(Hernandez et al., 2013) was used to model fibers. As discussed in the previous section, the T1 scan
used Statistical Parametric Mapping’s (SPM; https://www.fil.ion.ucl.ac.uk/spm/) unified normaliza-
tion and segmentation. This allowed warping of atlases from standard space to the patient’s space.
This was warped to native diffusion space by nonlinearly warping the T1 scan to the FA map (cual
has similar contrast). This allowed back-projection of our regions of interest (ROI) into the native
diffusion space. Finalmente, probtrackx (Hernandez-Fernandez et al., 2019) quantified connectivity.
This evaluated the connectivity between each and every region in the atlas.

DTI Connectome Creation

As described in Gleichgerrcht et al. (2017), a unique probabilistic DTI connectome was con-
structed for every participant using the 189 cortical regions defined by the Johns Hopkins
Universidad ( JHU) atlas (Faria et al., 2012; Mori et al., 2005; Wakana et al., 2004), resulting
en un 189 × 189 correlation matrix for each participant, where positive values signify greater
white matter connectivity of two regions. Every participant’s probabilistic white matter map
excluding the lesion was used as a mask for the estimation of their tractography. Para cada par
of regions, the number of streamlines arriving in one region when another region was seeded
was calculated, and the connectivity was defined as the average between the number of
streamlines arriving in region A when region B was seeded and the number of streamlines arriv-
ing in region B when region A was seeded. The connectivity between the regions was corrected
based on the sum of the volumes of the two regions to control for larger regions inherently
having a higher number of streamlines than smaller regions within the atlas. This resulted
en un 189 × 189 correlation matrix of weighted connections, which was used for CLSM.

Regions of Interest

Nine ROIs (Cifra 1), based on the JHU atlas (Faria et al., 2012; Mori et al., 2005; Wakana
et al., 2004), were used for lesion-deficit analysis. We analyzed several ROIs each from LIFC,

Neurobiology of Language

329

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Cifra 1. ROI (izquierda; mín.. 8, máximo. 33) and voxelwise lesion incidence map (mín.. 8, máximo. 40).

LATL, and LpT-iP. From LIFC/Broca’s area, we examined two ROIs: IFGoper and IFGtri. De
LATL, we examined three ROIs: middle and superior temporal poles (MTGpole, STGpole) y
the anterior portion of the inferior temporal gyrus (ITG). From LpT-iP, we examined four ROIs:
posterior middle and superior temporal gyri (pMTG, pSTG), supramarginal gyrus (SMG), y
angular gyrus (AG).

Experimental Design and Statistical Analysis

Behavioral data

Our primary aim was to inspect the relative impairment of the sentence comprehension task
(auditory sentence sensibility) factoring out the contribution of other tasks (visual lexical deci-
sión, visual semantic similarity judgment, auditory word comprehension, forward digit span).
For the auditory sentence sensibility and visual lexical decision tasks, d 0 was calculated as the
difference between the z transforms of the proportion of hits (es decir., when a sensible sentence
was identified as sensible) and the proportion of false alarms (es decir., when a nonsensible sen-
tence was identified as sensible). A d 0 score deviating from 0 for a participant reflects a sep-
aration between the number of correct hits and false alarms. For visual semantic similarity
judgment and auditory word comprehension, the proportion out of the total possible correct
was recorded. For forward digit span, the total number of digits correctly recalled, en orden,
fue usado.

Region- and voxel-based lesion-symptom mapping

RLSM was used within the nine ROIs (described in Regions of Interest) to identify damage
related to greater impairment of the sentence comprehension task as compared to the other
four tasks by regressing out performance in one condition from the other using NiiStat software
(www.nitrc.org/projects/niistat/). RLSM measures the relationship between percentage of vox-
els damaged within an ROI and a behavioral measure. Nuisance regression used the
Freedman-Lane method (Freedman & carril, 1983), allowing for permutation-based control
for family-wise error (Winkler et al., 2014). RLSM results were corrected for multiple compar-
isons using permutation analysis ( pag < 0.05, 1,000 permutations). Permutation analysis is a Neurobiology of Language 330 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d . / l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex nonparametric significance test that compares a test statistic to a null distribution that is created by randomly permuting the real data (Baldo & Dronkers, 2018; Baldo et al., 2012; Kimberg et al., 2007). Given the theoretical importance of LIFC, to increase power to detect effects VLSM was used within a restricted IFGoper and IFGtri ROI (combined into a single region for this analysis only) to investigate whether damage to voxels within this region was associated with worse sentence comprehension. VLSM binarily demarcates each voxel as either lesioned or unlesioned and tests the probability that damage to a voxel is associated with behavioral per- formance (Bates et al., 2003). VLSM results were thresholded at p < 0.001 voxel-wise and cluster-corrected to p < 0.05 using permutation analysis as correction for multiple comparisons (1,000 permutations). To improve power and minimize spatial bias, only voxels where at least 10% of patients had damage were considered (Baldo & Dronkers, 2018; Karnath et al., 2018). Region and voxelwise lesion incidence maps showed that we had sufficient coverage in all areas of interest (Figure 1). Connectivity-based lesion symptom mapping CLSM was used to investigate whether white matter connectivity between all regions included in the LIFC, LATL, and LpT-iP network of interest was specifically predictive of sentence com- prehension impairment compared to the other tasks using nuisance regression, as described above. Left-to-left and left-to-right connections (117 in total) between the regions were consid- ered to test for possible contributions from undamaged inter- or intra-hemispheric regions. White matter connectivity strengths were used in a general linear model to predict task per- formance. Alpha was set to 0.05, and significance was determined with permutation correc- tion for multiple comparisons (1,000 permutations). RESULTS RLSM A summary of the behavioral data can be found in Table 4. We first examined regions asso- ciated with the auditory sentence sensibility task, without including the other behavioral tasks as covariates. Auditory sentence sensibility performance was associated with percentage of voxels damaged in STGpole, SMG, AG, pSTG, and pMTG (Table 6; Figure 2). It was not sig- nificantly associated with damage in either of the LIFC ROIs. Next, we individually included data from each task as a covariate, partially accounting for potential contributions of auditory-verbal STM, lexical processing, executive function, seman- tics, and input modality (Table 6; Figure 2). Worse auditory sentence sensibility performance, controlling for forward digit span (auditory-verbal STM), was associated with percentage of voxels damaged in the SMG, AG, pMTG, and pSTG. Controlling for visual lexical decision (lexical processing with relatively low semantic and executive demands), it was associated with percentage of voxels damaged in the pSTG and pMTG. Controlling for visual semantic similarity judgment (lexical processing with relatively high semantic and executive demands), it was associated with percentage of voxels damaged in the pMTG. No areas were significantly associated with worse auditory sentence sensibility performance when controlling for auditory word comprehension (lexical task in the auditory modality with moderate executive and semantic demands). Neurobiology of Language 331 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d / . l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex Condition Auditory sentence sensibility Region STGpole z score −2.7 Table 6. Significant RLSM regions Auditory sentence sensibility controlling for forward digit span Auditory sentence sensibility controlling for visual lexical decision Auditory sentence sensibility controlling for visual semantic similarity judgment Note. Anatomical labels are according to the JHU atlas (Faria et al., 2012; Mori et al., 2005; Wakana et al., 2004). AG SMG pMTG pSTG AG SMG pMTG pSTG pMTG pSTG pMTG −3.0 −3.1 −3.4 −3.8 −2.7 −2.9 −2.9 −3.3 −3.2 −3.5 −2.5 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d . / l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Figure 2. RLSM results (color bar represents z scores). Areas where damage was associated with: (A) worse sentence comprehension, (B) worse sentence comprehension controlling for forward digit span, (C) worse sentence comprehension controlling for visual lexical decision, and (D) worse sentence comprehension controlling for visual semantic similarity judgment. Neurobiology of Language 332 Canonical sentence processing and the inferior frontal cortex Table 7. Significant VLSM peaks and Talairach coordinates Location IFGtri Cluster size (1 mm3 voxels) 364 Peak-z score −3.8 x −27 y 18 z 16 Condition Visual semantic similarity judgment controlling for auditory sentence sensibility VLSM VLSM, restricted to the LIFC for increased power, showed that no voxels were associated with worse auditory sentence sensibility performance by itself, or when including any other tasks as covariates. However, worse visual semantic similarity judgment performance, controlling for auditory sentence sensibility, was associated with a cluster of voxels in the LIFC (Table 7; Figure 3). CLSM Worse performance in the auditory sentence sensibility task, without including the other behavioral tasks as covariates, was associated with disruption of 10 white matter connections within the left hemisphere (Figure 4A). This included connections within and between LIFC, LATL, and LpT-iP. Worse auditory sentence sensibility performance, when including forward digit span, visual lexical decision, or auditory word comprehension as nuisance covariates, was asso- ciated with disruption of the same seven white matter connections within the left hemi- sphere. This included connections within and between LIFC, LATL, and LpT-iP (Figure 4B). l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d / . l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Figure 3. VLSM results for visual semantic similarity judgment controlling for auditory sentence sensibility. Blue voxels represent where damage is associated with worse visual semantic similarity judgment performance. Neurobiology of Language 333 Canonical sentence processing and the inferior frontal cortex (A) White matter connections where disruption was associated with worse auditory sentence sensibility performance. (B) White Figure 4. matter connections where disruption was associated with worse auditory sentence sensibility performance, controlling for forward digit span, visual lexical decision, and auditory word comprehension. (C) White matter connections where disruption was associated with worse auditory sentence sensibility performance, controlling for visual semantic similarity judgment. Colors are for display only. Worse auditory sentence sensibility performance, controlling for visual semantic similarity judgment, was associated with disruption of a single white matter connection within the left hemisphere; SMG to pMTG (Figure 4C). CLSM results for all analyses are summarized in Table 8. Table 8. Significant white matter connections. Disruption of these connections is associated with worse task performance. Condition Auditory sentence sensibility Connection IFGoper IFGtri SMG AG IFGoper IFGtri SMG SMG Auditory sentence sensibility controlling for forward digit span, visual lexical decision, and auditory word comprehension Auditory sentence sensibility controlling for visual semantic similarity judgment Neurobiology of Language SMG AG MTGpole SMG AG pMTG AG STGpole MTGpole pMTG SMG AG MTGpole SMG AG AG MTGpole pMTG z score 3.7 3.9 4.0 3.8 3.8 3.6 4.1 4.0 3.6 3.8 2.8 3.1 3.0 3.0 2.9 2.9 2.7 2.9 334 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d / . l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex DISCUSSION By using tasks that vary in their executive, lexical, and phonological demands, we interrogated the contributions of the LIFC, and two other language-related regions, in canonical sentence comprehension while controlling for these related cognitive processes. LIFC Focusing on pars opercularis and triangularis, we examined the effects of damage to LIFC, and also of lesions affecting the white matter connectivity of LIFC within the a priori language net- work. RLSM and VLSM provided little evidence supporting the contribution of the LIFC to canonical sentence comprehension. Damage to LIFC was not significantly associated with worse sentence comprehension in any of the V/RLSM analyses. Results here are similar to sev- eral prior studies. For example, in the classic investigation by Dronkers et al. (2004), damage to BA 44/45 was not associated with sentence comprehension deficits for any of the several included sentence types. Similar results were reported by Newhart et al. (2012) and Rogalsky et al. (2018). Here, we take these findings a step further by including a lexical task with relatively high semantic and executive demands, visual semantic similarity judgment. A more restrictive VLSM analysis, meant to increase power to detect effects within the LIFC, did not find a significant association between LIFC damage and sentence comprehension, even when not controlling for performance in the other tasks. However, damage to voxels within LIFC was associated with worse performance in the visual semantic similarity judgment task even after accounting for sentence comprehension performance. Visual semantic similarity judgment was one of the most semantically and executively demanding tasks included in our battery, suggesting a link between LIFC damage and impaired semantic access or control (Chiou et al., 2018; Jackson, 2021; Whitney et al., 2011). In contrast with the V/RLSM results, CLSM demonstrated that structural connectivity of the LIFC with other regions in the a priori language network was related to sentence comprehen- sion performance. Disruption of left-lateralized white matter connections within and between the LIFC, LATL, and LpT-iP was associated with worse sentence comprehension, not control- ling for other tasks. After controlling for visual lexical decision, forward digit span, and audi- tory word comprehension, seven white matter connections, again consisting of links between the left LIFC, LATL, and LpT-iP, remained significantly associated with sentence comprehen- sion. These findings demonstrate that connections between multiple left-hemisphere brain areas are important for canonical sentence comprehension, likely reflecting that successfully comprehending a sentence requires the coordination of numerous cognitive processes. Impor- tantly, the LIFC was part of this network after controlling for auditory word comprehension, lexical access, and auditory-verbal STM. This suggests that the role of the LIFC within the sen- tence comprehension network goes beyond those demands. These findings do not rule out the contribution of the LIFC to these subprocesses, but they do suggest that the LIFC contributes to canonical sentence comprehension in an additional way. Using the visual semantic similarity judgment task as a covariate allowed control for seman- tic and executive abilities, resulting in the absence of LIFC and LATL connections that were uniquely associated with sentence comprehension. This suggests that the LIFC and LATL connections are shared neural substrates for the sentence sensibility and semantic similarity judgment tasks. Considering this, the LIFC and LATL may contribute to canonical sentence comprehension via semantic knowledge/control and executive processes, likely associated with the task demands of the sentence sensibility task. The sentence sensibility task oriented Neurobiology of Language 335 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d / . l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex attention to the meaningfulness of the sentences, requiring participants to identify instances where the verb and subject/object fit together to create a coherent, sensible whole. Relating to the hypotheses discussed in the Introduction, our findings closely align with the neuroim- aging research suggesting that the LIFC is especially involved in sentence comprehension when the task explicitly focuses attention on semantics or is semantically demanding (Hagoort & Indefrey, 2014; Hasson et al., 2006; Love et al., 2006; Rogalsky & Hickok, 2009). In this framework, the LIFC contributes to canonical sentence comprehension via executive task- related demands, especially pertaining to detection of syntactic or semantic violations. An alternative hypothesis partially supported by the current results is that the LIFC contrib- utes to sentence processing via a variety of lexical-semantic operations, which are also needed by semantic similarity judgment (Fiebach et al., 2002; Heim et al., 2009; Kotz et al., 2002, 2010; Ruff et al., 2008). However, LIFC connectivity was still associated with sentence com- prehension after accounting for performance in two other lexical-semantic tasks that were comparatively less demanding (visual lexical decision, auditory word comprehension), sug- gesting that deeper task-related semantic analysis and control may be the main factors driving LIFC connectivity involvment in the current sentence comprehension task. The V/RLSM finding that damage to LIFC is not associated with canonical sentence com- prehension, when contrasted with the positive CLSM finding that structural disconnection of the LIFC from the language network is associated with worse performance, seems paradoxical at first. One possible explanation is that the anatomy-function relationship within the LIFC has been shown to be heterogenous between individuals (Amunts et al., 1999; Fedorenko & Blank, 2020), which may impact power to detect group-level effects in fine-grained analyses such as VLSM that require overlap at the voxelwise level. However, the negative results of the RLSM anlaysis, which should not require fine-grained anatomy-function homogeneity within a region to detect effects, makes this possibility unlikely. A more likely explanation is that, as discussed in the Introduction, V/RLSM can only detect areas of overlapping necrosis/gliosis but does not consider long-range white matter tract damage that can lead to cortical discon- nection and associated behavioral impairments (Gleichgerrcht et al., 2017). CLSM comple- ments V/RLSM by being able to detect effects of disrupted white matter connectivity resulting from damage anywhere along white matter tracts that connect two grey matter regions, even if those regions are spared by the lesion. In sum, the results here build upon previous studies suggesting that disrupted LIFC connectivity has negative consequences for sentence compre- hension (den Ouden et al., 2019; Fridriksson et al., 2018), and expand those findings by dem- onstrating that these consequences extend even to relatively simple canonical sentences and may be related to task-specific demands such as semantic analysis. LpT-iP Damage to the LpT-iP, especially the pMTG, was significantly associated with worse sentence comprehension, even after controlling for performance on tasks that measured auditory-verbal STM, semantic knowledge and control, and lexical access. This aligns with the previously dis- cussed neuropsychological evidence suggesting that the pMTG is involved in sentence com- prehension, even for canonical sentences (Magnusdottir et al., 2013; Rogalsky et al., 2018). The pMTG has been identified as an important area for sentence processing, with possible contributions being mapping wordforms to their meanings (“lexical interface;” Hickok & Poeppel, 2004, 2007), processing syntax (Griffiths et al., 2013; Snijders et al., 2009), and rep- resenting semantic knowledge (Binder & Desai, 2011; Binder et al., 2009). RLSM revealed that damage to pMTG was associated with worse sentence comprehension when controlling for all Neurobiology of Language 336 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d / . l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex tasks except for auditory single-word comprehension, a task with relatively high phonological demands. Additionally, CLSM revealed that pMTG connectivity was part of a larger network subserving sentence comprehension, but this pMTG connectivity disappeared when control- ling for performance in visual lexical decision, forward digit span, or auditory word compre- hension. When controlling for visual semantic similarity judgment, disruption of a single connection (pMTG to SMG) was associated with worse sentence comprehension. Considering that the pMTG did not survive any analysis using auditory word comprehension as a covariate, our results suggest that the pMTG may be a common neural substrate for auditory single-word and sentence comprehension. This aligns with the “lexical interface” hypothesis (Hickok & Poeppel, 2004, 2007), as well as studies suggesting that the pMTG may play a special role in specifically auditory language comprehension (Pillay et al., 2017). However, given the wealth of processes attributed to the pMTG, it is also possible that this area performs many different functions or that it serves as a general hub of connectivity within the language com- prehension network (Turken & Dronkers, 2011). Previous studies have demonstrated that the pSTG plays a vital role in phonological pro- cessing (Buchsbaum et al., 2001; Graves et al., 2008) and auditory-verbal STM (Leff et al., 2009; Richardson et al., 2011), both of which are related to sentence comprehension. RLSM in the current study found that pSTG damage was associated with worse sentence comprehen- sion, but this association did not survive when controlling for auditory word comprehension performance. This supports the well-established role of pSTG in acoustic-phonological analysis (Hickok & Poeppel, 2000, 2007, 2016; Robson et al., 2012). A somewhat surprising finding was that pSTG damage was associated with sentence comprehension even after con- trolling for forward digit span, a measure of auditory-verbal STM. We do not interpret this as evidence against a role of pSTG in auditory-verbal STM. Instead, this finding was possibly due to forward digit span being a production task, meaning that some participants likely have dis- played worse performance due to deficits in production despite relatively spared auditory- verbal STM abilities, limiting power to detect effects in brain areas related to auditory-verbal STM. Digit span is a widely used measure of auditory-verbal STM in neuropsychological stud- ies of language comprehension (Caplan et al., 2016; Leff et al., 2009; Newhart et al., 2012; Pettigrew & Hillis, 2014; Pisoni et al., 2019; Salis, 2012), but its production component is an inherent limitation when comparing it directly to comprehension tasks. In sum, it is likely that pSTG contributes to auditory sentence comprehension via both acoustic-phonological analy- sis and auditory-verbal STM. Similar to the pSTG, previous studies implicate the SMG in phonological processing, auditory-verbal STM, and the processing of lexical and sub-lexical cues in both visual and auditory modalities (Deschamps et al., 2014; Hartwigsen et al., 2010; Oberhuber et al., 2016; Sliwinska et al., 2012). Damage to SMG was associated with worse sentence compre- hension, and disrupted connectivity between SMG and AG, IFG, and MTGpole predicted worse sentence comprehension after controlling for auditory-verbal STM, auditory word comprehension, and visual lexical decision performance. We interpret the SMG connectivity as reflecting the coordination of complex acoustic-phonological analysis in SMG with distrib- uted lexical-semantic and executive processes required for successful completion of the sen- tence sensibility task. Further support for this interpretation was found when these connections to LATL, AG, and IFG did not survive after correcting for visual semantic similarity judgment, a task with relatively high lexical-semantic and executive demands. Instead, a single connection (SMG to pMTG) was associated with sentence comprehension. This likely reflects that the sen- tence comprehension task was auditory, while the semantic similarity judgment was visual, considering the previously mentioned studies indicating that SMG and pMTG play important Neurobiology of Language 337 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d / . l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex roles in acoustic-phonological processing and auditory language comprehension, respectively. Finally, the AG is a well-established part of the lexical-semantic system, being associated with the representation of semantic knowledge across many different tasks and concept types (Binder & Desai, 2011; Binder et al., 2009). RLSM revealed that damage to the AG was asso- ciated with worse sentence comprehension, but this did not survive after controlling for any of the lexical-semantic tasks (visual lexical decision, visual semantic similarity judgment, audi- tory word comprehension). CLSM showed that disconnection of the AG from IFG and SMG was associated with worse sentence comprehension after controlling for all tasks except for visual semantic similarity judgment. The RLSM findings demonstrate that the AG is a shared neural substrate for a variety of lexical-semantic tasks, while CLSM suggests that connectivity from AG to the larger language comprehension network may be shared neural correlates spe- cifically for explicitly semantic tasks (semantic similarity and sentence sensibility judgment). LATL RLSM revealed that damage to the STGpole was associated with worse sentence comprehen- sion when not controlling for the other behavioral measures. This finding lends support to theories that implicate the LATL in sentence processing (Brennan & Pylkkanen, 2017; Brennan et al., 2012; Humphries et al., 2001, 2005). However, the fact that LATL damage was not asso- ciated with worse sentence comprehension when adding the other behavioral measures as covariates could reflect that the LATL contributes to multiple functions in addition to sentence processing, such as lexical or semantic retrieval (Lambon Ralph et al., 2017; Mesulam et al., 2013). It is also possible that regions specialized for sentence-related syntactic processing are located more inferiorly in the LATL (Humphries et al., 2005), where relatively few patients had damage, limiting power to detect effects. CLSM revealed that disconnection of the MTGpole from the IFGoper and SMG was associated with worse sentence comprehension after control- ling for all tasks except for visual semantic similarity judgment. These RLSM and CLSM results, considering the already well-established role of the LATL in lexical-semantic access and retrieval (Lambon Ralph et al., 2017; Mesulam et al., 2013), suggest that the LATL contributes to canonical sentence comprehension through the representation or access of lexical-semantic knowledge. Limitations This study leveraged preexisting data to interrogate the contribution of various language- related areas to canonical sentence comprehension and related cognitive processes. The tasks were not explicitly designed for the current investigation, and the retrospective nature of the study inherently limits our control over the presence of certain psycholinguistic variables. For example, while the auditory word comprehension task is a widely used clinical measure of word recognition impairments, it contains words from semantic categories that were not included in the other tasks (e.g., colors) and lacks a verb component. While our analysis focused on core language-related areas that are expected to be involved in the processing of many different types of words, it is possible that some effects observed here are associated with those lexical-semantic differences between tasks. The current study consisted of canonical declarative sentences. Inclusion of other types of sentences, including noncanonical and more syntactically complex sentences would provide additional information about neural substrates of sentence comprehension when syntactic demands are manipulated. Additionally, it is possible that participants employed a number Neurobiology of Language 338 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d / . l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex of strategies to aid in their performance of the sentence sensibility task, such as using lexical- semantic associations between words within the sentences, as this was not explicitly con- trolled for, or by using animacy clues (e.g., does the subject noun typically perform the action described by the verb). However, half of the sensible sentences were figurative (The bank pulled the plug on the deal ). In these sentences, subject/object nouns and verbs are combined in ways that are not sensible literally, and the subject and object nouns usually have low lexical-semantic associations (e.g., bank and plug). This means that participants who heavily relied on the aforementioned strategies would still likely perform poorly overall on the task. Further, using the other lexical-semantic tasks included in the current study as covariates would also help control for this possibility. A single meaningfulness judgment task was used. Adding other tasks, such as sentence- picture matching, could allow for direct comparisons of task type and demands. Additionally, “executive control” involves many processes, and having multiple tasks that stress control in conjunction with visual semantic similarity judgment to probe those specific processes would be helpful in understanding the specific role of the LIFC. Finally, the spatial resolution of lesion studies is inherently limited, and a fine-grained (e.g., millimeter scale) organization of function is better studied with methods such as functional magnetic resonance imaging (fMRI). Conclusion VLSM and RLSM analyses suggested that the left pars opercularis and triangularis regions are not vital for canonical sentence comprehension in the context of a sensibility judgment task. LIFC damage was associated instead with impairments in a semantic similarity judgment task that had high semantic and executive demands. However, the complementary CLSM method revealed that disruption of left-lateralized white matter connections from LIFC to LATL and LpT-iP was associated with worse sentence comprehension after controlling for performance in tasks related to lexical access, auditory word comprehension, and auditory-verbal STM. Semantic similarity judgment on single words explained similar variance to sentence sensibil- ity judgment in LIFC connectivity, consistent with the hypothesis that LIFC’s contribution to sentence comprehension is related to task-related processes when semantic demands are high. Damage to the LpT-iP, especially pMTG, predicted worse sentence comprehension after con- trolling for lexical access, semantic knowledge, and auditory-verbal STM, but not auditory word comprehension. This supports previous studies demonstrating that the pMTG is a vital region for sentence comprehension and suggests that the pMTG may contribute to canonical sentence comprehension by acting as an auditory “lexical interface" or general hub for audi- tory language comprehension. ACKNOWLEDGMENTS This work was supported by NIH/ NIDCD grants R01DC010783 (Rutvik H. Desai), R56DC010783 (Rutvik H. Desai), R01DC017162 (Rutvik H. Desai), and P50DC014664 ( Julius Fridriksson). We thank three reviewers for their insightful comments. FUNDING INFORMATION Rutvik H. Desai, National Institute on Deafness and Other Communication Disorders (https:// dx.doi.org/10.13039/100000055), Award ID: R01DC010783. Rutvik H. Desai, National Insti- tute on Deafness and Other Communication Disorders (https://dx.doi.org/10.13039 /100000055), Award ID: R56DC010783. Rutvik H. Desai, National Institute on Deafness and Other Communication Disorders (https://dx.doi.org/10.13039/100000055), Award ID: Neurobiology of Language 339 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . e d u n o / l / l a r t i c e - p d f / / / / 3 2 3 1 8 2 0 0 8 4 8 2 n o _ a _ 0 0 0 6 7 p d . / l f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Canonical sentence processing and the inferior frontal cortex R01DC017162. Julius Fridriksson, National Institute on Deafness and Other Communication Disorders (https://dx.doi.org/10.13039/100000055), Award ID: P50DC014664. AUTHOR CONTRIBUTIONS Nicholas Riccardi: Conceptualization: Supporting; Formal analysis: Lead; Investigation: Lead; Visualization: Lead; Writing – original draft: Lead; Writing – review & editing: Equal. Chris Rorden: Data curation: Lead; Funding acquisition: Supporting; Methodology: Supporting; Resources: Equal; Software: Lead; Writing – review & editing: Equal. Julius Fridriksson: Conceptualization: Equal; Data curation: Supporting; Funding acquisition: Equal; Project administration: Lead; Resources: Lead; Supervision: Supporting; Writing – review & editing: Equal. Rutvik H. Desai: Conceptualization: Lead; Formal analysis: Supporting; Funding acquisition: Equal; Investigation: Equal; Project administration: Supporting; Supervision: Lead; Writing – original draft: Supporting; Writing – review & editing: Equal. REFERENCES Amunts, K., Schleicher, A., Burgel, U., Mohlberg, H., Uylings, H. B., & Zilles, K. (1999). Broca’s region revisited: Cytoarchitec- ture and intersubject variability. Journal of Comparative Neurol- ogy, 412(2), 319–341. https://doi.org/10.1002/(SICI)1096-9861 (19990920)412:2<319::AID-CNE10>3.0.CO;2-7, PubMed:
10441759

andersson, j. l. r., Skare, S., & Ashburner, j. (2003). How to correct
susceptibility distortions in spin-echo echo-planar images: Appli-
cation to diffusion tensor imaging. NeuroImagen, 20(2), 870–888.
https://doi.org/10.1016/S1053-8119(03)00336-7, PubMed:
14568458

andersson, j. l. r., & Sotiropoulos, S. norte. (2015). Non-parametric
representation and prediction of single- and multi-shell
diffusion-weighted MRI data using Gaussian processes. Neuro-
Image, 122, 166–176. https://doi.org/10.1016/j.neuroimage
.2015.07.067, PubMed: 26236030

Ashburner, J., & Friston, k. j. (2005). Unified segmentation. Neuro-
Image, 26(3), 839–851. https://doi.org/10.1016/j.neuroimage
.2005.02.018, PubMed: 15955494

Baldo, j. v., & Dronkers, norte. F. (2018). Lesion studies. In A. METRO. B.
de Groot & PAG. Hagoort (Editores.), Research methods in psycholin-
guistics and the neurobiology of language: A practical guide
(páginas. 310–329). wiley.

Baldo, j. v., wilson, S. METRO., & Dronkers, norte. (2012). Uncovering the
neural substrates of language: A voxel-based lesion-symptom
mapping approach. En m. Faust (Ed.), The handbook of the neu-
ropsychology of language (páginas. 582–594). Wiley-Blackwell.
https://doi.org/10.1002/9781118432501.ch28

Votar, D. A., Yap, METRO. J., Cortese, METRO. J., hutchison, k. A., Kessler,
B., Loftis, B., Neely, j. h., nelson, D. l., Simpson, GRAMO. B., &
Treiman, R. (2007). The English lexicon project. Behavior Research
Métodos, 39(3), 445–459. https://doi.org/10.3758/BF03193014,
PubMed: 17958156

Bates, MI., wilson, S. METRO., Saygin, A. PAG., Dick, F., Sereno, METRO. I.,
Caballero, R. T., & Dronkers, norte. F.
(2003). Voxel-based
lesion-symptom mapping. Neurociencia de la naturaleza, 6(5), 448–450.
https://doi.org/10.1038/nn1050, PubMed: 12704393

Binder, j. r., & Desai, R. h. (2011). The neurobiology of semantic
memory. Trends in Cognitive Science, 15(11), 527–536. https://
doi.org/10.1016/j.tics.2011.10.001, PubMed: 22001867

Binder, j. r., Desai, R. h., Tumbas, W.. w., & Conant, l. l. (2009).
Where is the semantic system? A critical review and

meta-analysis of 120 functional neuroimaging studies. Cerebral
Corteza, 19(12), 2767–2796. https://doi.org/10.1093/cercor
/bhp055, PubMed: 19329570

Bonilha, l., & Fridriksson, j. (2009). Subcortical damage and white
matter disconnection associated with non-fluent speech. Cerebro,
132(6), Article e108. https://doi.org/10.1093/ brain/awn200,
PubMed: 18723562

Bonilha, l., Gleichgerrcht, MI., Fridriksson, J., Rorden, C., Breedlove,
j. l., Nesland, T., Paulus, w., Helms, GRAMO., & Focke, norte. k. (2015).
Reproducibility of the structural brain connectome derived from
imágenes de tensor de difusión. MÁS UNO, 10(8), Article e0135247.
https://doi.org/10.1371/journal.pone.0135247, PubMed:
26332788

Bonilha, l., Nesland, T., Rorden, C., Fillmore, PAG., Ratnayake, R. PAG.,
& Fridriksson, j. (2014). Mapping remote subcortical ramifica-
tions of injury after ischemic strokes. Behavioral Neurology,
2014, Article 215380. https://doi.org/10.1155/2014/215380,
PubMed: 24868120

Bonilha, l., Rorden, C., & Fridriksson, j. (2014). Assessing the clin-
ical effect of residual cortical disconnection after ischemic
strokes. Stroke, 45, 988–993. https://doi.org/10.1161
/STROKEAHA.113.004137, PubMed: 24619391

Brennan, j. [r.], Nir, y., Hasson, Ud., Malach, r., Heeger, D. J., &
Pylkkanen, l. (2012). Syntactic structure building in the anterior
temporal lobe during natural story listening. Brain and Language,
120(2), 163–173. https://doi.org/10.1016/j.bandl.2010.04.002,
PubMed: 20472279

Brennan, j. r., & Pylkkanen, l. (2017). MEG evidence for incre-
mental sentence composition in the anterior temporal lobe. Cog-
nitive Science, 41(S6), 1515–1531. https://doi.org/10.1111/cogs
.12445, PubMed: 27813182

Buchsbaum, B. r., Hickok, GRAMO., & Humphries, C. (2001). Role of left
posterior superior temporal gyrus in phonological processing for
speech perception and production. Ciencia cognitiva, 25(5),
663–678. https://doi.org/10.1207/s15516709cog2505_2

Caplan, D., Michaud, J., Hufford, r., & Makris, norte. (2016). Deficit-
lesion correlations in syntactic comprehension in aphasia. Cerebro
and Language, 152, 14–27. https://doi.org/10.1016/j.bandl.2015
.10.005, PubMed: 26688433

Caramazza, A., Capasso, r., Capitani, MI., & Miceli, GRAMO. (2005). Pat-
terns of comprehension performance in agrammatic Broca’s
aphasia: A test of the trace deletion hypothesis. Brain and

Neurobiology of Language

340

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Idioma, 94(1), 43–53. https://doi.org/10.1016/j.bandl.2004.11
.006, PubMed: 15896382

Catani, METRO., Dell’Acqua, F., Bizzi, A., Forkel, S. J., williams, S. C.,
Simmons, A., Murphy, D. GRAMO., & Thiebaut de Schotten, METRO. (2012).
Beyond cortical localization in clinico-anatomical correlation.
Corteza, 48(10), 1262–1287. https://doi.org/10.1016/j.cortex
.2012.07.001, PubMed: 22995574

Catani, METRO., & ffytche, D. h. (2005). The rises and falls of discon-
nection syndromes. Cerebro, 128(10), 2224–2239. https://doi.org
/10.1093/brain/awh622, PubMed: 16141282

Chiou, r., Humphreys, GRAMO. F., Jung, J., & Lambon Ralph, METRO. A.
(2018). Controlled semantic cognition relies upon dynamic and
flexible interactions between the executive ‘semantic control’
and hub-and-spoke ‘semantic representation’ systems. Corteza,
103, 100–116. https://doi.org/10.1016/j.cortex.2018.02.018,
PubMed: 29604611

den Ouden, D. B., Malyutina, S., Basilakos, A., Bonilha, l.,
Gleichgerrcht, MI., Yourganov, GRAMO., Hillis, A. MI., Hickok, GRAMO.,
Rorden, C., & Fridriksson, j. (2019). Cortical and structural-
connectivity damage correlated with impaired syntactic process-
ing in aphasia. Mapeo del cerebro humano, 40(7), 2153–2173.
https://doi.org/10.1002/hbm.24514, PubMed: 30666767

Desai, R. h., Herter, T., Riccardi, NORTE., Rorden, C., & Fridriksson, j.
(2015). Concepts within reach: Action performance predicts
action language processing in stroke. Neuropsicología, 71,
217–224. https://doi.org/10.1016/j.neuropsychologia.2015.04
.006, PubMed: 25858602

Desai, R. h., & Riccardi, norte. (2021). Cognitive neuroscience of
idioma. In W. Xu & j. R. taylor (Editores.), The Routledge handbook
of cognitive linguistics (páginas. 615–642). Routledge. https://doi.org
/10.4324/9781351034708-42

Deschamps, I., Baum, S. r., & Gracco, V. l. (2014). On the role of
the supramarginal gyrus in phonological processing and verbal
working memory: Evidence from rTMS studies. Neuropsycholo-
gia, 53, 39–46. https://doi.org/10.1016/j.neuropsychologia.2013
.10.015, PubMed: 24184438

Dronkers, norte. F., Plaisant, o., Iba-Zizen, METRO. T., & Cabanis, mi. A.
(2007). Paul Broca’s historic cases: High resolution MR imaging
of the brains of Leborgne and Lelong. Cerebro, 130(5), 1432–1441.
https://doi.org/10.1093/brain/awm042, PubMed: 17405763

Dronkers, norte. F., Wilkins, D. PAG., Van Valin, R. D., Jr., Redfern, B. B.,
& Jaeger, j. j. (2004). Lesion analysis of the brain areas involved
in language comprehension. Cognición, 92(1–2), 145–177.
https://doi.org/10.1016/j.cognition.2003.11.002, PubMed:
15037129

Fadiga, l., Craighero, l., & D’Ausilio, A. (2009). Broca’s area in
idioma, acción, and music. Annals of the New York Academy
of Sciences, 1169(1), 448–458. https://doi.org/10.1111/j.1749
-6632.2009.04582.X, PubMed: 19673823

Faria, A. v., joel, S. MI., zhang, y., Oishi, K., van Zjil, PAG. C., Molinero,
METRO. I., Pekar, j. J., & Mori, S. (2012). Atlas-based analysis of
resting-state functional connectivity: Evaluation for reproducibil-
ity and multi-modal anatomy-function correlation studies. nuevo-
roImage, 61(3), 613–621. https://doi.org/10.1016/j.neuroimage
.2012.03.078, PubMed: 22498656

Fedorenko, MI., & Blank, I. A. (2020). Broca’s area is not a natural
kind. Trends in Cognitive Science, 24(4), 270–284. https://doi.org
/10.1016/j.tics.2020.01.001, PubMed: 32160565

Fernandino, l., Conant, l. l., Binder, j. r., Blindauer, K., Hiner, B.,
Spangler, K., & Desai, R. h. (2013). Where is the action? Acción
sentence processing in Parkinson’s disease. Neuropsicología,
51(8), 1510–1517. https://doi.org/10.1016/j.neuropsychologia
.2013.04.008, PubMed: 23624313

Fiebach, C. J., Friederici, A. D., Muller, K., & por cramon, D. Y.
(2002). fMRI evidence for dual routes to the mental lexicon in
visual word recognition. Revista de neurociencia cognitiva,
14(1), 11–23. https://doi.org/10.1162/089892902317205285,
PubMed: 11798383

Freedman, D. J., & carril, D. (1983). A nonstochastic interpretation
of reported significance levels. Revista de negocios & Económico
Estadísticas, 1(4), 292–298. https://doi.org/10.2307/1391660

Fridriksson, J., Bonilha, l., & Rorden, C. (2007). Severe Broca’s
aphasia without Broca’s area damage. Behavioural Neurology,
18(4), Article 785280. https://doi.org/10.1155/2007/785280,
PubMed: 18430982

Fridriksson, J., den Ouden, D. B., Hillis, A. MI., Hickok, GRAMO., Rorden,
C., Basilakos, A., Yourganov, GRAMO., & Bonilha, l. (2018). Anatomy
of aphasia revisited. Cerebro, 141(3), 848–862. https://doi.org/10
.1093/brain/awx363, PubMed: 29360947

Fridriksson, J., Fillmore, PAG., guo, D., & Rorden, C. (2015). Chronic
Broca’s aphasia is caused by damage to Broca’s and Wernicke’s
areas. Corteza cerebral, 25(12), 4689–4696. https://doi.org/10
.1093/cercor/bhu152, PubMed: 25016386

Friederici, A. D. (2012). The cortical language circuit: From audi-
tory perception to sentence comprehension. Tendencias en Cognitivo
Ciencia, 16(5), 262–268. https://doi.org/10.1016/j.tics.2012.04
.001, PubMed: 22516238

Gleichgerrcht, MI., Fridriksson, J., Rorden, C., & Bonilha, l. (2017).
Connectome-based lesion-symptom mapping (CLSM): A novel
approach to map neurological function. NeuroImagen: Clinical,
16, 461–467. https://doi.org/10.1016/j.nicl.2017.08.018,
PubMed: 28884073

Graessner, A., Zaccarella, MI., & Hartwigsen, GRAMO. (2021). Differential
contributions of left-hemispheric language regions to basic seman-
tic composition. Brain Structure and Function, 226(2), 501–518.
https://doi.org/10.1007/s00429-020-02196-2, PubMed: 33515279
Tumbas, W.. w., Grabowski, t. J., Mehta, S., & Gupta, PAG. (2008). El
left posterior superior temporal gyrus participates specifically in
accessing lexical phonology. Revista de neurociencia cognitiva,
20(9), 1698–1710. https://doi.org/10.1162/jocn.2008.20113,
PubMed: 18345989

Griffiths, j. D., Marslen-Wilson, W.. D., Stamatakis, mi. A., & tyler,
l. k. (2013). Functional organization of the neural language sys-
tema: Dorsal and ventral pathways are critical for syntax. Cerebral
Corteza, 23(1), 139–147. https://doi.org/10.1093/cercor/bhr386,
PubMed: 22275482

Hagoort, PAG., & Indefrey, PAG. (2014). The neurobiology of language
beyond single words. Revisión anual de neurociencia, 37,
347–362. https://doi.org/10.1146/annurev-neuro-071013
-013847, PubMed: 24905595

Hartwigsen, GRAMO., Baumgaertner, A., Precio, C. J., Koehnke, METRO., Ulmer,
S., & Siebner, h. R. (2010). Phonological decisions require both
the left and right supramarginal gyri. Actas del Nacional
Academia de Ciencias, 107(38), 16494–16499. https://doi.org/10
.1073/pnas.1008121107, PubMed: 20807747

Hasson, Ud., Nusbaum, h. C., & Pequeño, S. l. (2006). Repetition sup-
pression for spoken sentences and the effect of task demands.
Revista de neurociencia cognitiva, 18(12), 2013–2029. https://
doi.org/10.1162/jocn.2006.18.12.2013, PubMed: 17129188
Heim, S., Eickhoff, S. B., Friederici, A. D., & Amunts, k. (2009). Left
cytoarchitectonic area 44 supports selection in the mental lexi-
con during language production. Brain Structure and Function,
213(4–5), 441–456. https://doi.org/10.1007/s00429-009-0213-9,
PubMed: 19621242

Hernández, METRO., Guerrero, GRAMO. D., Cecilia, j. METRO., Garcia, j. METRO.,
Inuggi, A., Jbabdi, S., Behrens, t. MI., & Sotiropoulos, S. norte.

Neurobiology of Language

341

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

(2013). Accelerating fibre orientation estimation from diffusion
weighted magnetic resonance imaging using GPUs. PLOS
ONE, 8(4), Article e61892. https://doi.org/10.1371/journal.pone
.0061892, PubMed: 23658616

Hernandez-Fernandez, METRO., Reguly, I., Jbabdi, S., Giles, METRO., Herrero, S.,
& Sotiropoulos, S. norte. (2019). Using GPUs to accelerate computa-
tional diffusion MRI: From microstructure estimation to tractogra-
phy and connectomes. NeuroImagen, 188, 598–615. https://doi
.org/10.1016/j.neuroimage.2018.12.015, PubMed: 30537563
Hickok, GRAMO., & Poeppel, D. (2000). Towards a functional neuroanat-
omy of speech perception. Trends in Cognitive Science, 4(4),
131–138. https://doi.org/10.1016/s1364-6613(00)01463-7,
PubMed: 10740277

Hickok, GRAMO., & Poeppel, D. (2004). Dorsal and ventral streams: A
framework for understanding aspects of the functional anatomy
of language. Cognición, 92(1–2), 67–99. https://doi.org/10.1016/j
.cognition.2003.10.011, PubMed: 15037127

Hickok, GRAMO., & Poeppel, D. (2007). The cortical organization of
speech processing. Naturaleza Reseñas Neurociencia, 8(5), 393–402.
https://doi.org/10.1038/nrn2113, PubMed: 17431404

Hickok, GRAMO., & Poeppel, D. (2016). The neural basis of speech
percepción. In G. Hickok & S. Pequeño (Editores.), Neurobiology of
idioma (páginas. 299–310). Elsevier. https://doi.org/10.1016/B978
-0-12-407794-2.00025-0

Humphries, C., Binder, j. r., Medler, D. A., & Liebenthal, mi. (2006).
Syntactic and semantic modulation of neural activity during
auditory sentence comprehension. Journal of Cognitive Neuro-
ciencia, 18(4), 665–679. https://doi.org/10.1162/jocn.2006.18.4
.665, PubMed: 16768368

Humphries, C., Amar, T., Swinney, D., & Hickok, GRAMO. (2005).
Response of anterior temporal cortex to syntactic and prosodic
manipulations during sentence processing. Human Brain Map-
ping, 26(2), 128–138. https://doi.org/10.1002/ hbm.20148,
PubMed: 15895428

Humphries, C., Willard, K., Buchsbaum, B., & Hickok, GRAMO. (2001).
Role of anterior temporal cortex in auditory sentence compre-
hension: Un estudio de resonancia magnética funcional. NeuroReport, 12(8), 1749–1752.
https://doi.org/10.1097/00001756-200106130-00046, PubMed:
11409752

Jackson, R. l. (2021). The neural correlates of semantic control
revisited. NeuroImagen, 224, Article 117444. https://doi.org/10
.1016/j.neuroimage.2020.117444, PubMed: 33059049

Johari, K., Riccardi, NORTE., Malyutina, S., Modi, METRO., & Desai, R. h.
(2021). HD-tDCS over motor cortex facilitates figurative and lit-
eral action sentence processing. Neuropsicología, 159, Article
107955. https://doi.org/10.1016/j.neuropsychologia.2021
.107955, PubMed: 34252418

Karnath, h. o., Sperber, C., & Rorden, C. (2018). Mapping human
brain lesions and their functional consequences. NeuroImagen,
165, 180–189. https://doi.org/10.1016/j.neuroimage.2017.10
.028, PubMed: 29042216

Kellner, MI., Dhital, B., Kiselev, V. GRAMO., & Reisert, METRO. (2016). Gibbs-
ringing artifact removal based on local subvoxel-shifts. Magnético
Resonancia en Medicina, 76(5), 1574–1581. https://doi.org/10
.1002/mrm.26054, PubMed: 26745823

Kemmerer, D. (2021). Cognitive neuroscience of language (2y ed.).

Prensa de Psicología.

Kertesz, A. (2007). Western Aphasia Battery—Revised. Pearson.

https://doi.org/10.1037/t15168-000

Kimberg, D. y., Coslett, h. B., & Schwartz, METRO. F. (2007). Power in
voxel-based lesion-symptom mapping. Journal of Cognitive Neu-
roscience, 19(7), 1067–1080. https://doi.org/10.1162/jocn.2007
.19.7.1067, PubMed: 17583984

Kinno, r., Muragaki, y., Hori, T., Maruyama, T., Kawamura, METRO., &
Sakai, k. l. (2009). Agrammatic comprehension caused by a gli-
oma in the left frontal cortex. Brain and Language, 110(2), 71–80.
https://doi.org/10.1016/j.bandl.2009.05.001, PubMed:
19573900

(2014). Differential

Kinno, r., Ohta, S., Muragaki, y., Maruyama, T., & Sakai,
k. l.
three syntax-
related networks induced by a left frontal glioma. Cerebro, 137(4),
1193–1212. https://doi.org/10.1093/ brain/awu013, PubMed:
24519977

reorganization of

Kotz, S. A., Cappa, S. F., por cramon, D. y., & Friederici, A. D.
(2002). Modulation of the lexical-semantic network by auditory
semantic priming: An event-related functional MRI study. Neuro-
Image, 17(4), 1761–1772. https://doi.org/10.1006/nimg.2002
.1316, PubMed: 12498750

Kotz, S. A., D’Ausilio, A., Raettig, T., Begliomini, C., Craighero, l.,
Fabbri-Destro, METRO., Zingales, C., Haggard, PAG., & Fadiga, l. (2010).
Lexicality drives audio-motor transformations in Broca’s area.
Brain and Language, 112(1), 3–11. https://doi.org/10.1016/j
.bandl.2009.07.008, PubMed: 19698980

Kristinsson, S., Thors, h., Yourganov, GRAMO., Magnusdottir, S., Hjaltason,
h., Rígido, B. C., Basilakos, A., den Ouden, D. B., Bonilha, l.,
Rorden, C., Hickok, GRAMO., Hillis, A., & Fridriksson, j. (2020). Cerebro
damage associated with impaired sentence processing in acute
aphasia. Revista de neurociencia cognitiva, 32(2), 256–271.
https://doi.org/10.1162/jocn_a_01478, PubMed: 31596169

Lambon Ralph, METRO. A., Jefferies, MI., Patterson, K., & Rogers, t. t.
(2017). The neural and computational bases of semantic cogni-
ción. Naturaleza Reseñas Neurociencia, 18(1), 42–55. https://doi.org
/10.1038/nrn.2016.150, PubMed: 27881854

Leff, A. PAG., Schofield, t. METRO., Crinion, j. T., Seghier, METRO. l., Grogan, A.,
Verde, D. w., & Precio, C. j. (2009). The left superior temporal
gyrus is a shared substrate for auditory short-term memory and
speech comprehension: Evidencia de 210 patients with stroke.
Cerebro, 132(12), 3401–3410. https://doi.org/10.1093/ brain
/awp273, PubMed: 19892765

Amar, T., Haist, F., Nicol, J., & Swinney, D. (2006). A functional
neuroimaging investigation of the roles of structural complexity
and task-demand during auditory sentence processing. Corteza,
42(4), 577–590. https://doi.org/10.1016/s0010-9452(08)70396-4,
PubMed: 16881268

Lukic, S., Thompson, C. K., Barbieri, MI., Chiappetta, B., Bonakdarpour,
B., Kiran, S., Rapp, B., Parrish, t. B., & Caplan, D. (2021).
Common and distinct neural substrates of sentence production
and comprehension. NeuroImagen, 224, Article 117374. https://
doi.org/10.1016/j.neuroimage.2020.117374, PubMed:
32949711

Magnusdottir, S., Fillmore, PAG., den Ouden, D. B., Hjaltason, h.,
Rorden, C., Kjartansson, o., Bonilha, l., & Fridriksson, j.
(2013). Damage to left anterior temporal cortex predicts impair-
ment of complex syntactic processing: A lesion-symptom map-
ping study. Mapeo del cerebro humano, 34(10), 2715–2723. https://
doi.org/10.1002/hbm.22096, PubMed: 22522937

Matchin, w., Hammerly, C., & Lau, mi. (2017). The role of the IFG
and pSTS in syntactic prediction: Evidence from a parametric
study of hierarchical structure in fMRI. Corteza, 88, 106–123.
https://doi.org/10.1016/j.cortex.2016.12.010, PubMed:
28088041

Mazoyer, B. METRO., Tzourio, NORTE., Frak, v., Syrota, A., Murayama, NORTE.,
Levrier, o., Salamon, GRAMO., Dehaene, S., cohen, l., & Mehler, j.
(1993). The cortical representation of speech. Journal of Cogni-
tive Neuroscience, 5(4), 467–479. https://doi.org/10.1162/jocn
.1993.5.4.467, PubMed: 23964919

Neurobiology of Language

342

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

Mesulam, METRO. METRO., Wieneke, C., Hurley, r., Rademaker, A., Thompson,
C. K., Weintraub, S., & Rogalski, mi. j. (2013). Words and objects at
the tip of the left temporal lobe in primary progressive aphasia.
Cerebro, 136(2), 601–618. https://doi.org/10.1093/brain/aws336,
PubMed: 23361063

Mori, S., Wakana, S., van Zijl, PAG. C. METRO., & Nagae-Poetscher, l. METRO.

(2005). MRI atlas of human white matter (1st ed.). Elsevier.

Nachev, PAG., Coulthard, MI., Jager, h. r., Kennard, C., & Husain, METRO.
(2008). Enantiomorphic normalization of focally lesioned brains.
NeuroImagen, 39(3), 1215–1226. https://doi.org/10.1016/j
.neuroimage.2007.10.002, PubMed: 18023365

Newhart, METRO., Trupe, l. A., Gómez, y., Cloutman, l., Molitoris, j. J.,
davis, C., Leigh, r., Gottesman, R. F., Carrera, D., & Hillis, A. mi.
(2012). Asyntactic comprehension, working memory, and acute
ischemia in Broca’s area versus angular gyrus. Corteza, 48(10),
1288–1297. https://doi.org/10.1016/j.cortex.2011.09.009,
PubMed: 22079684

Oberhuber, METRO., Hope, t. METRO. h., Seghier, METRO. l., Parker Jones, o.,
Prejawa, S., Verde, D. w., & Precio, C. j. (2016). Four functionally
distinct regions in the left supramarginal gyrus support word pro-
cesando. Corteza cerebral, 26(11), 4212–4226. https://doi.org/10
.1093/cercor/bhw251, PubMed: 27600852

Pettigrew, C., & Hillis, A. mi. (2014). Role for memory capacity in
sentence comprehension: Evidence from acute stroke. Aphasiol-
ogia, 28(10), 1258–1280. https://doi.org/10.1080/02687038.2014
.919436, PubMed: 25221377

Pexman, PAG. METRO., Muraki, MI., Sidhu, D. METRO., Siakaluk, PAG. D., & Yap,
METRO. j. (2019). Quantifying sensorimotor experience: Body-object
interaction ratings for more than 9,000 English words. Comportamiento
Research Methods, 51(2), 453–466. https://doi.org/10.3758
/s13428-018-1171-z, PubMed: 30484218

Pillay, S. B., Binder, j. r., Humphries, C., Bruto, W.. l., & Book,
D. S. (2017). Lesion localization of speech comprehension defi-
cits in chronic aphasia. Neurología, 88(10), 970–975. https://
doi.org/10.1212/ WNL.0000000000003683, PubMed:
28179469

Pisoni, A., Mattavelli, GRAMO., Casarotti, A., Comi, A., Riva, METRO., Bello,
l . , & P a p a g n o , C .
( 2 0 1 9 ) . T h e n e u r a l c o r r e l a t e s o f
a u d i t o r y – v e r b a l s h o r t – t e r m m e m o r y : A v o x e l – b a s e d
lesion-symptom mapping study on 103 patients after glioma
removal. Brain Structure and Function, 224(6), 2199–2211.
https://doi.org/10.1007/s00429-019-01902-z, PubMed:
31177297

Riccardi, NORTE., Yourganov, GRAMO., Rorden, C., Fridriksson, J., & Desai,
R. h. (2019). Dissociating action and abstract verb comprehen-
sion post-stroke. Corteza, 120, 131–146. https://doi.org/10.1016/j
.cortex.2019.05.013, PubMed: 31302507

Riccardi, NORTE., Yourganov, GRAMO., Rorden, C., Fridriksson, J., & Desai, R.
(2020). Degradation of praxis brain networks and impaired com-
prehension of manipulable nouns in stroke. Revista de Cognitivo
Neurociencia, 32(3), 467–483. https://doi.org/10.1162/jocn_a
_01495, PubMed: 31682566

Richardson, F. METRO., Ramsden, S., Ellis, C., Burnett, S., Megnin, o.,
Catmur, C., Schofield, t. METRO., Leff, A. PAG., & Precio, C. j. (2011).
Auditory short-term memory capacity correlates with gray mat-
ter density in the left posterior STS in cognitively normal and
dyslexic adults. Revista de neurociencia cognitiva, 23(12),
3746–3756. https://doi.org/10.1162/jocn_a_00060, PubMed:
21568634

Robson, h., Sage, K., & Lambon Ralph, METRO. A. (2012). Wernicke’s
aphasia reflects a combination of acoustic-phonological and
semantic control deficits: A case-series comparison of Wer-
nicke’s aphasia, semantic dementia and semantic aphasia.

Neuropsicología, 50(2), 266–275. https://doi.org/10.1016/j
.neuropsychologia.2011.11.021, PubMed: 22178742

Rogalsky, C., & Hickok, GRAMO. (2009). Selective attention to semantic
and syntactic features modulates sentence processing networks
in anterior temporal cortex. Corteza cerebral, 19(4), 786–796.
https://doi.org/10.1093/cercor/bhn126, PubMed: 18669589
Rogalsky, C., & Hickok, GRAMO. (2011). The role of Broca’s area in sen-
tence comprehension. Revista de neurociencia cognitiva, 23(7),
1664–1680. https://doi.org/10.1162/jocn.2010.21530, PubMed:
20617890

Rogalsky, C., LaCroix, A. NORTE., Chen, k. h., anderson, S. w., Damasio,
h., Amar, T., & Hickok, GRAMO. (2018). The neurobiology of agram-
matic sentence comprehension: A lesion study. Journal of Cog-
nitive Neuroscience, 30(2), 234–255. https://doi.org/10.1162
/jocn_a_01200, PubMed: 29064339

Rogalsky, C., Matchin, w., & Hickok, GRAMO. (2008). Broca’s area, sen-
tence comprehension, and working memory: An fMRI Study.
Frontiers in Human Neuroscience, 2, Article 14. https://doi.org
/10.3389/neuro.09.014.2008, PubMed: 18958214

Rorden, C., Bonilha, l., Fridriksson, J., Bender, B., & Karnath, h. oh.
(2012). Age-specific CT and MRI templates for spatial normaliza-
ción. NeuroImagen, 61(4), 957–965. https://doi.org/10.1016/j
.neuroimage.2012.03.020, PubMed: 22440645

Fallar, I., Blumstein, S. MI., miers, mi. B., & hutchison, mi. (2008).
R e c r u i t m e n t o f a n t e r i o r a n d p o s t e r i o r s t r u c t u r e s i n
lexical-semantic processing: An fMRI study comparing implicit
and explicit tasks. Brain and Language, 105(1), 41–49. https://
doi.org/10.1016/j.bandl.2008.01.003, PubMed: 18279947

Salis, C. (2012). Short-term memory treatment: Patterns of learning
and generalisation to sentence comprehension in a person with
aphasia. Neuropsychological Rehabilitation, 22(3), 428–448.
https://doi.org/10.1080/09602011.2012.656460, PubMed:
22300398

Schell, METRO., Zaccarella, MI., & Friederici, A. D. (2017). Differential
cortical contribution of syntax and semantics: Un estudio de resonancia magnética funcional
on two-word phrasal processing. Corteza, 96, 105–120. https://
doi.org/10.1016/j.cortex.2017.09.002, PubMed: 29024818

Sliwinska, METRO. w., Khadilkar, METRO., Campbell-Ratcliffe, J., Quevenco,
F., & Devlin, j. t. (2012). Early and sustained supramarginal gyrus
contributions to phonological processing. Frontiers in Psychol-
ogia, 3, Article 161. https://doi.org/10.3389/fpsyg.2012.00161,
PubMed: 22654779

Snijders, t. METRO., Vosse, T., Kempen, GRAMO., Van Berkum, j. J., Petersson,
k. METRO., & Hagoort, PAG. (2009). Retrieval and unification of syntactic
structure in sentence comprehension: An fMRI study using
word-category ambiguity. Corteza cerebral, 19(7), 1493–1503.
https://doi.org/10.1093/cercor/bhn187, PubMed: 19001084
Thothathiri, METRO., Kimberg, D. y., & Schwartz, METRO. F. (2012). The neu-
ral basis of reversible sentence comprehension: Evidencia de
voxel-based lesion symptom mapping in aphasia. Journal of Cog-
nitive Neuroscience, 24(1), 212–222. https://doi.org/10.1162
/jocn_a_00118, PubMed: 21861679

Turken, A. Ud., & Dronkers, norte. F. (2011). The neural architecture of
the language comprehension network: Converging evidence
from lesion and connectivity analyses. Frontiers in Systems Neu-
roscience, 5, Article 1. https://doi.org/10.3389/fnsys.2011.00001,
PubMed: 21347218

Veraart, J., Fieremans, MI., & Nóvikov, D. S. (2016). Diffusion MRI
noise mapping using random matrix theory. Resonancia magnetica
En medicina, 76(5), 1582–1593. https://doi.org/10.1002/mrm
.26059, PubMed: 26599599

Wakana, S., Jiang, h., Nagae-Poetscher, l. METRO., van Zijl, PAG. C. METRO., &
Mori, S. (2004). Fiber tract-based atlas of human white matter

Neurobiology of Language

343

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu
norte
oh

yo
/

a
r
t
i
C
mi
–
pag
d

F
/

3
2
3
1
8
2
0
0
8
4
8
2
norte
oh
_
a
_
0
0
0
6
7
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Canonical sentence processing and the inferior frontal cortex

anatomy. Radiology, 230(1), 77–87. https://doi.org/10.1148
/radiol.2301021640, PubMed: 14645885

Walenski, METRO., Europa, MI., Caplan, D., & Thompson, C. k. (2019).
Neural networks for sentence comprehension and production:
An ALE-based meta-analysis of neuroimaging studies. Humano
Mapeo cerebral, 40(8), 2275–2304. https://doi.org/10.1002/hbm
.24523, PubMed: 30689268

Whitney, C., Kirk, METRO., O'Sullivan, J., Lambon Ralph, METRO. A., & Jefferies,
mi. (2011). The neural organization of semantic control: TMS evi-
dence for a distributed network in left inferior frontal and posterior
middle temporal gyrus. Corteza cerebral, 21(5), 1066–1075. https://
doi.org/10.1093/cercor/bhq180, PubMed: 20851853

wilson, S. METRO., DeMarco, A. T., Henry, METRO. l., Gesierich, B., Babiak,
METRO., Molinero, B. l., & Gorno-Tempini, METRO. l. (2016). Variable disrup-
tion of a syntactic processing network in primary progressive

aphasia. Cerebro, 139(11), 2994–3006. https://doi.org/10.1093
/brain/aww218, PubMed: 27554388

Winkler, A. METRO., Ridgway, GRAMO. r., Webster, METRO. A., Herrero, S. METRO., &
Nichols, t. mi. (2014). Permutation inference for the general linear
modelo. NeuroImagen, 92, 381–397. https://doi.org/10.1016/j
.neuroimage.2014.01.060, PubMed: 24530839

Zaccarella, MI., & Friederici, A. D. (2015). Merge in the human
cerebro: A sub-region based functional investigation in the left pars
opercularis. Fronteras en psicología, 6, Article 1818. https://doi
.org/10.3389/fpsyg.2015.01818, PubMed: 26640453

Zaccarella, MI., Schell, METRO., & Friederici, A. D. (2017). Reviewing the
functional basis of the syntactic merge mechanism for language: A
coordinate-based activation likelihood estimation meta-analysis.
Neurociencia & Revisiones de biocomportamiento, 80, 646–656. https://doi
.org/10.1016/j.neubiorev.2017.06.011, PubMed: 28743620