Brain Topogr (2016) 29:94–107
DOI 10.1007/s10548-015-0427-5
ORIGINAL PAPER
Involvement of the Motor System in Comprehension of Non-
Literal Action Language: A Meta-Analysis Study
Jie Yang • Hua Shu
Received: 5 December 2013 / Accepted: 5 February 2015 / Published online: 14 February 2015
Ó Springer Science+Business Media New York 2015
Abstract Although numerous studies have shown that
the sensory-motor system is involved in semantic pro-
cessing of language stimuli, it is still unclear whether
comprehension of abstract concepts is embodied, and
whether the involvement of the sensory-motor system is
context-dependent. Investigation of how the motor system
is activated during comprehension of non-literal action
languages can help address these issues. So far several
studies have reported brain activations during non-literal
action language comprehension, but the findings are highly
inconsistent because of different types of non-literal action
language stimuli. To clarify how the motor system is in-
volved in comprehension of different types of non-literal
languages, the current study conducted quantitative meta-
analyses on fMRI findings about comprehension of sen-
tences describing fictive motions, metaphoric actions, and
idiomatic actions. Results showed that fictive motion sen-
tences elicited activation in the right parahippocampal
gyrus, an area important for spatial processing. For meta-
phoric actions, the left precentral gyrus (BA 6) was
strongly activated, suggesting a link between metaphoric
and literal meanings. For idiomatic actions, activity was
found in the left inferior frontal gyrus (BA 44/45), high-
lighting semantic selection and inhibition. No premotor or
motor activity was found in idiom condition. These results
together suggest that the involvement of the sensory-motor
J. Yang (&)
ARC Center of Excellence in Cognition and its Disorders,
Department of Cognitive Science, Macquarie University,
Sydney, Australia
e-mail: j.yang@mq.edu.au
H. Shu
National Key Laboratory of Cognitive Neuroscience and
Learning, Beijing Normal University, Beijing, China
123
system in abstract concepts processing is flexible, de-
pending on semantic features of the language stimuli and
links between abstract and literal meanings.
Keywords The motor system
Fictive motion

Metaphoric action
Idiomatic action
Meta-analysis
Introduction
The involvement of the sensory-motor system during lan-
guage comprehension has attracted increasing theoretical
and experimental interests (e.g., Barsalou 2008; Gallese
and Lakoff 2005; Kiefer and Pulvermüller 2012; Pul-
vermüller and Fadiga 2010). Studies using neuroimaging
techniques have found that comprehension of language
stimuli that contain sensory or action-related semantic
features could elicit activations in brain regions involved in
sensory and motor processes (e.g., Hauk et al. 2004;
Kemmerer et al. 2008; Pulvermüller et al. 2005, but see de
Zubicaray et al. 2010; Fernandino and Iacoboni 2010;
Postle et al. 2008). These findings have evoked a hot debate
about whether and how language comprehension is
grounded in the sensory-motor system (e.g., Caramazza
et al. 2014; Hauk and Tschentscher 2013; Mahon and
Caramazza 2008, 2009; Meteyard et al. 2012; Tomasino
and Rumiati 2013).
The view that the sensory-motor system is involved in
the comprehension of concrete concepts has been sup-
ported not only by fMRI findings, but also by studies that
use neurophysiological methods (EEG, MEG, and TMS)
which indicate that the motor system is involved in a time
window of early lexical-semantic processing (e.g., for a
review, see Hauk et al. 2008; but see Papeo et al. 2009,
2013, 2014). However, whether comprehension of abstract
Brain Topogr (2016) 29:94–107
concepts far away from sensory and motor experiences also
relies on the sensory-motor system is still unclear.
Another unresolved issue is about how the sensory-
motor system is involved in language comprehension, i.e.
whether conceptual processing evokes a complete recre-
ation of perceptual or motor experiences, or it just high-
lights the selective perceptual and motor attributes
depending on the context (Chatterjee 2010). A complete
recreation of sensory and motor experiences would imply
an automatic recruitment of the sensory and motor regions
independent of context (Boulenger et al. 2009; Hauk et al.,
2004; Pulvermüller 2005; Shtyrov et al. 2004; Tettamanti
et al. 2005), while a selective activation of sensory and
motor experiences would imply the flexibility of sensory-
motor activity during conceptual processing (e.g., Hoenig
et al. 2008; Raposo et al. 2009; Rüschemeyer et al. 2007;
van Dam et al. 2010; Van Elk et al. 2008, 2010). Since the
evidence about the embodiment of abstract conceptual
processing is so scarce, it is possible that comprehension of
concrete and abstract concepts might involve the sensory-
motor system with different degrees.
If we put the two issues together, it is quite clear that
they are tightly related to each other. A graded embodiment
for language comprehension means contexts can influence
the involvement of the sensory-motor system. For instance,
if the comprehension of abstract concepts is less embodied
than the comprehension of concrete concepts, then it shows
that the involvement of the sensory-motor system is flex-
ible and context-dependent.
A valid method to address the two issues together is to
examine whether the sensory-motor system is involved in
the comprehension of non-literal language containing
words relating to actions or motions (e.g, grasp the idea). In
this example, the hand verb ‘‘grasp’’ is presented, but the
context does not support its literal meaning. Instead, the
phrase expresses an abstract meaning ‘‘understand the
idea’’. Is the motor system involved when we comprehend
such non-literal language? If yes, then which part of the
motor system is involved? So far there is increasing evi-
dence about this topic, but the findings are highly
inconsistent.
A main factor causing the inconsistences is the various
types of non-literal action language stimuli used in these
studies, including sentences describing fictive motion,
metaphoric action, and idiomatic action. Fictive motion is
an expression in which a motion verb is used to convey a
non-concrete, static state of objects or abstract concepts
through space (Matlock 2004; Talmy 2000). In a fictive
motion sentence, the subject is usually inanimate and in-
capable of performing the motion (e.g., the road runs along
the coast). Since fictive motions express spatial relations
which are linked with the described motions, one might
expect brain activity in regions involved in spatial
95
encoding, such as parahippocampal gyrus (Epstein and
Kanwisher 1998). In addition, regions involved during vi-
sual perception of motions, such as MT area, might also be
activated (e.g., Saygin et al. 2010; Wallentin et al. 2005a).
Activity in the premotor and motor areas observed during
human action processing might not be necessary, because
verbs in fictive motion sentences can express both human
motions and animal motions (i.e. animate motions, Wal-
lentin et al. 2005a).
So far, several studies have examined neural substrates
of fictive motion comprehension. For example, Wallentin
et al. (2005b) found that sentences about real motions (e.g.,
the man goes through the house) and fictive motions (e.g.,
the trail goes through the house) elicited strong activity in
the bilateral posterior regions, such as the fusiform gyrus,
the parahippocampal gyrus, the occipital-temporal-parietal
junction, the left precuneus, and the right postcentral gyrus.
Comparison between the two types of sentences revealed
an effect in the right parietal lobe related to motion pro-
cessing, but did not reach the significance level. This
suggests that processing sentences about fictive motions
and normal motions might involve similar mechanisms
related to mental navigation and spatial memory. In Saygin
et al. (2010), region of interest analysis showed that both
normal motion sentences (e.g., I drove from Modesto to
Fresno) and fictive motion sentences (e.g., the highway
runs from Modesto to Fresno) elicited higher activity in the
bilateral MT areas than static sentences (e.g., Modesto and
Fresno are in California). Furthermore, normal motion
sentences elicited a higher effect than fictive motion sen-
tences did. This finding indicates that semantic processing
of fictive motion still involves simulation of visual per-
ception of motion, but to a lesser extent compared with
semantic processing of normal motion sentences.
Other studies showed that fictive motion comprehension
involves motor cortex. For instance, Cacciari et al. (2011)
found that when single pulse-transcranial magnetic
stimulation (TMS) was delivered at the end of each sen-
tence over the leg area in the primary motor cortex, motor
evoked potentials (MEPs) from the right gastrocnemius and
tibialis anterior muscles for fictive motion sentences (e.g.,
the road runs along the impetuous river) were not different
from that for literal motion sentences (e.g., the man runs in
the beautiful country). This suggests that fictive sentences
still preserve the motor component of the motion verbs that
involve the motor system during comprehension.
Different from fictive motions that express static spatial
relations, metaphoric actions use action verbs to indicate
‘‘a higher level of abstraction to refer to any instance of
goal-driven conjoint motion’’ (Cacciari et al. 2011). For
instance, in the sentence ‘‘the woman runs with her fantasy
often’’, the verb ‘‘run’’ is used metaphorically to express a
mental state. The figurative meaning is more abstract than
123
96
the literal meaning, but there is still a link between the two
domains. Thus, metaphoric actions could maintain the
original meaning of the constituent words, and the figura-
tive meaning can be flexible, depending on context infor-
mation (Cacciari et al. 2011). In addition, the subjects of
metaphoric action sentences are often human beings (e.g.,
Desai et al. 2011, 2013; Lacey et al. 2012; Lauro et al.
2013). Based on these features, one might expect that
comprehending metaphoric actions would activate premo-
tor and motor cortex.
Findings about metaphoric action comprehension are
inconsistent. Some studies indicate motor activation during
comprehension of metaphoric actions. Lauro et al. (2013)
indicated that the left precentral gyrus was activated during
comprehension of metaphoric actions (e.g., Matilde throws
her sadness far away). Cacciari et al. (2011) found that
MEPs recorded in the right gastrocnemius and tibialis an-
terior muscles were larger when participants were pre-
sented with metaphorical sentences than with idiomatic and
mental-state sentences. No difference was found between
the MEPs in metaphoric sentences and in literal sentences.
However, some studies did not find premotor or motor
activity for comprehension of metaphoric actions. For in-
stance, Aziz-Zadeh et al. (2006) studied comprehension of
metaphor and literal phrases containing hand, mouth and
foot action verbs. Results showed that within the premotor
cortex in the left hemisphere, a somatotopic activation
pattern was found for literal phrases, but not for metaphor
phrases. Chen et al. (2008) showed that compared with
non-motive sentences, predictive metaphors (e.g., the man
fell under her spell) indicated a stronger effect in the left
posterior middle temporal gyrus (MTG). Some researchers
found that higher-level motor-related regions are activated
during comprehension of metaphoric actions. For instance,
Desai et al. (2011) indicated that metaphoric actions (e.g.,
the public grasped the idea) activated bilateral anterior
inferior partial lobule (aIPL), which is involved in action
planning. In another study, Desai and colleagues again
found that comprehension of metaphoric actions elicited
stronger effects in the left aIPL, the left supramarginal
gyrus, and the left superior occipital gyrus (SOG) (Desai
et al. 2013).
One factor that might influence the motor activity during
metaphoric action comprehension is familiarity. Aziz-
Zadeh and Damasio (2008) suggested that the lack of the
motor activity in the metaphor condition in Aziz-Zadeh
et al. (2006) could be cause by the over-practice of meta-
phor reading. Once a metaphor becomes well established,
the same network it may have initially might not be acti-
vated any more. For instance, when participants read
‘‘grasping the idea’’ for the first time, the hand motor area
is activated because the motor component of the verb
‘‘grasp’’ is used to access the figurative meaning. When the
123
Brain Topogr (2016) 29:94–107
metaphor is learned, it might no longer rely on these rep-
resentations and thus the involvement of the motor system
is unnecessary. This view is support by Desai et al. (2011),
which indicated that activations in the bilateral central
sulcus, bilateral supplementary area, and the left posterior
superior temporal sulcus (STS) were inversely correlated
with familiarity. This finding suggests that the involvement
the sensory-motor system during metaphor comprehension
is in a graded way, with strong simulation in comprehen-
sion of unfamiliar metaphors. As metaphors become more
familiar, the simulation becomes weaker. However, other
researchers have found that familiar metaphors can activate
sensory and motor areas during comprehension. For in-
stance, in Lacey et al. (2012), participants listened to
sentences containing conventional texture metaphors (e.g.,
she had a rough day), and the authors found that texture
metaphors activated somatosensory cortex in the bilateral
parietal operculum and right inferior precentral sulcus as
compared with literal control sentences (e.g., she had a bad
day). This finding suggests that metaphor comprehension is
grounded in the sensory-motor system, even for the highly
familiar metaphors.
A third type of non-literal action expressions is idioms,
such as ‘‘kick the bucket’’. Different from fictive motions
and metaphoric actions, idiomatic actions usually convey a
unique standardized meaning independent of context in-
formation. Because the figurative meaning is standardized,
the link between the figurative meaning and the literal
meaning (i.e. the motor component of the action verb) is
lost. For instance, it is hard to find the connection between
the action of kicking the bucket and death. Thus, during
idiom comprehension, little or no activity in the premotor
and motor cortex should be observed.
Some studies found motor activity during idiomatic
action comprehension. Boulenger et al. (2009) found so-
matotopic activation along the motor strip (premotor and
motor cortex) was elicited by both literal sentences and
idiomatic sentences (e.g., she kicks the habit). The se-
mantic somatotopy was most significant after the sentence
endings, reflecting that the motor system is involved during
sentence-level processing. However, this finding has been
criticized because it cannot distinguish online motor ac-
tivity from motor imagery after comprehension. To clarify
the effect, Boulenger et al. (2012) used MEG to investigate
the time course of cortical activation during comprehension
of literal and idiomatic sentences. Results showed that
compared with comprehension of literal sentences, com-
prehension of idioms changed activity in anterior fronto-
temporal regions at 150–250 ms after the onset of the
critical words (kick the bucket). In parallel, somatotopic
activation in precentral regions was found for action verbs
in the sentences. The authors claim that both compositional
and abstract context-driven semantic processes contribute
Brain Topogr (2016) 29:94–107
to idiom comprehension. The above evidence strongly
supports an automatic involvement of the motor system
during idiom comprehension. However, many studies did
not find motor activity during idiom processing (Cacciari
et al. 2011; Desai et al. 2013; Raposo et al. 2009; Lauro
et al. 2013; Schuil et al. 2013). For instance, Raposo et al.
(2009) reported motor activity for hand- and foot-action
verbs when verbs were presented in isolation, and to a
lesser extent in literal sentential contexts. When verbs were
presented in idiomatic context, no activation was found in
motor and premotor regions. Schuil et al. (2013) found that
compared with non-word sentences, idiomatic action sen-
tences indicated stronger effects in the left IFG, the left
superior frontal gyrus, and the left cingulate gyrus, but not
in motor areas.
Several studies compared comprehension of different
types of non-literal action sentences, and they showed that
while sentences of fictive motion and metaphoric actions
elicit activity in sensory-motor areas, sentences of id-
iomatic actions do not. Cacciari et al. (2011) found that
when single-TMS was applied to leg motor area, MEPs
were larger for literal action, fictive motion, and meta-
phoric action sentences, but not for idiomatic action and
abstract sentences. Lauro et al. (2013) found that the left
precentral gyrus was activated for verbs in literal and
metaphoric action sentences, but not for verbs in idiomatic
sentences. This is consistent with Desai et al. (2013), which
indicated activity in the left aIPL for metaphoric action
sentences, but not for idiomatic action sentences and ab-
stract sentences. These findings suggest that involvement
of the sensory-motor systems during language compre-
hension could be modulated by the figurativeness of lan-
guage stimuli.
To summarize, different types of non-literal action lan-
guage stimuli have different features, and thus compre-
hension of these language stimuli can involve the motor
system differently. This might cause the inconsistent
findings of brain activity in the previous studies. In the
current study, we conduct meta-analyses on the existing
fMRI findings of comprehension of fictive motions, meta-
phoric actions, and idiomatic actions. The results will help
to clarify how the motor system is involved in the com-
prehension of non-literal action language.
If abstract concepts are grounded in the sensory-motor
system and evoke a complete recreation of perceptual or
motor experiences during comprehension, all three types of
language stimuli will show significant effects in sensory
and motor areas. Specifically, we predict that fictive motion
sentences will show effects in areas related to spatial en-
coding and visual perception of motions (e.g., parahip-
pocampal gyrus and MT), and that metaphoric action
sentences and idiomatic action sentences will show effects
97
in areas related to motor processing (e.g., premotor cortex,
primary motor cortex, and aIPL).
If abstract concepts are grounded in the sensory-motor
system in a flexible way and the embodied representation is
modulated by figurativeness or abstractness, the three types
of language stimuli should elicit different effects in sensory
and motor areas. Specifically, fictive motion sentences will
elicit effects in areas involved in spatial encoding and vi-
sual perception of motions, and metaphoric action sen-
tences will elicit effects in motor areas, but idiomatic
action sentences may elicit little or no effect in sensory and
motor areas. This is because in idiomatic action sentences
the link between the figurative meaning and the literal
meaning (i.e. action conceptual knowledge) is lost. In other
words, idiomatic action sentences have higher abstractness
than the other two types of language stimuli.
Finally, if the representation of abstract concepts is not
embodied, no effects in sensory and motor areas should be
found in all three types of language stimuli.
Methods and Materials
Paper Selection
We searched for papers from the PubMed database (http://
www.ncbi.nlm.nih.gov/pubmed/) using several keywords,
including ‘fMRI’ to indicate neuroimaging technique;
‘fictive motion’, ‘metaphoric action’ and ‘idiomatic action’
to indicate experimental stimuli; and ‘comprehension’ to
indicate task demand. Thirteen papers were identified. The
inclusion criteria are listed as follows:
1. All studies included in the meta-analysis used fMRI
technique and collected data from healthy adults. Studies
that employed other neuroimaging techniques (PET, MEG
and EEG) were excluded to make sure that all included
studies have approximately comparable spatial resolution.
Additionally, studies recruiting a patient group and a
control group were included if they reported the brain ac-
tivity of the control group. Two studies were excluded
based on this criterion (Cacciari et al. 2011; Boulenger
et al. 2012).
2. Studies that included at least one condition of non-
literal action language comprehension were selected. Ele-
ven studies were chosen based on this criterion.
3. Studies that reported brain activity related to com-
prehension of non-literal action language from whole-brain
analyses and gave standard coordinates for activation foci
in either Talairach (Talairach and Tournoux, 1988) or
Montreal Neurologic Institute (MNI) spaces were selected.
Three studies were excluded based on this criterion (Desai
et al. 2011; Raposo et al. 2009; Wallentin et al. 2005a).
123
98
Eight studies met all the criteria and were included in
the meta-analysis. The time span of these studies is from
2005 to 2013. Table 1 indicates information of all the
studies.
Selection of Contrasts
In all the included studies, contrasts between a non-literal
action language condition and a control condition (e.g., a
literal action condition, rest, an abstract literal condition, or
a non-word condition) were selected as critical contrasts.
These critical contrasts were divided into three groups. The
first group contains contrast between a fictive motion
condition and a control condition. Contrasts from three
studies were in this group and 11 foci were found. The
second group contains contrast between a metaphoric ac-
tion condition and a control condition. Contrasts from four
studies were in this group and 13 foci were found. The third
group contains contrast between anidiomatic action con-
dition and a control condition. Contrasts from four studies
were in this group and 56 foci were found.
ALE Meta-Analyses
In the current study, ALE meta-analysis was performed
with Ginger ALE 2.1.1 software (http://www.brainmap.
org/ale/) (Eickhoff et al. 2009) and conducted in the stan-
dard Talairach space. If coordinates of foci were reported
in the standard MNI space, we transformed the coordinates
into standard Talairach space using a linear transformation
(Lancaster et al. 2007).
ALE method considers each included focus as a center
of probability distribution, rather than a point in the voxel-
scaled brain. In this way, each reported coordinate indi-
cates activation information in a dimensional way (i.e. with
some possibilities) rather than in a binary way (i.e. the
activation is here or not).
First, ALE modelled the included foci as centers of 3D
Gaussian probability distribution that could capture two
types of spatial uncertainty associated with each focus
(Eickhoff et al. 2009). One type of uncertainty is caused by
between-subject variances, and the other type of uncer-
tainty is caused by between-laboratory variances. Different
from the original ALE approach in which the included foci
were modelled by Gaussian probability distributions of
identical, user specified width (e.g., Laird et al. 2005;
Turkeltaub et al. 2002), the current method to select the
parameters of the probability distribution (i.e. width) was
empirically determined based on the two types of vari-
ances. Furthermore, the width of the spatial uncertainty of
each focus was adjusted by the number of participants in
each study: the more participants in a study, the bigger the
123
Brain Topogr (2016) 29:94–107
likelihood of true activation located in the foci reported in
the study.
Second, the probability distributions of all included foci
in each individual study were merged to generate a mod-
elled activation map. After that, a final ALE map con-
taining voxel-wise ALE scores was generated by merging
the activation maps of all included studies. This analysis is
restricted to voxels with a minimum 10 % probability for
grey matter, based on ICBM tissue probability maps
(Evans et al. 1994). A random-effects analysis (i.e. testing
for the convergence of all included experiments) was
conducted to identify the significant ALE effect (Eickhoff
et al. 2009). This random-effects analysis allows conclu-
sions to be generalized to studies not included in the meta-
analysis (Penny and Holmes 2006).
Third, the computed ALE map was compared with a
null distribution to distinguish true and random conver-
gences (i.e. a random spatial association between the
modelled activation maps). The comparison identified
voxels in which activation maps converged more robustly
than expected in the null distribution. The threshold for the
ALE map was calculated based on the generated p-values
with an false discovery rate (FDR) algorithm from Tom
Nichols’s website (http://www.sph.umich.edu/*nichols/
FDR/) and Genovese et al. (2002). An FDR was chosen for
a significance level of p \ 0.05 with minimum volume of
50 mm3. A cluster analysis was conducted on the thresh-
olded ALE map and the anatomical label of each cluster
location was provided.
Results
Brain Activations for Fictive Motion Comprehension
A total of seven clusters were identified for fictive motion
comprehension by the ALE meta-analysis. The clusters
were located in the left middle frontal gyrus (MFG) (BA 6),
left middle temporal gyrus (MTG) (BA 19), left middle
occipital gyrus (MOG) (BA 19), left posterior cingulate
cortex (PCC) (BA 30), right superior occipital gyrus (SOG)
(BA 19), right PCC (BA 30), and right parahippocampal
gyrus (BA 36) (Table 2; Fig. 1a).
Brain Activations for Metaphoric Action
Comprehension
A total of four clusters were identified for metaphoric ac-
tion comprehension. The clusters were all located in the
left hemisphere, including the precentral gyrus (BA 6),
MTG (BA 21), inferior parietal lobule (IPL) (BA 40), and
superior parietal lobule (SPL) (BA 7) (Table 2; Fig. 1b).
 Table 1 A list of studies that investigate brain responses during comprehension of non-literal action languages
Study Technique Stimuli Task Baseline Contrast involving non-literal Subject
conditions

Fictive motion
Cacciari et al. (2011) TMS Metaphoric action sentences Silent reading Abstract sentences N/A 9
Idiomatic action sentences
Fictive motion sentences
Literal action sentences
Brain Topogr (2016) 29:94–107

Abstract sentences
Lauro et al. (2013)* fMRI Metaphoric action sentences
Idiomatic action sentences
Fictive motion sentences
Literal action sentences
Abstract sentences
Saygin et al. (2010)* fMRI Literal motion sentences
Fictive motion sentences
Static sentences
Anomalous sentences
Wallentin et al. (2005a) fMRI Literal action sentences
Fictive action sentences
Metaphoric action sentences
Anomalous sentences
Wallentin et al. fMRI Animate motion sentences
(2005b)* Fictive motion sentences
Animate static sentences
Inanimate static sentences
Metaphoric action
Aziz-Zadeh et al. fMRI Metaphoric action phrases
(2006)* Literal action phrases
Chen et al. (2008)* fMRI Metaphoric action sentences
congruency judgment
Sentence comprehension
Sentence comprehension
Sentence comprehension
Silent reading
Plausibility judgment
Abstract sentences All experimental versus abstract 24
Idiom versus abstract
Metaphor versus abstract
Fictive versus
(idiom ? metaphor)
Literal versus
(idiom ? metaphor)
(Idiom ? metaphor ? fictive)
versus (literal ? Abstract)
Idiom versus metaphor versus
fictive
(Idiom ? metaphor) versus
fictive
Static sentences Fictive versus static 16
N/A (metaphor ? anomalous) versus 18
(literal ? fictive)
Meaningless sentences Motion versus control 15
Rest Metaphor versus rest 12
Non-motive sentences Metaphor versus non-motive 14
99

123
 Table 1 continued
Study Technique Stimuli
123
Literal action sentences
Non-motive sentences
Desai et al. (2011) fMRI Metaphoric action sentences
Literal action sentences
Abstract sentences
Desai et al. (2013)* fMRI Metaphoric action sentences
Idiomatic action sentences
Literal action sentences
Abstract sentences
Cacciari et al. (2011) TMS Metaphoric action sentences
Idiomatic action sentences
Fictive motion sentences
Literal action sentences
Abstract sentences
Lauro et al. (2013)* fMRI Metaphoric action sentences
Idiomatic action sentences
Fictive motion sentences
Literal action sentences
Abstract sentences
Idiomatic action
Boulenger et al. (2009)* fMRI Idiomatic action sentences
Literal action sentences
Boulenger et al. (2012) MEG Idiomatic action sentences
Literal action sentences
Raposo et al. (2009) fMRI Action words
100
Task Baseline Contrast involving non-literal Subject
conditions
Metaphor versus literal
Meaningfulness judgment Abstract sentences Metaphor versus abstract 22
Metaphor versus literal
Meaningfulness judgment Abstract sentences Metaphor versus abstract 27
Idiom versus abstract
Silent reading Abstract sentences N/A 9
Congruency judgment Abstract sentences All experimental versus abstract 24
Idiom versus abstract
Metaphor versus abstract
Fictive versus
(idiom ? metaphor)
Literal versus
(idiom ? metaphor)
(Idiom ? metaphor ? fictive)
versus (literal ? Abstract)
Idiom versus metaphor versus
fictive
(Idiom ? metaphor) versus
fictive
Answering sentence-related Hash-marks Idiom versus hash-marks 18
questions
Idiom versus literal
Answering sentence-related N/A N/A 18
questions
Passive listening (words) Signal -correlated noise Literal versus idiom 22
Brain Topogr (2016) 29:94–107
 Table 1 continued
Study Technique Stimuli
Literal action sentences
Idiomatic action sentences
Non-action words
Non-action sentences
Schuil et al. (2013)* fMRI Literal action sentences
Idiomatic action sentences
Desai et al. (2013)* fMRI Metaphoric action sentences
Idiomatic action sentences
Literal action sentences
Abstract sentences
Cacciari et al. (2011) TMS Metaphoric action sentences
Idiomatic action sentences
Fictive motion sentences
Literal action sentences
Abstract sentences
Lauro et al. (2013)* fMRI Metaphoric action sentences
Idiomatic action sentences
Fictive motion sentences
Literal action sentences
Abstract sentences
* The studies included in the current meta-analysis
123
Task Baseline Contrast involving non-literal Subject
conditions
Sentence-word association judgment Non-action words All sentences versus noise
(sentence)
Non-action sentences
Brain Topogr (2016) 29:94–107
Word-sentence judgement Non-word sentences Idiom versus non-word 20
Idiom versus literal
Meaningfulness judgment Abstract sentences Metaphor versus abstract 27
Idiom versus abstract
Silent reading Abstract sentences N/A 9
Congruency judgment Abstract sentences All experimental versus abstract 24
Idiom versus abstract
Metaphor versus abstract
Fictive versus
(idiom ? metaphor)
Literal versus
(idiom ? metaphor)
(Idiom ? metaphor ? fictive)
versus (literal ? Abstract)
Idiom versus metaphor versus
fictive
(Idiom ? metaphor) versus
fictive
101
102 Brain Topogr (2016) 29:94–107

Table 2 Meta-analysis results Volume (mm3) X Y Z Hemisphere Region BA

of three types of non-literal
action languages Fictive motion
56 -20 0 48 Left Middle frontal gyrus 6
56 -52 -58 -6 Left Middle temporal gyrus 19
64 -34 -78 26 Left Middle temporal gyrus 19
64 -14 -58 14 Left Posterior cingulate 30
56 40 -74 26 Right Superior occipital gyrus 19
64 26 -30 -16 Right Parahippocampal gyrus 36
56 10 -54 10 Right Posterior cingulate 30
Metaphoric action
72 -30 -10 56 Left Precentral gyrus 6
96 -27 -65 48 Left Superior parietal lobule 7
56 -42 -44 46 Left Inferior parietal lobule 40
80 -52 -30 0 Left Middle temporal gyrus 21
Idiomatic action
2096 -46 24 12 Left Inferior frontal gyrus 45
The coordinates indicate the 424 -52 10 18 Left Inferior frontal gyrus 44
location of the maximum ALE 216 -58 -56 4 Left Middle temporal gyrus 21
value

Brain activations for idiomatic action comprehension
Three clusters were found for idiomatic action compre-
hension. The clusters were located in the left inferior
frontal gyrus (IFG) (BA 44/45), and the left MTG (BA 21)
(Table 2; Fig. 1c).
Discussion
The current study aims to clarify whether the motor system
is involved in comprehension of non-literal action language
stimuli. To achieve the goal, we conducted meta-analyses
on previous findings about comprehension of fictive mo-
tions, metaphoric actions, and idiomatic actions. Results
indicated that different types of non-literal language stimuli
elicited different brain activations.
For fictive motion, one interesting result is the activity in
the right parahippocampus gyrus. Parahippocampal gyrus
is a grey matter cortical region surrounding the hip-
pocampus. Previous studies have found that this region is
critically involved in encoding and recognition of spatial
information in humans (Aguirre et al. 1996, 1998;
Eichenbaum and Lipton 2008; Hötting et al. 2013; Mundy
et al. 2013). Particularly, a ‘‘parahippocampal place area’’
within the parahippocampal cortex has been found acti-
vated by spatial scenes but not objects (Epstein and Kan-
wisher 1998). Aguirre et al. (1996) using a virtual maze
task found that during the learning and recall of topo-
graphic information, the parahippocampal gyrus showed
strong activity. Mundy et al. (2013) found that posterior
hippocampus and posterior parahippocampal gyrus showed
stronger activity in scene pattern learning, but not in face
and dot pattern learning. Neuropsychological studies have
shown that damage in parahippocampal gyrus can cause
spatial impairments. For example, Nyffeler et al. (2005)
indicated that a patient with lesions in the right parahip-
pocampal gyrus had impairments in allocentric and ego-
centric visuospatial functions. Harvey and Rossit (2012)
suggested that visuomotor neglect-specific deficits emerge
when tasks require scene-based coordinates or spatial re-
lational metrics, and that these deficits are associated with
damage to occipitotemporal and parahippocampal cortex.
As referred in the introduction, fictive motion sentences
can express static spatial relations between objects and the
environment. During comprehension of fictive motions,
people tend to have visual imagery and spatial navigation
(e.g., Talmy 2000). The activation in the right parahip-
pocampal gyrus revealed by the meta-analysis supports this
view. This result indicates that conceptual processing of
fictive motion is grounded in the sensory-motor system
involved in spatial information encoding and recognition.
In the fictive motion condition, no activity in motor
areas or motion-related visual perception areas was found.
Although the results showed two clusters in the left MTG,
none of them are overlapped with the MT area. In the
frontal lobe, the only activation was found in the left MFG,
which is anterior to the premotor cortex. These results
might suggest that during comprehension of fictive motion
sentences, the motor components of verbs are used to
construct the spatial relations conveyed by the sentences.
Thus, instead of having motor simulation, participants
might conduct simulation of spatial processing. It is also
possible that the inanimate subjects in fictive motion

123
Brain Topogr (2016) 29:94–107
Fig. 1 Results of the meta-
analyses for three types of non-
literal action sentences. a ALE
result for the comprehension of
fictive motion sentences. b ALE
result for the comprehension of
metaphoric action sentences.
c ALE result for the
comprehension of idiomatic
action sentences
sentences prevent participants from conducting motor
simulation. Whether participants conduct motor simulation
within a short time window after the verb onset during
fictive motion comprehension needs further investigations
using techniques with high temporal resolution.
The results of meta-analysis showed that comprehension
of metaphoric action elicited activity in the left precentral
gyrus (BA 6). This indicates that premotor cortex is in-
volved in understanding the figurative meaning, and sug-
gests that in metaphoric actions, a link between a concrete
domain (e.g., motor domain) and an abstract domain (e.g.,
mental-state) exists. This is consistent with the view that
metaphoric actions could maintain the original meaning of
the constituent words, and the figurative meaning can be
flexible, depending on context information (Cacciari et al.
2011). One interesting finding is that the left SPL and IPL
are also significantly activated. Previous studies have found
that IPL and provides a neural mechanism for higher order
cognitive motor functions, including understanding action
intentions (e.g., Fogassi and Luppino 2005). The involve-
ment of the IPL in comprehension of metaphoric actions
might suggest that processing of abstract figurative mean-
ing highlights higher order motor functions, which could be
the link between the motor domain and the abstract do-
main. For example, in the metaphor grasping the idea,
103
knowing the intention of the action grasp (i.e. to get
something the actor needs) might be important for ac-
cessing the figurative meaning (i.e. to understand the idea).
The meta-analysis showed that comprehension of id-
iomatic actions mainly activated the left IFG (BA 44/45)
and the left MTG. No significant effect in premotor or
motor areas was found. In idioms, the figurative meaning is
highly conventional, fixed, and independent of context
(e.g., kick the bucket). Thus, the links between the literal
meaning (i.e. the motor component of the verbs) and the
figurative meaning are lost, and this could be the reason
why no motor activation was observed. Since in idioms the
literal meaning and the figurative meaning can be inde-
pendent of each other, comprehension of idioms requires
selection of figurative meaning and inhibition of literal
meaning. The left IFG (BA 44/45) has been found strongly
involved in such semantic processes (O’Reilly 2010). The
current meta-analysis result is consistent with previous
fMRI findings (e.g., Lauro et al. 2008), which suggest that
idiom comprehension requires a selection process between
literal and figurative meanings, and that the IFG is likely to
support this process.
The brain effects in the fictive motion condition,
metaphoric action condition, and idiomatic action condi-
tion imply that as abstractness of the language stimuli
123
104
increases, the involvement of the sensory-motor system
decreases. This graded effect is consistent with recent
evidence from fMRI studies (Desai et al. 2013; Lauro
et al. 2013), a TMS study (Cacciari et al. 2011), and a
meta-analysis about action comprehension in words and
images (Watson et al. 2013). In the fMRI studies, the
authors find that idiomatic action sentences and abstract
sentences elicited less activation in the action-related ar-
eas than did other non-literal action languages (e.g.,
metaphoric action sentences, fictive motion sentences, and
literal action sentences). In the TMS study, the result
indicates that MEPs recorded from leg muscles were
larger when participants were presented with literal, fic-
tive and metaphorical motion sentences than with id-
iomatic motion sentences and mental sentences. Both
fMRI and TMS evidence show that language stimuli with
higher abstractness involve the sensory-motor system less
as compared with language stimuli with lower abstract-
ness. In the meta-analysis study, the authors found no
significant effects in the premotor and primary motor
areas during lexical-semantic processing of action con-
cepts. Moreover, the comprehension of action images
elicited stronger effects in the bilateral occipital areas,
whereas action verb comprehension showed stronger ef-
fects in the left posterior MTG. The authors suggest that
the differences between action images and action words
reflect that the embodied representation of action concepts
is modulated by a gradient of abstraction. The studies
above and the current meta-analysis together indicate that
the comprehension of abstract concepts is less embodied
than the comprehension of concrete concepts, and that the
involvement of the sensory-motor system is context-
dependent (Chatterjee 2010; Willems and Casasanto
2011).These findings question the view that conceptual
processing evokes a complete recreation of perceptual or
motor experiences independent of language context (e.g.,
Boulenger et al. 2009).
The meta-analysis of all three conditions showed sig-
nificant effects in the left MTG. This result was consistent
with recent research that found that the left MTG plays an
important role in the comprehension of non-literal lan-
guage stimuli (e.g., Rapp et al. 2012). Rapp et al. (2012)
suggest that the left MTG has numerous connections to
other cortical areas, such as the left IFG (BA 45/47), MFG
(BA 46), and IPL (BA 39), and that this region can be
considered as a multimodal association area that holds a
key position in the network for non-literal language com-
prehension. The current meta-analysis effects support this
view. One possible role of the left MTG in non-literal
language processing is that it is involved in selecting cor-
rect meanings (e.g., abstract non-action meanings of fictive
motions, metaphoric actions, and idiomatic actions), sus-
taining them in the working memory during comprehension,
123
Brain Topogr (2016) 29:94–107
and integrating the meanings with context (Turken and
Dronkers 2011). However, it is worth noting that the left
MTG effects in three conditions are different: for fictive
motion sentences and idiomatic action sentences, the effects
were located in the left posterior MTG, whereas for meta-
phoric action sentences, the effect was located in the left
middle MTG. Thus, the specific roles of the left MTG in
each language condition might be slightly different, which
require further investigation.
One limitation of the study is that the number of studies
included in the meta-analysis is relatively small as com-
pared with other meta-analysis studies (Hétu et al. 2013;
McNorgan 2012; Rapp et al. 2012). One reason is that the
embodied representation of non-literal concepts at the brain
level is a new research topic and the neuroimaging evi-
dence is little. Another reason is that the topic focuses on
how the motor system is involved in the comprehension of
non-literal language. The small numbers of individual
studies might weaken the effects of meta-analysis, and one
needs to be cautious when generalizing the conclusions of
the current finding. Future meta-analysis with larger
numbers of studies is needed to provide stronger effects.
Another limitation is that in the contrast selection, the
control conditions are different in the included studies.
They are different in semantic information and other lan-
guage information, and these differences might cause in-
consistency in the reported findings. The complexity of
baseline condition has been investigated by previous meta-
analysis studies in which the numbers of studies permitted
such an analysis (e.g., McNorgan 2012). However, the
limited number of studies in the current meta-analysis
precluded the analysis of non-literal action language pro-
cessing within each baseline condition. Future research is
needed to address the issue.
Taken together, the current meta-analyses indicate that
comprehension of different types of non-literal action
languages involve different neural substrates, depending on
the semantic features of language stimuli. These results
provide answers for the two questions mentioned in the
introduction. The first question is whether semantic pro-
cessing of abstract concepts involves the sensory-motor
system. The current study shows that it depends on the
links between the abstract meaning and the concrete
meaning of the constitutional words. If the links still exist,
such as in fictive motion sentences and metaphoric action
sentences, then the sensory-motor system will be activated;
if the links are lost, such as in idiomatic action sentences,
then the sensory-motor system will not be involved. This
also answers the second question: whether the involvement
of the sensory-motor system is flexible. On the one hand,
the involvement of the sensory-motor system depends on
semantic features of language stimuli. In fictive motion
sentences, the spatial feature is highlighted, and hence the
Brain Topogr (2016) 29:94–107
parahippocampal gyrus is significantly activated. In meta-
phoric action sentences, the figurative meaning is related to
motor features, and thus the premotor cortex is activated.
On the other hand, abstractness and conventionality can
influence the sensory-motor activations. During idiomatic
action comprehension, no activation in sensory or motor
areas was found. This suggests that a higher degree of
abstractness or conventionality will involve the sensory-
motor system to a less extent. It also implies that learning
novel metaphoric or idiomatic actions might elicit dynamic
changes of activity in premotor and motor cortex. If this is
the case, then the functional links between sensory and
motor areas and language areas might not be automatic but
have dynamic changes during language processing. One
direction for future meta-analyses is to examine what fac-
tors can influence the sensory-motor activity during the
comprehension of a certain type of non-literal language
stimuli, which we are not able to do currently due to the
limited number of studies. For instance, it might be im-
portant to clarify whether the comprehension of novel and
conventional metaphoric actions will elicit different brain
activity in sensory and motor areas. Examining how lan-
guage context influences embodied representation will ul-
timately help understand the causality issue in embodied
semantics (i.e. whether the sensory-motor system makes a
contribution to conceptual processing) (Caramazza et al.
2014; Hauk and Tschentscher 2013; Mahon and Caramazza
2008, 2009; Tomasino and Rumiati 2013) and how and
why the causality might change.
References
Aguirre GK, Detre JA, Alsop DC, D’Esposito M (1996) The
parahippocampus subserves topographical learning in man.
Cereb Cortex 6:823–829
Aguirre GK, Zarahn E, D’Esposito M (1998) Neural components of
topographical representation. Proc Natl Acad Sci USA
95:839–846
Aziz-Zadeh L, Damasio A (2008) Embodied semantics for actions:
findings from functional brain imaging. J Physiol Paris
102:35–39. doi:10.1016/j.jphysparis.2008.03.012
Aziz-Zadeh L, Wilson SM, Rizzolatti G, Iacoboni M (2006)
Congruent embodied representations for visually presented
actions and linguistic phrases describing actions. Curr Biol
16:1818–1823
Barsalou LW (2008) Grounded cognition. Annu Rev Psychol
59:617–645
Boulenger V, Hauk O, Pulvermüller F (2009) Grasping ideas with the
motor system: semantic somatotopy in idiom comprehension.
Cereb Cortex 19:1905–1914. doi:10.1093/cercor/bhn217
Boulenger V, Shtyrov Y, Pulvermüller F (2012) When do you grasp
the idea? MEG evidence for instantaneous idiom understanding.
Neuroimage 59:3502–3513. doi:10.1016/j.neuroimage.2011.11.
011
Cacciari C, Bolognini N, Senna I, Pellicciari MC, Miniussi C,
Papagno C (2011) Literal, fictive and metaphorical motion
105
sentences preserve the motion component of the verb: a TMS
study. Brain Lang 119:149–157. doi:10.1016/j.bandl.2011.05.
004
Caramazza A, Anzellotti S, Strnad L, Lingnau A (2014) Embodied
cognition and mirror neurons: a critical assessment. Annu Rev
Neurosci 37:1–15. doi:10.1146/annurev-neuro-071013-013950
Chatterjee A (2010) Disembodying cognition. Lang Cogn 2:79–116
Chen E, Widick P, Chatterjee A (2008) Functional-anatomical
organization of predicate metaphor processing. Brain Lang
107:194–202. doi:10.1016/j.bandl.2008.06.007
de Zubicaray G, Postle N, McMahon K, Meredith M, Ashton R
(2010) Mirror neurons, the representation of word meaning, and
the foot of the third left frontal convolution. Brain Lang
112:77–84
Desai RH, Binder JR, Conant LL, Mano QR, Seidenberg MS (2011)
The neural career of sensory-motor metaphors. J Cogn Neurosci
23:2376–2386. doi:10.1162/jocn.2010.21596
Desai RH, Conant LL, Binder JR, Park H, Seidenberg MS (2013) A
piece of the action: modulation of sensory-motor regions by
action idioms and metaphors. Neuroimage 83:862–869. doi:10.
1016/j.neuroimage.2013.07.044
Eichenbaum H, Lipton PA (2008) Towards a functional organization
of the medial temporal lobe memory system: role of the
parahippocampal and medial entorhinal cortical areas. Hip-
pocampus 18:1314–1324. doi:10.1002/hipo.20500
Eickhoff SB, Laird AR, Grefkes C, Wang LE, Zilles K, Fox PT
(2009) Coordinate-based activation likelihood estimation meta-
analysis of neuroimaging data: a random-effects approach based
on empirical estimates of spatial uncertainty. Hum Brain Mapp
30:2907–2926. doi:10.1002/hbm.20718
Epstein R, Kanwisher N (1998) A cortical representation of the local
visual environment. Nature 392:598–601
Evans AC, Kamber M, Collins DL, MacDonald D (1994) An MRI
based probabilistic atlas of neuroanatomy. In: Shorvon S, Fish D,
Andermann F, Bydder GM (eds) Magnetic resonance scanning
and epilepsy. Springer, pp 263–274
Fernandino L, Iacoboni M (2010) Are cortical motor maps based on
body parts or coordinated actions? Implications for embodied
semantics. Brain Lang 112:44–53
Fogassi L, Luppino G (2005) Motor functions of the parietal lobe.
Curr Opin Neurobiol 15:626–631
Gallese V, Lakoff G (2005) The Brain’s concepts: the role of the
Sensory-motor system in conceptual knowledge. Cogn Neu-
ropsychol 22:455–479. doi:10.1080/02643290442000310
Genovese CR, Lazar NA, Nichols T (2002) Thresholding of statistical
maps in functional neuroimaging using the false discovery rate.
Neuroimage 15:870–878. doi:10.1006/ning.2001.1037
Harvey M, Rossit S (2012) Visuospatial neglect in action. Neuropsy-
chologia 50:1018–1028. doi:10.1016/j.neuropsychologia.2011.
09.030
Hauk O, Tschentscher N (2013) The body of evidence: what can
neuroscience tell us about embodied semantics? Front Psychol
4:50. doi:10.3389/fpsyg.2013.00050
Hauk O, Johnsrude I, Pulvermüller F (2004) Somatotopic represen-
tation of action words in human motor and premotor cortex.
Neuron 41:301–307
Hauk O, Shtyrov Y, Pulvermüller F (2008) The time course of action
and action-word comprehension in the human brain as revealed
by neurophysiology. J Physiol Paris 102:50–58. doi:10.1016/j.
jphysparis.2008.03.013
Hétu S, Grégoire M, Saimpont A, Coll MP, Eugène F, Michon PE,
Jackson PL (2013) The neural network of motor imagery: an
ALE meta-analysis. Neurosci Biobehav Rev 37(5):930–949.
doi:10.1016/j.neubiorev.2013.03.017
Hoenig K, Sim EJ, Bochev V, Herrnberger B, Kiefer M (2008)
Conceptual flexibility in the human brain: dynamic recruitment
123
106
of semantic maps from visual, motor, and motion-related areas.
J Cogn Neurosci 20:1799–1814. doi:10.1162/jocn.2008.20123
Hötting K, Holzschneider K, Stenzel A, Wolbers T, Röder B (2013)
Effects of a cognitive training on spatial learning and associated
functional brain activations. BMC Neurosci 14:73. doi:10.1186/
1471-2202-14-73
Kemmerer D, Castillo JG, Talavage T, Patterson S, Wiley C (2008)
Neuroanatomical distribution of five semantic components of
verbs: evidence from fMRI. Brain Lang 107:16–43
Kiefer M, Pulvermüller F (2012) Conceptual representations in mind
and brain: theoretical developments, current evidence and future
directions. Cortex 48:805–825. doi:10.1016/j.cortex.2011.04.006
Lacey S, Stilla R, Sathian K (2012) Metaphorically feeling: compre-
hending textural metaphors activates somatosensory cor-
tex. Brain Lang 120:416–421
Laird AR, Fox PM, Price CJ, Glahn DC, Uecker AM, Lancaster JL,
Turkeltaub PE, Kochunov P, Fox PT (2005) ALE meta-analysis:
controlling the false discovery rate and performing statistical
contrasts. Hum Brain Mapp 25:155–164
Lancaster JL, Tordesillas-Gutiérrez D, Martinez M, Salinas F, Evans
A, Zilles K, Mazziotta JC, Fox PT (2007) Bias between MNI and
Talairach coordinates analyzed using the ICBM-152 brain
template. Hum Brain Mapp 28:1194–1205
Lauro LJR, Mattavelli G, Papagno C, Tettamanti M (2013) She runs,
the road runs, my mind runs, bad blood runs between us: literal
and figurative motion verbs: an fMRI study. Neuroimage
83:361–371. doi:10.1016/j.neuroimage.2013.06.050
Lauro LJR, Tettamanti M, Cappa SF, Papagno C (2008) Idiom
comprehension: a prefrontal task? Cereb Cortex 18:162–170
Mahon BZ, Caramazza A (2008) A critical look at the embodied
cognition hypothesis and a new proposal for grounding concep-
tual content. J Physiol Paris 102:59–70. doi:10.1016/j.jphysparis.
2008.03.004
Mahon BZ, Caramazza A (2009) Concepts and categories: a cognitive
neuropsychological perspective. Annu Rev Psychol 60:27–51.
doi:10.1146/annurev.psych.60.110707.163532
Matlock T (2004) Fictive motion as cognitive simulation. Mem
Cognit 32:1389–1400
McNorgan C (2012) A meta-analytic review of multisensory imagery
identifies the neural correlates of modality-specific and mod-
ality-general imagery. Front Hum Neurosci 6:285. doi:10.3389/
fnhum.2012.00285
Meteyard L, Cuadrado SR, Bahrami B, Vigliocco G (2012) Coming
of age: a review of embodiment and the neuroscience of
semantics. Cortex 48:788–804. doi:10.1016/j.cortex.2010.11.002
Mundy ME, Downing PE, Dwyer DM, Honey RC, Graham KS (2013)
A critical role for the hippocampus and perirhinal cortex in
perceptual learning of scenes and faces: complementary findings
from amnesia and FMRI. J Neurosci 33:10490–10502. doi:10.
1523/JNEUROSCI.2958-12.2013
Nyffeler T, Gutbrod K, Pflugshaupt T, von Wartburg R, Hess CW,
Müri RM (2005) Allocentric and egocentric spatial impairments
in a case of topographical disorientation. Cortex 41:133–143
O’Reilly RC (2010) The What and How of prefrontal cortical
organization. Trends Neurosci 33:355–361. doi:10.1016/j.tins.
2010.05.002
Papeo L, Vallesi A, Isaja A, Rumiati RI (2009) Effects of TMS on
different stages of motor and non-motor verb processing in the
primary motor cortex. PLoS One 4:e4508. doi:10.1371/journal.
pone.0004508
Papeo L, Pascual-Leone A, Caramazza A (2013) Disrupting the brain
to validate hypotheses on the neurobiology of language. Front
Hum Neurosci 7:148. doi:10.3389/fnhum.2013.00148
Papeo L, Lingnau A, Agosta S, Pascual-Leone A, Battelli L,
Caramazza A (2014). The origin of word-related motor activity.
Cereb Cortex
123
Brain Topogr (2016) 29:94–107
Penny W, Holmes A (2006) Random effects analysis. In: Friston K,
Ashburner J, Kiebel S, Nichols T, Penny W (eds) Statistical
parametric mapping: the analysis of functional brain images.
Elsevier, London, pp 156–165
Postle N, McMahon KL, Ashton R, Meredith M, de Zubicaray GI
(2008) Action word meaning representations in cytoarchitec-
tonically defined primary and premotor cortices. Neuroimage
43:634–644
Pulvermüller F (2005) Brain mechanisms linking language and
action. Nat Rev Neurosci 6:576–582
Pulvermüller F, Fadiga L (2010) Active perception: sensorimotor
circuits as a cortical basis for language. Nat Rev Neurosci
11:351–360. doi:10.1038/nrn2811
Raposo A, Moss HE, Stamatakis EA, Tyler LK (2009) Modulation of
motor and premotor cortices by actions, action words and action
sentences. Neuropsychologia 47:388–396. doi:10.1016/j.neurop
sychologia.2008.09.017
Rapp AM, Mutschler DE, Erb M (2012) Where in the brain is
nonliteral language? A coordinate-based meta-analysis of func-
tional magnetic resonance imaging studies. Neuroimage
63:600–610. doi:10.1016/j.neuroimage.2012.06.022
Rüschemeyer SA, Brass M, Friederici AD (2007) Comprehending
prehending: neural correlates of processing verbs with motor
stems. J Cogn Neurosci 19:855–865
Saygin AP, McCullough S, Alac M, Emmorey K (2010) Modulation
of BOLD response in motion-sensitive lateral temporal cortex by
real and fictive motion sentences. J Cogn Neurosci
22:2480–2490. doi:10.1162/jocn.2009.21388
Schuil KDI, Smits M, Zwaan RA (2013) Sentential context modulates
the involvement of the motor cortex in action language
processing: an FMRI study. Front Hum Neurosci 7:100.
doi:10.3389/fnhum.2013.00100
Shtyrov Y, Hauk O, Pulvermüller F (2004) Distributed neuronal networks
for encoding category-specific semantic information: the mismatch
negativity to action words. Eur J Neurosci 19:1083–1092
Talairach J, Tournoux P (1988) Co-planar stereotaxic atlas of the
human brain. Thieme, New York
Talmy L (2000) Toward a cognitive semantics (2 voll). MIT Press,
Cambridge
Tettamanti M, Buccino G, Saccuman MC, Gallese V, Danna M, Scifo
P, Fazio F, Rizzolatti G, Cappa SF, Perani D (2005) Listening to
action-related sentences activates fronto-parietal motor circuits.
J Cogn Neurosci 17:273–281
Tomasino B, Rumiati RI (2013) At the mercy of strategies: the role of
motor representations in language understanding. Front Psychol
4:27. doi:10.3389/fpsyg.2013.00027
Turkeltaub PE, Eden GF, Jones KM, Zeffiro TA (2002) Meta-analysis
of the functional neuroanatomy of single-word reading: method
and validation. Neuroimage 16:765–780
Turken AU, Dronkers NF (2011) The neural architecture of the
language comprehension network: converging evidence from
lesion and connectivity analyses. Front Syst Neurosci 5:1
van Dam WO, Rüschemeyer SA, Lindemann O, Bekkering H (2010)
Context effects in embodied lexical-semantic processing. Front
Psychol 1:150. doi:10.3389/fpsyg.2010.00150
van Elk M, van Schie HT, Bekkering H (2008) Semantics in action:
an electrophysiological study on the use of semantic knowledge
for action. J Physiol Paris 102:95–100. doi:10.1016/j.jphysparis.
2008.03.011
van Elk M, van Schie HT, Zwaan RA, Bekkering H (2010) The
functional role of motor activation in language processing: motor
cortical oscillations support lexical-semantic retrieval. Neuroim-
age 50:665–677. doi:10.1016/j.neuroimage.2009.12.123
Wallentin M, Lund TE, Ostergaard S, Ostergaard L, Roepstorff A
(2005a) Motion verb sentences activate left posterior middle
temporal cortex despite static context. Neuroreport 16:649–652
Brain Topogr (2016) 29:94–107 107

Wallentin M, Ostergaard S, Lund TE, Ostergaard L, Roepstorff A Willems RM, Casasanto D (2011) Flexibility in embodied language
(2005b) Concrete spatial language: see what I mean? Brain Lang understanding. Front Psychol 2(116):2011. doi:10.3389/fpsyg.
92:221–233 00116
Watson CE, Cardillo ER, Ianni GR, Chatterjee A (2013) Action
concepts in the brain: an activation likelihood estimation meta-
analysis. J Cogn Neurosci 25:1191–1205. doi:10.1162/jocn-a-
00401

123
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.