Abstract

Distributed brain systems contribute to representation of semantic knowledge. Whether sensory and motor systems of the brain are causally involved in representing conceptual knowledge is an especially controversial question. Here, we tested 57 chronic left-hemisphere stroke patients using a semantic similarity judgment task consisting of manipulable and nonmanipulable nouns. Three complementary methods were used to assess the neuroanatomical correlates of semantic processing: voxel-based lesion–symptom mapping, resting-state functional connectivity, and gray matter fractional anisotropy. The three measures provided converging evidence that injury to the brain networks required for action observation, execution, planning, and visuomotor coordination are associated with specific deficits in manipulable noun comprehension relative to nonmanipulable items. Damage or disrupted connectivity of areas such as the middle posterior temporal gyrus, anterior inferior parietal lobe, and premotor cortex was related specifically to the impairment of manipulable noun comprehension. These results suggest that praxis brain networks contribute especially to the comprehension of manipulable object nouns.

INTRODUCTION

Over two decades of research have shown that conceptual knowledge is widely distributed in the brain. Putative “hub” regions, such as the angular gyrus and anterior temporal lobe, are commonly found to contribute to processing a wide variety of concepts (Fernandino, Humphries, Conant, Seidenberg, & Binder, 2016; Binder & Desai, 2011; Binder, Desai, Graves, & Conant, 2009). At the same time, a number of modality-specific areas also appear to contribute to concepts that load heavily on that modality (Kiefer & Pulvermüller, 2012; Meteyard, Cuadrado, Bahrami, & Vigliocco, 2012; Binder & Desai, 2011). This is consistent with the view that sensory–motor systems play an important role in representing conceptual knowledge (Barsalou, 2008; Gallese & Lakoff, 2005). However, causal evidence directly linking disruption of sensory–motor areas to specific conceptual deficits has been somewhat limited, especially in poststroke patient populations (Meteyard et al., 2012). This gap in the literature has led some to argue that sensory–motor networks do not directly contribute to conceptual processing and that they instead play an auxiliary or epiphenomenal role (Mahon, 2015; Caramazza, Anzellotti, Strnad, & Lingnau, 2014).

Brain stimulation methods are one way in which this question has been tackled. TMS studies have found that stimulating motor or premotor areas affects action-related word processing compared with other word types (Reilly, Howerton, & Desai, 2019; Vukovic, Feurra, Shpektor, Myachykov, & Shtyrov, 2017; Cacciari et al., 2011; Willems, Labruna, D'Esposito, Ivry, & Casasanto, 2011; Pulvermüller, Hauk, Nikulin, & Ilmoniemi, 2005). Notably, Oliveri et al. (2004) and Gough et al. (2012) found that TMS to the primary motor cortex affected motor-evoked potentials when retrieving action and graspable words, respectively, compared with nonaction or ungraspable words. These data substantiate a role for motor areas in the processing of action-related words and manipulable objects specifically. However, a recent study using a novel combination of repetitive and single-pulse TMS demonstrated that the effects observed in primary motor cortices may be driven by other brain areas, such as the posterior middle temporal gyrus (pMTG; Papeo et al., 2015). Although the effect was observed only relative to the no-stimulation sham and no effect was observed relative to control occipital stimulation, it is suggestive that pMTG may also play a role in action semantics. As such, the exact contributions of these areas to action-related semantic processing is still an open question.

Studies of patients with neurodegenerative motor disorders have also revealed greater relative deficits to action-related compared with nonaction concepts (Cardona et al., 2014; Fernandino et al., 2013a, 2013b; Ibáñez et al., 2013; Péran et al., 2013; Bak & Chandran, 2012; Grossman et al., 2008; Bak & Hodges, 2004; Péran, Démonet, Pernet, & Cardebat, 2003; Bak, O'Donovan, Xuereb, Boniface, & Hodges, 2001). These studies provide evidence that, as cortical and subcortical action networks deteriorate, processing of action-related concepts are disproportionately impaired. However, some of these studies have directly compared verbs to nouns, leaving open the possibility that the effects could be due to grammatical class and not action or motor properties. Comparing within grammatical class is vital for addressing this possible confound. Furthermore, these studies largely focused on the processing of verbs specifically; the effects of damage to motor networks on manipulable noun comprehension requires further investigation.

In addition, some lesion-deficit association studies have found evidence that damage to specific sensory–motor systems impair comprehension of corresponding concept types (e.g., action or sound related; Riccardi, Yourganov, Rorden, Fridriksson, & Desai, 2019; Desai, Herter, Riccardi, Rorden, & Fridriksson, 2015; Dreyer et al., 2015; Trumpp, Kliese, Hoenig, Haarmeier, & Kiefer, 2013; Arévalo, Baldo, & Dronkers, 2012; Bonner & Grossman, 2012; Kemmerer, Rudrauf, Manzel, & Tranel, 2012; Arévalo et al., 2007; Buxbaum & Saffran, 2002). For example, Desai et al. (2015) used an exoskeleton robot to measure reaching performance in left-hemisphere (LH) stroke patients and found that greater motor impairment was associated with greater relative deficits to manipulable compared with nonmanipulable noun comprehension, measured by a semantic similarity judgment (SSJ) task. Relatedly, Buxbaum and Saffran (2002) demonstrated that, compared with nonapraxic stroke patients, individuals with apraxia were more impaired in semantic knowledge of tool manipulation and body parts, but not other categories, as measured by word and picture versions of an SSJ task. These studies provide evidence for a close relationship between praxis brain networks and knowledge of manipulable objects. However, because these studies associated performance on one domain (manual reaching/tool use) to another (SSJ) without fine-grained anatomical data, the neuroanatomical overlap between praxis and manipulable object comprehension still requires delineation.

Conversely, other lesion studies have not found a relationship between action or manipulable object comprehension and degradation of praxis brain networks (Papeo, Negri, Zadini, & Rumiati, 2010; Mahon et al., 2007; Negri, Rumiati, et al., 2007; Rosci, Chiesa, Laiacona, & Capitani, 2003; Halsband et al., 2001; Rumiati, Zanini, Vorano, & Shallice, 2001). For example, Papeo et al. (2010) demonstrated double dissociations between tool use and a tool word–picture matching task in LH stroke patients. Similarly, Rosci et al. (2003) found that apraxia was not associated with deficits to the processing of manipulable objects, measured by word–picture matching and picture-naming tasks. Although these observations provide evidence of anatomical dissociations between praxis and manipulable object comprehension, some alternative explanations must be considered.

For example, tasks such as word–picture matching and picture naming, although requiring semantics access at some level, do not require explicit retrieval of conceptual features. It is possible that, in some of these cases, deeper probing of semantic knowledge and associated concepts, such as through SSJ tasks, would reveal deficits for manipulable objects. However, even deeper probing of semantic knowledge has occasionally revealed dissociations between semantics and praxis (Mahon, 2015; Garcea, Dombovy, & Mahon, 2013; Negri, Lunardelli, Reverberi, Gigli, & Rumiati, 2007). For these cases, it is important to note that the brain networks underlying semantic knowledge and praxis are complex, and behavioral or anatomical dissociations can be observed if nonoverlapping parts of these networks are damaged. This issue is further complicated by the possibility of poststroke compensation from undamaged parts of these networks (Hillis, Rorden, & Fridriksson, 2017; Price & Friston, 2002), such as perilesional areas or right-hemisphere (RH) homologues. This compensation can mask brain–behavior relations, resulting in Type II error. Therefore, additional measures of network integrity, such as fractional anisotropy (FA) or resting-state functional connectivity (RSFC), could be considered alongside traditional lesion-deficit measures such as voxel-based lesion–symptom mapping (VLSM). To the best of our knowledge, no previous studies have used these measures in conjunction to examine the neural correlates of manipulable object noun comprehension.

Here, we used VLSM, RSFC, and FA analyses in chronic LH stroke patients to examine semantic knowledge of manipulable nouns. Nonmanipulable nouns matched for a variety of psycholinguistic variables served as a control condition to account for global executive/linguistic impairments. We used VLSM and FA to measure macro- and microstructural gray matter integrity, respectively. We also used RSFC in an a priori network of interest (NOI) composed of brain areas commonly associated with praxis to investigate the role of intra- and interhemispheric functional connectivity.

METHODS

Participants

Fifty-seven participants (19 women; Table 1) in the chronic stage (>6 months) of LH unilateral stroke had imaging data for at least one of the three measures used (VLSM, RSFC, FA) and met our inclusion criterion (see the Experimental Design: Behavioral Data section). Participants were at least 6 months poststroke (M = 4.4 years, Min = 0.67 years, Max = 16.83 years), with a mean age at time of testing of 58.92 years (SD = 9.1 years). The mean aphasia quotient (AQ) was 78.84 (SD = 20.74), as measured by the Western Aphasia Battery (WAB). Participants signed informed consent, and the University of South Carolina institutional review board approved the research.

Table 1. 
Patient Information
Participant IDSexAge (years)Time Poststroke (years)WAB-AQManip AccNonManip Acc
LM1001 56.42 5.67 64.6 0.675 0.77 
LM1002** 69.58 4.58 63.6 0.77 0.82 
LM10032 38.57 7.07 55.2 0.61 0.55 
LM10042 54.58 3.92 76.2 0.71 0.89 
LM10052 56.67 10.75 83.2 0.52 0.64 
LM1006 64.75 9.42 80.4 0.68 0.64 
LM1009 59.42 15.42 86.2 0.72 0.89 
LM1010 79.75 6.92 99.1 0.95 0.93 
LM1011 59.67 3.58 99.2 0.98 0.95 
LM1014 56.22 6.72 51.5 0.62 0.47 
LM1015 62.25 5.83 94 0.75 0.58 
LM1016 55.58 6.08 59.4 0.92 0.88 
LM1017 65.33 1.50 52.7 0.77 0.88 
LM1018** 54.33 5.42 88.8 0.92 0.87 
LM1019 69.58 7.83 99.2 0.95 0.98 
LM1021 65.42 2.33 99.6 0.98 0.95 
LM1022 65.50 4.42 93.2 0.89 
LM1023 37.00 7.17 98.5 0.92 
LM1024 41.58 1.67 94.2 0.92 0.82 
LM10252 61.58 4.58 58.2 0.68 0.56 
LM10262 52.00 3.00 57.5 0.78 0.78 
LM1027 75.58 0.83 72.1 0.79 0.87 
LM1029 61.67 1.67 97.3 0.95 0.85 
LM1031 55.33 2.25 91.1 0.58 0.66 
LM1032 59.42 0.92 25.3 0.63 0.69 
LM1034 64.25 1.92 98.6 0.95 0.93 
LM1037 61.00 2.00 93.1 0.95 0.93 
LM1039 48.58 3.25 87.6 0.65 0.68 
LM10402 71.83 0.83 73.5 0.88 0.73 
LM1041 67.25 4.75 77.8 
LM1044 47.75 1.67 87.5 0.81 0.98 
LM1045 53.08 8.58 74.8 0.8 0.74 
LM10462 69.00 16.83 48.9 0.63 0.77 
LM1047 62.75 2.25 98.9 0.95 0.93 
LM1048 69.33 1.08 67.2 0.91 0.94 
LM10492 62.42 7.83 57.2 0.78 0.82 
LM1050 38.92 7.25 96.7 0.68 0.69 
LM1052 51.42 1.08 99.6 0.95 0.95 
LM1053* 72.08 0.92 99.1 0.95 0.95 
LM1054 49.08 8.42 96.6 0.95 0.84 
LM1055 52.08 8.67 64.6 0.62 0.92 
LM1057 59.58 0.67 98.4 0.94 0.91 
LM10582 61.33 4.58 31.2 0.86 0.9 
LM1059 50.75 11.75 43.4 0.69 0.68 
LM1061 64.92 0.83 82.9 0.93 0.95 
LM1063 68.17 1.67 97 0.97 0.92 
LM1064 60.75 2.50 90.1 0.92 0.95 
LM1066 55.33 2.17 91.3 0.85 0.93 
LM10672 67.25 1.00 47.8 0.58 0.65 
LM1071 52.42 2.08 93.4 0.925 
LM10722 48.92 5.92 43 0.63 0.71 
LM1073* 56.08 5.67 74.6 0.77 0.92 
LM10742 48.58 0.92 97.5 0.83 0.85 
LM10752 54.17 0.92 99.2 0.88 0.93 
LM10762 58.67 0.75 99.2 0.98 0.93 
LM10782 63.58 0.83 98.1 0.91 0.97 
LM10832 69.17 2.50 44.7 0.71 0.65 
Participant IDSexAge (years)Time Poststroke (years)WAB-AQManip AccNonManip Acc
LM1001 56.42 5.67 64.6 0.675 0.77 
LM1002** 69.58 4.58 63.6 0.77 0.82 
LM10032 38.57 7.07 55.2 0.61 0.55 
LM10042 54.58 3.92 76.2 0.71 0.89 
LM10052 56.67 10.75 83.2 0.52 0.64 
LM1006 64.75 9.42 80.4 0.68 0.64 
LM1009 59.42 15.42 86.2 0.72 0.89 
LM1010 79.75 6.92 99.1 0.95 0.93 
LM1011 59.67 3.58 99.2 0.98 0.95 
LM1014 56.22 6.72 51.5 0.62 0.47 
LM1015 62.25 5.83 94 0.75 0.58 
LM1016 55.58 6.08 59.4 0.92 0.88 
LM1017 65.33 1.50 52.7 0.77 0.88 
LM1018** 54.33 5.42 88.8 0.92 0.87 
LM1019 69.58 7.83 99.2 0.95 0.98 
LM1021 65.42 2.33 99.6 0.98 0.95 
LM1022 65.50 4.42 93.2 0.89 
LM1023 37.00 7.17 98.5 0.92 
LM1024 41.58 1.67 94.2 0.92 0.82 
LM10252 61.58 4.58 58.2 0.68 0.56 
LM10262 52.00 3.00 57.5 0.78 0.78 
LM1027 75.58 0.83 72.1 0.79 0.87 
LM1029 61.67 1.67 97.3 0.95 0.85 
LM1031 55.33 2.25 91.1 0.58 0.66 
LM1032 59.42 0.92 25.3 0.63 0.69 
LM1034 64.25 1.92 98.6 0.95 0.93 
LM1037 61.00 2.00 93.1 0.95 0.93 
LM1039 48.58 3.25 87.6 0.65 0.68 
LM10402 71.83 0.83 73.5 0.88 0.73 
LM1041 67.25 4.75 77.8 
LM1044 47.75 1.67 87.5 0.81 0.98 
LM1045 53.08 8.58 74.8 0.8 0.74 
LM10462 69.00 16.83 48.9 0.63 0.77 
LM1047 62.75 2.25 98.9 0.95 0.93 
LM1048 69.33 1.08 67.2 0.91 0.94 
LM10492 62.42 7.83 57.2 0.78 0.82 
LM1050 38.92 7.25 96.7 0.68 0.69 
LM1052 51.42 1.08 99.6 0.95 0.95 
LM1053* 72.08 0.92 99.1 0.95 0.95 
LM1054 49.08 8.42 96.6 0.95 0.84 
LM1055 52.08 8.67 64.6 0.62 0.92 
LM1057 59.58 0.67 98.4 0.94 0.91 
LM10582 61.33 4.58 31.2 0.86 0.9 
LM1059 50.75 11.75 43.4 0.69 0.68 
LM1061 64.92 0.83 82.9 0.93 0.95 
LM1063 68.17 1.67 97 0.97 0.92 
LM1064 60.75 2.50 90.1 0.92 0.95 
LM1066 55.33 2.17 91.3 0.85 0.93 
LM10672 67.25 1.00 47.8 0.58 0.65 
LM1071 52.42 2.08 93.4 0.925 
LM10722 48.92 5.92 43 0.63 0.71 
LM1073* 56.08 5.67 74.6 0.77 0.92 
LM10742 48.58 0.92 97.5 0.83 0.85 
LM10752 54.17 0.92 99.2 0.88 0.93 
LM10762 58.67 0.75 99.2 0.98 0.93 
LM10782 63.58 0.83 98.1 0.91 0.97 
LM10832 69.17 2.50 44.7 0.71 0.65 

A superscript of 2 denotes patient data acquired using the second scanning sequence described in the MRI Data Acquisition section. A single asterisk (*) denotes a patient without resting-state data (n = 2), whereas two asterisks (**) denote a patient without resting-state and FA data (n = 2). F = female; M = male.

Materials

An SSJ task was used (Figure 1), consisting of 120 manipulable (“the pen”) and 120 nonmanipulable nouns (“the basement”). The conditions were organized into 40 triplets each, such that in every triplet two of the nouns had similar meanings and a less similar noun served as a distractor. For all 80 trials, three nouns were shown in a triangular format simultaneously. Every triplet was made entirely of manipulable or nonmanipulable nouns. For example, a manipulable triplet would consist of a target item on top (“the shovel”), with the bottom two items being the distractor (“the rake”) and the correct answer (“the spade”).

Figure 1. 

Left: Manipulable SSJ example. Right: Nonmanipulable SSJ example.

Figure 1. 

Left: Manipulable SSJ example. Right: Nonmanipulable SSJ example.

Manipulable and nonmanipulable conditions differed according to their manipulability and body–object interaction (BOI) ratings (Tillotson, Siakaluk, & Pexman, 2008), with the manipulable condition having significantly higher BOI (p < .0001).1 Studies have shown that concepts with high BOI are more graspable than those with low BOI (Heard, Madan, Protzner, & Pexman, 2019; Pexman, Muraki, Sidhu, Siakaluk, & Yap, 2019), demonstrating that BOI is an appropriate measure of manipulability. Noun triads were constructed, based on pilot testing, to have similar task difficulty between conditions. The main goal of the design was to match on the aggregate effect of all variables at the triad level, resulting in similar task difficulty between conditions. The 120 nouns from each condition were still matched for numerous variables, including number of letters, phonemes, age of acquisition, imageability, and semantic diversity (Table 2). They were also matched for lemma frequency, naming RTs, lexical decision RT, and lexical decision accuracy (Acc; Balota et al., 2007).

Table 2. 
Mean (and SD) of the Measures for Both Conditions of the SSJ
 Manipulable Mean (SD)Nonmanipulable Mean (SD)p
Length 5.7 (1.67) 5.68 (1.54) 
No. phonemes 4.51 (1.34) 4.62 (1.55) .48 
No. syllables 1.58 (0.62) 1.66 (0.75) .30 
Lexical decision RT 661.9 (73.08) 655.73 (64.02) .49 
Lexical decision Acc 0.94 (0.09) 0.95 (0.08) .61 
Naming RT 637.75 (62.59) 630.05 (54.41) .29 
Imageability 564.65 (45.77) 570.81 (44.68) .38 
Age of acquisition 332.1 (75.77) 302.84 (80.1) .17 
CobLog frequency 0.99 (0.54) 1.07 (0.49) .20 
Semantic diversity 1.51 (0.23) 1.49 (0.22) .40 
BOI 5.14 (1) 3.91 (1.30) <.0001 
 Manipulable Mean (SD)Nonmanipulable Mean (SD)p
Length 5.7 (1.67) 5.68 (1.54) 
No. phonemes 4.51 (1.34) 4.62 (1.55) .48 
No. syllables 1.58 (0.62) 1.66 (0.75) .30 
Lexical decision RT 661.9 (73.08) 655.73 (64.02) .49 
Lexical decision Acc 0.94 (0.09) 0.95 (0.08) .61 
Naming RT 637.75 (62.59) 630.05 (54.41) .29 
Imageability 564.65 (45.77) 570.81 (44.68) .38 
Age of acquisition 332.1 (75.77) 302.84 (80.1) .17 
CobLog frequency 0.99 (0.54) 1.07 (0.49) .20 
Semantic diversity 1.51 (0.23) 1.49 (0.22) .40 
BOI 5.14 (1) 3.91 (1.30) <.0001 

Frequencies were gathered from the WebCelex database (celex.mpi.nl). SemD values were gathered from Hoffman, Ralph, and Rogers (2013). BOI ratings were obtained from Tillotson et al. (2008). The English Lexicon (elexicon.wustl.edu; Balota et al., 2007) and MRC (Coltheart, 1981) databases provided all other measures. The t test column shows the p values of the test between manipulable and nonmanipulable nouns. Boldface indicates statistical significance.

Procedure

The SSJ task was administered in a separate session (within 2–3 days) from the fMRI and was part of a neuropsychological test battery to assess language and speech abilities. The SSJ task was presented on a laptop PC running E-Prime software (Version 1.2, Psychology Software Tools, Inc.). Trials were presented in random order in a single block of testing. Participants pressed one of two response buttons to indicate their response. Position of the bottom two words was counterbalanced across participants. Participants were instructed to respond as accurately and quickly as possible using whichever hand they preferred. Words remained on the screen for 5 sec, and no response within that time would result in the next triplet appearing.

MRI Data Acquisition

MRI data were gathered using a Siemens 3T system with initial participants scanned using the Trio configuration (using a 12-channel head coil), and more recent individuals scanned after the system were upgraded to the Prisma configuration (with a 20-channel coil). Scanning included two anatomical MRI sequences: (i) T1-weighted imaging sequence using an magnetization-prepared rapid-gradient echo turbo field echo sequence with voxel size = 1 mm3, field of view (FOV) = 256 × 256 mm, 192 sagittal slices, 9° flip angle, repetition time (TR) = 2250 msec, inversion time = 925 msec, echo time (TE) = 4.15 msec, generalized autocalibrating partial parallel acquisition (GRAPPA) = 2, and 80 reference lines; and (ii) T2-weighted MRI with a 3-D sampling perfection and application of optimized contrasts by using a flip angle evolutions protocol with the following parameters: voxel size = 1 mm3, FOV = 256 × 256 mm, 160 sagittal slices, variable flip angle, TR = 3200 msec, TE = 212 msec, and GRAPPA = 2 (80 reference lines). The same slice center and angle were used as in the T1 sequence.

Functional connectivity was assessed using resting-state fMRI imaging. For 37 participants (scanned on the Trio), images were acquired via an EPI sequence with FOV = 208 × 208 mm, 64 × 64 matrix size of 3.25 mm voxels, 75° flip angle, 34 axial slices (3 mm thick with 20% gap yielding 3.6 mm between slice centers), TR = 1850 msec, TE = 30 msec, GRAPPA = 2, 32 reference lines, sequential descending acquisition, 196 volumes acquired. For 16 participants (scanned on the Prisma), images were acquired via a multiband sequence (×2) with FOV = 216 × 216 mm, 90 × 90 matrix size of 2.4 mm voxels, 72° flip angle, 50 axial slices (2 mm thick with 20% gap yielding 2.4 mm between slice centers), TR = 1650 msec, TE = 35 msec, GRAPPA = 2, 44 reference lines, interleaved ascending acquisition, and 427 volumes acquired.

FA was computed using diffusion tensor imaging (DTI). For 42 participants (scanned in the Trio), we used a monopolar sequence with 82 isotropic (2.3 mm) volumes (×10 B = 0, ×72 B = 1000), TR = 4987 msec, TE = 79.2 msec, 90 × 90 matrix, with parallel imaging GRAPPA = 2, and 50 contiguous slices. For 13 participants (scanned the Prisma), we used a monopolar sequence with 86 isotropic (1.5 mm) volumes (×14 B = 0, ×72 B = 1000), TR = 5250 msec, TE = 80 msec, 140 × 140 matrix, 80 contiguous slices. For all participants, the sequence was acquired in two series (41 and 43 volumes in each series, respectively) with opposite phase encoding allowing us to spatially undistort the images with TOPUP.

Preprocessing of Structural Images

Lesions were defined in native space by a neurologist in MRIcron (Rorden, Bonilha, Fridriksson, Bender, & Karnath, 2012) on individual T2-weighted images. Preprocessing started with coregistration of the T2-weighted images to match the T1-weighted images, permitting alignment of the lesions to native T1 space. Images were converted to standard space using the enantiomorphic (Nachev, Coulthard, Jager, Kennard, & Husain, 2008) segmentation–normalization (Ashburner & Friston, 2005) custom Matlab script (https://github.com/rordenlab/spmScripts/blob/master/nii_enat_norm.m) to warp the images to an age-appropriate template image contained within the Clinical Toolbox (Rorden et al., 2012). The lesion was resliced into standard space with linear interpolation, and the resulting lesion maps were stored at 1 × 1 × 1-mm resolution and binarized using a threshold of 50% (because interpolation can result in fractional likelihoods, this precaution guarantees that each voxel is categorically either lesioned or not without biasing total lesion volume). The tissue segmentation maps from the unified normalization–segmentation we used to create brain-extracted examples of the individuals T1 and T2 scans, which are leveraged for normalizing the other lower resolution modalities (as described below). Preprocessing quality of the normalized images was confirmed by visual inspection.

Preprocessing of RSFC Data

Motion correction for fMRI data was achieved using the SPM12 “realign and unwarp” procedure with default settings. Slice timing correction was performed with SPM12. Brain extraction was completed using the SPM12 script pm_brain_mask with default settings. The extracted mean fMRI volume for each subject was aligned to the equivalent extracted T2-weighted image to calculate the spatial change between the fMRI data and the lesion mask. The fMRI data were spatially smoothed with a Gaussian kernel with an FWHM of 6 mm.

The process outlined in Yourganov, Fridriksson, Stark, and Rorden (2018) was used to eliminate artifacts driven by lesions. FSL MELODIC package was used to decompose the data into independent components and to calculate the Z-scored spatial maps for each component. The resulting maps were thresholded at p < .05 and juxtaposed with the lesion mask for that patient. If the overlap (measured with Jaccard index) between the lesion mask and the thresholded IC map was greater than 5%, the corresponding component was considered to have significant overlap with the lesion mask. Any components meeting this criterion were regressed out of the fMRI data using the fsl_regfilt script from the FSL package.

RSFC Connectome Creation

A unique RSFC connectome was constructed for every participant using the subsequent steps: (1) determination of the probabilistic gray matter map from T1-weighted images; (2) segmentation of the gray matter map into 189 ROIs according to the Johns Hopkins University (JHU) atlas (Faria et al., 2012; Mori, Wakana, Nagae-Poetscher, & van Zijl, 2005; Wakana, Jiang, Nagae-Poetscher, van Zijl, & Mori, 2004); (3) calculation of ROI-specific time courses of the BOLD signal by averaging across all voxels within each individual ROI; and (4) creating a 189 × 189 correlation matrix for each participant, where positive values signify greater time-locked activation of two regions.

FA Preprocessing

The processing of diffusion-weighted images used the same pipeline described by (Peters et al., 2018). Specifically, the diffusion images were undistorted using FSL’s TOPUP and Eddy tools (Andersson & Sotiropoulos, 2016; Andersson, Skare, & Ashburner, 2003) with excess scalp removed using the FSL BET tool. FSL’s dtifit tool was used to compute an FA map. To improve registration between T1 and DTI spaces, the scalp-stripped (based on segmentation estimates) T1 image was nonlinearly normalized (using SPM12's “old normalization” function) to match the undistorted FA image. This leverages the similarity of the image intensity in the T1 scan and the FA map, with the high resolution and tissue contrast allowing the T1 scan to achieve superior normalization Acc. The same transformation matrix was applied to the map of segmented cortical ROIs and the probabilistic white matter map (which were in T1 space) to transform these maps into DTI space (using nearest neighbor interpolation to preserve discrete regions).

Experimental Design and Statistical Analysis

Behavioral Data

Our primary aim was to inspect the relative impairment of the manipulable compared with nonmanipulable condition. The average Acc was calculated for the manipulable and nonmanipulable conditions for each patient. Seven participants performed below chance (less than 60 % Acc, p < .05) in both conditions, indicating general impairments (e.g., lexical processing, executive function), and were omitted from the analysis. This safeguards against including participants who were unable to perform either task condition at an above-chance level. Condition-specific differences in such globally impaired patients are difficult to interpret meaningfully. It also guarantees that patients with chance performance in one condition but relatively spared (i.e., above chance) performance in the other would be included, as these patients provide the most information about the differences between the conditions. Trials with missing responses were excluded from analysis. The results of central interest are the residuals for each condition when the performance in other condition is regressed out (ResidManip, ResidNonManip). By regressing the conditions out from each other, condition-relative deficits can be assessed while controlling for confounds such as global impairment.

Voxel-based Lesion–Symptom Mapping

Whole-brain VLSM was used to identify damage related to greater relative impairment of manipulable (ResidManip) noun comprehension compared with nonmanipulable (ResidNonManip) by regressing out performance in one condition from the other using NiiStat software (www.nitrc.org/projects/niistat/). This examines condition-relative deficits while controlling for linguistic/executive confounds. VLSM marks each voxel as either lesioned or unlesioned and tests the probability that damage to a voxel is associated with performance on a behavioral measure (Bates et al., 2003). Nuisance regression used the Freedman–Lane method (Freedman & Lane, 1983), allowing permutation-based control for family-wise error, as described by Winkler, Ridgway, Webster, Smith, and Nichols (2014). VLSM results were thresholded at p < .0005 voxel-wise and cluster-corrected to p < .05 using permutation analysis as correction for multiple comparisons (1000 permutations). Permutation analysis is a nonparametric significance test comparing a test statistic to a null distribution that is derived from randomly permuting existing data. Permutation testing relies on minimal assumptions, approaches exact control of false positives, and is one of the most common statistical methods for conducting VLSM (Baldo & Dronkers, 2018; Baldo, Wilson, & Dronkers, 2012; Kimberg, Coslett, & Schwartz, 2007). Only voxels where at least five patients had damage were considered (improves power and minimizing spatial bias; see Karnath, Sperber, and Rorden (2017)), according to the cutoff recommendation of 5–10% of the patient sample (Baldo & Dronkers, 2018).

RSFC: Praxis NOI

A praxis NOI was extracted from the original 189 × 189 connectome for RSFC analysis. This NOI included the supramarginal gyrus (SMG), precentral gyrus and postcentral gyrus (PoC), pMTG and posterior superior temporal gyrus (pSTG), and the lateral inferior occipital gyrus. These areas were chosen due to being consistently implicated in praxis across a variety of methodologies (Borra & Luppino, 2019; Martin et al., 2017; Lingnau & Downing, 2015; Orban & Caruana, 2014; Buxbaum & Kalénine, 2010; Hermsdörfer, Terlinden, Mühlau, Goldenberg, & Wohlschläger, 2007). Although arguments could be made for the inclusion of other areas in this network, to maintain power to detect effects we wished to limit the NOI to areas consistently and especially associated with praxis. This meant excluding some areas, such as the angular gyrus, which may function as putative semantic “hubs” and are involved in processing concepts of many types (Fernandino, Binder, et al., 2016; Binder & Desai, 2011). We hypothesized that disruption of RSFC links between areas within this NOI would result in worse performance in the manipulable (ResidManip) compared with nonmanipulable (ResidNonManip) condition, demonstrating a connection between praxis network integrity and manipulable object semantics specifically. Left-to-left, left-to-right, and right-to-right connections (66 in total) were considered to test possible contributions from undamaged homologues in the RH. Functional connectivity strengths were used in a general linear model predicting ResidManip. Alpha was set to .05, and significance was determined with permutation correction for multiple comparisons (1000 permutations).

Fractional Anisotropy

FA is a measure of directional water diffusion that can be used to assess microscopic (as opposed to macroscropic) brain matter integrity, with intact neural material having more directional diffusion and higher FA (Beaulieu, 2002). Although white matter has naturally more directional diffusion than gray matter, gray matter does indeed have directional diffusion, and FA has been used to study microscopic gray matter integrity in studies of aging, Alzheimer’s disease, and other neurodegenerative conditions (Mayer, Hanlon, & Ling, 2015; Weston, Simpson, Ryan, Ourselin, & Fox, 2015; Cappellani et al., 2014; Bouix et al., 2013; Hansen, Jespersen, Leigland, & Kroenke, 2013; Rossi et al., 2008; Osuka et al., 2012). Gray matter FA provides a promising additional measure to reveal brain–behavior relationships that might not be found using VLSM or RSFC for multiple reasons.

First, FA is a continuous, region-wide measure compared with VLSM’s binary, voxel-wide approach, allowing for detection of microstructural changes to region-wide gray matter that may not overlap at a voxel level between patients. Second, evidence suggests that gray matter distant from the lesion site can undergo structural changes following stroke (Wang et al., 2018; Abela et al., 2015), and these changes could remain undetected if using VLSM alone. Third, measures of gray matter diffusion can detect microstructural changes to the integrity of tissue, which cannot be detected by traditional macroanatomical scans (Weston et al., 2015). Fourth, although structural integrity of gray matter is related to RSFC, this relationship is not straightforward. Reduced gray matter integrity has been associated with unpredictable changes to RSFC, including up or down regulation in areas both near and far from the gray matter dysfunction (Marstaller, Williams, Rich, Savage, & Burianova, 2015; van Tol et al., 2014; Verstraete et al., 2010). Taken together, these points mean that measures of micro- and macrostructural integrity (FA and VLSM) can each make unique contributions. FA analysis was used for all 55 gray matter areas identified by the JHU atlas to detect areas where lower FA (i.e., compromised structural integrity) predicted ResidManip in a general linear model using NiiStat. Alpha was set to .05, and significance was determined via permutation correction for multiple comparisons (1000 permutations).

RESULTS

Behavioral

There was no significant difference between Acc in the manipulable (M = 81.8%, SD = 13.8%) and nonmanipulable (M = 83%, SD = 13.2%) conditions, t(56) = 0.67, p = .33. WAB-AQ (Apahsia Quotient) scores did not correlate with the difference scores, r(56) = .2, p = .14, and neither did the verbal comprehension subtest of the WAB, r(56) = .17, p = .21. These data indicate that the conditions were well matched for difficulty and that anatomical differences observed between the conditions are unlikely to be driven by general effects of difficulty or due to general comprehension deficits.

Voxel-based Lesion–Symptom Mapping

The lesion overlay map can be viewed in Figure 2. Whole-brain VLSM for ResidManip revealed a cluster of voxels to which damage predicted worse manipulable noun comprehension compared with nonmanipulable (peak z = −5.138, p < .0001). The main body of this cluster extended from the MTG and STG posteriorly and into anterior inferior parietal areas such as the SMG and small portions of the cluster reached medially into white matter such as the superior longitudinal fasciculus and subcortical areas such as the posterior insula (Table 3; Figure 3). No clusters were significant for ResidNonManip.

Figure 2. 

Lesion overlay map after excluding voxels in less than five patients. The color scale range represents the areas of least lesion coverage (n = 5; dark blue) to the areas with the greatest lesion overlap (n = 27; dark red).

Figure 2. 

Lesion overlay map after excluding voxels in less than five patients. The color scale range represents the areas of least lesion coverage (n = 5; dark blue) to the areas with the greatest lesion overlap (n = 27; dark red).

Table 3. 
VLSM Results for ResidManip
Volume (1 × 1 × 1 mm)Region
37,924 STG 
pSTG 
MTG 
pMTG 
SMG 
Angular gyrus 
Middle occipital gyrus 
Superior longitudinal fasciculus 
Posterior insula 
Thalamus 
Volume (1 × 1 × 1 mm)Region
37,924 STG 
pSTG 
MTG 
pMTG 
SMG 
Angular gyrus 
Middle occipital gyrus 
Superior longitudinal fasciculus 
Posterior insula 
Thalamus 

Regions are labeled according to the JHU atlas.

Figure 3. 

Left: Whole-brain VLSM results for ResidManip, indicating where damage predicted worse performance on the manipulable condition of SSJ compared with nonmanipulable (voxelwise p < .0005; cluster thresholded p < .05). Right: VLSM correlates for impairments in actual tool-use for comparison (Brunner-Munzel p < .05, false discovery rate corrected Z > 2.32; reprinted from Salazar-López et al. [2016] with permission from Elsevier).

Figure 3. 

Left: Whole-brain VLSM results for ResidManip, indicating where damage predicted worse performance on the manipulable condition of SSJ compared with nonmanipulable (voxelwise p < .0005; cluster thresholded p < .05). Right: VLSM correlates for impairments in actual tool-use for comparison (Brunner-Munzel p < .05, false discovery rate corrected Z > 2.32; reprinted from Salazar-López et al. [2016] with permission from Elsevier).

Praxis RSFC Network

RSFC within the praxis NOI revealed three connections associated with ResidManip, with decreased functional connectivity in these regions being associated with worse relative performance on the manipulable condition (Table 4; Figure 4). These connections were (1) left postcentral to left supramarginal (z = 3.22, p = .0006), (2) left supramarginal to right supramarginal (z = 3.37, p = .0004), and (3) left posterior superior temporal to right posterior middle temporal (z = 3.24, p = .0006).

Table 4. 
Significant Functional Connections for ResidManip
Connection Z score
Left SMG Left postcentral gryus 3.22 
Right SMG 3.37 
Left pSTG Right pMTG 3.24 
Connection Z score
Left SMG Left postcentral gryus 3.22 
Right SMG 3.37 
Left pSTG Right pMTG 3.24 
Figure 4. 

Colored regions and corresponding links indicate where disrupted RSFC predicted worse performance on manipulable compared with nonmanipulable conditions of SSJ (Z > 3.17). Different colors are used for display purposes only. Left: LH areas include PoC (dark orange), SMG (light orange), and pSTG (yellow). Middle: Superior view. Right: RH areas include SMG (yellow) and pMTG (orange).

Figure 4. 

Colored regions and corresponding links indicate where disrupted RSFC predicted worse performance on manipulable compared with nonmanipulable conditions of SSJ (Z > 3.17). Different colors are used for display purposes only. Left: LH areas include PoC (dark orange), SMG (light orange), and pSTG (yellow). Middle: Superior view. Right: RH areas include SMG (yellow) and pMTG (orange).

Fractional Anisotropy

FA of gray matter revealed eight cortical areas significantly associated with ResidManip, with decreased FA in these regions being associated with worse relative performance on the manipulable condition (Table 5; Figures 5 and 6). These areas were the posterior middle frontal gyrus (pMFG), inferior frontal gyrus pars opercularis and pars triangularis (IFGoper and IFGtri, respectively), SMG, angular gyrus, STG, pMTG, and pSTG. Five subcortical areas were also found, consisting of the thalamus, hypothalamus, red nucleus, substantia nigra, and midbrain (all ps < .05). No areas were significant for ResidNonManip.

Table 5. 
Regions Where FA Predicted ResidManip
RegionZ score
pMFG 3.44 
IFGoper 3.36 
IFGtri 3.30 
SMG 3.35 
Angular gyrus 3.40 
STG 3.33 
pMTG 4.05 
pSTG 3.05 
Thalamus 3.50 
Hypothalamus 4.02 
Red nucleus 3.70 
Substantia nigra 3.77 
Midbrain 3.58 
RegionZ score
pMFG 3.44 
IFGoper 3.36 
IFGtri 3.30 
SMG 3.35 
Angular gyrus 3.40 
STG 3.33 
pMTG 4.05 
pSTG 3.05 
Thalamus 3.50 
Hypothalamus 4.02 
Red nucleus 3.70 
Substantia nigra 3.77 
Midbrain 3.58 
Figure 5. 

Cortical regions where lower FA predicted worse performance on manipulable compared with nonmanipulable SSJ (Z = 3.04): (1) IFGtri, (2) IFGoper, (3) pMFG, (4) STG, (5) SMG, (6) angular gyrus, (7) pSTG, (8) pMTG. Different colors are used for display purposes only.

Figure 5. 

Cortical regions where lower FA predicted worse performance on manipulable compared with nonmanipulable SSJ (Z = 3.04): (1) IFGtri, (2) IFGoper, (3) pMFG, (4) STG, (5) SMG, (6) angular gyrus, (7) pSTG, (8) pMTG. Different colors are used for display purposes only.

Figure 6. 

Left: Conjunction of VLSM and FA where reduced structural integrity of cortical areas resulted in poorer comprehension of manipulable compared with nonmanipulable nouns. The significant VLSM cluster is in dark red. For FA, orange: pMFG, yellow: IFGtri, red: IFGoper. Right: Likely regions of the human “lateral grasping system” (reprinted from Borra & Luppino [2019] with permission from Elsevier).

Figure 6. 

Left: Conjunction of VLSM and FA where reduced structural integrity of cortical areas resulted in poorer comprehension of manipulable compared with nonmanipulable nouns. The significant VLSM cluster is in dark red. For FA, orange: pMFG, yellow: IFGtri, red: IFGoper. Right: Likely regions of the human “lateral grasping system” (reprinted from Borra & Luppino [2019] with permission from Elsevier).

DISCUSSION

The results provide evidence for the contribution of praxis-related brain areas to manipulable noun comprehension specifically (Figure 6). The three neuroanatomical measures (VLSM, FA, RSFC) revealed a relationship between the degradation of praxis networks and worse relative performance on the manipulable condition of the SSJ task, with the nonmanipulable condition serving as a control. Each of the three measures provided complementary, additive evidence that would not have been revealed by using any single method in isolation, highlighting the importance of using multiple measures of network integrity in patient studies when possible.

Damage to the Posterior Praxis Network and Manipulable Noun Comprehension

VLSM revealed that damage to anterior parietal (SMG) and posterior temporal (pMTG, pSTG) regions is associated with worse comprehension of manipulable compared with nonmanipulable nouns. This swath of damage corresponds to the posterior portion of the praxis network (Borra & Luppino, 2019; Buxbaum & Kalénine, 2010), and it displays a remarkable resemblance to the results of lesion studies of impaired tool use (Martin et al., 2017; Salazar-López, Schwaiger, & Hermsdörfer, 2016).

The left anterior inferior parietal lobe (aIPL) is well established as serving goal-directed movement such as reaching to grasp and tool use (Frey, 2008; Johnson-Frey, 2004). For example, lesions to the aIPL are associated with deficits in actual or pantomimed tool use (Salazar-López et al., 2016; Goldenberg, 2009; Haaland, Harrington, & Knight, 2000) and the production/imitation of object-related gestures (Buxbaum, Kyle, Grossman, & Coslett, 2007). TMS to the SMG, especially anterior portions, results in impaired judgment of whether objects are manipulated similarly (Pelgrims, Olivier, & Andres, 2011) and delayed grasp orientation for manipulable objects (Potok, Maskiewicz, Króliczak, & Marangon, 2019; McDowell, Holmes, Sunderland, & Schurmann, 2018). Neuroimaging studies also implicate the SMG and adjacent IPL with actions such as tool use and orienting grip to grasp novel objects (Brandi, Wohlschläger, Sorg, & Hermsdörfer, 2014; Peeters, Rizzolatti, & Orban, 2013; Hermsdörfer et al., 2007; Frey, Vinton, Norlund, & Grafton, 2005; Johnson-Frey, Newman-Norlund, & Grafton, 2005) and the processing of action semantics (Desai, Choi, Lai, & Henderson, 2016; Desai, Conant, Binder, Park, & Seidenberg, 2013; Desai, Binder, Conant, & Seidenberg, 2010; Binder et al., 2009).

Similarly, the posterior temporal lobe is associated with action observation, gesture recognition, and retrieval of action- and tool-related knowledge (Johnson-Frey, 2004). For example, a neuroimaging meta-analysis of action observation and imitation revealed a significant cluster in the pMTG (Caspers, Zilles, Laird, & Eickhoff, 2010). Lesion and TMS evidence has indicated that the pMTG is critical in retrieving the meanings of actions (Papeo et al., 2015; Kalénine, Buxbaum, & Coslett, 2010) and to the recognition of meaningful actions and their kinematics (Martin et al., 2017). This evidence, when considered alongside proposals that the pMTG serves as an integrative zone (Binder et al., 2009; Willems, Ozyurek, & Hagoort, 2009), suggests that this region may be involved in assimilating/coordinating multimodal information, such as motor and visuospatial properties from distributed sensory–motor cortices, to support the meanings of purposeful actions or action-related language.

Thus, it is possible that the posterior temporal lobe is involved in linking motor-planning representations from the aIPL with other, multimodal information such as visual motion to (1) extract meaning during action observation or imitation and (2) form associations to related concepts, such as objects or communicative intentions that correspond to the action in question. The current results demonstrate that damage to these areas is associated with impairments to semantic knowledge of manipulable nouns, substantiating an important role for these higher order action observation, execution, planning, and visuomotor coordination cortices in conceptual representations of manipulable objects specifically.

It is important to note, however, that there is also evidence that damage to posterior temporal regions results in lexical impairment (Hillis et al., 2017). More specifically, pMTG damage has been associated with word comprehension difficulties even after factoring out object recognition. This suggests that the posterior temporal region may be involved in linking words to their conceptual content, which are then represented elsewhere in the cortex (Bonilha et al., 2017). This word-to-concept matching hypothesis for the pMTG is compatible with the current results under the interpretation that (1) damage to the pMTG alone would impair the manipulable and nonmanipulable conditions roughly equally due to lexical impairment but (2) further damage to the distributed action execution, observation, planning, and visuomotor coordination regions implicated here would then selectively impair understanding of manipulable nouns due to impoverished conceptual representations. More research is needed to clarify the lexical versus conceptual nature of representations in the posterior temporal lobe.

Functional Connectivity within the Posterior Praxis Network Is Associated with Manipulable Noun Comprehension

Although tool use is a largely left-lateralized process (Frey, 2008), evidence suggests that connectivity from the LH to homologues in the RH can play a role in praxis following LH stroke (Watson, Gotts, Martin, & Buxbaum, 2018). As a whole, RSFC network integrity is efficacious for predicting performance in a variety of tasks poststroke (Siegel et al., 2016). RSFC analysis was used to examine how inter- and intrahemispheric connectivity within an a priori praxis NOI may contribute to manipulable noun comprehension specifically.

Disruptions of one intrahemispheric (left PoC to left SMG) and two interhemispheric (left SMG to right SMG; left pSTG to right pMTG) connections were associated with worse comprehension of manipulable nouns compared with nonmanipulable. Regarding the intrahemispheric connection, the PoC is the primary sensory cortex and is principally responsible for processing somatotopic tactile sensations (Dreyer, Loe, Metz, & Whitsel, 1975; Whitsel, Dreyer, & Roppolo, 1971). Lesions to the PoC are associated with apraxia (Weiss et al., 2016), and neuroimaging studies have implicated the PoC in tool use (Hermsdörfer et al., 2007) and action observation (Caspers et al., 2010). Additionally, the PoC displays strong connectivity to the SMG (Ruschel et al., 2014) providing evidence that these areas form part of a coherent network in healthy brains. In the context of the current study, it is possible that connectivity between the left PoC and SMG helps integrate tactile and hand orientation representations that serve manipulable noun comprehension. When this connection is disrupted, performance in the manipulable condition declines to a greater degree than in the nonmanipulable condition due to the greater sensory–motor associations for manipulable object concepts.

Additionally, the interhemispheric connections align well with those found by Watson et al. (2018), who demonstrated that patients with worse interhemispheric connectivity between parietal and posterior temporal regions display more severe symptoms of apraxia compared with those with better connectivity. It is possible that the connections in the current study represent compensatory connectivity between undamaged RH homologues and surviving portions of the LH to aid comprehension. Studies of poststroke language abilities have found that increased activity or connectivity to RH homologues is associated with better behavioral outcomes (Riccardi et al., 2019; Skipper-Kallal, Lacey, Xing, & Turkeltaub, 2017; Richter, Miltner, & Straube, 2008; Saur et al., 2006). Similarly, poststroke motor recovery has been shown to be reliant on connectivity between contralesional homologues and surviving portions of the ipsilesional cortex (Kantak, Stinear, Buch, & Cohen, 2012; Bestmann et al., 2010; Johansen-Berg et al., 2002). Taken together, the current results suggest that disruption of connectivity within the LH or to RH homologues of the praxis network is associated with manipulable noun processing especially, even after accounting for general executive or cognitive decline via the nonmanipulable comparison condition.

Integrity of Frontal and Subcortical Motor Structures Predicts Worse Manipulable Noun Comprehension

Praxis is a widely distributed process that involves contributions from cortical and subcortical structures. Although the VLSM and RSFC results highlighted the contributions of posterior cortical areas to manipulable noun comprehension, FA provided complementary evidence by identifying microstructural changes to gray matter cannot be established via traditional VLSM. The FA results (1) added evidence suggesting that compromised integrity of subcortical and frontal structures also disproportionately impairs manipulable noun understanding compared with nonmanipulable and (2) replicated VLSM and RSFC results regarding the importance of the posterior praxis network (pMTG, SMG) for manipulable noun comprehension.

The FA analysis revealed that reduced integrity of subcortical areas, including the thalamus and midbrain structures, is associated with worse comprehension of manipulable nouns compared with nonmanipulable. Although the specific contributions of these areas to semantic processing is understudied, evidence from Parkinson’s disease (PD) suggests that degradation of the basal ganglia-thalamo-cortical network can selectively impair action-related language processing (Bocanegra et al., 2015; Fernandino et al., 2013a, 2013b). PD is associated with the degeneration of dopaminergic neurons in the midbrain, especially in the substantia nigra, causing multiple motor and cognitive impairments (Chinta & Andersen, 2005; Hoehn & Yahr, 1967). These midbrain areas, including the red nucleus, are important parts of the motor system (Herter, Takei, Munoz, & Scott, 2015; Hosp, Pekanovic, Rioult-Pedotti, & Luft, 2011) and display connectivity to cortical motor areas as part of the corticospinal tract (Habas & Cabanis, 2006; Humphrey, Gold, & Reed, 1984; Hartmann-von Monakow, Akert, & Kunzle, 1979). The current results demonstrate that, following stroke, reduced structural integrity of these subcortical motor areas is associated with impaired processing of manipulable nouns compared with nonmanipulable. This is an addition to the literature regarding the possible contributions of the midbrain and other subcortical structures to action-related language processing, as most previous patient studies about the linguistic role of these structures have focused on verb processing in PD (see Buccino et al., 2017, for a recent exception).

Regarding frontal cortices, the analysis revealed that reduced FA in the pMFG (corresponding to the premotor cortex [PMc]), IFGtri, and IFGoper was associated with worse relative comprehension of manipulable nouns. These areas have been shown to be involved in motor planning, execution, observation, and visuomotor coordination, especially pertaining to hand-related actions (Caspers et al., 2010; Binkofski & Buccino, 2006; Rizzolatti, Fogassi, & Gallese, 2001). PMc and IFG have also been implicated in the production and comprehension of action language; words referring to face, leg, or arm actions are associated with somatotopically organized activations in the PMc (Aziz-Zadeh, Wilson, Rizzolatti, & Iacoboni, 2006; Hauk, Johnsrude, & Pulvermüller, 2004). Furthermore, deterioration of the IFG from amyotrophic lateral sclerosis is associated with selective impairments to action verbs, supporting a causal role for the IFG in the comprehension of action-related language (Bak & Chandran, 2012; Bak & Hodges, 2004; Bak et al., 2001). Finally, evidence from stroke patients implicates these frontal areas in lexical and conceptual knowledge of actions (Kemmerer et al., 2012; Tranel, Kemmerer, Adolphs, Damasio, & Damasio, 2003; Tranel, Adolphs, Damasio, & Damasio, 2001). Indeed, a recent lesion–symptom mapping investigation found that damage to gray matter underlying the IFG, as well as reduced connectivity of the left IFG to distributed sensory–motor cortices, was associated with worse action verb comprehension compared with nonaction verbs (Riccardi et al., 2019). However, many of these past studies focused specifically on action verbs or concrete objects without specifically addressing manipulability, leaving gaps in the patient literature regarding the neural substrates of nouns that are highly action related. Our results help address this, demonstrating that reduced integrity of the PMc and IFG is associated with worse relative comprehension of manipulable objects compared with the nonmanipulable condition. It is important to note that the IFG is known to have many functions, such as cognitive control (Fadiga, Craighero, & D'Ausilio, 2009; Badre, 2008; Grodzinsky & Santi, 2008). However, the effect observed here is unlikely to be driven by these more general processes due to the matching of difficulty and psycholinguistic variables, such as semantic diversity, between the manipulable and nonmanipulable conditions. We propose that the IFG plays many roles and that these roles can be differentially recruited depending on task characteristics. In this case, it appears that the IFG is especially involved in the comprehension of manipulable compared with nonmanipulable nouns, over and above its domain-general functions. Thus, it seems likely that these frontal portions of the motor network are especially involved in representing concepts with high motor or BOI properties.

Two possible caveats should be noted concerning the FA results. First, stroke is associated with ventricular enlargement (Hijdra & Verbeeten, 1991), and it is possible that low FA measures in subcortical areas, especially areas directly adjacent to the ventricles such as the thalamus, are driven by this enlargement. For this reason, the subcortical FA results should be treated with caution, and future studies should investigate the role of subcortical motor areas in action-related noun processing beyond the typical PD patient model. Second, the FA analysis did not survive lesion volume correction. However, it is important to note that, because the independent variable was the residuals of the closely matched manipulable versus nonmanipulable conditions within the same patients, lesion size is partially controlled for (i.e., lesion size is held constant between the two conditions). Regardless, the possibility remains that there is a Lesion Size × Location interaction where large lesions that extend to affect the areas indicated by our FA analysis are associated with the manipulable noun impairments observed here.

Negative Results for Nonmanipulable Nouns

The analyses presented here did not reveal significant results for the ResidNonManip condition. This is likely due to the nonmanipulable stimuli being built on a criterion of exclusion (i.e., nouns that are not manipulable but equally imageable) instead of being constructed under a unifying feature, a tradition that has been common to many past studies of language processing (Barsalou, Dutriaux, & Scheepers, 2018; Desai, Reilly, & van Dam, 2018). This was done to match the two conditions on imageability, thereby dissociating the dimension of imageability from that of manipulability, which are correlated (Heard et al., 2019). Because of this, the nonmanipulable condition serves best as a control for general cognitive, lexical, or semantic decline and not as an overt exploration of the anatomical substrates of the sensory–motor, emotional, or other sui generis properties of the nonmanipulable words. Future lesion studies could construct stimuli that are specifically meant to test salient features other than manipulability.

Limitations

This work has limitations. Damage from middle cerebral artery stroke does not usually extend to some areas that are implicated in action, such as the lateral occipital cortex (Lingnau & Downing, 2015). This could limit the ability to test the contribution of these areas to semantic processing in middle cerebral artery (MCA) stroke patients. Furthermore, the nature of MCA stroke means that certain regions are often damaged together, making it difficult to draw fine-grained anatomical boundaries for brain–behavior relationships. For example, STG damage was predictive of ResidManip in both the FA and VLSM analysis, but this region is commonly associated with general auditory and speech processing (Liebenthal, Desai, Humphries, Sabri, & Desai, 2014). This is likely due to the superior temporal region and praxis-related ventral frontoparietal regions being affected together in MCA stroke. In our sample, only three patients had damage to the STG without corresponding damage to the SMG, meaning that there were not enough patients with this pattern of damage to dissociate the function of these areas. This natural limitation of MCA stroke investigations is one possible reason why apraxia and general language difficulties often co-occur (Weiss et al., 2016; Goldenberg & Randerath, 2015). Future studies can make finer distinctions via more focal methods such as TMS or examining patients with relatively localized lesions.

An argument could be made that the network identified here reflects motor imagery processes, which display tight anatomical correspondences to the praxis network (Filimon, Nelson, Hagler, & Sereno, 2007). However, this is not problematic, as our results demonstrate that whatever processes that are affected by the damage, be they related to action performance, planning, imagery, or visuomotor coordination, are necessary for performing the manipulable SSJ task and are not irrelevant components. Furthermore, our network consists of well-established action-related areas. Although the aIPL may be activated for motor imagery tasks (Filimon et al., 2007), this is consistent with the role of the aIPL in action planning and hand–object interaction systems, especially considering that motor imagery has been shown to have functional consequences for motor planning and execution (Eaves, Behmer, & Vogt, 2016; Eaves, Riach, Holmes, & Wright, 2016).

Finally, we note that the praxis network was defined here purely based on anatomical grounds, using brain areas known to be involved in simple and complex action. Behavioral deficits in action performance were not examined. The study does not address the issue of the relationship between action performance and semantic processing, only between brain regions related to action performance and semantics. The former issue was addressed in the study of Desai et al. (2015), who showed that a decline in action performance also predicts a selective deficit in action word processing. A future direction of research can involve examining relationship between apraxia and measures of action performance with action semantics.

Conclusions

Here, using three complementary methods, we found evidence demonstrating that degradation of praxis brain networks especially impairs semantic processing of manipulable nouns compared with nonmanipulable. First, VLSM revealed that damage to the posterior temporal lobe and aIPL is associated with impaired manipulable noun comprehension. Second, RSFC analysis revealed that disrupted connectivity to RH homologues of this network was associated with worse relative performance on the manipulable condition, suggesting a compensatory role of RH homologues with an impaired LH. Finally, FA confirmed the VLSM results while also revealing that microstructural integrity of inferior frontal cortex and PMc, as well as subcortical structures, is likely important for the comprehension of manipulable nouns. Overall, these results suggest that brain areas underlying action planning, observation, execution, and visuomotor coordination contribute to the representation of semantic information pertaining to words with high BOI properties, providing novel support for a close and causal link between action–perception and conceptual systems.

Acknowledgments

This work was supported by National Institutes of Health/National Institute on Deafness and Other Communication Disorders grants R01 DC010783 (R. H. D.), R56 DC010783 (R. H. D.), and P50-DC014664 (J. F.). We thank Jessica Green and two anonymous reviewers for helpful comments and suggestions.

Reprint requests should be sent to Nicholas Riccardi, University of South Carolina, 915 Greene St., Columbia, SC, or via e-mail: riccardn@email.sc.edu.

Note

1. 

BOI ratings measure interaction with the whole body. In addition to manipulable items such as spoon or crutch, words such as sofa also receive high BOI ratings. Here, we only used objects that are typically manipulated with hand/arm and did not include items such as furniture that are not typically manipulated.

REFERENCES

REFERENCES
Abela
,
E.
,
Seiler
,
A.
,
Missimer
,
J. H.
,
Federspiel
,
A.
,
Hess
,
C. W.
,
Sturzenegger
,
M.
, et al
(
2015
).
Grey matter volumetric changes related to recovery from hand paresis after cortical sensorimotor stroke
.
Brain Structure & Function
,
220
,
2533
2550
.
Andersson
,
J. L.
,
Skare
,
S.
, &
Ashburner
,
J.
(
2003
).
How to correct susceptibility distortions in spin-echo echo-planar images: Application to diffusion tensor imaging
.
Neuroimage
,
20
,
870
888
.
Andersson
,
J. L.
, &
Sotiropoulos
,
S. N.
(
2016
).
An integrated approach to correction for off-resonance effects and subject movement in diffusion MR imaging
.
Neuroimage
,
125
,
1063
1078
.
Arévalo
,
A.
,
Baldo
,
J. V.
, &
Dronkers
,
N. F.
(
2012
).
What do brain lesions tell us about theories of embodied semantics and the human mirror neuron system?
Cortex
,
48
,
242
254
.
Arévalo
,
A.
,
Perani
,
D.
,
Cappa
,
S. F.
,
Butler
,
A.
,
Bates
,
E.
, &
Dronkers
,
N.
(
2007
).
Action and object processing in aphasia: From nouns and verbs to the effect of manipulability
.
Brain and Language
,
100
,
79
94
.
Ashburner
,
J.
, &
Friston
,
K. J.
(
2005
).
Unified segmentation
.
Neuroimage
,
26
,
839
851
.
Aziz-Zadeh
,
L.
,
Wilson
,
S. M.
,
Rizzolatti
,
G.
, &
Iacoboni
,
M.
(
2006
).
Congruent embodied representations for visually presented actions and linguistic phrases describing actions
.
Current Biology
,
16
,
1818
1823
.
Badre
,
D.
(
2008
).
Cognitive control, hierarchy, and the rostro-caudal organization of the frontal lobes
.
Trends in Cognitive Sciences
,
12
,
193
200
.
Bak
,
T. H.
, &
Chandran
,
S.
(
2012
).
What wires together dies together: Verbs, actions and neurodegeneration in motor neuron disease
.
Cortex
,
48
,
936
944
.
Bak
,
T. H.
, &
Hodges
,
J. R.
(
2004
).
The effects of motor neurone disease on language: Further evidence
.
Brain and Language
,
89
,
354
361
.
Bak
,
T. H.
,
O'Donovan
,
D. G.
,
Xuereb
,
J. H.
,
Boniface
,
S.
, &
Hodges
,
J. R.
(
2001
).
Selective impairment of verb processing associated with pathological changes in Brodmann areas 44 and 45 in the motor neurone disease-dementia-aphasia syndrome
.
Brain
,
124
,
103
120
.
Baldo
,
J. V.
, &
Dronkers
,
N.
(
2018
).
Lesion studies
. In
A. M. B.
de Groot
&
P.
Hagoort
(Eds.),
Research methods in psycholinguistics and the neurobiology of language: A practical guide
(pp.
310
329
).
Pondicherry, India
:
Wiley
.
Baldo
,
J. V.
,
Wilson
,
S. M.
, &
Dronkers
,
N.
(
2012
).
Uncovering the neural substrates of language: A voxel-based lesion–symptom mapping approach
. In
M.
Faust
(Ed.),
The handbook of the neuropsychology of language
(1st ed., pp.
582
594
).
West Sussex, UK
:
Blackwell Publishing
.
Balota
,
D. A.
,
Yap
,
M. J.
,
Cortese
,
M. J.
,
Hutchison
,
K. A.
,
Kessler
,
B.
,
Loftis
,
B.
, et al
(
2007
).
The English lexicon project
.
Behavior Research Methods
,
39
,
445
459
.
Barsalou
,
L. W.
(
2008
).
Cognitive and neural contributions to understanding the conceptual system
.
Current Directions in Psychological Science
,
17
,
91
95
.
Barsalou
,
L. W.
,
Dutriaux
,
L.
, &
Scheepers
,
C.
(
2018
).
Moving beyond the distinction between concrete and abstract concepts
.
Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences
,
373
,
20170144
.
Bates
,
E.
,
Wilson
,
S. M.
,
Saygin
,
A. P.
,
Dick
,
F.
,
Sereno
,
M. I.
,
Knight
,
R. T.
, et al
(
2003
).
Voxel-based lesion–symptom mapping
.
Nature Neuroscience
,
6
,
448
450
.
Beaulieu
,
C.
(
2002
).
The basis of anisotropic water diffusion in the nervous system—A technical review
.
NMR in Biomedicine
,
15
,
435
455
.
Bestmann
,
S.
,
Swayne
,
O.
,
Blankenburg
,
F.
,
Ruff
,
C. C.
,
Teo
,
J.
,
Weiskopf
,
N.
, et al
(
2010
).
The role of contralesional dorsal premotor cortex after stroke as studied with concurrent TMS-fMRI
.
Journal of Neuroscience
,
30
,
11926
11937
.
Binder
,
J. R.
, &
Desai
,
R. H.
(
2011
).
The neurobiology of semantic memory
.
Trends in Cognitive Sciences
,
15
,
527
536
.
Binder
,
J. R.
,
Desai
,
R. H.
,
Graves
,
W. W.
, &
Conant
,
L. L.
(
2009
).
Where is the semantic system? A critical review and meta-analysis of 120 functional neuroimaging studies
.
Cerebral Cortex
,
19
,
2767
2796
.
Binkofski
,
F.
, &
Buccino
,
G.
(
2006
).
The role of ventral premotor cortex in action execution and action understanding
.
Journal of Physiology: Paris
,
99
,
396
405
.
Bocanegra
,
Y.
,
Garcia
,
A. M.
,
Pineda
,
D.
,
Buriticá
,
O.
,
Villegas
,
A.
,
Lopera
,
F.
, et al
(
2015
).
Syntax, action verbs, action semantics, and object semantics in Parkinson's disease: Dissociability, progression, and executive influences
.
Cortex
,
69
,
237
254
.
Bonilha
,
L.
,
Hillis
,
A. E.
,
Hickok
,
G.
,
den Ouden
,
D. B.
,
Rorden
,
C.
, &
Fridriksson
,
J.
(
2017
).
Temporal lobe networks supporting the comprehension of spoken words
.
Brain
,
140
,
2370
2380
.
Bonner
,
M. F.
, &
Grossman
,
M.
(
2012
).
Gray matter density of auditory association cortex relates to knowledge of sound concepts in primary progressive aphasia
.
Journal of Neuroscience
,
32
,
7986
7991
.
Borra
,
E.
, &
Luppino
,
G.
(
2019
).
Large-scale temporo-parieto-frontal networks for motor and cognitive motor functions in the primate brain
.
Cortex
,
118
,
19
37
.
Bouix
,
S.
,
Pasternak
,
O.
,
Rathi
,
Y.
,
Pelavin
,
P. E.
,
Zafonte
,
R.
, &
Shenton
,
M. E.
(
2013
).
Increased gray matter diffusion anisotropy in patients with persistent post-concussive symptoms following mild traumatic brain injury
.
PLoS One
,
8
,
e66205
.
Brandi
,
M. L.
,
Wohlschläger
,
A.
,
Sorg
,
C.
, &
Hermsdörfer
,
J.
(
2014
).
The neural correlates of planning and executing actual tool use
.
Journal of Neuroscience
,
34
,
13183
13194
.
Buccino
,
G.
,
Dalla Volta
,
R.
,
Arabia
,
G.
,
Morelli
,
M.
,
Chiriaco
,
C.
,
Lupo
,
A.
, et al
(
2017
).
Processing graspable object images and their nouns is impaired in Parkinson's disease patients
.
Cortex
,
100
,
32
39
.
Buxbaum
,
L. J.
, &
Kalénine
,
S.
(
2010
).
Action knowledge, visuomotor activation, and embodiment in the two action systems
.
Annals of the New York Academy of Sciences
,
1191
,
201
218
.
Buxbaum
,
L. J.
,
Kyle
,
K.
,
Grossman
,
M.
, &
Coslett
,
H. B.
(
2007
).
Left inferior parietal representations for skilled hand–object interactions: Evidence from stroke and corticobasal degeneration
.
Cortex
,
43
,
411
423
.
Buxbaum
,
L. J.
, &
Saffran
,
E. M.
(
2002
).
Knowledge of object manipulation and object function: Dissociations in apraxic and nonapraxic subjects
.
Brain and Language
,
82
,
179
199
.
Cacciari
,
C.
,
Bolognini
,
N.
,
Senna
,
I.
,
Pellicciari
,
M. C.
,
Miniussi
,
C.
, &
Papagno
,
C.
(
2011
).
Literal, fictive and metaphorical motion sentences preserve the motion component of the verb: A TMS study
.
Brain and Language
,
119
,
149
157
.
Cappellani
,
R.
,
Bergsland
,
N.
,
Weinstock-Guttman
,
B.
,
Kennedy
,
C.
,
Carl
,
E.
,
Ramasamy
,
D. P.
, et al
(
2014
).
Subcortical deep gray matter pathology in patients with multiple sclerosis is associated with white matter lesion burden and atrophy but not with cortical atrophy: A diffusion tensor MRI study
.
American Journal of Neuroradiology
,
35
,
912
919
.
Caramazza
,
A.
,
Anzellotti
,
S.
,
Strnad
,
L.
, &
Lingnau
,
A.
(
2014
).
Embodied cognition and mirror neurons: A critical assessment
.
Annual Review of Neuroscience
,
37
,
1
15
.
Cardona
,
J. F.
,
Kargieman
,
L.
,
Sinay
,
V.
,
Gershanik
,
O.
,
Gelormini
,
C.
,
Amoruso
,
L.
, et al
(
2014
).
How embodied is action language? Neurological evidence from motor diseases
.
Cognition
,
131
,
311
322
.
Caspers
,
S.
,
Zilles
,
K.
,
Laird
,
A. R.
, &
Eickhoff
,
S. B.
(
2010
).
ALE meta-analysis of action observation and imitation in the human brain
.
Neuroimage
,
50
,
1148
1167
.
Chinta
,
S. J.
, &
Andersen
,
J. K.
(
2005
).
Dopaminergic neurons
.
International Journal of Biochemistry and Cell Biology
,
37
,
942
946
.
Coltheart
,
M.
(
1981
).
The MRC psycholinguistic database
.
Quarterly Journal of Experimental Psychology Section A
,
33
,
497
505
.
Desai
,
R. H.
,
Binder
,
J. R.
,
Conant
,
L. L.
, &
Seidenberg
,
M. S.
(
2010
).
Activation of sensory–motor areas in sentence comprehension
.
Cerebral Cortex
,
20
,
468
478
.
Desai
,
R. H.
,
Choi
,
W.
,
Lai
,
V. T.
, &
Henderson
,
J. M.
(
2016
).
Toward semantics in the wild: Activation to manipulable nouns in naturalistic reading
.
Journal of Neuroscience
,
36
,
4050
4055
.
Desai
,
R. H.
,
Conant
,
L. L.
,
Binder
,
J. R.
,
Park
,
H.
, &
Seidenberg
,
M. S.
(
2013
).
A piece of the action: Modulation of sensory–motor regions by action idioms and metaphors
.
Neuroimage
,
83
,
862
869
.
Desai
,
R. H.
,
Herter
,
T.
,
Riccardi
,
N.
,
Rorden
,
C.
, &
Fridriksson
,
J.
(
2015
).
Concepts within reach: Action performance predicts action language processing in stroke
.
Neuropsychologia
,
71
,
217
224
.
Desai
,
R. H.
,
Reilly
,
M.
, &
van Dam
,
W.
(
2018
).
The multifaceted abstract brain
.
Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences
,
373
,
20170122
.
Dreyer
,
F. R.
,
Frey
,
D.
,
Arana
,
S.
,
von Saldern
,
S.
,
Picht
,
T.
,
Vajkoczy
,
P.
, et al
(
2015
).
Is the motor system necessary for processing action and abstract emotion words? Evidence from focal brain lesions
.
Frontiers in Psychology
,
6
,
1661
.
Dreyer
,
D.
,
Loe
,
P.
,
Metz
,
C.
, &
Whitsel
,
B.
(
1975
).
Representation of head and face in postcentral gyrus of the macaque
.
Journal of Neurophysiology
,
38
,
714
733
.
Eaves
,
D. L.
,
Behmer
,
L. P.
, Jr.
, &
Vogt
,
S.
(
2016
).
EEG and behavioural correlates of different forms of motor imagery during action observation in rhythmical actions
.
Brain and Cognition
,
106
,
90
103
.
Eaves
,
D. L.
,
Riach
,
M.
,
Holmes
,
P. S.
, &
Wright
,
D. J.
(
2016
).
Motor imagery during action observation: A brief review of evidence, theory and future research opportunities
.
Frontiers in Neuroscience
,
10
,
514
.
Fadiga
,
L.
,
Craighero
,
L.
, &
D'Ausilio
,
A.
(
2009
).
Broca's area in language, action, and music
.
Annals of the New York Academy of Sciences
,
1169
,
448
458
.
Faria
,
A. V.
,
Joel
,
S. E.
,
Zhang
,
Y.
,
Oishi
,
K.
,
van Zjil
,
P. C.
,
Miller
,
M. I.
, et al
(
2012
).
Atlas-based analysis of resting-state functional connectivity: Evaluation for reproducibility and multi-modal anatomy-function correlation studies
.
Neuroimage
,
61
,
613
621
.
Fernandino
,
L.
,
Binder
,
J. R.
,
Desai
,
R. H.
,
Pendl
,
S. L.
,
Humphries
,
C. J.
,
Gross
,
W. L.
, et al
(
2016
).
Concept representation reflects multimodal abstraction: A framework for embodied semantics
.
Cerebral Cortex
,
26
,
2018
2034
.
Fernandino
,
L.
,
Conant
,
L. L.
,
Binder
,
J. R.
,
Blindauer
,
K.
,
Hiner
,
B.
,
Spangler
,
K.
, et al
(
2013a
).
Parkinson's disease disrupts both automatic and controlled processing of action verbs
.
Brain and Language
,
127
,
65
74
.
Fernandino
,
L.
,
Conant
,
L. L.
,
Binder
,
J. R.
,
Blindauer
,
K.
,
Hiner
,
B.
,
Spangler
,
K.
, et al
(
2013b
).
Where is the action? Action sentence processing in Parkinson's disease
.
Neuropsychologia
,
51
,
1510
1517
.
Fernandino
,
L.
,
Humphries
,
C. J.
,
Conant
,
L. L.
,
Seidenberg
,
M. S.
, &
Binder
,
J. R.
(
2016
).
Heteromodal cortical areas encode sensory–motor features of word meaning
.
Journal of Neuroscience
,
36
,
9763
9769
.
Filimon
,
F.
,
Nelson
,
J. D.
,
Hagler
,
D. J.
, &
Sereno
,
M. I.
(
2007
).
Human cortical representations for reaching: Mirror neurons for execution, observation, and imagery
.
Neuroimage
,
37
,
1315
1328
.
Freedman
,
D.
, &
Lane
,
D.
(
1983
).
A nonstochastic interpretation of reported significance levels
.
Journal of Business & Economic Statistics
,
1
,
292
298
.
Frey
,
S. H.
(
2008
).
Tool use, communicative gesture and cerebral asymmetries in the modern human brain
.
Philosophical Transactions of the Royal Society of London, Series B: Biologial Sciences
,
363
,
1951
1957
.
Frey
,
S. H.
,
Vinton
,
D.
,
Norlund
,
R.
, &
Grafton
,
S. T.
(
2005
).
Cortical topography of human anterior intraparietal cortex active during visually guided grasping
.
Cognitive Brain Research
,
23
,
397
405
.
Gallese
,
V.
, &
Lakoff
,
G.
(
2005
).
The brain's concepts: The role of the Sensory–motor system in conceptual knowledge
.
Cognitive Neuropsychology
,
22
,
455
479
.
Garcea
,
F. E.
,
Dombovy
,
M.
, &
Mahon
,
B. Z.
(
2013
).
Preserved tool knowledge in the context of impaired action knowledge: Implications for models of semantic memory
.
Frontiers in Human Neuroscience
,
7
,
120
.
Goldenberg
,
G.
(
2009
).
Apraxia and the parietal lobes
.
Neuropsychologia
,
47
,
1449
1459
.
Goldenberg
,
G.
, &
Randerath
,
J.
(
2015
).
Shared neural substrates of apraxia and aphasia
.
Neuropsychologia
,
75
,
40
49
.
Gough
,
P. M.
,
Riggio
,
L.
,
Chersi
,
F.
,
Sato
,
M.
,
Fogassi
,
L.
, &
Buccino
,
G.
(
2012
).
Nouns referring to tools and natural objects differentially modulate the motor system
.
Neuropsychologia
,
50
,
19
25
.
Grodzinsky
,
Y.
, &
Santi
,
A.
(
2008
).
The battle for Broca's region
.
Trends in Cognitive Sciences
,
12
,
474
480
.
Grossman
,
M.
,
Anderson
,
C.
,
Khan
,
A.
,
Avants
,
B.
,
Elman
,
L.
, &
McCluskey
,
L.
(
2008
).
Impaired action knowledge in amyotrophic lateral sclerosis
.
Neurology
,
71
,
1396
1401
.
Haaland
,
K. Y.
,
Harrington
,
D. L.
, &
Knight
,
R. T.
(
2000
).
Neural representations of skilled movement
.
Brain
,
123
,
2306
2313
.
Habas
,
C.
, &
Cabanis
,
E. A.
(
2006
).
Cortical projections to the human red nucleus: A diffusion tensor tractography study with a 1.5-T MRI machine
.
Neuroradiology
,
48
,
755
762
.
Halsband
,
U.
,
Schmitt
,
J.
,
Weyers
,
M.
,
Binkofski
,
F.
,
Grützner
,
G.
, &
Freund
,
H. J.
(
2001
).
Recognition and imitation of pantomimed motor acts after unliateral parietal and premotor lesions: A perspective on apraxia
.
Neuropsychologia
,
39
,
200
216
.
Hansen
,
M. B.
,
Jespersen
,
S. N.
,
Leigland
,
L. A.
, &
Kroenke
,
C. D.
(
2013
).
Using diffusion anisotropy to characterize neuronal morphology in gray matter: The orientation distribution of axons and dendrites in the NeuroMorpho.org database
.
Frontiers in Integrative Neuroscience
,
7
,
31
.
Hartmann-von Monakow
,
K.
,
Akert
,
K.
, &
Kunzle
,
H.
(
1979
).
Projections of precentral and premotor cortex to the red nucleus and other midbrain areas in macaca fascicularis
.
Experimental Brain Research
,
34
,
91
105
.
Hauk
,
O.
,
Johnsrude
,
I.
, &
Pulvermüller
,
F.
(
2004
).
Somatotopic representation of action words in human motor and premotor cortex
.
Neuron
,
41
,
301
307
.
Heard
,
A.
,
Madan
,
C. R.
,
Protzner
,
A. B.
, &
Pexman
,
P. M.
(
2019
).
Getting a grip on sensorimotor effects in lexical–semantic processing
.
Behavior Research Methods
,
51
,
1
13
.
Hermsdörfer
,
J.
,
Terlinden
,
G.
,
Mühlau
,
M.
,
Goldenberg
,
G.
, &
Wohlschläger
,
A. M.
(
2007
).
Neural representations of pantomimed and actual tool use: Evidence from an event-related fMRI study
.
Neuroimage
,
36(Suppl. 2)
,
T109
T118
.
Herter
,
T. M.
,
Takei
,
T.
,
Munoz
,
D. P.
, &
Scott
,
S. H.
(
2015
).
Neurons in red nucleus and primary motor cortex exhibit similar responses to mechanical perturbations applied to the upper-limb during posture
.
Frontiers in Integrative Neuroscience
,
9
,
29
.
Hijdra
,
A.
, &
Verbeeten
,
B.
(
1991
).
Leukoaraiosis and ventricular enlargement in patients with ischemic stroke
.
Stroke
,
22
,
447
450
.
Hillis
,
A. E.
,
Rorden
,
C.
, &
Fridriksson
,
J.
(
2017
).
Brain regions essential for word comprehension: Drawing inferences from patients
.
Annals of Neurology
,
81
,
759
768
.
Hoehn
,
M. M.
, &
Yahr
,
M. D.
(
1967
).
Parkinsonism: Onset, progression, and mortality
.
Neurology
,
17
,
427
442
.
Hoffman
,
P.
,
Ralph
,
M. A. L.
, &
Rogers
,
T. T.
(
2013
).
Semantic diversity: A measure of semantic ambiguity based on variability in the contextual usage of words
.
Behavior Research Methods
,
45
,
718
730
.
Hosp
,
J. A.
,
Pekanovic
,
A.
,
Rioult-Pedotti
,
M. S.
, &
Luft
,
A. R.
(
2011
).
Dopaminergic projections from midbrain to primary motor cortex mediate motor skill learning
.
Journal of Neuroscience
,
31
,
2481
2487
.
Humphrey
,
D. R.
,
Gold
,
R.
, &
Reed
,
D. J.
(
1984
).
Sizes, laminar and topographic origins of cortical projections to the major divisions of the red nucleus in the monkey
.
Journal of Comparative Neurology
,
225
,
75
94
.
Ibáñez
,
A.
,
Cardona
,
J. F.
,
Dos Santos
,
Y. V.
,
Blenkmann
,
A.
,
Aravena
,
P.
,
Roca
,
M.
, et al
(
2013
).
Motor-language coupling: Direct evidence from early Parkinson's disease and intracranial cortical recordings
.
Cortex
,
49
,
968
984
.
Johansen-Berg
,
H.
,
Rushworth
,
M. F.
,
Bogdanovic
,
M. D.
,
Kischka
,
U.
,
Wimalaratna
,
S.
, &
Matthews
,
P. M.
(
2002
).
The role of ipsilateral premotor cortex in hand movement after stroke
.
Proceedings of the National Academy of Sciences, U.S.A.
,
99
,
14518
14523
.
Johnson-Frey
,
S. H.
(
2004
).
The neural bases of complex tool use in humans
.
Trends in Cognitive Sciences
,
8
,
71
78
.
Johnson-Frey
,
S. H.
,
Newman-Norlund
,
R.
, &
Grafton
,
S. T.
(
2005
).
A distributed left hemisphere network active during planning of everyday tool use skills
.
Cerebral Cortex
,
15
,
681
695
.
Kalénine
,
S.
,
Buxbaum
,
L. J.
, &
Coslett
,
H. B.
(
2010
).
Critical brain regions for action recognition: Lesion symptom mapping in left hemisphere stroke
.
Brain
,
133
,
3269
3280
.
Kantak
,
S. S.
,
Stinear
,
J. W.
,
Buch
,
E. R.
, &
Cohen
,
L. G.
(
2012
).
Rewiring the brain: Potential role of the premotor cortex in motor control, learning, and recovery of function following brain injury
.
Neurorehabilitation and Neural Repair
,
26
,
282
292
.
Karnath
,
H. O.
,
Sperber
,
C.
, &
Rorden
,
C.
(
2017
).
Mapping human brain lesions and their functional consequences
.
Neuroimage
,
165
,
180
189
.
Kemmerer
,
D.
,
Rudrauf
,
D.
,
Manzel
,
K.
, &
Tranel
,
D.
(
2012
).
Behavioral patterns and lesion sites associated with impaired processing of lexical and conceptual knowledge of actions
.
Cortex
,
48
,
826
848
.
Kiefer
,
M.
, &
Pulvermüller
,
F.
(
2012
).
Conceptual representations in mind and brain: Theoretical developments, current evidence and future directions
.
Cortex
,
48
,
805
825
.
Kimberg
,
D. Y.
,
Coslett
,
H. B.
, &
Schwartz
,
M. F.
(
2007
).
Power in voxel-based lesion–symptom mapping
.
Journal of Cognitive Neuroscience
,
19
,
1067
1080
.
Liebenthal
,
E.
,
Desai
,
R. H.
,
Humphries
,
C.
,
Sabri
,
M.
, &
Desai
,
A.
(
2014
).
The functional organization of the left STS: A large scale meta-analysis of PET and fMRI studies of healthy adults
.
Frontiers in Neuroscience
,
8
,
289
.
Lingnau
,
A.
, &
Downing
,
P. E.
(
2015
).
The lateral occipitotemporal cortex in action
.
Trends in Cognitive Sciences
,
19
,
268
277
.
Mahon
,
B. Z.
(
2015
).
The burden of embodied cognition
.
Canadian Journal of Experimental Psychology
,
69
,
172
178
.
Mahon
,
B. Z.
,
Milleville
,
S. C.
,
Negri
,
G. A.
,
Rumiati
,
R. I.
,
Caramazza
,
A.
, &
Martin
,
A.
(
2007
).
Action-related properties shape object representations in the ventral stream
.
Neuron
,
55
,
507
520
.
Marstaller
,
L.
,
Williams
,
M.
,
Rich
,
A.
,
Savage
,
G.
, &
Burianová
,
H.
(
2015
).
Aging and large-scale functional networks: White matter integrity, gray matter volume, and functional connectivity in the resting state
.
Neuroscience
,
290
,
369
378
.
Martin
,
M.
,
Dressing
,
A.
,
Bormann
,
T.
,
Schmidt
,
C. S. M.
,
Kummerer
,
D.
,
Beume
,
L.
, et al
(
2017
).
Componential network for the recognition of tool-associated actions: Evidence from voxel-based lesion–symptom mapping in acute stroke patients
.
Cerebral Cortex
,
27
,
4139
4152
.
Mayer
,
A. R.
,
Hanlon
,
F. M.
, &
Ling
,
J. M.
(
2015
).
Gray matter abnormalities in pediatric mild traumatic brain injury
.
Journal of Neurotrauma
,
32
,
723
730
.
McDowell
,
T.
,
Holmes
,
N. P.
,
Sunderland
,
A.
, &
Schürmann
,
M.
(
2018
).
TMS over the supramarginal gyrus delays selection of appropriate grasp orientation during reaching and grasping tools for use
.
Cortex
,
103
,
117
129
.
Meteyard
,
L.
,
Cuadrado
,
S. R.
,
Bahrami
,
B.
, &
Vigliocco
,
G.
(
2012
).
Coming of age: A review of embodiment and the neuroscience of semantics
.
Cortex
,
48
,
788
804
.
Mori
,
S.
,
Wakana
,
S.
,
Nagae-Poetscher
,
L. M.
, &
van Zijl
,
P. C.
(
2005
).
MRI atlas of human white matter
(1st ed.).
Amsterdam
:
Elsevier
.
Nachev
,
P.
,
Coulthard
,
E.
,
Jager
,
H. R.
,
Kennard
,
C.
, &
Husain
,
M.
(
2008
).
Enantiomorphic normalization of focally lesioned brains
.
Neuroimage
,
39
,
1215
1226
.
Negri
,
G. A.
,
Lunardelli
,
A.
,
Reverberi
,
C.
,
Gigli
,
G. L.
, &
Rumiati
,
R. I.
(
2007
).
Degraded semantic knowledge and accurate object use
.
Cortex
,
43
,
376
388
.
Negri
,
G. A.
,
Rumiati
,
R. I.
,
Zadini
,
A.
,
Ukmar
,
M.
,
Mahon
,
B. Z.
, &
Caramazza
,
A.
(
2007
).
What is the role of motor simulation in action and object recognition? Evidence from apraxia
.
Cognitive Neuropsychology
,
24
,
795
816
.
Oliveri
,
M.
,
Finocchiaro
,
C.
,
Shapiro
,
K.
,
Gangitano
,
M.
,
Caramazza
,
A.
, &
Pascual-Leone
,
A.
(
2004
).
All talk and no action: A transcranial magnetic stimulation study of motor cortex activation during action word production
.
Journal of Cognitive Neuroscience
,
16
,
374
381
.
Orban
,
G. A.
, &
Caruana
,
F.
(
2014
).
The neural basis of human tool use
.
Frontiers in Psychology
,
5
,
310
.
Osuka
,
S.
,
Matsushita
,
A.
,
Ishikawa
,
E.
,
Saotome
,
K.
,
Yamamoto
,
T.
,
Marushima
,
A.
, et al
(
2012
).
Elevated diffusion anisotropy in gray matter and the degree of brain compression
.
Journal of Neurosurgery
,
117
,
363
371
.
Papeo
,
L.
,
Lingnau
,
A.
,
Agosta
,
S.
,
Pascual-Leone
,
A.
,
Battelli
,
L.
, &
Caramazza
,
A.
(
2015
).
The origin of word-related motor activity
.
Cerebral Cortex
,
25
,
1668
1675
.
Papeo
,
L.
,
Negri
,
G. A.
,
Zadini
,
A.
, &
Rumiati
,
R. I.
(
2010
).
Action performance and action-word understanding: Evidence of double dissociations in left-damaged patients
.
Cognitive Neuropsychology
,
27
,
428
461
.
Peeters
,
R. R.
,
Rizzolatti
,
G.
, &
Orban
,
G. A.
(
2013
).
Functional properties of the left parietal tool use region
.
Neuroimage
,
78
,
83
93
.
Pelgrims
,
B.
,
Olivier
,
E.
, &
Andres
,
M.
(
2011
).
Dissociation between manipulation and conceptual knowledge of object use in the supramarginalis gyrus
.
Human Brain Mapping
,
32
,
1802
1810
.
Péran
,
P.
,
Démonet
,
J. F.
,
Pernet
,
C.
, &
Cardebat
,
D.
(
2003
).
Verb and noun generation tasks in Huntington's disease
.
Movement Disorders
,
19
,
565
571
.
Péran
,
P.
,
Nemmi
,
F.
,
Méligne
,
D.
,
Cardebat
,
D.
,
Peppe
,
A.
,
Rascol
,
O.
, et al
(
2013
).
Effect of levodopa on both verbal and motor representations of action in Parkinson's disease: A fMRI study
.
Brain and Language
,
125
,
324
329
.
Peters
,
D. M.
,
Fridriksson
,
J.
,
Stewart
,
J. C.
,
Richardson
,
J. D.
,
Rorden
,
C.
,
Bonilha
,
L.
, et al
(
2018
).
Cortical disconnection of the ipsilesional primary motor cortex is associated with gait speed and upper extremity motor impairment in chronic left hemispheric stroke
.
Human Brain Mapping
,
39
,
120
132
.
Pexman
,
P. M.
,
Muraki
,
E.
,
Sidhu
,
D. M.
,
Siakaluk
,
P. D.
, &
Yap
,
M. J.
(
2019
).
Quantifying sensorimotor experience: Body–object interaction ratings for more than 9,000 English words
.
Behavior Research Methods
,
51
,
453
466
.
Potok
,
W.
,
Maskiewicz
,
A.
,
Króliczak
,
G.
, &
Marangon
,
M.
(
2019
).
The temporal involvement of the left supramarginal gyrus in planning functional grasps: A neuronavigated TMS study
.
Cortex
,
111
,
16
34
.
Price
,
C. J.
, &
Friston
,
K. J.
(
2002
).
Degeneracy and cognitive anatomy
.
Trends in Cognitive Sciences
,
6
,
416
421
.
Pulvermüller
,
F.
,
Hauk
,
O.
,
Nikulin
,
V. V.
, &
Ilmoniemi
,
R. J.
(
2005
).
Functional links between motor and language systems
.
European Journal of Neuroscience
,
21
,
793
797
.
Reilly
,
M.
,
Howerton
,
O.
, &
Desai
,
R. H.
(
2019
).
Time-course of motor involvement in literal and metaphoric action sentence processing: A TMS study
.
Frontiers in Psychology
,
10
,
371
.
Riccardi
,
N.
,
Yourganov
,
G.
,
Rorden
,
C.
,
Fridriksson
,
J.
, &
Desai
,
R. H.
(
2019
).
Dissociating action and abstract verb comprehension poststroke
.
Cortex
,
120
,
131
146
.
Richter
,
M.
,
Miltner
,
W. H.
, &
Straube
,
T.
(
2008
).
Association between therapy outcome and right-hemispheric activation in chronic aphasia
.
Brain
,
131
,
1391
1401
.
Rizzolatti
,
G.
,
Fogassi
,
L.
, &
Gallese
,
V.
(
2001
).
Neurophysiological mechanisms underlying the understanding and imitation of action
.
Nature Reviews Neuroscience
,
2
,
661
670
.
Rorden
,
C.
,
Bonilha
,
L.
,
Fridriksson
,
J.
,
Bender
,
B.
, &
Karnath
,
H. O.
(
2012
).
Age-specific CT and MRI templates for spatial normalization
.
Neuroimage
,
61
,
957
965
.
Rosci
,
C.
,
Chiesa
,
V.
,
Laiacona
,
M.
, &
Capitani
,
E.
(
2003
).
Apraxia is not associated to a disproportionate naming impairment for manipulable objects
.
Brain and Cognition
,
53
,
412
415
.
Rossi
,
C.
,
Boss
,
A.
,
Steidle
,
G.
,
Martirosian
,
P.
,
Klose
,
U.
,
Capuani
,
S.
, et al
(
2008
).
Water diffusion anisotropy in white and gray matter of the human spinal cord
.
Journal of Magnetic Resonance Imaging
,
27
,
476
482
.
Rumiati
,
R. I.
,
Zanini
,
S.
,
Vorano
,
L.
, &
Shallice
,
T.
(
2001
).
A form of ideational apraxia as a delective deficit of contention scheduling
.
Cognitive Neuropsychology
,
18
,
617
642
.
Ruschel
,
M.
,
Knösche
,
T. R.
,
Friederici
,
A. D.
,
Turner
,
R.
,
Geyer
,
S.
, &
Anwander
,
A.
(
2014
).
Connectivity architecture and subdivision of the human inferior parietal cortex revealed by diffusion MRI
.
Cerebral Cortex
,
24
,
2436
2448
.
Salazar-López
,
E.
,
Schwaiger
,
B. J.
, &
Hermsdörfer
,
J.
(
2016
).
Lesion correlates of impairments in actual tool use following unilateral brain damage
.
Neuropsychologia
,
84
,
167
180
.
Saur
,
D.
,
Lange
,
R.
,
Baumgaertner
,
A.
,
Schraknepper
,
V.
,
Willmes
,
K.
,
Rijntjes
,
M.
, et al
(
2006
).
Dynamics of language reorganization after stroke
.
Brain
,
129
,
1371
1384
.
Siegel
,
J. S.
,
Ramsey
,
L. E.
,
Snyder
,
A. Z.
,
Metcalf
,
N. V.
,
Chacko
,
R. V.
,
Weinberger
,
K.
, et al
(
2016
).
Disruptions of network connectivity predict impairment in multiple behavioral domains after stroke
.
Proceedings of the National Academy of Sciences, U.S.A.
,
113
,
E4367
E4376
.
Skipper-Kallal
,
L. M.
,
Lacey
,
E. H.
,
Xing
,
S.
, &
Turkeltaub
,
P. E.
(
2017
).
Right hemisphere remapping of naming functions depends on lesion size and location in poststroke aphasia
.
Neural Plasticity
,
2017
,
8740353
.
Tillotson
,
S. M.
,
Siakaluk
,
P. D.
, &
Pexman
,
P. M.
(
2008
).
Body–object interaction ratings for 1,618 monosyllabic nouns
.
Behavior Research Methods
,
40
,
1075
1078
.
Tranel
,
D.
,
Adolphs
,
R.
,
Damasio
,
H.
, &
Damasio
,
A. R.
(
2001
).
A neural basis for the retrieval of words for actions
.
Cognitive Neuropsychology
,
18
,
655
674
.
Tranel
,
D.
,
Kemmerer
,
D.
,
Adolphs
,
R.
,
Damasio
,
H.
, &
Damasio
,
A. R.
(
2003
).
Neural correlates of conceptual knowledge for actions
.
Cognitive Neuropsychology
,
20
,
409
432
.
Trumpp
,
N. M.
,
Kliese
,
D.
,
Hoenig
,
K.
,
Haarmeier
,
T.
, &
Kiefer
,
M.
(
2013
).
Losing the sound of concepts: Damage to auditory association cortex impairs the processing of sound-related concepts
.
Cortex
,
49
,
474
486
.
van Tol
,
M.-J.
,
Li
,
M.
,
Metzger
,
C. D.
,
Hailla
,
N.
,
Horn
,
D. I.
,
Li
,
W.
, et al
(
2014
).
Local cortical thinning links to resting-state disconnectivity in major depressive disorder
.
Psychological Medicine
,
44
,
2053
2065
.
Verstraete
,
E.
,
van den Heuvel
,
M. P.
,
Veldink
,
J. H.
,
Blanken
,
N.
,
Mandl
,
R. C.
,
Pol
,
H. E. H.
, et al
(
2010
).
Motor network degeneration in amyotrophic lateral sclerosis: A structural and functional connectivity study
.
PLoS One
,
5
,
e13664
.
Vukovic
,
N.
,
Feurra
,
M.
,
Shpektor
,
A.
,
Myachykov
,
A.
, &
Shtyrov
,
Y.
(
2017
).
Primary motor cortex functionally contributes to language comprehension: An online rTMS study
.
Neuropsychologia
,
96
,
222
229
.
Wakana
,
S.
,
Jiang
,
H.
,
Nagae-Poetscher
,
L. M.
,
van Zijl
,
P. C.
, &
Mori
,
S.
(
2004
).
Fiber tract-based atlas of human white matter anatomy
.
Radiology
,
230
,
77
87
.
Wang
,
P.
,
Jia
,
X.
,
Zhang
,
M.
,
Cao
,
Y.
,
Zhao
,
Z.
,
Shan
,
Y.
, et al
(
2018
).
Correlation of longitudinal gray matter volume changes and motor recovery in patients after pontine infarction
.
Frontiers in Neurology
,
9
,
312
.
Watson
,
C. E.
,
Gotts
,
S. J.
,
Martin
,
A.
, &
Buxbaum
,
L. J.
(
2018
).
Bilateral functional connectivity at rest predicts apraxic symptoms after left hemisphere stroke
.
Neuroimage: Clinical
,
21
,
101526
.
Weiss
,
P. H.
,
Ubben
,
S. D.
,
Kaesberg
,
S.
,
Kalbe
,
E.
,
Kessler
,
J.
,
Liebig
,
T.
, et al
(
2016
).
Where language meets meaningful action: A combined behavior and lesion analysis of aphasia and apraxia
.
Brain Structure & Function
,
221
,
563
576
.
Weston
,
P. S.
,
Simpson
,
I. J.
,
Ryan
,
N. S.
,
Ourselin
,
S.
, &
Fox
,
N. C.
(
2015
).
Diffusion imaging changes in grey matter in Alzheimer's disease: A potential marker of early neurodegeneration
.
Alzheimer's Research & Therapy
,
7
,
47
.
Whitsel
,
B.
,
Dreyer
,
D.
, &
Roppolo
,
J.
(
1971
).
Determinants of body representation in postcentral gyrus of macaques
.
Journal of Neurophysiology
,
34
,
1018
1034
.
Willems
,
R. M.
,
Labruna
,
L.
,
D'Esposito
,
M.
,
Ivry
,
R.
, &
Casasanto
,
D.
(
2011
).
A functional role for the motor system in language understanding: Evidence from theta-burst transcranial magnetic stimulation
.
Psychological Science
,
22
,
849
854
.
Willems
,
R. M.
,
Ozyurek
,
A.
, &
Hagoort
,
P.
(
2009
).
Differential roles for left inferior frontal and superior temporal cortex in multimodal integration of action and language
.
Neuroimage
,
47
,
1992
2004
.
Winkler
,
A. M.
,
Ridgway
,
G. R.
,
Webster
,
M. A.
,
Smith
,
S. M.
, &
Nichols
,
T. E.
(
2014
).
Permutation inference for the general linear model
.
Neuroimage
,
92
,
381
397
.
Yourganov
,
G.
,
Fridriksson
,
J.
,
Stark
,
B.
, &
Rorden
,
C.
(
2018
).
Removal of artifacts from resting-state fMRI data in stroke
.
Neuroimage: Clinical
,
17
,
297
305
.