Action Verb Processing Research Articles

Semantic priming in the motor cortex

Reading action verbs is associated with activity in the motor cortices involved in performing the corresponding actions. Here, we present new evidence that the motor cortex is involved in semantic processing of bodily action verbs. In contrast to previous studies, we used a direct, nonbehavioural index of semantic processing after repetitive transcranial magnetic stimulation (rTMS). Participants saw pairs of hand-related (e.g. to grab-to point) or mouth-related (e.g. to speak-to sing) verbs, whereas semantic priming was assessed using event-related potentials. Presentation of the first verb coincided with rTMS over the participant's cortical-left hand area and event-related brain potentials were analysed time-locked to the presentation onset of the second verb. Semantic integration - indexed by the N400 brain potential - was impaired for hand-related but not for mouth-related verb pairs after rTMS. This finding provides strong evidence that the motor cortex is involved in semantic encoding of action verbs, and supports the 'embodied semantics' hypothesis.

Action-verb processing in Parkinson’s disease: new pathways for motor–language coupling

Recent studies suggest that action-verb processing is particularly affected in early stage Parkinson's disease (PD), highlighting the potential role of subcortical areas in language processing and in the semantic integration of actions. However, this disorder-related language impairment is frequently unrecognized by clinicians and often remains untreated. Early detection of action-language processing deficits could be critical for diagnosing and developing treatment strategies for PD. In this article, we review how action-verb processing is affected in PD and propose a model in which multiple and parallel frontotemporal circuits between the cortex and the basal ganglia provide the anatomic substrate for supporting action-language processing. We hypothesize that contextual coupling of action-language networks are partially dependent on cortical-subcortical integration, and not only on somatotopic motor cortical organization or in a mirror neuron system. This hypothesis is supported by both experimental and clinical evidence. Then, we identify further research steps that would help to determine the reliability of action-language impairments as an early marker of PD. Finally, theoretical implications for clinical assessment and for models of action-language interaction (action-perception cycle theories, mirror system models of language, and embodied cognition approaches to language) are discussed.

10 Modulation of arm reaching movements during hand-related verb processing in Parkinson’s patients

Background and methods Several studies have demonstrated that action verbs processing is selectively impaired in Parkinson’s disease (PD) patients. These results suggest that the integrity of the motor system is needed to properly elaborate action related verbs. To further explore this topic we exploited a go/no-go paradigm where 15 non-demented idiopathic PD patients were required to perform arm reaching movements toward a target when verbs expressing either hand or foot actions were shown and to refrain from moving when abstract verbs were presented. Patients were tested once on dopaminergic medication (ON-state) and once after a washout period of 12 h (OFF-state). Experimental conditions were counterbalanced across patients and administered in two different experimental sessions occurring on different days. Results Preliminary results showed that reaction times (RT) and percentage of errors increased when the verb involved the same effector used to give the response. This interference effect tended to be greater in the OFF-state than in the ON-state. Furthermore, a control experiment showed that when the color of the printed verb and not its meaning was the cue for movement execution, the differences between RTs and errors percentage between verb categories disappeared. Thus the interference occurs only when the semantic content of a verb has to be retrieved. Conclusions The described modulation of planning and execution of arm reaching movements induced by hand-related verbs suggests that the mechanism underlying processing of action-verbs is modulated by dopaminergic medications.

The effects of rTMS over the primary motor cortex: The link between action and language

Is the primary motor cortex (M1) necessary for language comprehension? The present study investigates the role of the primary motor cortex during verbs comprehension, within the framework of the embodied theories of language. We applied rTMS over the right and left hand portion of M1 and tested the effects of the stimulation toward the processing of hand-related action verbs versus abstract verbs. Results underlined a specific inhibition effect following left stimulation, only with hand-related action verbs. These findings seem to corroborate the hypothesis of a functional role of M1 in action verbs comprehension.

Parkinson’s disease disrupts both automatic and controlled processing of action verbs

The problem of how word meaning is processed in the brain has been a topic of intense investigation in cognitive neuroscience. While considerable correlational evidence exists for the involvement of sensory-motor systems in conceptual processing, it is still unclear whether they play a causal role. We investigated this issue by comparing the performance of patients with Parkinson’s disease (PD) with that of age-matched controls when processing action and abstract verbs. To examine the effects of task demands, we used tasks in which semantic demands were either implicit (lexical decision and priming) or explicit (semantic similarity judgment). In both tasks, PD patients’ performance was selectively impaired for action verbs (relative to controls), indicating that the motor system plays a more central role in the processing of action verbs than in the processing of abstract verbs. These results argue for a causal role of sensory-motor systems in semantic processing.

Effects of stimulation of the subthalamic nucleus on naming and reading nouns and verbs in Parkinson's disease

An impairment for verbs has been described in patients with Parkinson's disease (PD), suggesting that a disruption of frontal–subcortical circuits may result in dysfunction of the neural systems involved in action–verb processing. A previous study suggested that deep brain stimulation (DBS) of the subthalamic nucleus (STN) during verb generation may affect the ability to select from many competing lexical alternatives.In this study, 12 PD patients who had undergone bilateral STN DBS and completed an 8-year follow-up and 14 matched normal controls were administered action and object naming tasks and verb and noun reading tasks. Their responses were recorded using a microphone, resulting in a signal that marked the onset of the verbal response and allowed to measure response times (RTs). Accuracy was scored manually. ResultsOverall performance in naming (independently of stimulation): In naming task controls were faster and more accurate than PD patients. In both groups, performance (accuracy and RTs) was worse on action naming than object naming. PD patients were significantly slower than controls in naming actions. Effect of stimulation: Compared with the OFF stimulation condition, in ON stimulation condition PD patients showed improved performance on object and action naming tasks (increased accuracy, faster RTs), with a decreased number of semantic errors. Some evidence also emerged that action naming in the ON stimulation condition improved more than object naming. On noun and verb reading tasks, although accuracy was at ceiling in both groups and no significant difference was observed in RTs for nouns and verbs, PD patients were slower than controls. ConclusionsOur findings suggest that STN DBS may improve lexical search in PD patients. We hypothesize that STN stimulation may facilitate the motor components involved in naming and reading tasks (increased speed of speech onset), resulting in shorter RTs in both naming and reading and, to some extent, in increased accuracy in naming due to fewer omissions (no response within the 7500ms time limit). However, to account for greater accuracy in naming due to decreased number of semantic errors in the ON stimulation condition, we hypothesize that STN stimulation restores the activity of the corticostriatal circuits involved in selection processes of a target word among different alternatives.

Motor-language coupling: Direct evidence from early Parkinson’s disease and intracranial cortical recordings

Language and action systems are functionally coupled in the brain as demonstrated by converging evidence using Functional magnetic resonance imaging (fMRI), electroencephalography (EEG), transcranial magnetic stimulation (TMS), and lesion studies. In particular, this coupling has been demonstrated using the action-sentence compatibility effect (ACE) in which motor activity and language interact. The ACE task requires participants to listen to sentences that described actions typically performed with an open hand (e.g., clapping), a closed hand (e.g., hammering), or without any hand action (neutral); and to press a large button with either an open hand position or closed hand position immediately upon comprehending each sentence. The ACE is defined as a longer reaction time (RT) in the action-sentence incompatible conditions than in the compatible conditions. Here we investigated direct motor-language coupling in two novel and uniquely informative ways. First, we measured the behavioural ACE in patients with motor impairment (early Parkinson’s disease – EPD), and second, in epileptic patients with direct electrocorticography (ECoG) recordings. In experiment 1, EPD participants with preserved general cognitive repertoire, showed a much diminished ACE relative to non-EPD volunteers. Moreover, a correlation between ACE performance and action-verb processing (kissing and dancing test – KDT) was observed. Direct cortical recordings (ECoG) in motor and language areas (experiment 2) demonstrated simultaneous bidirectional effects: motor preparation affected language processing (N400 at left inferior frontal gyrus and middle/superior temporal gyrus), and language processing affected activity in movement-related areas (motor potential at premotor and M1). Our findings show that the ACE paradigm requires ongoing integration of preserved motor and language coupling (abolished in EPD) and engages motor-temporal cortices in a bidirectional way. In addition, both experiments suggest the presence of a motor-language network which is not restricted to somatotopically defined brain areas. These results open new pathways in the fields of motor diseases, theoretical approaches to language understanding, and models of action-perception coupling.

Motor Simulation during Action Word Processing in Neurosurgical Patients

The role that human motor areas play in linguistic processing is the subject of a stimulating debate. Data from nine neurosurgical patients with selective lesions of the precentral and postcentral sulcus could provide a direct answer as to whether motor area activation is necessary for action word processing. Action-related verbs (face-, hand-, and feet-related verbs plus neutral verbs) silently read were processed for (i) motor imagery by vividness ratings and (ii) frequency ratings. Although no stimulus- or task-dependent modulation was found in the RTs of healthy controls, patients showed a task × stimulus interaction resulting in a stimulus-dependent somatotopic pattern of RTs for the imagery task. A lesion affecting a part of the cortex that represents a body part also led to slower RTs during the creation of mental images for verbs describing actions involving that same body part. By contrast, no category-related differences were seen in the frequency judgment task. This task-related dissociation suggests that the sensorimotor area is critically involved in processing action verbs only when subjects are simulating the corresponding movement. These findings have important implications for the ongoing discussion regarding the involvement of the sensorimotor cortex in linguistic processing.

Action verb understanding in first-episode schizophrenia: Is there evidence for a simulation deficit?

Schizophrenia is often associated with deficits in the domain of language, which are thought to be closely related to deficits in the structure of semantic knowledge. The main aim of the present study was to behaviorally investigate whether semantic impairments in schizophrenia are present also at the very basic level of action verb processing, in particular at the level of motor simulation. We used a go-no go paradigm both for a semantic decision task (with either an early, EGD, or a delayed go-signal delivery, DGD) and for a lexical decision task (control task). Only the first task requires motor simulation to be solved. We found that first-episode schizophrenia (FES) patients, like healthy control (HC) participants, use motor simulation as a basic strategy to semantically judge action verbs. In the EGD condition, both motor simulation and action verb understanding seem to be preserved in FES. However, differently from HC participants, FES patients kept on using the simulation strategy also with the DGD condition, whereas, simultaneously, task performance during this condition appeared to be less efficient and sensitive. Voxel-based morphometry analysis suggested that this altered performance in FES patients could be related to structural brain abnormalities in the right dorsolateral prefrontal cortex. We propose that a prolonged motor simulation in FES may serve as a compensatory strategy for impairments in the selection of action representation and/or for memory deficits disclosed by the DGD condition during the semantic decision task investigated in the present study.

Language-induced modulation during the prediction of others’ actions

Processing of action words has been shown to influence the perception of the actions the words refer to. Specifically, the accuracy with which people predict the future course of actions observed in another individual seems to be affected by verbal primes. Two processes may be involved in action prediction; dynamic simulation (updating) and static matching. The present study examined this issue by testing the impact of action verb processing on action prediction performance using a masked priming paradigm. Evidence of dynamic updating was revealed after prime verbs expressing dynamic actions (e.g., 'to catch') but not those expressing static actions (e.g., 'to lean'). In contrast to previous work, the primes were masked and did not require any response at all. Hence, our results indicate that implicit action-related linguistic processing may trigger action simulation that in turn might affect action prediction (see also Liepelt, Dolk, & Prinz, Psychological Research, 2012, in this issue).

The role of volleyball expertise in motor simulation

We explored the impact of motor e xperience on the interaction between implicit motor simulation and language-processing. In an action familiarity judgment task, expert volleyball players, fans and novices were presented with semantically correct sentences describing possible and not possible motor actions, all as negative or positive contexts, e.g., “Don't shank!” or “Assist!”. As processing negated action-phrases is known to reduce simulation states, exposure to negative or positive contexts was used here to test how simulation varies according to motor feasibility (possible, impossible) and experience (experts and fans). A significant group × stimulus × context interaction showed that athletes and fans, took longer to process negative than positive contexts for possible actions, compared to action-impossible sentences. In addition, experts were significantly faster and more accurate than fans and, in turn, they were both more accurate than novices. Thus, implicit motor simulation impacts on action-verb processing depending on (i) the domain-relevant expertise, (ii) the feasibility of the actions, and (iii) on whether scenes are presented in a negated context. These results suggest that the implicit triggering of motor representations is modulated by the context and it is tuned to people's motor repertoire, even when actions are described linguistically.

Action Semantics Modulate Action Prediction

Previous studies have demonstrated that action prediction involves an internal action simulation that runs time-locked to the real action. The present study replicates and extends these findings by indicating a real-time simulation process (Graf et al., 2007), which can be differentiated from a similarity-based evaluation of internal action representations. Moreover, results showed that action semantics modulate action prediction accuracy. The semantic effect was specified by the processing of action verbs and concrete nouns (Experiment 1) and, more specifically, by the dynamics described by action verbs (Experiment 2) and the speed described by the verbs (e.g., "to catch" vs. "to grasp" vs. "to stretch"; Experiment 3). These results propose a linkage between action simulation and action semantics as two yet unrelated domains, a view that coincides with a recent notion of a close link between motor processes and the understanding of action language.

To move or not to move: imperatives modulate action-related verb processing in the motor system

It has been suggested that the processing of action-related words involves activation of the motor circuitry. Using fMRI (functional magnetic resonance imaging), the current study further explored the interaction between action and language by investigating whether the linguistic context, in which an action word occurs, modulates motor circuitry activity related to the processing of action words. To this end, we examined whether the presentation of hand action-related verbs as positive or negative imperatives, for example, “Do grasp” or “Don't write,” modulates neural activity in the hand area of primary motor cortex (M1) or premotor cortex (Pm). Subjects ( n = 19) were asked to read silently the imperative phrases, in which both meaningful action verbs and meaningless pseudo-verbs were presented, and to decide whether they made sense (lexical decision task). At the behavioral level, response times in the lexical decision task were significantly longer for negative, compared to positive, imperatives. At the neural level, activity was differentially decreased by action verbs presented as negative imperatives for the premotor and the primary motor cortex of both hemispheres. The data suggest that context (here: positive vs. negative imperatives), in which an action verb is encountered, modulates the neural activity within key areas of the motor system. The finding implies that motor simulation (or motor planning) rather than semantic processing per se may underlie previously observed motor system activation related to action verb processing. Furthermore, the current data suggest that negative imperatives may inhibit motor simulation or motor planning processes.

The functional role of motor activation in language processing: Motor cortical oscillations support lexical-semantic retrieval

There is increasing experimental evidence that processing action-related language results in the automatic activation of associated regions of the motor and premotor cortex. However, the functional significance of motor activation in language processing is still under debate. In the present EEG study, we set out to investigate if language-induced motor activation primarily reflects the retrieval of lexical-semantic information or post-lexical motor imagery. The processing of action verbs was found accompanied by an early activation of motor-related brain areas, as reflected by a desynchronization in the mu- and beta-frequency bands which was localized to motor and premotor areas. A stronger motor activation was observed for verbs presented in an animal context (e.g. “The deer jumped over the stream”) compared to a human context (e.g. “The athlete jumped over the fence”) and motor resonance was directly modulated by the cloze probability of the noun–verb pairs. The onset of the motor effects preceded classical measures of semantic integration (i.e. the N400 component) and the strength of motor activation was found inversely related to the size of the N400 effect. These findings support the hypothesis that motor activation in language processing primarily supports the retrieval and integration of lexical-semantic information.

From hand to mouth: Verb embodiment in action naming

Introduction: A growing body of research suggests that the conceptual representations of actions are partly grounded in motor representations and, therefore, rely on the motor cortex for processing (Pulvermuller, 2005). This embodiment theory further suggests that lesions to motor areas of the brain may lead to action-verb processing deficits in aphasia (Arevalo et al., 2007). Research on embodiment has typically considered verb processing, not production. In the present fMRI study, we investigated whether lexical retrieval of action verbs in overt speech production involves motor areas that are associated with these verbs.

Effects of TMS on different stages of motor and non-motor verb processing in the primary motor cortex.

The embodied cognition hypothesis suggests that motor and premotor areas are automatically and necessarily involved in understanding action language, as word conceptual representations are embodied. This transcranial magnetic stimulation (TMS) study explores the role of the left primary motor cortex in action-verb processing. TMS-induced motor-evoked potentials from right-hand muscles were recorded as a measure of M1 activity, while participants were asked either to judge explicitly whether a verb was action-related (semantic task) or to decide on the number of syllables in a verb (syllabic task). TMS was applied in three different experiments at 170, 350 and 500 ms post-stimulus during both tasks to identify when the enhancement of M1 activity occurred during word processing. The delays between stimulus onset and magnetic stimulation were consistent with electrophysiological studies, suggesting that word recognition can be differentiated into early (within 200 ms) and late (within 400 ms) lexical-semantic stages, and post-conceptual stages. Reaction times and accuracy were recorded to measure the extent to which the participants' linguistic performance was affected by the interference of TMS with M1 activity. No enhancement of M1 activity specific for action verbs was found at 170 and 350 ms post-stimulus, when lexical-semantic processes are presumed to occur (Experiments 1–2). When TMS was applied at 500 ms post-stimulus (Experiment 3), processing action verbs, compared with non-action verbs, increased the M1-activity in the semantic task and decreased it in the syllabic task. This effect was specific for hand-action verbs and was not observed for action-verbs related to other body parts. Neither accuracy nor RTs were affected by TMS. These findings suggest that the lexical-semantic processing of action verbs does not automatically activate the M1. This area seems to be rather involved in post-conceptual processing that follows the retrieval of motor representations, its activity being modulated (facilitated or inhibited), in a top-down manner, by the specific demand of the task.

Language-Induced Motor Perturbations during the Execution of a Reaching Movement

In a recent study Boulenger et al. (2006) found that processing action verbs assisted reaching movement when the word was processed prior to movement onset and interfered with the movement when the word was processed at movement onset. The present study aimed to further corroborate the existence of such cross-talk between language processes and overt motor behaviour by demonstrating that the reaching movement can be disturbed by action words even when the words are presented delayed with respect to movement onset (50 ms and 200 ms). The results are compared to studies that show language-motor interaction in conditions where the word is presented prior to movement onset and are discussed within the context of embodied theories of language comprehension.

Action verbs and the primary motor cortex: A comparative TMS study of silent reading, frequency judgments, and motor imagery

Single pulse transcranial magnetic stimulation (TMS) was applied to the hand area of the left primary motor cortex or, as a control, to the vertex (STIMULATION: TMSM1 vs. TMSvertex) while right-handed volunteers silently read verbs related to hand actions. We examined three different tasks and time points for stimulation within the same experiment: subjects indicated with their left foot when they (i) had finished reading, (ii) had judged whether the corresponding movement involved a hand rotation after simulating the hand movement, and (iii) had judged whether they would frequently encounter the action verb in a newspaper (TASK: silent reading, motor imagery, and frequency judgment). Response times were compared between TMSM1 and TMSvertex, both applied at different time points after stimulus onset (DELAY: 150, 300, 450, 600, and 750ms).TMSM1 differentially modulated task performance: there was a significant facilitatory effect of TMSM1 for the imagery task only (about 88ms), with subjects responding about 10% faster (compared to TMSvertex). In contrast, response times for silent reading and frequency judgments were unaffected by TMSM1. No differential effect of the time point of TMSM1 was observed.The differential effect of TMSM1 when subjects performed a motor imagery task (relative to performing silent reading or frequency judgments with the same set of verbs) suggests that the primary motor cortex is critically involved in processing action verbs only when subjects are simulating the corresponding movement. This task-dependent effect of hand motor cortex TMS on the processing of hand-related action verbs is discussed with respect to the notion of embodied cognition and the associationist theory.

The influence of rTMS over prefrontal and motor areas in a morphological task: Grammatical vs. semantic effects

We investigated the differential role of two frontal regions in the processing of grammatical and semantic knowledge. Given the documented specificity of the prefrontal cortex for the grammatical class of verbs, and of the primary motor cortex for the semantic class of action words, we sought to investigate whether the prefrontal cortex is also sensitive to semantic effects, and whether the motor cortex is also sensitive to grammatical class effects. We used repetitive transcranial magnetic stimulation (rTMS) to suppress the excitability of a portion of left prefontal cortex (first experiment) and of the motor area (second experiment). In the first experiment we found that rTMS applied to the left prefrontal cortex delays the processing of action verbs’ retrieval, but is not critical for retrieval of state verbs and state nouns. In the second experiment we found that rTMS applied to the left motor cortex delays the processing of action words, both name and verbs, while it is not critical for the processing of state words.These results support the notion that left prefrontal and motor cortex are involved in the process of action word retrieval. Left prefrontal cortex subserves processing of both grammatical and semantic information, whereas motor cortex contributes to the processing of semantic representation of action words without any involvement in the representation of grammatical categories.

Comprehending Prehending: Neural Correlates of Processing Verbs with Motor Stems

The interaction between language and action systems has become an increasingly interesting topic of discussion in cognitive neuroscience. Several recent studies have shown that processing of action verbs elicits activation in the cerebral motor system in a somatotopic manner. The current study extends these findings to show that the brain responses for processing of verbs with specific motor meanings differ not only from that of other motor verbs, but, crucially, that the comprehension of verbs with motor meanings (i.e., greifen, to grasp) differs fundamentally from the processing of verbs with abstract meanings (i.e., denken, to think). Second, the current study investigated the neural correlates of processing morphologically complex verbs with abstract meanings built on stems with motor versus abstract meanings (i.e., begreifen, to comprehend vs. bedenken, to consider). Although residual effects of motor stem meaning might have been expected, we see no evidence for this in our data. Processing of morphologically complex verbs built on motor stems showed no differences in involvement of the motor system when compared with processing complex verbs with abstract stems. Complex verbs built on motor stems did show increased activation compared with complex verbs built on abstract stems in the right posterior temporal cortex. This result is discussed in light of the involvement of the right temporal cortex in comprehension of metaphoric or figurative language.