Language–motor interference reflected in MEG beta oscillations

Abstract

The involvement of the brain's motor system in action-related language processing can lead to overt interference with simultaneous action execution. The aim of the current study was to find evidence for this behavioural interference effect and to investigate its neurophysiological correlates using oscillatory MEG analysis. Subjects performed a semantic decision task on single action verbs, describing actions executed with the hands or the feet, and abstract verbs. Right hand button press responses were given for concrete verbs only. Therefore, longer response latencies for hand compared to foot verbs should reflect interference. We found interference effects to depend on verb imageability: overall response latencies for hand verbs did not differ significantly from foot verbs. However, imageability interacted with effector: while response latencies to hand and foot verbs with low imageability were equally fast, those for highly imageable hand verbs were longer than for highly imageable foot verbs. The difference is reflected in motor-related MEG beta band power suppression, which was weaker for highly imageable hand verbs compared with highly imageable foot verbs. This provides a putative neuronal mechanism for language-motor interference where the involvement of cortical hand motor areas in hand verb processing interacts with the typical beta suppression seen before movements. We found that the facilitatory effect of higher imageability on action verb processing time is perturbed when verb and motor response relate to the same body part. Importantly, this effect is accompanied by neurophysiological effects in beta band oscillations. The attenuated power suppression around the time of movement, reflecting decreased cortical excitability, seems to result from motor simulation during action-related language processing. This is in line with embodied cognition theories.

Effortful verb retrieval from semantic memory drives beta suppression in mesial frontal regions involved in action initiation

“…Comparison of their results highlights the advantage of MEG over EEG in capturing verb-related beta oscillations (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) and their modulations that were previously reported for synchronous MEG and EEG recordings (Nikulin, Nikulina, Yamashita, Rossi, & Kähkönen, 2005). MEG studies (Klepp et al, 2015;Niccolai et al, 2014) demonstrated that action verb processing was accompanied by suppression of sensorimotor oscillations within beta frequency band (15-25 Hz), starting 250 ms after verb presentation. Whereas in EEG studies, the changes in sensorimotor oscillatory activity during action verb presentation were reported mainly for the mu/alpha frequency range (van Elk et al, 2010), with no reliable beta suppression in the EEG electrodes positioned over the motor cortex (Schaller et al, 2017).…”

“…Recent EEG/MEG studies of motor-language interaction in the brain employed beta ERD to characterize brain activation elicited by action verbs (Klepp, Niccolai, Buccino, Schnitzler, & Biermann-Ruben, 2015;Niccolai et al, 2014;Schaller, Weiss, & Müller, 2017;van Elk, van Schie, Zwaan, & Bekkering, 2010). Comparison of their results highlights the advantage of MEG over EEG in capturing verb-related beta oscillations (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) and their modulations that were previously reported for synchronous MEG and EEG recordings (Nikulin, Nikulina, Yamashita, Rossi, & Kähkönen, 2005).…”

“…The topology of observed beta suppressions can be influenced by anticipatory motor preparation, as the above MEG studies required responses given by using the right hand (Klepp et al, 2015) or by changing the gaze direction (Niccolai et al, 2014). The topology of observed beta suppressions can be influenced by anticipatory motor preparation, as the above MEG studies required responses given by using the right hand (Klepp et al, 2015) or by changing the gaze direction (Niccolai et al, 2014).…”

EEG beta-power changes reflect motor involvement in abstract action language processing

“…However, they further discuss that later EEG components might either reflect a second comprehension step or postcomprehension processes . In studies by Pulvermüller, Härle, and Hummel (2001) and Klepp, Niccolai, Buccino, Schnitzler, and Biermann-Ruben (2015), effects concerning brain responses to action language stimuli were also reported as early as 250 ms and 400-600 ms post action verb onset. It needs to be taken into account, though, that these studies firstly presented single action verbs not embedded in a sentence context.…”

“…Secondly, Pulvermüller et al (2001) did not analyze a specific frequency range but focused on ERP differences between electrodes. Thirdly, only the study by Klepp et al (2015) used abstract verbs as a comparative condition. These methodological differences need to be considered when comparing the current results to those of earlier studies.…”