Learning to associate novel words with motor actions: Language-induced motor activity following short training

Abstract

Action words referring to face, arm or leg actions activate areas along the motor strip that also control the planning and execution of the actions specified by the words. This electroencephalogram (EEG) study aimed to test the learning profile of this language-induced motor activity. Participants were trained to associate novel verbal stimuli to videos of object-oriented hand and arm movements or animated visual images on two consecutive days. Each training session was preceded and followed by a test-session with isolated videos and verbal stimuli. We measured motor-related brain activity (reflected by a desynchronization in the μ frequency bands; 8-12 Hz range) localized at centro-parietal and fronto-central electrodes. We compared activity from viewing the videos to activity resulting from processing the language stimuli only. At centro-parietal electrodes, stable action-related μ suppression was observed during viewing of videos in each test-session of the two days. For processing of verbal stimuli associated with motor actions, a similar pattern of activity was evident only in the second test-session of Day 1. Over the fronto-central regions, μ suppression was observed in the second test-session of Day 2 for the videos and in the second test-session of Day 1 for the verbal stimuli. Whereas the centro-parietal μ suppression can be attributed to motor events actually experienced during training, the fronto-central μ suppression seems to serve as a convergence zone that mediates underspecified motor information. Consequently, sensory-motor reactivations through which concepts are comprehended seem to differ in neural dynamics from those implicated in their acquisition.

Language–motor interference reflected in MEG beta oscillations

“…Despite a possible role for alpha oscillations in embodied language processing (van Elk et al, 2010;Fargier et al, 2012), no clusters in the alpha range were identified in the comparison between conditions nor were suggested by alpha power topographies. This indicates that oscillations in the beta band are the specific neurophysiological mechanism associated with language-motor interference.…”

“…Taken together, the role of beta oscillations in embodied language processing makes it a feasible candidate for a functional mechanism of language-motor interference. Similarly, alpha band (8-13 Hz) oscillations have been associated with action execution (Salmelin et al, 1995;Sebastiani et al, 2014), observation (Caetano et al, 2007;Avanzini et al, 2012), motor imagery (Pfurtscheller et al, 2006;de Lange et al, 2008), spoken language processing (Strauß et al, 2014), and action language processing (Alemanno et al, 2012;Fargier et al, 2012). Since alpha band oscillations may be more related to sensory than motor processing (Salmelin et al, 1995;Brinkman et al, 2014;Sebastiani et al, 2014;Coll et al, 2015), the focus in the current study is on the beta band, but alpha oscillations are also investigated.…”

Brain dynamics in the comprehension of action-related language. A time-frequency analysis of mu rhythms

“…One drawback of the study, however, is that the absence of non-action sentences as a contrasting condition does not allow the conclusion that these timeresolved mu and beta rhythms are exclusively modulated by action language. A different research strategy was employed by Fargier et al (2012). They trained participants to associate new words with action related videos or with visual motion videos in two sessions.…”

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

“…Further, b-oscillations are very sensitive towards dissociation between concrete and abstract concepts (Weiss & Müller, 2013;Weiss & Rappelsberger, 1996. Recently, in the context of the embodied language theory, studies revealed that power decreases in the b-as well as the l-band are correlated with the processing of concrete action related language (Alemanno et al, 2012;Fargier et al, 2012;Moreno, de Vega, & León, 2013;Moreno et al, 2015;van Elk, van Schie, Zwaan, & Bekkering, 2010). This finding strongly sup-ports the theory of embodiment, which, in the case of language processing, indicates that understanding linguistic input requires a mental neural simulation of respective sensorimotor information in somatosensory and motor areas (Barsalou, 1999(Barsalou, , 2008Gallese & Lakoff, 2005;Glenberg & Kaschak, 2003; for a review see also Buccino, Colagè, Gobbi, & Bonaccorso, 2016).…”