The role of beta oscillations in retrieval of verbs from semantic memory

The involvement of motor brain mechanisms in a verb’s conceptual processing is a hotly debated topic in cognitive neuroscience. Here we examined the claim of the embodied approach that the retrieval of verb meaning critically depends on the re-use of neural circuits that subserve the corresponding actions. For this purpose we employed a measure, which reflects engagement of cortical motor areas – suppression of MEG beta oscillations. Beta oscillations (15-30 Hz) are generated in cortical-basal ganglia-cortical circuitry whenever there is a need for maintaining a current motor program. Preparation for a new action or action sequence leads to beta suppression in the cortical motor regions involved in action initiation and programming. 

Using the verb generation task and MEG beta suppression index, we tested two predictions of the embodied theory. First, if the re-use of motor representations is a requisite part of verb semantic retrieval, the beginning of beta suppression should coincide with extraction of a target verb from semantic memory and, thus, substantially precede the overt production of a target verb. Second, considering functional contribution of motor circuitry to semantic retrieval, more difficult search for an appropriate verb response should correlate with greater beta suppression. 

With a full accord of the embodied approach, we observed sustained beta suppression that emerged long before the response onset and engaged a widespread left-hemispheric network including premotor and supplementary motor areas. Crucially, when the verb retrieval was hindered by the absence of strong noun-verb associations, the greater beta suppression was observed specifically in the supplementary and pre-supplementary motor areas, as well as the premotor area of the left frontal lobe (Fig. 1). Thus, our results support the view that the action-initiation brain circuitry contributes to verb retrieval from semantic memory.

Fig. 1. Cortical sources that show a stronger beta suppression
during difficult (weak association) vs easy (strong association) search
for a suitable verb in semantic memory.