Poster A40, Thursday, August 16, 10:15 am – 12:00 pm, Room 2000AB

The neural network of action language: a comparative meta-analysis

Melody Courson1,2, Pascale Tremblay1,2;1Université Laval, Faculty of Medicine, Department of Rehabilitation, Québec, Canada, 2CERVO research center, Québec, Canada

Accumulating empirical evidence shows a response of the cerebral motor system during the processing of action language (AL) [1] [2]. It has been suggested that this motor activation could represent motor imagery [3] or motor planning/execution processes, such as motor program selection [4], but the underlying processes remain unknown. Knowing whether activation patterns of AL resemble those of other motor-related processes could help clarify the role of the motor system in AL. The objective of this systematic review and meta-analysis was thus to determine whether the AL network is comparable to that of action observation, motor imagery or motor execution. To this aim, a set of activation likelihood estimation (ALE) meta-analyses was conducted to provide comparative maps of the brain structures involved in AL. A literature search on Pubmed was conducted for each process. Studies published in peer-reviewed journals in English were included if they used fMRI, included at least one group of healthy right-handed participants aged up to 60 years and reported Talairach or MNI coordinates. A total of 91 studies, including 1628 participants (range: 18-56 years), were selected: 16 in AL, 25 in action observation, 20 in motor imagery and 30 in motor execution. Coordinates for all the contrasts of interest were entered in Ginger Ale (Version 2.3.6) which computed activation maps for each motor-related process individually. A conjunction analysis was then conducted between the activation map of AL and the activation map of each of the other motor-related processes. Significant clusters were identified at an uncorrected p-value threshold of .001 and with a minimum volume of 120 mm3 [5]. Results showed that the AL network includes the pre-supplementary motor area (pre-SMA), inferior frontal gyrus (IFG), ventral premotor cortex (PMv), post-central gyrus, posterior middle temporal gyrus (MTGp), anterior superior temporal gyrus and superior frontal gyrus. The conjunction analyses showed that the AL network shares the IFG, PMv, postcentral gyrus and MTGp with the action observation network and the MTGp and pre-SMA with the motor imagery network. No significant cluster emerged from the conjunction of AL and motor execution. To further understand the mechanisms underlying AL, an additional set of ALE analyses was conducted. Observation and imagery studies were separated into those using action-related tasks (e.g. object grasping) and those using movement (e.g. finger tapping). Conjunction analyses were then conducted between AL and each of these categories. Results showed that the common activation sites between AL and action observation were mostly linked to action. This meta-analysis is the first to provide quantitative comparative motor-related activation maps of AL. The AL network partially overlaps that of action observation and motor imagery, with a larger overlap with observation. These shared activation clusters may result from a complex interaction between motor task (observation vs. imagery) and motor content (action vs. movement). [1] Tremblay & Small (2010) Cerebral Cortex. [2] Hauk et al. (2004) Neuron. [3] Yang & Shu (2014) Experimental Brain Research. [4] Courson et al. (2017) Cortex. [5] Hétu et al. (2013) Neuroscience and Biobehavioral Reviews

Topic Area: Meaning: Lexical Semantics

Back