Are brain stimulation studies suggesting the involvement of the motor cortex in language comprehension reliable?
According to the embodiment theory, processing language that refers to motor actions (e.g., verbs such as "grasp" or "step") recruits the activity of brain regions involved in controlling different parts of the body (hands, feet...). This idea has gained support from a multitude of studies, especially in recent years those using brain stimulation techniques such as Transcranial Magnetic Stimulation (TMS) or Transcranial Direct Current Stimulation (tDCS). Recently, CIMCYC researchers Pablo Solana and Julio Santiago have evaluated the reliability of this literature through a meta-analytical study published in the journal Neuroscience and Biobehavioral Reviews.
The authors began by conducting a literature search that led them to locate 43 studies where TMS or tDCS was used to modulate motor cortex activity during action language processing. These studies were then subjected to a novel meta-analytic technique known as p-curve analysis, which allows them to obtain reliability indicators for a set of studies based on how the p-values associated with their main findings are distributed. The results showed that it is not possible to affirm that the studies currently published explore real effects. In addition, the estimated statistical power of this set of studies was found to be approximately 30%. This means that only 30% of the findings of these studies are replicable, while the remaining 70% could be false positives. Finally, additional oversignificance tests revealed that there is an excess of significant findings in this literature, indicating that these studies are "too good to be true," suggesting the existence of publication bias.
The authors of the study propose that future embodiment studies employing TMS and tDCS should urgently adopt practices such as using adequate sample sizes, preregistration, or conducting replications of previous studies. Only then will we be able to reliably assess whether or not the motor system is actually involved in language comprehension.
Contact:
Pablo Solana: @email
Julio Santiago: @email
Full reference:
Solana, P. & Santiago, J. (2022). Does the involvement of motor cortex in embodied language comprehension stand on solid ground? A p-curve analysis and test for excess significance of the TMS and tDCS evidence. Neuroscience and Biobehavioral Reviews, 141, 104834. https://doi.org/10.1016/j.neubiorev.2022.104834