Structural and functional motor-network disruptions predict selective action-concept deficits: Evidence from frontal lobe epilepsy

Sebastian Moguilner, Agustina Birba, Daniel Fino, Roberto Isoardi, Celeste Huetagoyena, Raúl Otoya, Viviana Tirapu, Fabián Cremaschi, Lucas Sedeño, Agustín Ibáñez, Adolfo M. García

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


Built on neurodegenerative lesions models, the disrupted motor grounding hypothesis (DMGH) posits that motor-system alterations selectively impair action comprehension. However, major doubts remain concerning the dissociability, neural signatures, and etiological generalizability of such deficits. Few studies have compared action-concept outcomes between disorders affecting and sparing motor circuitry, and none has examined their multimodal network predictors via data-driven approaches. Here, we first assessed action- and object-concept processing in patients with frontal lobe epilepsy (FLE), patients with posterior cortex epilepsy (PCE), and healthy controls. Then, we examined structural and functional network signatures via diffusion tensor imaging and resting-state connectivity measures. Finally, we used these measures to predict behavioral performance with an XGBoost machine learning regression algorithm. Relative to controls, FLE (but not PCE) patients exhibited selective action-concept deficits together with structural and functional abnormalities along motor networks. The XGBoost model reached a significantly large effect size only for action-concept outcomes in FLE, mainly predicted by structural (cortico-spinal tract, anterior thalamic radiation, uncinate fasciculus) and functional (M1-parietal/supramarginal connectivity) motor networks. These results extend the DMGH, suggesting that action-concept deficits are dissociable markers of frontal/motor (relative to posterior) disruptions, directly related to the structural and functional integrity of motor networks, and traceable beyond canonical movement disorders.

Original languageEnglish
Pages (from-to)43-55
Number of pages13
StatePublished - Nov 2021
Externally publishedYes


  • Action semantics
  • Diffusion tensor imaging
  • Frontal lobe epilepsy
  • Functional connectivity
  • Machine learning


Dive into the research topics of 'Structural and functional motor-network disruptions predict selective action-concept deficits: Evidence from frontal lobe epilepsy'. Together they form a unique fingerprint.

Cite this