Search in this site

Grenoble Institut des Neurosciences Grenoble Institut des Neurosciences

  • Linkedin
  • Twitter
  • Youtube

Home > About us > News > Team david News

See the complete sitemap

Direct recordings from human anterior insula reveal its leading role within the error-monitoring network

on the March 16, 2016

The activity of the anterior insular cortex increases very rapidly after an error and this structure sends these signals towards the prefrontal cortex


The ability to monitor our own errors is a critical aspect of learning that is mediated by coordinated activity in the dorsal medial prefrontal cortex (dMPFC, including anterior cingulate cortex and pre-supplementary motor area) and anterior insula. However, the functional hierarchy and precise spatio-temporal dynamics of error-monitoring signals in the human brain remains unclear.

In the present study (Bastin et al., in press), we directly recording neural signals with depth electrodes implanted in key structures of the error-monitoring network. To this end, we recorded intracerebral electroencephalography (stereotactic electroencephalography, sEEG) in six epileptic patients while they performed a stop-signal task. This task has been used in numerous previous studies to probe the neural correlates of the error-monitoring network. Using sEEG, we were able to assess both local activity and inter-areal connectivity modulations during error processing with both high spatial and temporal precision. Key to this study was the opportunity to simultaneously record AI and dmPFC structures (ACC or pre-SMA) in most patients.


Depth electrode local field potentials recordings in the Anterior Insula (AI) and dorso-medial prefrontal cortex revealed a striking power increase in oscillatory activity as well as a reversal of information flow within the error-monitoring circuitry.

This allowed us to probe the role of AI by measuring error-related changes in (a) its local activity and (b) its long-range effective connectivity within the error-monitoring network. We found that error commission was associated with an increase in broadband gamma activity (BGA, 50-150 Hz) in AI as well as a remarkable reversal of information flow within the error-monitoring circuitry with AI driving dmPFC activity on error trials but dmPFC driving AI on correct responses. Taken together, our findings reveal the broadband electrophysiological underpinnings of human AI activity following behavioural errors and, most importantly, they posit AI as the key error signalling component within the error-monitoring network.

Bastin J, Deman P, David O, Gueguen M, Benis D, Minotti L, Hoffman D, Combrisson E, Kujala J, Perrone-Bertolotti M, Kahane P, Lachaux JP, Jerbi K (2016). Direct Recordings from Human Anterior Insula Reveal its Leading Role within the Error-Monitoring Network. Cerebral Cortex. doi: 10.1093/cercor/bhv352.

Updated on December 8, 2017

Associés renforcés
Associés simples