You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

 
Poster E43, Friday, November 10, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Spatio-temporal granularity of dorsal stream processing during word production

F.-X. Alario1, C. Liegeois-Chauvel2,3, A.-S. Dubarry4, I. Wang3, S. Alomar3, I. Najm3, J. Gonzalez-Martinez3;1Aix Marseille Univ, CNRS, LPC, Marseille, France, 2Aix Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France, 3Cleveland Clinic Foundation, Cleveland (OH), USA, 4Aix Marseille Univ, CNRS, LPL, Aix-en-Provence, France

**Introduction** Cognitive models of language processing have been grounded in neurobiological descriptions of brain activity, giving a primary role to the segregation of functional brain activity into distinct streams of processing [1]. The dorsal stream interfaces temporo-parietal areas with frontal areas, particularly supra-marginal gyrus, angular gyrus, etc. with inferior frontal gyrus [2]. During word production, a left lateralized dorsal pathway is thought to map auditory-phonological information with articulatory motor programs [3]. We used functional data recorded from intra-cerebral electrodes during a picture naming task to quantify dorsal stream activity [4] at a finer spatial and temporal granularity than is currently available. **Methods** Participants were epileptic patients that underwent a stereo-electro encephalography (SEEG) as pre-surgical assessment of pharmaco-resistant epilepsy [5]. They were enrolled under criteria approved by the local Institutional Review Board. In all 17 participants, the implantation included a parieto-frontal network that was either unilateral (9 left, 2 right hemisphere), or bilateral (6 patients). One patient was discarded due to poor behavioral performance. The experimental task consisted in naming out-loud simple drawings of familiar objects. The procedure was organized in experimental blocks, each with 30 trials during which five different items were presented and repeated randomly. Due to clinical constraints (fatigue, availability), the number of trials recorded varied between patients. Electrode localization was achieved by co-registration of the CT to the MRI for each individual patient then projected to a common space (MNI). Brain area labels were attributed based on Talairach coordinates. Regions sampled in only one patient were not considered. For each electrode contact, we computed intra-cerebral event related potentials (iERP) and high gamma activity (HGA) as indexes of focal processing within the area. Significant activity was detected per contact, then combined across patients. **Results** In total, 490 contacts were available (381 left, 109 right) in 22 different brain areas. From these, 398 contacts showed significant iERPs in the relevant time window (315 left, 83 right), confirming the involvement of a widespread network. Significant iERPs were detected in Supramarginal Gyrus and Angular Gyrus, peaking on average 320 ms after stimulus. Responses in Angular Gyrus were much more consistent across patients than in Supramarginal Gyrus. Significant iERPs and HGA were also detected in left IFG, peaking around 350 ms. They were most clear and consistent in Pars opercularis, compared to Triangularis and Orbitalis. **Conclusion** Dorsal stream activity during word production simultaneously involves specific sub-regions within temporo-parietal and inferior frontal areas. **References** 1. Poeppel, D. et al. (2012). J. Neurosci. 32, 14125–14131. 2. Hickok, G. (2014). Lang. Cogn. Neurosci. 29, 2–20. 3. Schwartz, M. F., et al. (2012). Brain 135, 3799–814. 4. Flinker, A. et al. Proc. Natl. Acad. Sci. (2015). 112, 2871-2875. 5. Talairach, J. & Bancaud, J. (1973). Prog. Neurol. Surg. 5, 297–354.

Topic Area: Meaning: Lexical Semantics

Back to Poster Schedule