Language processing from the perspective of electrical stimulation mapping

Abstract

Electrical Stimulation (ES) is a neurostimulation technique that is used to localize language functions in the brain of people with intractable epilepsy and/or brain tumors. We reviewed 25 ES articles published between 1984 and 2018 and interpreted them from a cognitive neuropsychological perspective. Our aim was to highlight ES as a tool to further our understanding of cognitive models of language. We focused on associations and dissociations between cognitive functions within the framework of two non-neuroanatomically specified models of language. Also, we discussed parallels between the ES and the stroke literatures and showed how ES data can help us to generate hypotheses regarding how language is processed. A good understanding of cognitive models of language is essential to motivate task selection and to tailor surgical procedures, for example, by avoiding testing the same cognitive functions and understanding which functions may be more or less relevant to be tested during surgery.

Small Semantic Networks in Individuals with Autism Spectrum Disorder Without Intellectual Impairment: A Verbal Fluency Approach

“…Binder and Desai 2011;Kiefer and Pulvermüller 2012;Labache et al 2019;Mirman et al 2015;Patterson et al 2007;Tremblay and Steven 2017). Separable sub-systems are suggested to serve language production vs. recognition (Mirman et al 2015), semantic vs. phonologic processing (Hickok and Poeppel 2004;Mirman et al 2015) as well as syntactic (Rofes et al 2019) and orthographic (Peleg et al 2016;Seidenberg and McClelland 1989) processing. In terms of language production, neurolinguistic models have thus suggested multi-level processing including conceptualisation, lexical-semantic access, syntactic and phonological encoding, and articulatory preparation (e.g.…”

“…These featural representations have been proposed to be "stored in semantic memory" (Binder and Desai 2011) in a fashion that modality specific information is connected (Patterson et al 2007) or gradually converges to form more abstract concepts eventually allowing for lexical-semantic operations (Binder and Desai 2011;Damasio et al 1994;Rofes et al 2019). This network structure could account for semantic associations (Rofes et al 2019) between distinct items which thus share common "conceptual memory traces" (Kiefer and Pulvermüller 2012), including perceptual features (e.g. the feature 'furry' could be shared by 'cat' and 'rabbit') as well as individual values (e.g.…”

A Vector Space Model for Neural Network Functions: Inspirations From Similarities Between the Theory of Connectivity and the Logarithmic Time Course of Word Production

“…There is, however, wide agreement on core processes encompassing conceptual preparation, lexical selection, and form encoding (e.g., Dell, 1986 ; Levelt, 1999 ; Indefrey and Levelt, 2004 ; for a review see Henry and Crawford, 2005 ; Walker and Hickok, 2016 ) which especially involve the activation of left lateralized frontotemporal cortical networks (Indefrey and Levelt, 2000 ; Binder and Desai, 2011 ; Robinson et al, 2012 ; Mirman et al, 2015 ; Conner et al, 2019 ). Moreover, a wide-spread cortical system (Indefrey and Levelt, 2000 ; Riès et al, 2017 ) seems to be involved in the initial activation of semantic concepts, i.e., non-verbal representations of an object's sensory, motor, and affective features (encompassing, e.g., shape, use, familiarity, and relationships with other objects; Levelt, 1999 ; Pulvermüller, 1999 ; Binder and Desai, 2011 ; Kiefer and Pulvermüller, 2012 ; Rofes et al, 2019 ). Gradual convergence (Damasio et al, 1994 ), possibly involving connective hubs (Patterson et al, 2007 ), was proposed to connect the modality specific information thus forming more abstract semantic concepts (Binder and Desai, 2011 ; Rofes et al, 2019 ).…”

Direct electrical stimulation mapping of cognitive functions in the human brain

“…DES can be applied via a hand held bipolar or monopolar stimulator, or through implanted grids or strips; the former technique is more frequently used during awake craniotomies, while the latter technique is used in the setting of long-term monitoring and extra-operative mapping of the margins separating epileptogenic from eloquent tissue. The awake craniotomy was the clinical preparation in which DES was developed for use in humans, by pioneers such as Wilder Penfield and George Ojemann (Ojemann, 1979(Ojemann, , 1981(Ojemann, , 1983a(Ojemann, , 1983b(Ojemann, , 1983c(Ojemann, , 1986(Ojemann, , 1987(Ojemann, , 1988Penfield, 1954Penfield, , 1956Penfield, , 1961Penfield & Boldrey, 1937; for reviews, see Mazurek & Schieber, 2019;Rofes et al, 2018). During awake craniotomies, patients are titrated off of general anesthesia during their surgery (if general anesthesia was used), with local anesthetic applied at the site of incision, and are thus comfortable and able to carry out cognitive testing (for a video overview of the procedures involved in an awake craniotomy, see Mahon et al, 2019).…”

“…For instance, for motor mapping, disruptions in voluntary movement are outwardly observable by the clinical team, and thus the patient "merely" needs to stay on task (e.g., tapping fingers). Similarly, in the setting of mapping of speech production, patients are engaged in a task, such as picture naming, word reading, counting, or sentence production (inter alia-see Rofes, de Aguiar, & Miceli, 2015;Rofes, Spena, Miozzo, Fontanella, & Miceli, 2015;Rofes et al, 2018). When patients make errors (i.e., errors of commission, failure to respond) due to DES, the effect of DES is observed directly by the clinical team.…”