Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Elementalism: a clinical framework for testing the neurobiology of language
Poster Session D, Saturday, October 26, 10:30 am - 12:00 pm, Great Hall 3 and 4
T R Williamson1,2,3,4; 1University of the West of England, 2Southmead Hospital, North Bristol NHS Trust, 3University of Oxford, 4University of Southern California
Studying the neurobiology of language has produced a wealth of insights into the features, functions, and dynamics of language-specific systems in the brain (Fedorenko et al., 2024; Matchin & Hickok, 2020). The fruits of these endeavours are clear, illustrating the possibility for non-invasive semantic decoding (Tang et al., 2023), the apparent subcomponents of syntactic abilities (Grodzinsky et al., 2021), and the cross-linguistic patterns of language-related representations in the brain (Malik-Moraleda et al., 2022). This research is invaluable for general characterisations, but it alone provides insufficient grounds to motivate the investigation of the neural substrates of core subcomponents of language within the context of designing neurosurgical language tests. This is because its primary goal is not to (or cannot; Kuchcinski et al., 2015) establish what those core subcomponents are, nor is it to determine precise structure-function relationships of those subcomponents with certain neural parcellations. Knowledge of both of these is essential for the design of tests administered on neuro-oncology patients during awake craniotomy with direct electrical stimulation. While clinical protocols for intraoperative linguistic procedures exist (De Witte et al., 2015), they are not based in contemporary or complete understanding of the neurobiology of language and undertake little pre-clinical validation of location-function claims. Moreover, they presuppose that neurolinguistic explananda are adequately encapsulated by descriptions of linguistic systems suitable for the surface or output level (Luppi et al., 2022), such as in being delimited into natural kinds from behaviour like ‘semantics’, ‘syntax’, etc. that were never designed to engage the reality of neurocognitive function. This talk proposes ‘Elementalism’ as an approach to better determine the core subcomponents, or ‘elements’, of linguistic and cognitive behaviour with associated neural substrates for the purposes of neurosurgical language test design. It starts by rejecting the segmentation of language into components that descriptions of behaviour suggest (semantics, syntax, etc.) as unsuitable for correspondence to the reality of neurological function. Instead, elements should be defined mathematically with respect to their contribution to a particular neuropsychological process, where any element ought to be necessary or probable for that process as determined only through neurostimulation (e.g., TMS or DES) inhibiting that process, which enables psychometric standard-setting for neurosurgical language tasks. Elements and processes together determine the ‘minimal competence’ required to exhibit a given level of communicative complexity which is modelled in a hierarchy from one-word to discursive utterances that is essential for locating a neuro-oncology patient’s idiolectal competence and subsequently selecting tasks appropriate to them. Reviewing literature into the neurobiology of language and neurosurgical language testing enables the prospective identification of possible elements: switching (Sierpowska et al., 2018), retrieval (Damasio et al., 2004), and merge/composition (Chang et al., 2018; Li et al., 2024), which are discussed in the context of ongoing and future research within the Elementalism framework. Concluding remarks are offered tentatively about (a) what other elements may be discoverable, (b) the possibility of Elementalism enabling new avenues for uncovering the neurobiological ontology of language, (c) characterising the kinds of language that impaired, developing, non-human, and artificial cognitive systems have.
Topic Areas: Disorders: Acquired, Methods