Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Neural and cognitive predictors of individual variability in speech comprehension
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster A9 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Lucy J MacGregor1, Fritz Peters1,2, Matthew H Davis1; 1University of Cambridge, 2University College London
Individuals vary in the accuracy and ease with which they understand spoken language. We are identifying and explaining the neural and cognitive sources of this variability, focusing on a common challenge that arises during comprehension – how to interpret intended meaning when there is no one-to-one mapping between spoken words and their meaning (e.g. “BALL” refers to a spherical object and a dance party). Successful comprehension of semantic ambiguity requires a listener to use multiple cues to access and select the appropriate word meaning, and suppress the alternative(s), which sometimes involves reinterpretation (Rodd, 2020). These cognitive operations are supported by inferior frontal and posterior temporal cortical regions (Rodd et al. 2005), although the exact timing of their engagement and specific functional contributions remain underspecified (MacGregor et al., 2020). Moreover, we do not yet understand how comprehension success may be related to functional and anatomical differences in fronto-temporal brain regions within the language-selective and domain-general networks, or to language-selective (e.g. quality of lexical knowledge) and/or domain-general (e.g. flexible selection and inhibition) cognition. In an online behavioural study, we measured semantic ambiguity comprehension ability with a Meaning Definitions Task: volunteers (n=71, 19-59 years) listened to 122 sentences containing an AMBIGUOUS word disambiguated by later context to an unexpected (non-dominant) meaning (Sally worried the BALL would be too crowded). After each sentence, listeners heard the ambiguous word in isolation and defined the word as used in the sentence (mean accuracy = .84, SD = .12, split-half reliability, rs = .82, p < .001). Volunteers also completed the Mill Hill Vocabulary Test (mean accuracy = .57, SD = .11), the Spot the Word Task (mean = .79, SD = .10) and the four-part Cattell 2a Culture Fair Test (mean = .72, SD = .14). A PCA on test scores revealed two components explaining 71% of variance (p < .001), reflecting vocabulary knowledge (a language-specific ability) and non-verbal IQ (a domain-general ability). Correlational analyses showed that as age increased, factor scores increased for vocabulary-knowledge (r = 0.26, p <.05) but decreased for non-verbal-IQ (r = -0.27, p < .05), consistent with previous research (Hartshorne & Germine, 2015). Multiple linear regression showed comprehension success was predicted by vocabulary knowledge and non-verbal IQ, not age. We have revised the Meaning Definitions Task to make a four-alternative forced-choice version, using IRT to reduce the item set (58 sentences). The task is included in the CamCAN Re-Scan Study, a 10-year follow up of several hundred healthy volunteers across the lifespan (see Shafto et al., 2014 for the original study protocol). In this ongoing study we will test whether semantic ambiguity comprehension ability is predicted by behavioural measures of language-specific and/or domain-general function. Making use of structural MRI and resting state MEG data, we will explore how comprehension is predicted by grey matter density, plus anatomical and functional connectivity in fronto-temporal regions. Multivariate statistical models will enable us to relate comprehension ability to variability in neural and cognitive structure and function, to better understand predictors of successful spoken language comprehension.
Topic Areas: Meaning: Discourse and Pragmatics, Control, Selection, and Executive Processes