EEG phase synchronization during semantic unification relates to individual differences in children’s vocabulary skill

Abstract

As we listen to speech, our ability to understand what was said requires us to retrieve and bind together individual word meanings into a coherent discourse representation. This so-called semantic unification is a fundamental cognitive skill, and its development relies on the integration of neural activity throughout widely distributed functional brain networks. In this proof-of-concept study, we examine, for the first time, how these functional brain networks develop in children. Twenty-six children (ages 4-17) listened to well-formed sentences and sentences containing a semantic violation, while EEG was recorded. Children with stronger vocabulary showed N400 effects that were more concentrated to centroparietal electrodes and greater EEG phase synchrony (phase lag index; PLI) between right centroparietal and bilateral frontocentral electrodes in the delta frequency band (1-3 Hz) 1.27-1.53 s after listening to well-formed sentences compared to sentences containing a semantic violation. These effects related specifically to individual differences in receptive vocabulary, perhaps pointing to greater recruitment of functional brain networks important for top-down semantic unification with development. Less skilled children showed greater delta phase synchrony for violation sentences 3.41-3.64 s after critical word onset. This later effect was partly driven by individual differences in nonverbal reasoning, perhaps pointing to non-verbal compensatory processing to extract meaning from speech in children with less developed vocabulary. We suggest that functional brain network communication, as measured by momentary changes in the phase synchrony of EEG oscillations, develops throughout the school years to support language comprehension in different ways depending on children's verbal and nonverbal skill levels.

Temporal and topographical changes in theta power between middle childhood and adolescence during sentence comprehension

“…Similar developmental differences in the distribution and duration of the N400 in response to semantic violations have been reported and related to children's age and/or vocabulary skill (e.g., Khalifian, 2016 ; Schneider and Maguire, 2018 ; Panda et al, 2021 ). While informative, the N400 is calculated at the single word level in many of these studies, limiting our understanding of how differences at the word level manifest over the course of the entire sentence.…”

“…While informative, the N400 is calculated at the single word level in many of these studies, limiting our understanding of how differences at the word level manifest over the course of the entire sentence. Of the few studies which have examined developmental differences in theta engagement over the course of a sentence, many focus on processing demands surrounding semantic anomalies ( Schneider and Maguire, 2018 ; Panda et al, 2021 ), and fail to examine whether differences in theta band engagement exist based on retrieval versus unification processes ( Schneider et al, 2016 ). Building on these past studies, we were able to pinpoint two unique ways in which theta activation occurred, and how the localization and timing of these two responses varied, as a function of age.…”

Oscillatory activity and EEG phase synchrony of concurrent word segmentation and meaning-mapping in 9-year-old children

“…This result suggests that attentional modulation induced by consistent visual cues may facilitate learning. This idea fits well with recent results obtained during an audio-visual movie visualization task and showing increased frontal-occipital phase synchrony in the sensor space when children listened to well-formed sentences compared to semantically incongruous sentences ( Panda et al, 2020 ). Previous studies using similar methodologies have also shown that the neural tracking of a specific talker in higher-order auditory regions differs between attended and non-attended speech input ( Ding and Simon, 2012 ; Lakatos et al, 2013 ).…”

“…Further research shows that neural synchronization to the stimulus only occurs for attended stimuli ( Ding and Simon, 2012 ; Lakatos et al, 2013 ), whereas when the tracking of the stimulus is not possible, speech perception and comprehension is impoverished ( Ahissar et al, 2001 ; Luo and Poeppel, 2007 ). These findings suggest that top-down attentional mechanisms modulate speech processing in children ( Wang et al, 2014 ; Panda et al, 2020 ; Ríos‐López et al, 2020 ). One way to investigate oscillatory brain activity is to use frequency-tagging analysis, which shows the fine neural synchronization in response to both auditory and visual stimuli ( Doesburg et al, 2012 ; Kim et al, 2007 ; Keitel et al, 2014 ; Covic et al, 2017 ; Nozaradan et al, 2012 ).…”

Dynamic configuration of large-scale cortical networks during an inhibitory task accounts for heterogeneity in attention-deficit/hyperactivity disorder traits

“…Gratton and Coles (1983) method was used to correct for eye movements (50), and trials with an amplitude difference of 200μV over a 200ms interval, or with an amplitude >±200μV were rejected, resulting in 42.18 (SD = 5.89) Go trials and 44.35 (SD = 5.41) No-Go trials entering analysis. Connectivity between EEG sensors was calculated as in Panda et al (51). Specifically, trial-by-trial data (from 3s pre-stimulus to 3s post-stimulus) from each participant were z-scored and filtered into canonical frequency-bands: (1-3Hz; 4-7Hz; 8-13Hz; 14-29Hz; 30-90Hz).…”