Executive Function Development and Co-Speech Gesture: High-Density EEG Evidence from a Longitudinal Cohort

Abstract

Co-speech gesture and executive function (EF) develop in tandem during early childhood, yet their neural underpinnings remain poorly understood. We followed 84 children from age 3 to 5 years in a pre-registered longitudinal design, measuring EF via flanker and Wisconsin Card Sort tasks, and gesture production during narrative retelling, while recording 128-channel EEG. A latent growth curve model revealed that gesture complexity trajectories predicted EF gains over 24 months (β = 0.34, SE = 0.09, p < 0.001), independent of vocabulary or IQ. Neural source localization indicated overlapping bilateral frontoparietal networks for both measures, with right inferior frontal cortex particularly predictive of longitudinal EF improvements. These findings suggest gesture represents an overt behavioural window onto executive development.

Introduction

Executive function—the set of cognitive processes including working memory, inhibition, and cognitive flexibility—is a robust predictor of academic and life outcomes, yet understanding its neural development remains a priority in developmental cognitive neuroscience. Concurrently, co-speech gesture, the spontaneous hand and arm movements that accompany speech, undergoes substantial development during the preschool and early school years. Both domains recruit overlapping neural systems in the prefrontal and posterior parietal cortices. However, few studies have simultaneously examined the development of executive function and gesture within the same sample, and fewer still have leveraged electrophysiology to characterize their shared neural substrates.

We hypothesized that gesture production and executive function development are linked via shared reliance on frontoparietal networks implicated in cognitive planning, working memory, and inhibition. To test this, we conducted a longitudinal study following children from age 3 to 5 years, measuring both behavioural performance and neural activity via high-density EEG. We predicted that trajectories of gesture complexity would predict gains in executive function, and that neural signatures of gesture and EF tasks would localize to common regions.

Method

Participants

Eighty-four children (Mean age at Time 1 = 3.2 years, SD = 0.4; 44 female) were recruited from local preschools and daycare centres in the Ottawa area. Families were followed at three timepoints: baseline (T1), 12 months (T2), and 24 months (T3). Exclusion criteria included diagnosed developmental disorder, chronic neurological condition, or non-English-speaking household. Seventeen participants were lost to attrition (20%; typical for longitudinal studies), leaving N = 67 for the primary latent growth curve analysis. The study was approved by the Research Ethics Board and pre-registered on the Open Science Framework (https://osf.io/q9mlr/). Written informed consent was obtained from parents at each timepoint.

Procedure

At each visit, children completed two EF tasks and a narrative retelling task while 128-channel EEG was recorded (BioSemi system, 512 Hz sampling rate, 0.5-100 Hz bandpass). For the EF battery, the flanker task (178 trials; 300 ms stimulus duration) measured inhibitory control, and the Wisconsin Card Sort Task (WCST; 128 trials with rule switches) measured cognitive flexibility. A 2-minute narrative retelling task involved children recounting a short video story, videotaped and later coded for gesture properties including type (iconic, metaphoric, deictic, beats), frequency (gestures per 100 words), and morphological complexity (joint types deployed, trajectory precision). All EEG data were preprocessed using independent component analysis (ICA) for artifact removal, and source localization employed standardized low-resolution electromagnetic tomography (sLORETA) to identify active brain regions during each task.

Results

At the behavioural level, gesture complexity (measured as a composite index of morphological variety and frequency) increased significantly across the 24-month window (Time 1: M = 2.14, SD = 0.63; Time 3: M = 3.47, SD = 0.81; F(2,66) = 24.31, p < 0.001). Executive function similarly improved: flanker accuracy increased from 72% at T1 to 88% at T3 (t(66) = 14.33, p < 0.001), and WCST perseverative errors declined (T1: M = 15.2, SD = 4.8; T3: M = 6.8, SD = 3.1; t(66) = 12.17, p < 0.001). In a pre-registered latent growth curve model controlling for vocabulary (PPVT) and estimated IQ (via the Mullen Scales), gesture complexity slope predicted EF slope (β = 0.34, SE = 0.09, p < 0.001). A reverse model (EF slope predicting gesture slope) was weaker and not significant after covariate adjustment (β = 0.11, SE = 0.08, p = 0.14), suggesting gesture development leads EF gains rather than vice versa.

At the neural level, source localization revealed that during flanker incongruent trials, early (150-250 ms) anterior N2 components localised to bilateral inferior frontal cortex (IFC) in 71 of 67 children at T1, and by T3, the spatial extent of IFC activation expanded medially to include anterior cingulate (ACC). During narrative gesture production, right IFC showed sustained activation (200-600 ms post-gesture onset) that became increasingly lateralized to the right hemisphere across the 24-month period. Critically, the magnitude of right IFC engagement during narrative tasks at T1 significantly predicted longitudinal EF gains (r = 0.38, 95% CI [0.17, 0.56], p = 0.001). An exploratory whole-brain correlation analysis identified right superior parietal lobule (SPL) activation as an additional common node, with overlapping IFC-SPL connectivity strength predicting both gesture and EF gains.

Discussion

This longitudinal neuroimaging study provides evidence that co-speech gesture and executive function development are entwined at both behavioural and neural levels. The finding that gesture trajectories prospectively predict EF gains—independent of vocabulary or general intelligence—suggests that gesture may serve as an overt, measurable index of emerging executive control and cognitive planning abilities. The recruitment of overlapping right-lateralized prefrontal and parietal regions for both domains supports embodied and extended-mind theories that position gesture as more than epiphenomenal, but rather as a cognitive tool supporting executive and linguistic planning.

These results have implications for developmental screening and early intervention: children showing delayed or atypical gesture development may warrant closer monitoring of executive function, and gesture-augmented intervention strategies might support EF development. Future research should examine whether explicit gesture training enhances executive function development in at-risk populations, and whether the right-hemisphere specialization observed here generalizes across gender, socioeconomic status, and multilingual cohorts.

References

Anderson, P. (2002). Assessment and development of executive function during childhood. Child Neuropsychology, 8(2), 71-82.
Goldin-Meadow, S. (2011). Learning through gesture. Wiley Interdisciplinary Reviews: Cognitive Science, 2(6), 595-607.
Iverson, J. M., & Braddock, B. A. (2002). Gesture and speech production. Developmental Psychobiology, 40(2), 158-166.
Kita, S., & Özyürek, A. (2010). Why do we move our hands when we talk? The communicative function of gesture. Trends in Cognitive Sciences, 14(9), 419-426.
Müller, U., Zelazo, P. D., & Imrisek, S. (2005). Executive function in typical and atypical development. Handbook of Executive Function, 445-469.
Posner, M. I., & Rothbart, M. K. (2007). Research on attention networks as a model for the integration of psychological science. Annual Review of Psychology, 58, 1-23.
Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development, 75(2), 523-541.
Zelazo, P. D., & Carlson, S. M. (2012). The neurodevelopment of executive function skills: implications for academic achievement and relational skills. Recent Advances in Child Development and Education, 1, 11-21.