EEG activity during a vigilance task reveals distinct frequency band patterns in gifted children

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record.

All legal disclaimers apply. To investigate whether electroencephalography (EEG) recorded during a brief Psychomotor Vigilance Task provides physiological information that can differentiate children previously identified as gifted from control participants in school settings. EEG was recorded in 158 schoolchildren aged 9–13 years (48 gifted, 110 controls) using a 32-channel semi-dry system (256 Hz).

Each session comprised a 2-minute eyes-closed baseline followed by a 7-minute child-adapted PVT. Temporal, spectral, entropy, geometry, asymmetry, and functional connectivity features were extracted across canonical frequency bands and normalized relative to each participant’s resting baseline. Group differences were examined using non-parametric statistics, and supervised classification was evaluated using repeated cross-validation.

Behavioural performance did not differ significantly between groups. EEG analyses revealed differences across several frequency bands, feature domains, and cortical regions. The highest classification performance was obtained in the delta band using logistic regression (macro-F1 = $0.745 \pm 0.066$, AUC-ROC = $0.815 \pm 0.049$), while alpha and beta bands also provided discriminative information.

EEG recorded during a brief cognitive task contains physiological information associated with the distinction between gifted and control participants. Combined with machine learning, this approach provides an objective and complementary source of information that may support the development of more standardized neurophysiological methods for giftedness identification. The publication is part of the project PID2022-137397NB-I00, funded by MCIN/AEI/10.13039/501100011033/ERDF, EU, where PID2022-137397NB-I00 is the reference indicated in the grant award resolution; MCIN is the acronym of the Ministry of Science and Innovation; AEI is the acronym of the State Research Agency; 10.13039/501100011033 is the DOI (Digital Object Identifier) of the Agency; and ERDF is the acronym of the European Regional Development Fund. 1.

Roberto Sánchez-Reolid, Juan Carlos Pastor-Vicedo and Alejandro L. Borja contributed equally to this work. 1. Escuela Técnica Superior de Ingeniería Industrial de Albacete, Universidad de Castilla-La Mancha, Albacete, 02071, Spain 2.