1 Child Study Center, School of Medicine, Yale University, New Haven, CT 06511, USA.
2 Brain Imaging Research Core, University of Connecticut, Storrs, CT 06269, USA.
3 Nathan Kline Institute for Psychiatric Research, Orangeburg, NY 10962, USA.
4 Hearts for Hearing Foundation, Oklahoma City, OK 73120, USA.
5 Department of Psychology, Concordia University, Montreal, QC H3G 1M8, Canada.
6 Oberkotter Foundation, Philadelphia, PA 19102, USA.
7 Department of Psychological Sciences, University of Connecticut, Storrs, CT 06269, USA.
8 Department of Neurology, Universitätsklinikum Würzburg, Würzburg 97080, Germany.
9 School of Communication Sciences and Disorders, McGill University, Montreal, QC H3A 0G4, Canada.

Early hearing intervention in children with congenital hearing loss is critical for improving auditory development, speech recognition and both expressive and receptive language, which translates into better educational outcomes and quality of life. In children receiving hearing aids or cochlear implants, both adaptive and potentially maladaptive neural reorganization can mitigate higher-level functions that impact reading. The focus of the present study was to dissect the neural underpinnings of the reading networks in children with cochlear implants and assess how these networks mediate the reading ability in children with cochlear implants. Cortical activity was obtained using naturalistic stimuli from 75 children (50 cochlear implant recipients, aged 7-17, and 25 age-matched children with typical hearing) using functional near-infrared spectroscopy. Assessment of basic reading skill was completed using the Reading Inventory and Scholastic Evaluation. We computed directed functional connectivity of the haemodynamic activity in reading-associated anterior and posterior brain regions using the time-frequency causality estimation method known as temporal partial directed coherence. The influence of the cochlear implant-related clinical measures on reading outcome and the extent to which neural connectivity mediated these effects were examined using structural equation modelling. Our findings reveal that the timing of intervention (e.g. age of first cochlear implants, age of first hearing aid) in children with cochlear implants significantly influenced their reading ability. Furthermore, reading-related processes (word recognition and decoding, vocabulary, morphology and sentence processing) were substantially mediated by the directed functional connectivity within reading-related neural circuits. Notably, the impact of these effects differed across various reading skills. Hearing age, defined as the age at which a participant received adequate access to sound, and age of initial implantation emerged as robust predictors of reading proficiency. The current study is one of the first to identify the influence of neural characteristics on reading outcomes for children with cochlear implants. The findings emphasize the importance of the duration of deafness and early intervention for enhancing outcomes and strengthening neural network connections. However, the neural evidence further suggested that such positive influences cannot fully offset the detrimental impact of early auditory deprivation. Consequently, additional, perhaps more specialized, interventions might be necessary to maximize the benefits of early prosthetic hearing intervention.