Search

Despite early bilateral cochlear implantation, children with congenital deafness do not develop accurate spatial hearing; we thus asked whether auditory brain networks are disrupted in these children. EEG responses were evoked unilaterally and bilaterally in 13 children with normal hearing and 16 children receiving bilateral cochlear implants simultaneously. Active cortical areas were estimated by the Time Restricted Artifact and Coherent source Suppression (TRACS) beamformer and connected cortical areas were identified by measuring coherence between source responses. A whole-brain analysis of theta band coherence revealed the strongest connections between the temporal areas in all conditions at early latencies. Stronger imaginary coherence in activity between the two auditory cortices to bilateral than unilateral input was found in children with normal hearing reflecting facilitation in the auditory network during bilateral hearing. The opposite effect, depressed coherence, was found during bilateral stimulation in children using cochlear implants. Children with cochlear implants also showed a unique auditory network in response to bilateral stimulation which was marked by increased connectivity between occipital and frontal areas. These findings suggest that cortical networks for sound processing are normally facilitated by bilateral input but are disrupted in children who hear through two independent cochlear implants. Efforts to improve hearing in children with congenital deafness must thus include corrections to potential mismatches in bilateral input to support brain development.

Objective: This research aims to characterize the cortical network involved in listening with bilateral cochlear implants (CIs) through measures of functional connectivity. Rationale: A better understanding of the underlying cortical network involved in hearing will help elucidate remaining challenges experienced by children who received bilateral cochlear implants early relative to their peers with normal hearing. This work extends our previous focus on plasticity in temporal regions to include multiple cortical areas and considers the phase relationships between neural source signals. Methods: The data were collected using 64-channel electroencephalography in response to click stimuli in a passive listening condition. Source reconstruction was applied using the TRACS beamformer in order to localize the underlying neural generators. Sources were mapped to regions in the Automated Anatomical Labeling atlas and connectivity analyses were performed to examine statistical relationships between the atlas regi...