The Benefit of Cross-Modal Reorganization on Speech Perception in Pediatric Cochlear Implant Recipients Revealed Using Functional Near-Infrared Spectroscopy

Abstract

Cochlear implants (CIs) are the most successful treatment for severe-to-profound deafness in children. However, speech outcomes with a CI often lag behind those of normally-hearing children. Some authors have attributed these deficits to the takeover of the auditory temporal cortex by vision following deafness, which has prompted some clinicians to discourage the rehabilitation of pediatric CI recipients using visual speech. We studied this cross-modal activity in the temporal cortex, along with responses to auditory speech and non-speech stimuli, in experienced CI users and normallyhearing controls of school-age, using functional near-infrared spectroscopy. Strikingly, CI users displayed significantly greater cortical responses to visual speech, compared with controls. Importantly, in the same regions, the processing of auditory speech, compared with non-speech stimuli, did not significantly differ between the groups. This suggests that visual and auditory speech are processed synergistically in the temporal cortex of children with CIs, and they should be encouraged, rather than discouraged, to use visual speech.

Analysis methods for measuring passive auditory fNIRS responses generated by a block-design paradigm

“…An alternative approach to account for differences between the model and the measured response is to include a derivative term in the design matrix. 13 , 75 However, since we observed good correspondence between the response morphology and the expected canonical response, we did not include derivative terms in our analysis. The response morphology, and appropriate boxcar length, may vary with stimulus duration, characteristics, and brain structure of interest.…”

Use of Functional Near-Infrared Spectroscopy to Predict and Measure Cochlear Implant Outcomes: A Scoping Review

“…There have also been several fNIRS studies that have been conducted involving adults and children with CIs [77][78][79][80]. This work with CI users demonstrated levels of cross-modal plasticity [78,80], overlap in the processing of the features of auditory speech between CI users and normally hearing children [78], and an overall neurotypical pattern of activation during auditory language tasks [77]. Crucially, all of this work has demonstrated the utility and tolerability of this technique for studies with CI-users.…”

“…Three records reported studies from the UK [78,86,87], three reported studies from Germany [88][89][90], one reported a study from the US [91], and one study was from Australia [92]. All of the included studies were published between 2016 and 2020.…”

“…fNIRS has been used to explore cross-modal plasticity in adults with hearing loss [76]. There have also been several fNIRS studies that have been conducted involving adults and children with CIs [77][78][79][80]. This work with CI users demonstrated levels of cross-modal plasticity [78,80], overlap in the processing of the features of auditory speech between CI users and normally hearing children [78], and an overall neurotypical pattern of activation during auditory language tasks [77].…”

Decoding Multiple Sound-Categories in the Auditory Cortex by Neural Networks: An fNIRS Study

“…According to fNIRS analyses, newborns consistently exhibit a strong hemodynamic response to universally preferred syllables, which suggests that the early acquisition and perception of language can be detected using categorical linguistic sounds (Gomez et al, 2014 ). The applications of this technology have the potential to provide feedback for speech therapy or in the tuning of hearing aid devices (e.g., cochlear implants) at an early stage of development based on brain recordings (Mushtaq et al, 2020 ). Several groups have demonstrated fNIRS use for measuring brain responses in deaf children with cochlear implants (Sevy et al, 2010 ; Pollonini et al, 2014 ).…”