An fNIRS dataset for Multimodal Speech Comprehension in Normal Hearing Individuals and Cochlear Implant Users

Understanding cortical processing in cochlear implant (CI) users is crucial for improving speech outcomes. Functional near-infrared spectroscopy (fNIRS) provides a non-invasive, implant-compatible method for assessing cortical activity during speech comprehension. However, existing studies suffer from methodological heterogeneity and a lack of standardized datasets, limiting cross-study comparisons and generalizability. To address this gap, we present a multimodal fNIRS dataset comprising 46 CI users and 26 normal hearing controls. Participants completed a clinically relevant speech comprehension task using the German Matrix Sentence Test (OLSA) under speech-in-quiet, speech-in-noise, audiovisual and visual speech (i.e., lipreading) conditions. fNIRS recordings covered key cortical regions involved in speech processing, including the prefrontal, temporal, and visual cortices. Additionally, we provide detailed metadata, including patient history, hearing tests, behavioral measures, and spatially registered probe positions. This data descriptor aims to provide a comprehensive resource for investigating multimodal speech understanding in CI users. It enables researchers to explore cortical adaptations in prosthetic hearing, contributing to the refinement of CI rehabilitation strategies and advancing the understanding of auditory neuroplasticity.