Cognitive Processing of Natural Sequences in Humans and Artificial Intelligence (AI)

The goal of this study is to identify electrical patterns in the cortex that enable the generation of predictions in natural auditory sequences such as speech and music. Between 500 and 700 participants in total (250 to 350 normally hearing participants (including 60 to 120 participants from West Africa and 60 to 120 participants from Western Europe) and 250 to 350 participants with hearing loss (no restrictions on origin)). Participants will listen to naturally recorded music and speech via headphones. They will be asked to rate these stimuli on a scale from 0 to 10 in terms of pleasure and comprehension, using a keyboard and mouse or voice input. Additionally, they will be required to sing or produce novel/improvised vocal responses using high-fidelity microphones. This entire procedure will be conducted under electroencephalography (EEG) to record neuronal responses throughout the process.

Inclusion Criteria: For hearing participants: * Average pure-tone thresholds ≤ 20 dB HL For participants with hearing loss: * Average pure-tone thresholds \> 20 dB HL For participants of West African origin: * Average pure-tone thresholds ≤ 20 dB HL * Declare that both parents are of West African origin, either by their own birth or by the birth of both of their parents. Exclusion Criteria: * No known neurological or audiological disorders * Participants under legal guardianship (tutelle, curatelle) * Participants deprived of legal liberty by judicial or administrative decision