Search

International audience; To follow a conversation in a noisy environment is a real challenge that affects the listening effort and the associated cognitive workload. In a previous work, a speech intelligibility task was performed while prefrontal cortex activity was recorded with a functional near infrared spectroscopy (f/NIRS) system. The conditions of the target-masker-ratio (TMR) were: Adverse (masker was louder than the target), Intermediate (masker and target were almost at the same level) and Favorable (masker was softer than the target). Participants were asked to follow only the target voice?s instructions neglecting that of the Masker voice. Two behavioral performers emerged across the 16 participants, respectively the U shape performers (U) and the Non U Shape performers (NU). The performance of NU decreased monotonically when the masker level increased with no specific differences in term of cognitive workload between conditions. On the contrary, U seemed to be able to use the TMR to discriminate the two talkers even when the masker was louder with a consistent prefrontal activity in the Adverse and in the Intermediate Condition. In the present work, we used analogous behavioral and fNIRS protocol.We aimed to evaluate the possibility to train NU to use the TMR in order to become U. Listeners were divided into two groups according to two different types of training strategies. Main results showed that both groups beneficiated from their training and became U. No more brain functional activity difference with the initial U group was observed after training of the NU groups. It seems that the ability to use the TMR does not refer to a specific individual perceptual auditory ability but it could rather be learned by training people. This study could provide insights to design more efficient communication systems useful in challenging environments.