RESUMO
Normal hearing listeners exploit the formant transition (FT) detection to identify place of articulation for stop consonants. Neuro-imaging studies revealed that short FT induced less cortical activation than long FT. To determine the ability of hearing impaired listeners to distinguish short and long formant transitions (FT) from vowels of the same duration, 84 mild to severe hearing impaired listeners and 5 normal hearing listeners were asked to detect 10 synthesized stimuli with long (200 ms) or short (40 ms) FT among 30 stimuli of the same duration without FT. Hearing impaired listeners were tested with and without hearing aids. The effect of the difficulty of the task (short/long FT) was analysed as a function of the hearing loss with and without hearing aids. Normal hearing listeners were able to detect every FT (short and long). For hearing impaired listeners, the detection of long FT was better than that of short ones irrespective of their degree of hearing loss. The use of hearing aids improved detection of both kinds of FT; however, the detection of long FT remained much better than the detection of the short ones. The length of FT modified the ability of hearing impaired patients to detect FT. Short FT had access to less cortical processing than long FT and cochlea damages enhanced this specific deficit in short FT brain processing. These findings help to understand the limit of deafness rehabilitation in the time domain and should be taken into account in future devices development.