Temporal fine structure in cochlear implants: preliminary speech perception results in Cantonese-speaking implant...

Area of Science:

• Audiology
• Neuroscience
• Speech-language pathology

Background:

• Continuous interleaved sampling (CIS) strategies used in cochlear implants (CI) do not fully represent temporal fine structure (TFS), which is vital for perceiving tonal languages.
• TFS coding aims to improve speech perception by encoding temporal acoustic information.

Purpose of the Study:

• To evaluate speech recognition in Cantonese-speaking CI users using a TFS coding strategy.
• To compare TFS coding against CIS in individuals with no prior TFS experience.

Main Methods:

• Twelve long-term CIS users with MED-EL cochlear implants participated.
• Speech recognition was assessed for Cantonese lexical tones and CHINT sentences comparing CIS and TFS stimulation (4 TFS channels).
• Additional TFS strategy variations and pulse rates were tested in a subset of participants.

Main Results:

• No significant differences in Cantonese lexical tone identification or CHINT sentence recognition were found between CIS and TFS strategies.
• Mean tone identification scores were 59.2% for both CIS and TFS.
• Mean CHINT sentence scores were 54.2% for CIS and 55.9% for TFS.

Conclusions:

• Acute comparison revealed no significant speech perception differences between CIS and TFS coding strategies in Cantonese-speaking CI users.
• The unfamiliar TFS strategy performed comparably to the established CIS strategy.
• Further research is recommended to explore potential benefits of extended TFS use in tonal languages.