Speech recognition by bilateral cochlear implant users in a cocktail-party setting

Area of Science:

• Audiology and Hearing Science
• Neuroscience
• Biomedical Engineering

Background:

• Prior studies on bilateral cochlear implant (CI) users often used simplified listening scenarios with single sound interferers.
• Realistic auditory environments for CI users typically involve multiple interfering sounds from various spatial locations.
• Understanding binaural benefits in complex acoustic scenes is crucial for improving CI efficacy.

Purpose of the Study:

• To investigate speech reception in bilateral CI users under realistic conditions with multiple interferers (noise or competing talkers).
• To quantify binaural benefits, including monaural advantage and binaural interaction, in varied spatial configurations.
• To compare performance with noise versus speech interferers and assess the ecological validity of noise as a masker.

Main Methods:

• Speech reception thresholds were measured in bilateral CI users in unilateral and bilateral listening conditions.
• Four spatial configurations were tested with one or three interferers (modulated noise or competing talkers).
• Data were analyzed to determine monaural advantage (better-ear listening) and binaural interaction benefits.

Main Results:

• A total spatial release of 2-5 dB was observed, maintained even with multiple interferers.
• The advantage was predominantly driven by monaural advantage (1-6 dB), especially with energetic interferers.
• No significant binaural interaction benefit was found; performance with speech interferers was notably poorer than with noise.

Conclusions:

• Bilateral CI users derive spatial benefits, but these are largely based on monaural processing, not binaural interaction.
• Current noise-based masking paradigms may not fully represent the challenges faced by CI users with competing speech.
• Bilateral CI users exhibit limitations in utilizing binaural cues, particularly under informational masking conditions.