Perceptual learning of interrupted speech.

Area of Science:

• Auditory perception
• Cognitive psychology
• Speech processing

Background:

• Speech intelligibility can be improved by filling silent gaps with noise bursts.
• This phenomenon is linked to phonemic restoration, a cognitive process aiding degraded speech comprehension.
• The study investigates whether filler noise impacts the learning and effectiveness of speech restoration.

Purpose of the Study:

• To test if interrupted speech with filler noise is learned differently than speech with silent gaps.
• To determine if the benefit of filler noise diminishes with training.
• To explore the cognitive mechanisms underlying speech restoration with and without noise.

Main Methods:

• Two groups of normal-hearing listeners trained on interrupted sentences, one with filler noise, one without.
• Training involved auditory playback feedback and visual text display.
• Performance and perceived mental effort were assessed before and after training.

Main Results:

• Both groups showed significant performance improvements after training.
• The restoration benefit did not diminish with training in either group.
• Intelligibility gains and mental effort reductions were similar across both training conditions.

Conclusions:

• Perceptual learning effectively enhances interrupted speech intelligibility, engaging top-down restoration mechanisms.
• Similar learning rates and effort reductions suggest shared cognitive processes for speech restoration, regardless of filler noise.
• Training may serve as a rehabilitative tool for hearing-impaired individuals, though further research is needed.