Computerized measurement of speech intelligibility. II. Normative data from a 'closed response' test

Area of Science:

• Audiology
• Speech Science
• Hearing Impairment Research

Background:

• Assessing speech recognition in hearing-impaired individuals is crucial.
• Background noise significantly impacts speech perception.
• Frequency-dependent amplification is a common hearing aid fitting strategy.

Purpose of the Study:

• To evaluate the validity and reliability of the Danish Four Alternative Forced Choice (4AFC) test.
• To investigate speech recognition performance in normal-hearing and cochlear hearing-impaired listeners.
• To analyze consonant confusions in different noise conditions.

Main Methods:

• A Danish Four Alternative Forced Choice (4AFC) test was administered.
• Participants included normal-hearing and cochlear hearing-impaired listeners.
• Two background noises were used: modulated speech-spectrum-shaped noise and 8-person babble.
• Hearing-impaired listeners received frequency-dependent amplification (POGO-fitting).

Main Results:

• The 4AFC test demonstrated validity despite non-homogeneous alternatives.
• A significant learning effect was observed, impacting test reliability.
• Computer-assisted analysis revealed a predominance of initial consonant confusions based on voicing.

Conclusions:

• The Danish 4AFC test is a valid tool for assessing speech recognition in hearing-impaired individuals.
• The identified learning effect necessitates careful consideration during test administration.
• Consonant confusion patterns provide insights into phonemic discrimination deficits in hearing impairment.