How much does language proficiency by non-native listeners influence speech audiometric tests in noise?

Area of Science:

• Audiology
• Speech-Language Pathology
• Psychoacoustics

Background:

• Assessing speech recognition in noise is crucial for audiological diagnosis.
• Non-native listeners may exhibit different speech recognition patterns compared to native speakers.
• Linguistic complexity of speech tests can impact performance, especially in noisy environments.

Purpose of the Study:

• To determine how linguistic complexity of speech recognition tests and second language proficiency affect speech recognition thresholds (SRTs) in noise for non-native listeners.
• To compare the performance of non-native and native listeners on different German speech recognition tests.

Main Methods:

• Speech recognition thresholds (SRTs) were measured in noise for 64 non-native and 8 native German listeners.
• Three German speech recognition tests were used: digit triplet test (DTT), Oldenburg sentence test (OLSA), and Göttingen sentence test (GÖSA).

Main Results:

• Non-native listeners achieved native-like SRTs only with the linguistically simple DTT.
• Phonemic-acoustic cue limitations in the DTT affected both native and non-native listeners similarly.
• For more complex tests (OLSA, GÖSA), non-native listeners required a 3-dB to 6-dB better signal-to-noise ratio than native listeners.

Conclusions:

• Closed-set speech tests (DTT for screening, OLSA for clinical use) are recommended for non-native listeners in audiology.
• Open-set tests (GÖSA, HINT) may overestimate difficulties in non-native speakers.
• Utilizing tests in the patient's native language, if available, is advisable for accurate assessment.