Authors: Tugba Lulaci¹; Pelle Söderström¹’²; Mikael Roll¹
¹Lund University; ²National Acoustic Laboratories
Background: Listeners with clinically normal audiograms often exhibit significant variability in spoken-word recognition. Standard pure-tone tests (up to 8 kHz) may fail to capture how variation in extended high-frequency (EHF; >8 kHz) hearing sensitivity contributes to these differences. This study investigated whether EHF hearing supports access to subtle acoustic cues that facilitate rapid, predictive word recognition.
Method: The study utilized an auditory gating task to examine the relationship between EHF thresholds and spoken-word recognition. Participants identified words from increasingly longer word-onset fragments. This allowed the researchers to assess how efficiently early acoustic cues—particularly those with high-frequency content—were used to identify words before the full stimulus was presented.
Results: Listeners with better EHF thresholds demonstrated significantly higher accuracy at early gates (as early as 15 ms). This advantage was specific to word onsets containing distinct, high-frequency energy (e.g., the fricative /s/). Conversely, no such relationship was observed for words beginning with weaker or more diffuse high-frequency cues (e.g., /f/).
Conclusion: EHF hearing sensitivity facilitates the rapid extraction and use of acoustic cues at word onset, supporting efficient speech recognition. These findings suggest that hearing sensitivity above 8 kHz contributes to subclinical listening differences and indicates that EHF assessment may provide more comprehensive insights into a listener’s real-world communication abilities than standard audiometry alone.
