Enlarging the pinna improves azimuth localization only for hearing-impaired listeners

Evidence

The main observations which can be made from Fig. 6 are: Substantially better (lower MAA) performance in the NH group. Large individual variation, especially in the HI group. Very small deviations between test conditions in the NH group. Some deviations between test conditions in the HI group, with better performance with W-cues than with N-cues (with T-cues in between), most pronounced for the 120° conditions. A mixed-model ANOVA was used to analyze the NH MAA data. Besides a significant test-subject effect (p = 0.002) (the random variable “test subject” was only included as a main effect), the various effects in the model came out as being non-significant, as already indicated by the rather similar mean values shown in Fig. 6 . Thus, the NH-group performance did not depend on either cue type or reference angle. Another mixed-model ANOVA was used to analyze the HI MAA data. As could be expected based on the large individual variance displayed in Fig. 6 , a highly significant test-subject effect (p < < 0.001) was observed. Furthermore, a significant effect of cue type was observed (p = 0.002), while the effect of reference angle was non-significant. No interactions were significant. Since the difference between normal and enhanced cues was the essence of the research question, the cue-type effect was further explored by conducting two planned comparisons, N vs. T and N vs. W (across the two reference angles). These analyses showed a significant difference between N and W cues (mean MAA difference of around 10°, p = 0.0005) and a non-significant difference between N and T cues (mean MAA difference of around 5°, p = 0.06). — Dijk, Pim van, Physiology, Psychoacoustics and…, 3105

Categorization

• QUESTION: Why did enlarged pinna cues not affect normal-hearing listeners at all, while clearly helping the hearing-impaired? The ceiling-effect interpretation is plausible but deserves scrutiny.