To confirm an increased susceptibility to informational masking among individuals with single-sided deafness (SSD). To demonstrate a reduction in informational masking when SSD is treated with an integrated bone conduction hearing aid (IBC). To identify the acoustic cues that contribute to IBC-aided masking release. To determine the effects of device experience on the IBC advantage.
Informational masking was evaluated with the coordinate-response measure. Participants performed the task by reporting color and number coordinates that changed randomly within target sentences. The target sentences were presented in free field accompanied by zero to three distracting sentences. Target and distracting sentences were spoken by different talkers and originated from different source locations, creating two sources of information for auditory streaming. Susceptibility to informational masking was inferred from the error rates of unaided SSD patients relative to normal controls. These baseline measures were derived by testing inexperienced IBC users without the device on the day of their initial fitting. The benefits of IBC-aided listening were assessed by measuring the aided performance of users who had at least 3 months’ device experience. The acoustic basis of the listening advantage was isolated by correlating response errors with the voice pitch and location of distracting sentences. The effects of learning on cue effectiveness were evaluated by comparing the error rates of experienced and inexperienced users.
Unaided SSD participants (inexperienced users) performed as well as normal controls when tested without distracting sentences but produced significantly higher error rates when tested with distracting sentences. Most errors involved responding with coordinates that were contained in distracting sentences. This increased susceptibility to informational masking was significantly reduced when experienced IBC users were tested with the device. The listening advantage was most strongly correlated with the availability of voice pitch cues, although performance was also influenced by the location of distracting sentences. Directional asymmetries appear to be dictated by location-dependent cues that are derived from the distinctive transmission characteristics of IBC stimulation. Experienced users made better use of these cues than inexperienced users.
These results suggest that informational masking is a significant source of communication impairment among individuals with SSD. Despite the lateralization of auditory function, unaided SSD subjects experience informational masking when distractors occur in either the deaf or normal spatial hemifield. Restoration of aural sensitivity in the deaf hemifield with an IBC enhances speech intelligibility under complex listening conditions, presumably by providing additional sound-segregation cues that are derived from voice pitch and spatial location. The optimal use of these cues is not immediate, but a significant listening advantage is observed after 3 months of unstructured use.
This study examined the effects of integrated bone conduction hearing aids (IBCs) on informational masking in individuals with single-sided deafness (SSD). Informational masking was quantified by measuring speech recognition in multi-talker babble. All SSD participants were more susceptible to informational masking than normal control subjects. Experienced IBC users showed significantly less informational masking than unaided SSD participants and inexperienced IBC users. Acoustic manipulations of the multi-talker background linked the IBC listening advantage to an enhanced processing of voice pitch and spatial location cues. SSD patients learned to make effective use of these information sources during the initial three months of unstructured device exposure.
1Department of Otolaryngology, Head and Neck Surgery, Johns Hopkins University, Baltimore, Maryland, USA; and 2Department of Otorhinolaryngology, Hartford Hospital, Hartford, Connecticut, USA.
John Niparko is an unpaid consultant to Cochlear Limited. The other authors declare no conflict of interest.
This research was supported by funding from Cochlear Limited and the National Institute on Deafness and Other Communication Disorders grant P30 DC005211.
Address for correspondence: Bradford J. May, Department of Otolaryngology-Head and Neck Surgery, Johns Hopkins University, 720 Rutland Avenue, Baltimore, MD 21205, USA. E-mail: email@example.com