Comparisons of the pitch perception abilities of adults and children using cochlear implants or hearing aids

Introduction This paper compares the pitch perception abilities of adults to children fitted with a cochlear implant (CI) or a hearing aid (HA). It is well reported that adult CI recipients score significantly lower than normally hearing (NH) listeners on pitch-based tasks (Looi et al., 2012; McDermott, 2004). However, there is a paucity of research for children or for HA users. Results from adults cannot be directly translated to children as pre-lingually deafened children: (i) have no NH exposure to music and pitch information so they do not have a NH auditory memory; (ii) lack of exposure to high-resolution pitch cues, potentially impeding the development of pitch-related central auditory processing skills; (iii) are often implanted at a young age when cortical plasticity is high, potentially enabling them to adapt differently to electric stimulation; and (iv) have greater neuroplasticity. In addition, most of the existing research compares CI users to NH. The question must be asked as to whether this is appropriate given that CI users have significant sensorineural hearing loss, which has been shown to affect auditory filter bandwidths and the perception of pitch (Moore, 1995). Looi et al. (2008a) compared adult CI users to HA users who met CI criteria (i.e. severe-to-profound hearing loss) and found that although CI users had poorer results compared to HA users, the HA users were significantly worse than NH listeners. For example, HA users scored 75% in ranking three-semitone intervals, a task which NH listeners typically score 100%. Hence, this study aimed to compare the pitchranking skills of adults and children with CIs and/or HAs. Materials and methods Four existing studies have utilized the same pitchranking task: (i) Looi et al. (2008a) compared 15 adult unilateral CI recipients to 15 adult HA users; (ii) Looi et al. (2008b) compared nine adults pre-topost implant (i.e. HA pre-surgery, then unilateral CI at 3 months post-mapping); (iii) Looi and King (2012) comparing 18 adult unilateral CI recipients to 13 adult HA users; and (iv) Looi and Radford (2010) involving seven unilaterally implanted children, eight children using bimodal stimulation (BMS), and six children with bilateral HAs. In all the four studies, participants undertook a two-alternative forced-choice (2AFC) pitch-ranking task, one-, half-, and a quarter-octave apart. Each test item consisted of two different sung /a/ vowels at the required pitch interval. The fundamental frequencies of the stimuli are given in Table 1. Each note had a linear rise/decay ramp of 30 ms, with the two notes being presented sequentially, ascending or descending, separated by 500 ms of silence. Each pitch pair was presented eight times – four ascending and four descending. The loudness levels were randomized, and the participants were required to state which of the two notes in each pair was higher in pitch, ignoring any difference in the loudness of the notes. Stimuli were presented at comfortable loudness levels, and the participants used their preferred listening programme for testing. The one-octave and halfoctave subtests provided scores out of 96, while the quarter-octave subtest was out of 128. With a 2AFC response format, the chance score was 50%. Participant demographics are presented in Table 2 (adult) and Table 3 (children). All CI participants used Cochlear Ltd Nucleus devices with the ACE or SPEAK speech processing strategy, and all HAs were digital behind-the-ear devices. All of the adults had >6 months experience with their device, and all paediatric participants had >12 months experience Correspondence to: Valerie Looi, Sydney Cochlear Implant Centre, Macquarie University, Australian Hearing Hub, Ground Level, 16 University Avenue, Sydney, NSW 2109, Australia. Email: valerie.looi@scic.org.au