Japanese pitch-accent identification accuracy by children with autism spectrum disorder

Abstract

A previous study demonstrated when Japanese speakers hear sine-wave speech carrying no fundamental frequency information, they rely on the first formant frequency for their pitch-accent perception, resulting in poor identification accuracy. However, when sine-wave speech is noise-vocoded, each formant becomes less audible, and Japanese speakers use other acoustic cues (e.g., duration and intensity), leading to better identification [Y. Shinohara, “Japanese pitch-accent perception of noise-vocoded sine-wave speech,” The 183rd Meeting of ASA, Nashville, TN (2022)]. The present study examined pitch-accent perception by children with autism spectrum disorder (ASD) for the same three stimulus types: natural recordings, sine-wave speech, and noise-vocoded sine-wave speech. Children with ASD often have enhanced auditory pitch perception for non-speech stimuli. However, since the present study used a pitch-accent word identification task, it was hypothesized that Japanese-speaking children with ASD would show less identification accuracy for all three stimulus types compared to age-matched typically developing children. Our preliminary results showed a significant difference in identification accuracy between the two groups. A detailed analysis of the results and the acoustic features of each stimulus type will be discussed.