Effect of listener training on perceptual judgement of hypernasality

Abstract

Reliable perceptual judgement is important for documenting the severity of hypernasality, but high reliability can be difficult to obtain. This study investigated the effect of practice and feedback on intra-judge and inter-judge reliability of hypernasality judgements. The judges were 36 speech-language therapy students, who were randomly assigned to three groups for training: (1) Exposure (simple exposure to hypernasal speech samples), (2) Practice-only (practice with hypernasality judgements without feedback), and (3) Practice-Feedback (practice with hypernasality judgements with feedback). After training, the judges rated hypernasality in non-nasal sentences produced by 20 speakers with hypernasality and two normal speakers, using direct magnitude estimation. Both practice groups showed fair-to-good inter-judge reliability for rating the female samples: had more listeners who showed significant intra-judge reliability, and had significantly larger range of hypernasality ratings than the exposure group. To conclude, practice (with or without feedback) is useful for improving the reliability of hypernasality ratings.