Phonetic properties of aphasic-apraxic speech: A modified narrow transcription analysis

Abstract

We used a modified narrow phonetic transcription procedure to examine a speech sample produced by 10 speakers with coexisting aphasia and apraxia of speech. The transcription protocol was limited to eight diacritic marks selected based on previous perceptual descriptions of phonetic distortion among speakers with pure apraxia of speech. Additionally, there was one general distortion category for perceived distortions not captured by the main diacritic marks. The results showed that distortion errors were as common as substitution errors, that vowel and consonant segments were equally vulnerable to misproduction, and that there was no difference between the frequency of consonants produced incorrectly in prevocalic and postvocalic syllable positions. Among distortion errors, 32% were classified as segment prolongations and 41% were classified as general distortions. An independent transcription that used a comprehensive system of diacritic marks was performed as a follow-up. Most general distortions were not further specified by the comprehensive transcription system. A phonetic contrast analysis of errors detected at the broad transcription level yielded different speech characteristics from the observed distortion errors, but were in general correspondence with previously reported error patterns in intelligibility testing of the same speech sample. The results are discussed relative to the role of single word intelligibility testing in aphasia and apraxia of speech assessment and the challenges associated with transcribing disordered speech.