The influence of phonetic dimensions on aphasic speech perception

Abstract

Individuals with aphasia have more problems detecting small differences between speech sounds than larger ones. This paper reports how phonemic processing is impaired and how this is influenced by speechreading. A non-word discrimination task was carried out with ‘audiovisual’, ‘auditory only’ and ‘visual only’ stimulus display. Subjects had to decide whether two presented stimuli were the same. Six aphasic subjects with speech sound processing difficulties and 14 non-brain-damaged control subjects participated in this study. It was found that the aphasic subjects have difficulties in discriminating pairs of non-words, which are more profound for small differences. Differences in ‘voicing’ were least often detected and therefore seem most difficult to perceive. This implies different processing of the phonetic dimensions in speech sound perception. Performance improved when speechreading was possible. As this improvement is not based on differences in place of articulation only, theories of audiovisual processing need to be revised.

• aphasia
• speech reading
• speech sound processing
• phonetic dimensions

Acknowledgements

This project was funded by an Ubbo Emmius Scholarship of the University of Groningen to the first author. Research of the second and third author was funded by the University of Groningen. The authors gratefully acknowledge Callista Jippes for help with the preparation of the material and students in Speech and Language Pathology for helping collecting the data of the control subjects. Furthermore, they would like to thank the aphasic and control subjects, speech therapists and rehabilitation centres for participating in this study.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Notes

1. We are aware that terminology is not consistent in the literature and both ‘speechreading’ and ‘lipreading’ have been used. In this paper we use the term ‘speechreading’ because the visual input received is not restricted to the lips, but rather covers the lower face, neck, and upper chest. This terminology has also been suggested by Campbell, Dodd, and Burnham (1998) in order to clearly state that more than just lip information is taken into account and to stress that what is read is indeed natural speech.