Recursive Self-feedback Improved Speech Fluency in Two Patients with Chronic Nonfluent Aphasia

ABSTRACT

Background

Previous studies have demonstrated that people with nonfluent aphasia (PWNA) improve their language production after repeating personalized scripts, modeled by speech-language pathologists (SLPs). If PWNA could improve by using their own self-feedback, relying less on external feedback, barriers to aphasia treatment, such as a dearth of clinicians and mobility issues, can be overcome. Here we examine whether PWNA improve their language production through an automated procedure that exposes them to playbacks of their own speech, which are updated recursively, without any feedback from SLPs.

Method

We tested if recursive self-feedback could improve speech fluency in two persons with chronic nonfluent aphasia. We compared two treatments: script production with recursive self-feedback (a new technique) and a non-self-feedback training. We administered the treatments remotely to the participants through their smartphones using two versions of a mobile app we developed. Each participant engaged in each treatment for about three weeks. We estimated clinical improvements of script production through a quantitative trend analysis and nonoverlap of all pairs.

Results

Recursive self-feedback improved speaking rate and speech initiation latency of trained and untrained scripts in both participants. The control (non-self-feedback) training was also effective, but it induced a somewhat weaker improvement in speaking rate, and improved speech initiation latency in only one participant.

Conclusion

Our findings provide preliminary evidence that PWNA can improve their speaking rate and speech initiation latency during production of scripts via fully automated recursive self-feedback. The beneficial effects of recursive self-feedback training suggest that speech unison and repeated exposures to written scripts may be optional ingredients of script-based treatments for aphasia.

KEYWORDS:

• Nonfluent Aphasia
• Recursive Self-feedback
• Self-feedback
• Mobile-based Treatment
• Language Production

Acknowledgements

This study was supported by the National Institutes of Health R01 DC04722 grant awarded to Ofer Tchernichovski, Ph.D. We thank the participants for engaging in our study especially during the pandemic. We also thank members of the Neurolinguistics labs at the Graduate Center and Lehman College, City University of New York. We thank the two anonymous reviewers for their helpful comments. This research paper is based on and derived from the original PhD dissertation authored by Gerald C. Imaezue, Ph.D. serving as the primary source of information, analysis, and findings presented herein.

Disclosure statement

We have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.