Speech Entrainment Treatment for People With Aphasia

The objective of this research is to experimentally delineate the direct effect of speech entrainment practice on independent speech production and identify practice conditions that enhance treatment benefits. The primary outcome measure (Correct Information Units per minute) tallies informativeness and efficiency of independent speech in treated stories.

Speech entrainment refers to speaking in unison with a model speaker by imitating the model in real time. The objective of the study is to (1) experimentally establish the direct effect of speech entrainment practice on independent speech production post-treatment, and (2) identify conditions that enhance treatment benefits. These aims are addressed in a within-subject efficacy study, where 40 people with aphasia produce different stories with entrainment support. Speaking without entrainment is evaluated one day before and one day after speech entrainment practice. Different practice stories will be randomized within participants to three experimental conditions to assess the effect of treatment (trained vs. untrained), training schedule (massed vs. distributed presentation of stories), and entrainment modality (practice with auditory-only or audiovisual model). Correct information units per minute for each story will be tallied to evaluate the differences between conditions and the associations with patient characteristics. For consistency with prior research, number of different words per minute will serve as a secondary outcome. The proposed research addresses a clinical need by testing and optimizing a promising treatment technique for enhancing aphasia rehabilitation. This proof-of-concept study is the first to evaluate the direct effect of speech entrainment practice on independent speech. Thus, the effect size for accurate power estimation is not known. The strategy for this study is to optimize experimental sensitivity by maximizing the number of observations per participants per condition within the constraints of feasibility. The results from this study will provide critical information for evaluating the effect size for a subsequent phase. Nevertheless, to estimate an optimal sample size, the investigators leveraged the data from our Pilot Study, which examines the speech entrainment effect on immediate performance rather than learning. With 13 participants, the investigators estimated 80% power to detect the effect of 0.8 (mean difference between groups divided by the pooled standard deviation) observed in our pilot study for speaking with audiovisual speech entrainment vs. independent speech (Aim 1). The calculation is based on a paired t-test with an alpha of .05 (two-tailed) and a correlation of 0.5 between groups. With 24 participants, there is 80% power to detect the observed effect of 0.6 for speaking with audiovisual speech entrainment vs. auditory-only speech entrainment (alpha is set to .05, two-tailed, correlation between groups is .57; Aim 2b). The investigators have substantially increased the sample size (N=40) with the consideration that learning effects (the focus of the present study) are expected to be smaller than performance effects (the focus of the Pilot Study).

Audiovisual Speech Entrainment includes repeated practice of speaking synchronously with a recorded model, which provides contextual, temporal, and audiovisual cues for speaking.

Auditory Speech Entrainment includes repeated practice of speaking synchronously with a recorded model, which provides contextual, temporal, and auditory cues for speaking.

Speech Entrainment Practice is an aphasia rehabilitation treatment based on imitation of recorded speech in real time. Treatment includes repeated practice of speaking along with a recorded model, which provides contextual, temporal, auditory, and visual cues for speaking

Speech Entrainment Practice is an aphasia rehabilitation treatment based on imitation of recorded speech in real time. Treatment includes repeated practice of speaking along with a recorded model, which provides contextual, temporal, and auditory cues for speaking

Correct Information Units per Minute will be tallied by blinded coders to measure the informativeness and efficacy of speech

Number of Different Words per Minute will be tallied by blinded coders to measure the amount and variability of speech

Inclusion Criteria: Between 18-80 years at the time of enrollment English as native language greater than six months post onset Diagnosis of aphasia based on Western Aphasia Battery - Revised (WAB-R) Score of 7 or less on the Fluency subsection of the WAB-R Demonstrate having the willingness and stamina to participate in the multiple-session protocol, either by traveling to the MRRI testing site or being tested at home Exclusion Criteria: History of learning disabilities History of other comorbid neurological impairments