Differences in Connected Speech Outcomes Across Elicitation Methods

EasyChair Preprint 6408

Abstract

Eliciting connected speech is useful for capturing many aspects of an individual’s language abilities (Gordon, 2006; Nicholas & Brookshire, 1993; Rochon et al., 2000; Saffran et al., 1989). Common connected speech elicitation methods include description of pictured scenes (e.g., the Western Aphasia Battery’s picnic picture; Kertesz, 2007) and storytelling (e.g., the Cinderella Story). In comparison to picture description, storytelling elicits more content and more lexically diverse speech in speakers with chronic aphasia (Alyahya et al., 2020; Stark et al., 2019). However, it is unknown how these two methods compare in measuring structural and syntactic aspects of connected speech. Here, we compared picture description and storytelling in a large group of participants following acute left hemisphere stroke. We tested the degree of agreement and consistency across elicitation methods for structural, syntactic, and lexical measures of connected speech, as well as the degree of concordance in classifying deficits across individuals. 71 native-English speaking participants (59 ± 13 years; 25 female) completed picnic picture description and Cinderella storytelling within an average 3.9 days from left hemisphere stroke onset. Storytelling elicited significantly more structurally complex, syntactically accurate, and increased and more lexically diverse speech output in comparison to picture description. Methods were inconsistent across individuals in measuring lexical selection and syntactic accuracy, but generally consistent classifying individuals as impaired or spared, save for some structural and syntactic measures. We conclude that storytelling is the better measure to elicit connected speech for analyses of individual differences across patients. However, when assessing whether an individual has impaired or spared connected speech, either elicitation method will be generally sufficient, but take care when assessing syntactic accuracy.

Keyphrases: acute stroke, connected speech production, discourse, quantitative production analysis

@booklet{EasyChair:6408,
  author    = {Sharon Wang and Tatiana Schnur},
  title     = {Differences in Connected Speech Outcomes Across Elicitation Methods},
  howpublished = {EasyChair Preprint 6408},
  year      = {EasyChair, 2021}}