PP1607 - Heritage Bilingual Speakers of Spanish: An Investigation of English Speech Recognition Performance

Abstract: This study explored speech recognition performance of heritage Spanish speakers compared to monolingual English speakers on English-language word and sentence tests. Participants underwent pure-tone audiometry, a language proficiency questionnaire, the Northwestern University Auditory Test No. 6 (NU-6), and Revised Speech Perception in Noise test (R-SPIN). Findings suggest that participants with normal hearing but who are heritage Spanish speakers may incorrectly recognize words. This research contributes to the evidence of differences in performance for bilingual listeners compared to monolingual speakers and suggests that assessment methods for Spanish-language speakers may need to be further studied when presenting speech recognition in noise.

Summary:

Purpose: Spanish is the most common non-English language in the United States (US Census Bureau, 2022). For professionals in speech language pathology and audiology, it can be challenging to determine the optimal language to assess speech understanding with, particularly for individuals who learned English later in life or who have differing levels of proficiency in English and Spanish (Shi, 2017). Heritage speakers are those who learned Spanish at home but who became more proficient in English as they became older. Studies have demonstrated that English-dominant bilingual listeners who were more proficient in English than in Spanish had poorer performance on Spanish-language speech recognition tests compared to listeners who had greater proficiency in Spanish than in English (Shi & Canizales, 2013; Shi & Sanchez, 2010; Weiss & Dempsey, 2008). Performance on speech recognition measures suggest that poorer performance on a test in one language does not equate to similar performance or an inability to be evaluated in the other language (Shi, 2017). The present Summary * study was a conceptual replication and expansion of the work by Shi (2017), which examined Spanish heritage speakers’ performance and phoneme errors on the Northwestern University Auditory Test No. 6 (NU-6; Tillman & Carhart, 1966) test. Our study also examined the errors made on the Revised Speech Perception in Noise test (R-SPIN; Wilson et al., 2012) to explore the effect of context on performance. The IRB at California State University, Sacramento approved this study. Procedures: Participants completed a language proficiency questionnaire and underwent pure-tone audiometry, the NU-6 test, and the R-SPIN test in a sound-treated booth. Testing was administered using supra-aural headphones presented at 40 dB SL binaurally. All listeners had normal hearing (defined as less than or equal to 25 dB HL) between 250 and 8000 Hertz. Speech recognition test lists were presented randomly, with one list selected for each participant. Participants verbally repeated and wrote down every word that was heard, and the researcher recorded the responses.

Results: Data collection is currently in progress but will be completed by the time of the poster presentation. Scoring of the words is at the word and phoneme level based on the comparison between the verbal and written responses. Results will be discussed on the poster.