Researching the Production of Specific Features
Charlie Nagle · 2024
Specific features refer to the production of individual sounds and prosodic features. Traditionally, specific features have been elicited using controlled production tasks such as word and sentence reading, and accuracy has been quantified using acoustic analysis. This methodology remains important, but researchers have begun to examine the production of specific features under more spontaneous conditions, using tasks such as timed picture description. They have also recruited listeners to evaluate the intelligibility of specific sounds, reflecting a shift from phonetic accuracy or nativeness to intelligibility. This chapter discusses the types of specific features that make for good target structures, data elicitation methods, and data analysis. Two longitudinal data sets are presented and analyzed. The first example deals with stop consonant production in second language Spanish, specifically the production of voice onset time, a temporal acoustic measure. This example is analyzed using analysis of variance and mixed-effects modeling. The second example deals with vowel production in English as a second language. In that example, the outcome variable is a listener-based intelligibility measure, and the statistical analysis is a logistic mixed-effects model appropriate for binomial data.