Speaker
-
Katie Von HolzenDr. at TU Braunschweig, Germany
Learning words form learner-directed registers, by Dr Katie Von Holzen
Learning words form learner-directed registers
When addressing language learners, adult L1 speakers modify their speech. The speech used with infants (infant-directed speech; IDS) and foreign-language learners (foreigner-directed speech; FDS) has been found to have greater vowel hyperarticulation and slower speech rate in comparison to adult-directed speech (ADS), but pitch modifications (e.g., higher pitch peaks and average pitch) have only been consistently found in IDS (Piazza et al., 2022; Soderstrom, 2007). This variability in acoustic features may be due to the role they play in supporting language learning. Vowel hyperarticulation and lower speech rate are thought to improve speech perception and comprehension, which would benefit both infant- and foreign-language learners, while pitch modifications are thought to draw infants’ attention to speech, which is unnecessary for older learners (Uther et al., 2007). Experimental evidence supports this hypothesis for IDS in both infancy (e.g., Ma et al., 2011; Thiessen & Saffran, 2005) and adult foreign language learners (Golinkoff & Alioto, 1995). Although FDS has been found to facilitate word learning in adult learners of English (Piazza et al., 2023), this evidence comes from learners who had already gained some proficiency in English. To better understand the didactic consequences of learner-directed registers at all stages of acquisition, the current study examines whether both IDS and FDS facilitate word learning in adult ab initio learners of Chinese.
I will present preliminary experimental evidence from 13 German-speaking adults with no prior experience learning Mandarin Chinese. Using a procedure similar to Golinkoff and Alioto (1995), they first viewed a series of 10 objects (e.g., bottle), each paired with two sentences produced in Mandarin Chinese that presented the label of the object in utterance-final position (e.g., Zhe shi yi ge PING). Then in the test phase, participants viewed three of the previously familiarized objects while one of the objects was labeled (e.g., PING) and were instructed to select the corresponding object (e.g., bottle).
In a between-subjects design, the sentences participants heard were produced using either an ADS, IDS, or FDS register. Acoustic analyses of the sentence stimuli confirmed previous evidence of a lower speech rate and higher pitch peaks and average pitch for both IDS and FDS in comparison to ADS (p’s < .01). The vowel triangle for FDS was also greater than ADS, showing vowel hyperarticulation. However, the vowel triangle for IDS was unexpectedly smaller, but overall showed more within-category variability, which has been found in some studies of IDS acoustic characteristics (e.g., Rosslund et al., 2024).
Participants’ accuracy in selecting the correct object was modeled using a general-linear mixed model, revealing a significant main effect of register (p < .01). Accuracy was greater for both participants familiarized with IDS (p < .001) and FDS (p < .01) in comparison to participants familiarized with ADS. Accuracy was significantly above chance (33%) for both IDS and FDS (p < .05), but not ADS. Taken together, these preliminary results suggest that learner-directed registers may facilitate foreign language word learning, even in the absence of prior knowledge of the language.