Producing and recognizing words with two pronunciation variants: Evidence from novel schwa words

Abstract

This study examined the lexical representations and psycholinguistic mechanisms underlying the production and recognition of novel words with two pronunciation variants in French. Participants first learned novel schwa words (e.g., /ʃənyk/), which varied in their alternating status (i.e., whether these words were learned with one or two variants) and, for alternating words, in the frequency of their variants. They were then tested in picture-naming (free or induced) and recognition memory tasks (i.e., deciding whether spoken items were learned during the experiment or not). Results for free naming show an influence of variant frequency on responses, more frequent variants being produced more often. Moreover, our data show an effect of the alternating status of the novel words on naming latencies, with longer latencies for alternating than for nonalternating novel words. These induced naming results suggest that both variants are stored as lexical entries and compete during the lexeme selection process. Results for recognition show an effect of variant frequency on reaction times and no effect of variant type (i.e., schwa versus reduced variant). Taken together, our findings suggest that participants both comprehend and produce novel French schwa words using two lexical representations, one for each variant.

Keywords:

• Word production
• Spoken word recognition
• Phonological variation
• Lexical representations
• Schwa
• Novel words

Acknowledgments

The authors wish to thank Marina Laganaro for the use of her laboratory in Neuchâtel, the members of Gareth Gaskell's lab for their useful comments on a previous version of this manuscript, and Ann Travis for her careful proofreading of the manuscript.

Notes

¹ While there is usually little discussion as to which variant has to be considered canonical, the criteria underlying this categorization are not homogeneous across variation processes. The canonical form usually corresponds to the citation form, to the word's spelling, or to both.

² In French (and unlike, for instance, in English), schwa refers to a full vowel rather than to a centralized/reduced vowel. Moreover, the alternation between schwa and reduced variants is generally described in categorical terms (e.g., Côté & Morrison, 2007; F. Dell, 1985). By contrast, in some other languages (e.g., prestress schwa words in English, Davidson, 2006), reduced variants are thought to result from a gradient phonetic reduction process.

³ We assume here that the prelexical automatic restoration process described by Spinelli and Gros-Balthazard (2007) has no cost.

⁴ Using novel words to examine frequency also allows us to avoid one limitation of frequency measures extracted from corpora (as used for instance in Connine et al., 2008, or Pitt et al., 2011)—that is, that they are highly dependent on the style of speech used in the corpus.

⁵ Even though the apostrophe is the accepted way to refer to the nonschwa pronunciation, readers do not come across this spelling very often. As a consequence, the reduced pronunciation's spelling is probably harder to process than that of the schwa variant. This difference is, however, not relevant to the task at hand in our design (induced picture-naming task), given that this task does not involve the online processing of the spelling and given that the orthographic information is not activated online during spoken production tasks (e.g., Alario, Perre, Castel, & Ziegler, 2007; Bi, Wei, Janssen, & Han, 2009; Roelofs, 2006). Note in addition that our results do not show the advantage for schwa variants that would be expected if their responses had been affected by a difference in spelling.