Metrical encoding for words with regular and irregular stress

Niels O. Schiller^1,2, Bernadette M. Schmitt¹, Judith Peters¹, and Willem J. M. Levelt²
n.schiller@psychology.unimaas.nl
1 Department of Neurocognition, Faculty of Psychology
Universiteit Maastricht, The Netherlands
2 Max Planck Institute for Psycholinguistics, Nijmegen, The Netherlands

This study investigated the encoding of metrical information during speech production. According to Levelt's theory (Levelt, Roelofs, & Meyer, 1999) the metrical frame of a word specifies the number of syllables and the position of the main stress. Furthermore, the theory assumes that for stress assigning languages, such as Dutch, no metrical frame is stored for lexical items with regular default stress. Regular stress in Dutch means the first full vowel is stressed. This is the case for more than 90% of the Dutch words. For these regular items, the metrical pattern is assigned by default.

In the first experiment, participants were asked to name pictures that had initial or final stress (KAno 'canoe' vs. kaNON 'cannon'; capital letters indicate stressed syllables). Picture names were all bisyllabic and matched for frequency. An object/non-object decision control experiment demonstrated that there was no difference in terms of visual recognizability between the two types of items. The picture naming experiment revealed that picture names with initial stress were faster to name than picture names with final stress. This stress effect may be located at the level at which the metrical frame is either computed by default or retrieved from the lexicon. Our result suggests that

(1)

there is a difference between default and non-default stress and that

(2)

computing default stress is faster than retrieving an irregular stress pattern from memory.

This initial vs. final stress advantage remains visible in a task that taps into a later level than metrical retrieval, namely the phonological word level. In earlier research, Wheeldon and Levelt (1995) used a translation-naming task to investigate the encoding of segments at this level and found that the segments of a word are encoded incrementally from beginning to end. Here, we used a modification of that task, namely implicit picture naming. In this task, participants are asked to judge whether bisyllabic picture names had stress on the first or second syllable. Participants saw the same pictures as in the naming experiment on a computer screen and decided for each picture whether its name had initial or final stress without overtly naming the pictures. Results showed significantly faster decision times for initially stressed targets than for targets with final stress. To our knowledge, this default/non-default difference has not been shown in production tasks before. The results are discussed in terms of models of phonological encoding in speech production.

References

Levelt, W. J. M., Roelofs, A, & Meyer, A. S. (1999). A theory of lexical access in speech production. Behavioral and Brain Sciences, 22, 1-75.

Wheeldon, L. & Levelt, W. J. M. (1995). Monitoring the time course of phonological encoding. Journal of Memory and Language, 34, 311-334.