engleski

The role of syllable-position of segments in phonological encoding in Croatian: evidence from speech errors

The aim of this research was to revise the slots- and-filler model of phonological encoding (Shattuck-Hufnagel, 1979) based on the analysis of experimentally induced segmental errors in Croatian. Segmental errors were elicited by 48 tongue twisters. Each tongue twister consisted of four monosyllabic consonant-vocal-consonant (CVC) Croatian words and had two target consonants which were: (1) both in the syllable-initial position, (2) both in the syllable-final position or (3) one target consonant in the syllable-initial and other one in the syllable-final position. The target pair appeared in ABAB or BAAB position. Phonological similarity between target consonants was also controlled. Twenty-four female participants first read each tongue twister aloud several times, and then repeated it from memory. The rate of reciting the stimulus was 180 bits per second and was controlled by the metronome. Tongue twisters were presented to participants in written form on a computer screen in random order. This experiment is based on experiments used by Shattuck-Hufnagel (1992) and Wilshire (1999) and adopted for Croatian. The results show that target consonants replace each other significantly more often if they are in the same syllable position than if they are not. On the other hand, no significant difference was found between the number of mutual substitutions of target consonants in the syllable-initial and syllable- final position. The obtained results confirm some of the assumptions on which the slots-and-fillers model is based, for example that the position of segments in a larger unit does play a role in phonological encoding, but also suggests that some other assumptions need to be revisited in the context of different languages and their phonological features.

speech errors, tongue twisters, phonological encoding

nije evidentirano