Linguistics Honors Projects

Spontaneous Phonetic Imitation Across Regional Dialects

Ellen E. Dossey — Fri, 04 May 2012 13:45:39 PDT

Speakers subconsciously alter their pronunciation towards that of their conversation partner through a process called spontaneous phonetic imitation. Previous research has shown that this imitation may not be a completely automatic process. One factor that may influence the extent of imitation is social biases based on gender, age, or region. The current study used a lexical shadowing task to determine if speakers would spontaneously imitate American English vowel variants that were not found in their native dialect. Results showed that in the non-interactive task, regionally distinct vowels were significantly imitated. Furthermore, the level of imitation was not related to previous familiarity or experience with the dialect in question.

Vowel Duration and Perceptions of the Gay Accent

Laura Panfili — Wed, 19 Oct 2011 12:14:35 PDT

This study aimed to examine a potential linguistic cue that signals a speaker's sexual orientation. I examined the relationship between vowel duration and perceived sexual orientation for male speakers of American English. Speakers recorded a passage that was heard by naive listeners and ranked according to perceived sexual orientation. There was no significant difference in vowel duration between men perceived to sound gay and men perceived to sound straight. However, the gay-sounding men produced their diphthongs with more variance in duration than did the straight-sounding men.

Event Segmentation and Memory Retrieval in Reading Comprehension

Charles Baker Brenner — Fri, 07 May 2010 07:08:10 PDT

Comprehending text involves the convergence of top-down, expectation-driven processes and bottom-up, stimulus-driven processes. The precise nature of this convergence, however, is not well understood. The current study used narrative time shifts and shifts in protagonist goal, both hypothesized to encourage event-segmented memory representations, to investigate the interaction between automatic and constructive memory processes during reading. The addition of time and goal shifts was found to have no effect on the automatic retrieval of information from memory. The results are interpreted as support for the bottom-up account of retrieval of information during reading, and for the idea that the top-down account is best applied to the integration of information after retrieval.

Effect of Visual Input on Vowel Production in English Speakers

Amanda C. Richardson — Wed, 05 May 2010 12:44:33 PDT

This study analyzes whether there should be a visual component to a model of speech perception and production by comparing the jaw opening, advancement, and rounding of American English and non-English vowels in the presence and absence of a visual stimulus. Surprisingly, jaw opening did not change production, but the presence of the visual stimulus was found to be a significant factor in participants’ vowel advancement for non-English vowels. This may be explained by lip rounding, but requires further research in order to develop a full understanding of the impact of visual input on vowel production to be used in teaching and learning languages.

Avatime Noun Classes

Ronny C. Watkins — Tue, 04 May 2010 07:03:48 PDT

Like many African languages, particularly Bantu languages, the Ghanaian language of Avatime organizes its nouns into a class system. Noun classes use affixes to indicate semantic category. For example, many languages with noun classes have a class for ‘people’ nouns, such as ‘child’ and ‘stranger.’ Previous research on Avatime has postulated between seven and nine classes. The current study is based on original field work done in the village of Vane in Ghana. The new data show that Avatime has seven noun classes, distinguished by singular and plural prefixes, definite article suffixes, and semantic similarities.

The Effect of Learning on Sentence Prosody in Japanese

Joanna Baldwin Clark — Fri, 30 Apr 2010 06:30:26 PDT

This study investigates the effect of learning on prosodic production competence in native English L2 speakers of Japanese. Intonation contour and speech rate as indicators of competency were examined. It was hypothesized that more experience with Japanese would lead to more native-like prosody. The study tested the production of fourteen L2 learners, ten non-learners and six native speakers. Participants recorded twenty-three sentences of Japanese. Acoustic data was analyzed for speech rate and fundamental frequency (F0). Results showed that experience is positively correlated with speech rate and not correlated with deviation from the Japanese mean intonation contour.

The Origins of Nonsense: An Analysis of Bo'ri'va:r Sap in Khmer

Stephanie Farmer — Thu, 25 Sep 2008 07:00:44 PDT

Ashkii Bizaad: Verbal Morphology Loss in One Young Speaker's Navajo

Eric W. Weisser — Tue, 10 Jun 2008 11:32:39 PDT

With a rich verbal morphology and an aging population of native speakers, Navajo offers a valuable opportunity to examine language attrition in detail. Few Navajo children grow up completely unexposed to their heritage language, yet the number raised as monolingual English speakers has risen sharply in the past thirty years. This thesis compares one young speaker's production of verbs with conservative, "textbook" forms, analyzes the patterns found within this comparison, and draws on similar processes in the dying languages Dyirbal and Romansch to place these Navajo data in the larger context of language attrition.