▼ Identifying those phonological factors that native listeners rely on most when perceiving non-native speech is critical for setting priorities in pronunciation instruction. The importance of accurate lexical stress production, particularly primary stress, has been explored. However, little is known about the role of Secondary-stressed (SS) syllables and Unstressed-unreduced (UU) syllables, and the importance of their accuracy for speech perception. These questions are of relevance for Russian learners of English, who often reduce English SS and UU vowels—a phenomenon which is arguably due to the fact that only one stressed syllable per word is allowed in Russian phonology. Moreover, second language research has not addressed the issue of vowel over-reduction, which is a pattern typical of Russian learners. Low-accuracy productions of SS and UU syllables are generally not expected to lead to unintelligibility; however, they might interfere with the ease and accuracy with which speech is perceived. An acoustic study first compared realization of SS and UU syllables in words produced in isolation by six Russian learners of English and six native English speakers. Words were selected to contain low vowels and specific UU and SS syllable positions to optimally reflect vowel reduction by Russian speakers. Acoustic analyses revealed significant vowel quality and duration reductions in Russian-spoken SS and UU vowels, which were only half the duration of native English productions and significantly centralized. A subsequent psycholinguistic perceptual study investigated the degree of interference that inaccurate productions of SS and UU syllables have on native listeners’ speech processing. A cross-modal phonological priming technique combined with a lexical decision task assessed speech processing of 28 native English speakers as they listened to (1) native English speech, (2) unmodified Russian speech, and (3) modified Russian speech with SS and UU syllables altered to match native productions. Unmodified UU vowels led to significant inhibition of lexical access, while unmodified SS vowels revealed less of such interference. Acoustically “improving” vowel quality and duration in UU and SS syllables greatly facilitated word recognition only for UU-syllable-containing words. A recommendation is made that UU syllables are incorporated into pronunciation instruction for Russian learners of English. Advisors/Committee Members: Dilley, Laura.

▼ The processes by which spoken language is comprehended are extremely complex, and the development of these processes is not fully understood. The comprehension of language is a difficult process that involves decoding the incoming signal, accessing stored semantic information in the mental lexicon (i.e., lexical access), and integrating that information into the overall context of the sentence, paragraph, or conversation. Several cues may be used to access semantic information in the mental lexicon, including sentential context and frequency information. Use of these cues can provide evidence for either an interactive or modular process of lexical access. Based on research with homophones, it appears that adults are able to use sentential context, in an interactive fashion, to aid in lexical access. Preschool children, however, are unable to use context in this way. It is unclear as to when children begin to use contextual cues in an interactive manner to facilitate lexical access. The purpose of this research was to investigate sixth-grade children’s use of context and frequency cues to resolve lexical ambiguities encountered in spoken language. Seventy-five sixth grade children participated in a lexical ambiguity resolution task in order to determine whether frequency information or sentential context aided in lexical access. The participants heard a sentence (spoken by a man) that ended in a homophone. The context of this sentence was biased toward the subordinate sense of the homophone. Immediately following the sentence, participants heard a word (spoken by a female). The participants were instructed to repeat the word spoken by the female as quickly as possible. The time to repeat this word was recorded. Participants were faster to repeat words related to the subordinate sense of a homophone than they were to repeat unrelated words. However, for the dominant sense of a homophone, no significant difference in naming times were found compared to unrelated words. The results indicated that the sixth grade children were using sentential context, not frequency information, to aid in lexical access. The results support an interactive, rather than modular, model of lexical access, such as the context sensitive model of lexical ambiguity resolution. Advisors/Committee Members: Small, Larry H.

▼ This study used a mixed methods approach to explore the perceptions of adolescents who stutter (AWS) between the ages of 12 and 17 related to communication with parents and other significant people about stuttering. In addition, this study sought to better understand the relationship between general communication levels between AWS and their parents and how general communication was related to discussing stuttering. Participants took part in a semi-structured interview and completed the Communication about Stuttering Inventory (CASI) and the Parent-Adolescent Communication Scale (PACS).Responses to these two questionnaires were used to supplement the thematic analysis for participants. Findings from the interviews revealed four major themes and ten minor themes. Major themes consisted of (1) discussions with parents; (2) decisions about speech therapy; (3) types of parental assistance; and (4) discussions with others. Minor themes were (1) preference to talk with mothers; (2) speech techniques and general information; (3) parents' idea to begin speech therapy or(4) participants' idea to begin speech therapy; (5) reminding to use techniques and providing advice; (6) practicing speech techniques together; (7) good listening skills and not interrupting; (8) not directly talked about with friends; (9) rarely discussed with siblings; and (10) sharing experiences with other family members who stutter. Group comparisons between mothers and fathers revealed that participants communicated more with mothers about stuttering versus fathers. In addition, open communication levels were related to levels of communication about stuttering among both mothers and fathers. Advisors/Committee Members: Gabel, Rodney.

▼ This study explored whether K-12 teachers report role entrapment of people who stutter (PWS) in the form of occupational stereotyping. The Vocational Advice Scale (VAS) (Gabel et al., 2004) was completed by 204 teachers. The teachers also completed an open-ended question and a demographic questionnaire to identify factors that influenced their choices on the VAS. The findings did not indicate the presence of role entrapment for most careers; however this group of teachers did perceive that 10 careers on the VAS were less advisable for PWS. Analysis of the responses to the open-ended question found three major themes that influenced their decisions, including: “Related to disorder,” “Supportive,” and “Related to career.” The findings also suggested that issues related to exposure to stuttering did not influence the participants’ judgments of which careers were more or less advisable for PWS. Advisors/Committee Members: Gabel, Rodney.

▼ Previous research has shown the following phonological and lexical characteristics of words to be important for word learning: word length, phonotactic probability, word frequency, and neighborhood density. Storkel (2004) analyzed the influence of word length, word frequency, and neighborhood density on the age of acquisition (AOA) of words of young children. Neighborhood density and word frequency were found to be significant contributors to AOA. This study extends Storkel’s work by including phonotactic probability in the analysis, developing predictive models, and assessing the accuracy of the models. Words from the normative database of the MacArthur-Bates CDI were analyzed for each characteristic. Two linear regressions identified separate models for predicting AOA, each with only one significant contributor (word length and neighborhood density, respectively). Both models were highly accurate at predicting the AOA of individual words, with neither being more precise than the other. These results fit within prior work that has shown word length and neighborhood density to be important for word learning early on, and phonotactic probability as growing in importance with age. Advisors/Committee Members: Brackenbury, Tim P.

▼ The purpose of this study was to describe the acoustic and aerodynamic characteristics of diatonic scale singing at different tempi in a professional female soprano. The classically trained singer sang ascending-descending ninth scales on /a/ in the F# major key at five different tempi – slow, moderately slow, moderately fast, fast, and fastest. All scales were performed with a moderate loudness level and with a pre-determined metronome marking. Tempo was controlled in order to determine whether differences existed between scales that were sung at a comfortable tempo and scales that were sung at an uncomfortable tempo. The slow (1 note = 104 beats / minute), moderately fast (2 notes = 92 beats / minute), and fastest (4 notes = 92 beats / minute) tempi were designated as comfortable because they were typically performed with integer multiple vibrato cycles per note. The moderately slow (2 notes = 72 beats / minute) and fast (4 notes = 69 beats / minute) tempi were designated as potentially uncomfortable because the note durations were not integer multiples of the singer’s typical vibrato cycle period. There were a total of 5 ascending-descending scales analyzed in this study, one for each tempo. The results of the study strongly suggested trends in the acoustic and aerodynamic measures when tempo and scale direction were controlled. The following observations were noted: 1)mean fundamental frequency (F0) was higher in the descending notes than the corresponding ascending notes, 2)average F0 vibrato extent decreased as tempo increased, 3)mean note intensity was greatest at the highest notes and lowest at the lowest notes of the scale, 4)secondary intensity peaks were observed at the third (A#4) or fourth (B4) notes of both the ascending and descending portions of the scale, 5)airflow was greatest at the lowest notes of the scale and lowest at the highest notes of the scale, and 6)average airflow vibrato extent decreased as tempo increased. Distinctive trends were not observed between the two tempo categories (comfortable and uncomfortable). However, except for the slowest tempo, scales sung at a comfortable tempo demonstrated greater regularity in the number of vibrato cycles per note than scales sung at an uncomfortable tempo. The secondary intensity peaks observed in the scales suggest a tuning of the second harmonic partial with the first formant of /a/. Higher airflows at lower notes and lower airflows at higher notes suggest greater laryngeal flow resistance and greater glottal adduction at the higher notes. It is hypothesized that F0 and airflow vibrato extent may have decreased with tempo increase because the singer applied a control strategy of less variation of subglottal pressure, cricothyroid contraction, and glottal adduction as the tempo increased. Advisors/Committee Members: Scherer, Ronald.