Researchers in second language acquisition (SLA) have increasingly discussed the role that attention plays in the learning of a second language (L2). This discussion has led to research on proposed pedagogical strategies aimed at directing L2 learners' attention to aspects of the L2 grammar that are difficult to learn or acquire. Research on one such strategy, Processing Instruction (Cadierno, 1995; VanPatten and Cadierno, 1993a, 1993b; VanPatten and Wong, 2004), suggests that focusing learners' attention on certain grammatical forms in the input can lead to the learners' improvement in both the comprehension and production of those grammatical forms. However, studies regarding the role of attention in SLA, including those on Processing Instruction, have focused primarily on morphology and syntax, while the relationship between attention and pronunciation learning has been mostly ignored. Other research has focused on the impact that explicit instruction on L2 phonetics and phonology has on L2 learners' improvement in pronunciation (e.g. Elliott, 1995b, 1997; Lord, 2005), but this research has not considered the role that attention may play in pronunciation learning. This dissertation connects research on pronunciation learning with other research in SLA by comparing a more typical type of pronunciation assignment with a pronunciation assignment designed to encourage students to attend to the fine-grained phonetic details of L2 speech sounds. Specifically, native English-speaking learners of Spanish were tested on their learning of the Spanish vowels /e/ and /o/. These sounds were considered specifically because previous research has focused on them less than other Spanish sounds. Moreover, earlier studies have not attempted to quantify diphthongization of these vowels to better analyze learners' improvement on them. Twenty-eight students of Spanish, enrolled in two separate sections of a Spanish conversation class, participated in the study. The two sections were taught identically, except that one section received the more typical pronunciation assignment (production assignment, N=15), while the other received the pronunciation assignment designed to focus students' attention on L2 speech sounds (perception assignment, N=13). In addition, all students were recorded reading single words at the beginning (pretest) and end (posttest) of the semester. The impact of each assignment on improvement in pronunciation was assessed by acoustically analyzing each student's pronunciation of the vowels /e/ and /o/ on both the pretest and posttest. Both sections were given explicit instruction on differences in Spanish and English speech sounds that have been identified in past studies as difficult for English learners of Spanish. The production assignment, administered five times, required students to submit recordings of themselves reading a dialogue, on which they received feedback regarding their pronunciation from the instructor each time. For the perception assignment, also administered five times, students listened to recordings of native English participants speaking Spanish and were asked to evaluate the speaker's pronunciation. All native English speakers were unknown to the students receiving the perception assignment. For all assignments in both sections, focus was given to the aspects of Spanish pronunciation covered in class to that point. To analyze students' improvement of the mid vowels, diphthongization of these vowels was quantified. From participants' productions on the pretest and posttest, the first (F1) and second (F2) formants were measured throughout the production of each critical mid vowel, and the change in F1, as well as the change in F2-F1 (F2 minus F1), was calculated. Diphthongization of the /o/ vowel was measured by the decrease in F1, while diphthongization of the /e/ vowel was measured by the increase in F2-F1. Improvement in production of the mid vowels was indicated by a reduction in these changes in F1 and F2-F1 from the pretest to the posttest. This technique allowed for a more in depth analysis of improvement than would be obtained from labeling each vowel as a diphthong or monophthong. Results show a significant improvement in the diphthongization of both /o/ and /e/ for the section receiving the perception assignment, but they show no improvement on either vowel for the section receiving the production assignment. This finding suggests that the perception assignment used here was superior to the production assignment inasmuch as the two assignments are compared in this study. I argue that the perception assignment led to more improvement in pronunciation because it was designed to direct students' attention to L2 pronunciation in a way that is not required of students to complete the production assignment. This implies that attention not only plays a role in the learning of morphology and syntax, but also in the learning of pronunciation. The findings in this dissertation suggest, then, that when teachers design assignments targeting pronunciation in L2 classrooms, they should consider how or whether the assignments direct students' attention to key aspects of the L2 speech sounds. (Abstract shortened by UMI.) [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]