It is widely accepted that L2 learners often face communication problems due to lack of competency in the target language and familiarity with its culture of origin. One way to resolve miscommunication problems is to seek clarification of the utterance; this process is called "repair negotiation" (Nakahama et al. 2001). Repair negotiation has been studied as a subset of Long's (1996) Interaction Hypothesis in second language acquisition (SLA). Existing research has solely focused on repair negotiation which is initiated by the native speaker or teacher rather than the L2 learner, thus resulting in "other-initiated repair" (OIR). However, it has also been found that a L2 learner can recognize mistakes while monitoring self utterance and initiate repair or self correction of the utterance, resulting in "self-initiated repair" (SIR). Unfortunately, little attention has been given to repair negotiation initiated by the L2 learner and to what degree SIR facilitates L2 language development; even less attention has been given to overarching or combining two types of repair negotiation as well as measuring to what extent different types of repair negotiation between NNS-NNS lead to improving development of the second language. Three research questions were proposed for this dissertation research: (1) How do different types of tasks relate to the use of NNS's repair organization? (2) Does repair negotiation between L2 learners lead to development of linguistic targets? (3) Does type of task affect development of morpho-syntactic features? In order to answer these questions, two experiments were conducted. In Experiment 1 in response to the first research question, twenty two high beginning to low advance learners of English participated in this study. The participants with the same proficiency levels were paired. The participants completed several tasks, including one-way information gap task, jigsaw task, and decision-making task. The results show that the production of SIR was elicited most frequently in the decision-making task whereas the production of OIR was elicited most frequently in the one-way information gap task. After the treatment, a diagnostic test was given to the subjects of Experiment 1 in order to determine the appropriate level of participants for Experiment 2. The results of the diagnostic test show that the subjects from levels 3 and 4 were the most appropriate. Based on the results of Experiment 1, two types of tasks (one-way information gap task vs. the decision-making task) and the intermediate level of participants were chosen to address the second and third research questions. In response to the second and third research questions, Experiment 2 was conducted with intermediate L2 English learners (n = 36) using a pretest-post test-delayed post test design. The linguistic targets chosen for were past tense and relative clause (subject, object, and oblique relative clause) in the morpho-syntactic area. The results of this research suggest that past tense is learned more effectively than relative clause; repair negotiation between L2 learners is particularly beneficial in recognizing past tense morpheme, leading to the learning of morphology of past tense. Additionally, types of task differently influenced the learning of the two linguistic targets; one-way information gap task was more effective for learners in the short term than was the decision-making task. This study gives support for the interaction hypothesis to repair negotiation between learners along with task-based instructions in the classroom. [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.]