Previous literature has shown that voicing contrast is a difficult feature for Chinese-English emergent bilingual learners of L3 Spanish (Bravo Diaz, 2020; Chen, 2007; Zhang, 2022). The production of voicing contrast in stops has been broadly investigated in second and third language studies (e.g., Zampini, 1998; Llama, 2010; Wrembel, 2014), but almost exclusively focusing on VOT as a unified correlate and studying one moment in the process of L2 development. Adopting a Complex and Dynamic Systems Theory framework (e.g., Larsen-Freeman, 2007), this dissertation investigates the roles of other acoustic correlates of voicing contrast in multilingual participants' stop production, including individual differences in developmental paths of VOT using a time-series approach.Two studies were conducted studying stop production of a group of L1 Mandarin -- L2 English learners of L3 Spanish (N = 30). In Study 1, tokens of intervocalic stops in a text-reading task were analyzed to measure C:V ratio of intensity, percentage of voicing, and onset F0 of the following vowel to examine and compare their realizations in the participants' three languages. In Study 2, VOT samples in a wordlist reading task were measured five times during a period of one semester to trace their developmental paths. The role of individual differences including language aptitude, language use, and metalinguistic knowledge were integrated in mixed effect models.Results showed that participants relied on all acoustic cues under examination to signal voicing contrast in all three languages, with differences of cue weights across languages. At the same time, there were significant individual differences in individual developmental paths of VOT; mixed effect models revealed that phonetic coding ability had a facilitative effect in pronunciation accuracy; sequence recognition ability interacts with time to show an effect in more advanced stages of learning. This dissertation has implications for research on L3 phonological development and individual variation. It sheds light on the relevance of acoustic cues other than VOT in stop production among multilingual learners, and highlights the roles of explicit and implicit language aptitude in the learning of segment pronunciation. Finally, it contributes to a complex and dynamic view of voicing contrast development in a third language. [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://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]