There is no one-to-one mapping between speech acoustics and individual speech sounds; instead, the acoustic cues produced for individual speech sounds show wide variability both within and across talkers. Nonetheless, listeners perceive the speech of familiar and novel talkers with ease. It is theorized that listeners achieve this feat by maintaining a degree of flexibility in how acoustics are mapped to speech sound categories, allowing listeners to dynamically modify the mapping to speech sounds to reflect structure in input statistics. Building on previous work demonstrating that listeners are sensitive to individual talker differences in speech production, we test the hypothesis that distributional learning for input statistics is contextually governed by talker identity. Listeners (n = 320) completed two blocks of phonetic identification for VOT input distributions specifying the /g/ and /k/ categories. In one block, the input was shifted towards relatively shorter VOTs; in the other block, the input was shifted towards relatively longer VOTs. Across listener groups, we (1) manipulated block order and (2) whether or not the talker remained constant across blocks. In this way, a change in input statistics was concomitant with a change in talker for some listeners but not for other listeners. Predictions for talker-specific vs. talker-agnostic distributional learning were derived through simulations performed with the Bayesian belief-updating model of speech adaptation, which yielded qualitatively different patterns of learning for the same-talker vs. different-talker simulations. Specifically, the simulations for the same-talker condition predicted that listeners in the two order groups would show a different VOT voicing boundary in Block 1 and then converge in their voicing boundary in Block 2, consistent with a cumulative registration of input statistics for the same talker heard across the two blocks. In contrast, the simulations for the different-talker condition predicted that listeners in the two order groups would show a difference in the VOT voicing boundary in both blocks, given a resetting of distributional learning in block two (i.e., a return to prior knowledge) triggered by a change in talker. The results showed (1) robust evidence of distributional learning in that listeners' voicing boundaries moved block-to-block in line with changes in the input statistics, (2) no difference between the same-talker and different-talker conditions, and (3) learning patterns that were consistent with cumulative integration of distributional input statistics across blocks. These patterns were replicated across two experiments. Collectively, the results suggest that distributional learning in this paradigm is not talker-specific, which may reflect the a priori informativity of VOT as a cue to talker identity. [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.]