This project has gone through 2 rounds of peer review for journal submission. The manuscript has been reorganized to address some suggestions and questions raised by the reviewers. Critically, the results and main conclusions remained unchanged. For the sake of transparency, we provide here a summary of the major changes made to the structure of the manuscript. ## The first journal submission ## The manuscript was first submitted to a journal in August 2023, under the title of "**Adaptation to unfamiliar nonnative talkers: from first exposure to cumulative effects over a month**". We reported two experiments: (1) a 1-day experiment conducted with a campus population at the University of California, Irvine (Exp1); and (2) a 5-day experiment conducted with participants recruited through Prolific (Exp2). Although Exp.1 replicated a previous study (Xie et al., 2017), Exp.2 did not. The discrepancy between Exp.1 and Exp.2 suggests new methodological challenges facing a repeated-exposure-test experiment. As part of the general discussion, we explored and proposed possible ways to mitigate these challenges. To focus on the innovative aspects of the repeated-exposure-test paradigm as well as the important methodological challenges that remain for future studies, we decided to focus the current manuscript on Exp.2 and repackage Exp.1 as a separate manuscript. Although we resubmitted the contracted manuscript to the same journal, the extensive revision led to an unfortunate misunderstanding on the part of the new reviewer, which resulted in the rejection of the manuscript. ## The second journal submission ## The current manuscript was then submitted to another journal in February 2024 under the title "**From first encounters to longitudinal exposure: A repeated-exposure-test paradigm for monitoring language adaptation**". The current manuscript focuses exclusively on the 5-day experiment and the methodological insights it brought to light. We have added a new (post-hoc) power simulation in the General Discussion to address one of the possible reasons why the current study could not replicate Xie et al. (2017).