This thesis aims to examine the impact of mind wandering on statistical language learning, a form of implicit learning that allows the detection of patterns and regularities from external input. Mind wandering refers to the shift of attention away from an external task toward internal thoughts, often occurring involuntarily. While mind wandering is traditionally associated with impaired performance in attention-demanding tasks such as learning, emerging research suggests that mind wandering may have a beneficial effect on certain instinctual cognitive processes, such as statistical learning and implicit learning (Vékony et al., 2025). Two experiments are conducted to investigate whether mind wandering influences performance and metacognitive confidence in artificial language learning tasks. In Experiment 1, participants listen to a continuous stream of trisyllabic pseudowords and complete a two-alternative forced choice (2AFC) task, providing confidence ratings for each decision. In Experiment 2, participants complete familiarity rating tasks after exposure to both paused and continuous speech streams with either front-vowels or back-vowels. In both experiments, mind wandering is assessed through a self-report questionnaire administered after each exposure. Results indicate no significant relationship between mind wandering and statistical language learning performance. However, no participants reported fully disengaging from the tasks, which likely affected the results. Due to the limited variability in mind wandering, as indicated by the mind wandering questionnaire scores that do not indicate high mind wandering, further research is needed. Future studies should explore whether more substantial or sustained mind wandering might enhance statistical language learning, as reported in emerging research. In addition, the study design could be applied to other cognitive domains.