Ubiquitous technologies change the way we learn new languages. They provide easy access to multilingual media content within everyday scenarios to enhance language skills or improve vocabulary. The detection of learners’ vocabulary deficiencies is an integral part of effective language learning. However, retrieving translations during everyday media consumption causes inattention and hinders fluent learning. In this paper, we investigate Electroencephalography (EEG) to assess single-word incomprehension within non-native learning contents. In a user study (N = 10), we employed Rapid Serial Visual Presentation (RSVP) to display text while recording Event-Related Potentials (ERPs) for different word difficulties. Our results show that incomprehension of words can be detected in participants’ neural responses, as is confirmed by post hoc ratings. We conclude with use case scenarios in which our approach can facilitate seamless vocabulary deficiency detection in everyday life.