Lexical acquisition extends beyond the learning of surface-level word forms and encompasses underlying cognitive characteristics as well as broader processes that involve the learner. Nevertheless, in Japanese, as in many other languages, word difficulty has been characterized primarily by frequency and surface-level properties. Far less is known about the cognitive and affective dimensions that influence whether words are easier or more difficult to process. To address this gap, this study introduces a novel dataset that incorporates six psycholinguistic dimensions—familiarity, affective valence, arousal, imageability, abstractness, and understandability—collected through large-scale surveys of Japanese second language learners. Preliminary analyses of responses from 536 participants across 15 countries demonstrated that the dataset is both theoretically coherent and empirically reliable, consistent with established theories and findings while also yielding new insights into lexical processing. In addition to supporting more accurate estimations of word difficulty, the dataset provides a resource for future research by enabling systematic exploration of how lexical processing is shaped through the interaction of visual, affective, cognitive, and contextual factors.