Send message to Author

Paper short abstract

This work presents quantitative-qualitative analysis of perceived kanji difficulty among native speakers and learners of Japanese. Stroke count, radicals, and lexical familiarity were rated high as difficulty predictors, but empirical results show that perception may not correspond to performance.

Paper long abstract

This study investigates how learners and native speakers perceive kanji difficulty, based on 65 detailed survey responses. Participants were presented with five groups of five kanji pairs, each differing in one trait: frequency, stroke count, number of readings, number of meanings, and concreteness of meaning. Before making pairwise judgments, participants indicated which factors they believed most influence difficulty. This design allows comparison between stated beliefs and actual selections, as well as differences between proficiency groups.

Results reveal two key contrasts. First, both learners and native speakers claim stroke count and radicals are primary determinants of difficulty. However, their choices tell a more nuanced story. Learners disproportionately judged kanji with unfamiliar components or multiple readings as harder, while native speakers were more sensitive to subtle visual distinctions and semantic nuance. Across all responses, recognisable components, stroke count, and lexical familiarity emerged as dominant predictors, with mean influence scores above 2.1 on a 0–3 scale. Visual similarity strongly shaped judgments, exemplified by 未/末, where 末 was chosen as harder in 63% of cases, despite being more frequent. This suggest that human perception of frequency may be unreliable. In contrast, formal indicators such as JLPT level (mean 1.03) and school grade of instruction (1.06) were weak predictors, indicating that curricular labels do not match intuitive difficulty.

Qualitative comments reinforce these findings. Learners emphasised memorisation strategies, radicals, and vocabulary exposure, while native speakers mentioned graphical balance and stroke order. Symmetry and overall shape were mentioned but ranked low in quantitative influence. Several participants noted that familiarity through repeated exposure significantly reduced perceived difficulty.

What learners and natives believe is difficult may not correspond to error patterns or acquisition speed. Therefore, the next step is to contrast these perceptual biases with empirical data to identify where intuition diverges from reality.T his research provides a foundation for understanding perceptual bias (mainly overestimation of stroke count importance, and lack of frequency awareness) and its role in shaping effective, data-driven approaches to kanji acquisition.