What is it about?
When you read the word book, you know it means one thing. When you read books, you know it means several. But is that second connection — between the -s ending and the concept of "more than one" — equally automatic? This study tested that question by having Japanese learners and native English speakers read sentences while looking at pictures, then quickly judge whether the sentence matched the picture. Crucially, sometimes the number in the sentence and the picture deliberately clashed: a sentence with onion (singular) appeared above a picture of three onions, or a sentence with birds (plural) appeared above a picture of one bird. If the word's number is processed automatically, these mismatches should slow people down even though they were told to ignore them. The result: learners were slowed when onion mismatched a plural picture, but not when birds mismatched a singular one. The connection from singular to "one" worked automatically; the connection from plural to "more than one" did not.
Featured Image
Why is it important?
Previous research on this topic only ever tested one direction of the mismatch — singular word, plural picture — and found that Japanese and Chinese learners showed no delay, which was taken as evidence that they had simply failed to acquire the plural morpheme. This study adds the reverse direction for the first time, and the asymmetry it reveals tells a more precise story: the problem is not a blanket failure to acquire number morphology. Rather, singular and plural forms differ fundamentally in their conceptual properties — singular always means exactly one, while plural is semantically underspecified and can in principle include one — and it is this underspecification, compounded by the absence of obligatory plural marking in Japanese, that makes the plural-to-concept link fragile. The finding reframes the Morphological Congruency Hypothesis: L1 incongruence may only create a processing barrier when the target feature is also conceptually vague.
Perspectives
This paper started as part of my doctoral dissertation and took an unusually long road to publication — long enough that I had to substantially update the theoretical framing as the field moved on. What I find most satisfying about the final result is the interaction plot: the crossing lines for the learner group are almost unnervingly clean, and they say something I did not expect when I first designed the study. I had predicted an asymmetry, but I thought it would be a matter of degree — a smaller effect for plurals, not a near-zero one. The complete absence of a plural mismatch cost forces a stronger theoretical claim than I originally intended to make, and I am still not entirely certain whether that reflects a genuine processing null or a sensitivity issue with the paradigm. That uncertainty is what I hope future work will resolve.
Yu Tamura
Kansai Daigaku
Read the Original
This page is a summary of: Singular–plural asymmetry in L2 English number processing: A sentence-picture matching study of Japanese learners of English, International Journal of Bilingualism, February 2026, SAGE Publications,
DOI: 10.1177/13670069261422017.
You can read the full text:
Contributors
The following have contributed to this page
