Do we notice our own mistakes? Attention to misspellings and convention-breaking grammar during reading

Katrine Falcon Søby, Department of Nordic Studies and Linguistics, University of Copenhagen
Byurakn Ishkhanyan, Department of Nordic Studies and Linguistics, University of Copenhagen
Line Burholt Kristensen, Department of Nordic Studies and Linguistics, University of Copenhagen


“Is some grammar errors or mispellings more disturbing than others and why?” Typos, misspellings and anomalous grammar are common phenomena – both in texts written by native and non-native speakers. The language of native speakers is often idealized within second language learning theories (Bokamba 1984), but native speakers also break grammar and spelling conventions (Blom et al. 2017). While some anomalies occur both in texts written by native speakers and by L2 learners, other anomalies are only characteristic for one of the two groups (Hansen et al. 2019; Søby & Kristensen 2019). Our corpus of L1 high school essays shows that the frequency of anomalies follows the pattern orthographic > morphological > syntactic. In L2 users with English as L1, the order is morphological > orthographic > syntactic. Within these three major categories there are also different patterns for L1 and L2 texts.

In a proofreading study, we investigated if native speakers are better at noticing anomalies that are common in L2 production than those that are common in L1 production (and which they potentially produce themselves).

211 Danish high school students (98 women; 17-20 years, M = 18.31 years, SD = 0.67 years) 1) read two texts for comprehension, while underlining misspellings and grammar anomalies and 2) did a basic grammar quiz. The texts contained syntactic (V2 word order), morphological (gender congruency in noun phrases; verbal inflection) and mixed orthographic anomalies. Some anomalies were typical of L1 usage (e.g. use of infinitive for present tense when the two forms are homophone), while others were typical of L2 usage (e.g. use of present tense verb form for the infinitive).

The anomaly detection rate was analyzed using a mixed effects model. In line with L1 production patterns, we found the following detection pattern syntactic > morphological > orthographic, i.e. readers tend to overlook common L1 anomalies. Detailed analysis of gender congruency and verbal inflection pattern also show a match between perception and production patterns, e.g. lower detection rates when an infinitive is used for a homophone present tense (a common L1 mistake). We discuss the degree to which our results show a link between production and perception of grammar.



Blom, J. N., M. Rathje, B. F. Jakobsen, A. Holsting, K. R. Hansen, J. T. Svendsen, T. W. Vildhøj, A. V. Lindø (2017). Linguistic deviations in the written academic register of Danish university students. Oslo Studies in Language 9(3). 169–190.

Bokamba, E. (1984). The fiction of the native speaker in L2 research. GURT on languages and linguistics, 243–252.

Hansen, J. J., L. B. Kristensen, K. F. Søby (2019). Verbal inflections in L1 and L2: what is the role of homophony? Poster presentation at Fonologi i Norden, University of Edinburgh, Scotland, the 22nd of February 2019.

Søby, K. F., & Kristensen, L. B. (2019). Hjælp! Jeg har mistede min yndlings rød taske. Et studie af grammatikafvigelser. Ny forskning i grammatik, (26), 89-104.

Lämna ett svar

Din e-postadress kommer inte publiceras. Obligatoriska fält är märkta *