Frequency Effects of Regular Past Tense Forms in English on Native Speakers’ and Second Language Learners’ Accuracy Rate and Reaction Time

There is substantial debate over the mental representation of regular past tense forms in both first language (L1) and second language (L2) processing. Specifically, the controversy revolves around the nature of morphologically complex forms such as the past tense –ed in English and how morphological structures of such forms are represented in the mental lexicon. This study focuses on the results of a speeded acceptability judgment task testing English regular past tense forms of high- and low frequencies. In this task, participants judged the acceptability of sentences as quickly and accurately as possible. Thirty-two intermediate-to-advanced L1 Persian and L1 Arabic speakers (L2ers) and twenty-two Native speakers (NSs) of English made acceptability judgments for regular past tense forms of high- and low-frequency verbs in sentential contexts. Considering participants’ reaction times (RTs) and accuracy rates as the dependent variables, the main results are as follows. Despite NSs’ faster RTs and higher accuracy scores, both groups demonstrated the same pattern of accuracy rates and RTs. Specifically, for accuracy data, regular verbs yielded a reverse frequency effect or anti-frequency effect in both groups (i.e., lower accuracy rates for high- than low-frequency regular forms in NSs as well as L2ers). For RT data, while the NSs exhibited a marginally anti-frequency effect, the L2ers displayed a nonsignificant trend for the anti-frequency effect. These results support the dual-mechanism models suggesting that the mental mechanisms and representations of inflectional morphology are the same in NSs and intermediate-to-advanced L2ers.