Grammaticality judgments reflect a compound product of both grammatical and processing factors. But because they interact in a symbiotic way, very often grammatical and processing constraints are difficult to separate. According to generally accepted grammatical theory, (a) Who do you think John told Mary he fell in love with? and (b) Who do you think John told Mary fell in love with Sue? are equally grammatical. We have observed, however, that native speakers strongly accept sentences like (a) as grammatical but react quite variably to sentences like (b). A possible explanation is that native English speakers exhibit a processing preference, in searching for the extraction site for the wh- word, for object position over subject position. Proficient nonnative judgmental data offer additional support for a processing account. Nonnatives whose L1 grammars do not bias them toward objects also show preferences similar to those of natives. We provide a processing account based on Frazier's Minimal Attachment principle.

This study investigated first language (L1) orthographic influence on cognitive processing involved in second language (L2) reading. Using a fundamental unit of representation, three orthographic systems (i.e., alphabet, syllabary, and logography) can be categorized into two types: morphography and phonography. Previous L1 reading research has suggested that different strategies are used for phonological recoding by morphographic and phonographic readers. This study tested the possibility that these L1 recoding strategies are transferred and utilized in L2 reading. A cross-linguistic experiment involving adult L2 learners of English with contrasting L1 orthographic backgrounds (Arabic, Japanese, Spanish, and English—for native control) was conducted. The results indicate that reading among phonographic readers (Arabic, Spanish, and English) is seriously impaired when essential phonological information is inaccessible. Similar phonological inaccessibility, in contrast, apparently does not affect the reading performance of Japanese, or morphographic, readers. Further, the study demonstrated, first, that phonological inaccessibility exerts differential effects on the reading processes of phonographic and morphographic readers and, second, that L2 readers from different L1 orthographic backgrounds utilize their L1 strategies in reading English as an L2. Hence, the findings of the study verify cognitive strategy transfer during L2 reading.

Topical structure analysis (TSA), a text-based approach to the study of topic in discourse, has been useful in identifying text-based features of coherence. It has also been used to distinguish between essays written by groups of native English speakers with varying degrees of writing proficiency (Witte, 1983a, 1983b). More recently, TSA has distinguished between higher and lower rated ESL essays, but with different results from those found with native speakers of English (Connor & Schneider, 1988). The present study replicated the previous ESL study of two groups of essays written for the TOEFL Test of Written English with three groups of essays. Findings indicate that two topical structure variables, proportions of sequential and parallel topics in the essays, differentiate the highest rated group from the two lower rated groups. We offer explanations for the results and propose that all occurrences of a particular type of topic progression do not contribute equally to the coherence of a text.

The developing oral English of approximately 100 second language learners (four intact classes) was examined in this study. The learners were native speakers of French (aged 10–12 years) who had received a 5-month intensive ESL course in either grade 5 or grade 6 in elementary schools in Quebec. A large corpus of classroom observation data was also analyzed.

Substantial between-class differences were found in the accuracy with which students used such English structures as progressive -ing and adjective–noun order in noun phrases. There was some evidence that these differences (which were not correlated with performance on listening comprehension tests) were due to differences in teachers' form-focused instruction. These findings are discussed in terms of current competing views of the role of form-focused instruction in second language learning.