The effects of linear distance and working memory on the processing of gender agreement in Spanish

L2 Processing of Plural Inflection in English

“…However, this study also found a quantitative difference that suggests a limitation of L2 processing: when reading partitives, the Korean ESL learners’ sensitivity to missing plural inflection emerged one word region later than for the native speakers; while the native speakers began to slow down at Region 7, the learners did not slow down until they reached Region 8 (see Table ). Such late sensitivity has been reported in previous studies as well (Keating, ; Sabourin & Stowe, ; see also Gillon Dowens et al., ). For example, in an eye‐tracking experiment on gender agreement in Spanish, Keating () found that Spanish native speakers showed their sensitivity to gender violations (i.e., increase in RT) both in first‐pass and second‐pass reading, while advanced anglophone learners did not show such sensitivity until second‐pass reading.…”

“…Existing literature seems to lend support to this possibility, although it is left to future research to test it. For example, Keating () showed that the linear distance between the noun and the adjective in gender agreement dependency affects advanced L2 learners’ sensitivity to violations of gender. The language of interest was Spanish, and the learners’ L1 was English.…”

“…Current literature, however, provides evidence that L2 learners—at least advanced ones—can successfully process not only local checking of grammatical features (e.g., Keating, ; Sagarra & Herschensohn, ), but also nonlocal checking (e.g., Foote, ; Jiang et al., ; Keating, ; Wen et al., ). For instance, in Keating's () study, advanced English‐speaking learners of Spanish were sensitive to gender agreement violations between a noun and a predicative adjective occurring across a VP (e.g., *[ IP La casa fem . [ VP es viejo masc .…”

Grammatical gender in L2: A production or a real-time processing problem?

“…These findings suggest that advanced L2 learners may perform like native speakers in their sensitivity to violations of gender agreement between adjacent words, indicating that gender agreement per se is operative in advanced L2 learners’ real-time language use. However, sensitivity to violations of gender agreement is apparently more susceptible to disruption in L2 learners when mismatches are non-local, suggesting that non-linguistic factors such as working memory may play a role (for further evidence on the contribution of working memory, in both non-native as well as native speakers, see also Keating, 2010). …”

“…Different answers have been proposed, including age-related insensitivity to grammatical features not instantiated in the L1 (Franceschina, 2005; Hawkins, 2009; Sabourin and Stowe, 2008), performance limitations specifically affecting L2 production (Alarcón, 2011; Montrul et al, 2008), and more general limitations on L2 processing due to increased demands on working memory (Gillon Dowens et al, 2010; Keating, 2009, 2010) or to different language learning environments (Lew-Williams and Fernald, 2010). Numerous studies have reported comparisons between native and non-native speakers on tasks manipulating grammatical gender, with varied outcomes, and conclusions that are not entirely convergent.…”