This study aims at investigating whether Persian native speakers highly advanced in English as a second language (L2ers) can switch to optimal processing strategies in the languages they know and whether working memory capacity (WMC) plays a role in this respect. To this end, using a self-paced reading task, we examined the processing strategies 62 Persian speaking proficient L2ers used to read sentences containing ambiguous relative clauses in their L1 and L2. The results showed that L2ers adopt the same strategy as that used by English native speakers in both of their languages, indicating a target-language like parsing pattern in their L2 and an attrition of L1 parsing routine. Additionally, their attachment preferences were not modulated by WMC in L2. This result highlights the “skill-through-experience” position adopted by researchers who question the role of WMC in L2 syntactic parsing. However, high-capacity L2ers' preferences in L1 had attrited (becoming English-like), and low-capacity ones had no preference. This modulation, too, can bear out the above position owing to the observation that L2ers failed to differentiate between their L1 and L2, and particularly that their differing WMCs did not contribute to native-like performance in their L1.
