The long-term effects of age of instruction on the L2 acquisition of English reflexive pronouns

Abstract

The present study investigated whether younger L2 instructed learners perform better at tasks that test a UG-related linguistic property than their older counterparts. A sample of 24 Arabic speaking L2 learners at the college level in Saudi Arabia was tested on English reflexive binding. They completed a Truth Value Judgment Task (TVJT) in a story format. Participants were divided into groups according to the year they began their L2 instruction – grade 4, 6, or 7 - and their performance on this task was compared. This study aims to add to the SLA literature concerning the long-term effects of age on L2 acquisition in the instructed context. The results showed no significant differences in performance between the three groups of learners. However, two conditions of reflexive binding showed significant differences in some age groups but not the others. These two conditions were: (1) mono-clausal object antecedents (e.g., the policeman questioned the suspect about himself) and (2) non-finite bi clausal long-distance antecedents (e.g., Bill asked the policeman to protect himself). No condition created differences in all the three groups.In light of previous research (e.g., White et. al., 1997), the within-group differences associated with these two conditions were understood to be caused by methodological difficulties rather the effect of age. These conditions were predicted to increase performance differences among native speakers as well as L2 learners. Hence, the findings are in line with studies such as Muñoz (2006a, 2006b, 2011, 2014) and Al-Thubaiti (2010, 2014) which indicated that age at the onset of instruction is not a predictor of success in L2 learning.