The acquisition of differential object marking in L2 Spanish learners

This dissertation tests a grammatical structure, differential object marking (DOM), which is particularly difficult for L2 learners to acquire. DOM is a phenomenon in which some direct objects are morphologically marked to distinguish them from subjects (Comrie, 1979). In Spanish, animate and specific direct objects are marked with the preposition ‘a,’ as in Juan ve a María ‘Juan sees DOM María.’ DOM in Spanish has been found to be problematic for second language (L2) learners whose first language is English, with errors persisting after instructional intervention including positive and/or negative evidence (Bowles & Montrul, 2008; Bowles & Montrul, 2009a; Farley & McCollam, 2004; Guijarro-Fuentes & Marinis, 2007; Wiebe, 2004). Structures such as DOM in Spanish are not acquired quickly by L2 learners, and as such this structure is ideal for testing learners who have some prior knowledge of a structure, but who still make errors on the structure. This was the primary purpose of the current study. One way to improve learner acquisition of difficult structures is by increasing their awareness of these structures. The noticing hypothesis (Schmidt 1990, 1993, 1995, among others), maintains that noticing is necessary to learn target forms. Several studies have tested the noticing hypothesis using think-aloud protocols, in which participants speak their thoughts aloud while carrying out a task. These studies have found that in general higher levels of awareness correlate with increased acquisition of targeted forms (Alanen, 1995; Leow, 1997a, 1998a, 2001a, 2001b; Rosa & O’Neill, 1999). One way to increase awareness of grammatical structures in SLA is through explicit instruction and feedback, which have both generally been found to be effective (Li, 2010; Norris & Ortega, 2001; Russell & Spada, 2006; Spada & Tomita, 2010). This study tested 58 L2 learners of Spanish (L1 English), 27 with no prior knowledge of DOM, as shown by the pretest, and 31 with some prior knowledge, who still made comprehension and production errors with DOM, as shown by the pretest. Half of the learners in each group carried out computer administered explicit instruction and feedback, followed by two tasks, and half completed only the two tasks (comparison group). The study consisted of a pretest, posttest, delayed posttest design, with the posttest taking place one week after the pretest, and the delayed posttest two weeks later. The tasks were an oral picture description task and an untimed grammaticality judgment task, during which a subset of the participants completed think-aloud protocols. Results indicated that both instructed groups improved significantly more after instruction on the two tasks than the uninstructed groups, although not equally, with the instructed group with prior knowledge maintaining increases better than the instructed group without prior knowledge on the oral picture description task. Think-aloud protocols indicated that participants who demonstrated some level of awareness tended to have higher accuracy rates than those with no awareness of the structure, and that both instruction and prior knowledge were related with higher levels of awareness