Morphosyntactic Processing, Cue Interaction, and the Effects of Instruction: An Investigation of Processing Instruction and the Acquisition of Case Markings in L2 German

Open Access
- Author:
- Henry, Nicholas Alan
- Graduate Program:
- German
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 02, 2015
- Committee Members:
- Carrie Neal Jackson, Dissertation Advisor/Co-Advisor
Barry Richard Page Jr., Committee Member
Michael Travis Putnam, Committee Member
Paola Eulalia Dussias, Committee Chair/Co-Chair
Bill Van Patten, Special Member - Keywords:
- second language acquisition
sentence processing
German
applied linguistics
Processing Instruction
Input Processing - Abstract:
- Sentence comprehension is among the most fundamental elements of second language (L2) acquisition, but it is also incredibly complex because L2 learners have to develop the ability to detect multiple types of cues in the input and interpret them in context. This is particularly difficult given that L2 learners have limited knowledge of these cues and have limited resources to process them. As a result, L2 learners employ a variety of strategies to help them make sense of complicated input. These strategies include attending to simpler or more salient cues in the input, such as lexical items, semantic cues (e.g., animacy, plausibility), or word order (VanPatten, 2004a). Consequently, other cues, such as morphosyntax (e.g., tense or case information) are not processed readily, and learners have difficulty acquiring these forms. Thus the presence of more salient forms that encode the same semantic information, may present a significant barrier to acquisition. On the other hand, there may be situations where certain cues support each other and promote acquisition. For example, current research (e.g., Grünloh, Lieven, & Tomasello, 2011) suggests that prosody (e.g., stress, pitch) may support the processing of morphosyntactic forms. The goal of this dissertation is to explore the psychological principles that underlie learners’ interactions with lexical-semantic, morphosyntactic, and prosodic cues in an input stream and to investigate how these processes impact the acquisition of morphosyntactic forms. Furthermore, this dissertation seeks to examine whether instructional interventions such as Processing Instruction can take advantage of these psychological principles in order to change processing behaviors and promote the acquisition of morphosyntactic cues. In order to address these issues, the present research tests the offline and online (i.e., real-time) effects of Processing Instruction on the acquisition and subsequent processing of German accusative case markers, which signal subject-first and object-first word orders as in (1): (1a) Der Kellner küsst die Frau im kleinen Restaurant. (Subject-First) TheNOM waiter kisses theACC woman in.the small restaurant. "The waiter kisses the woman in the small restaurant." (1b) Den Kellner küsst die Frau im kleinen Restaurant. (Object-First) TheACC waiter kisses theNOM woman in.the small restaurant. "The woman kisses the waiter in the small restaurant." In Experiment 1, two groups of third semester learners of German received two types of training: Processing Instruction (PI), in which lexical-semantic and reliable word order cues were explicitly removed from the input; and Traditional Instruction (TI), in which cues were not manipulated. Offline effects of the treatments were measured by sentence interpretation and picture description tasks following a traditional pre-test/post-test design. Changes in real-time processing strategies were assessed via a self-paced reading (SPR) task using sentences like those in (1). During this task, reading times for individual phrases were measured, and it was expected that learners would read object-first sentences slower if they processed the accusative case-markings. The results indicated that the PI group outperformed the TI group on the posttest offline comprehension task and equaled their gains on the offline production task. However, results from the SPR task provided no evidence that either group was slower to read object-first sentences. The goal of Experiment 2 was to improve the methodology of Experiment 1 and investigate whether the inclusion of prosodic cues in the training could facilitate acquisition and processing of the target form. Thus, three training groups were under investigation: TI, standard PI, and PI that includes prosodic cues (PI+P). The effects of instruction were again measured using the online and offline tasks used in Experiment 1. The results indicated that the PI and PI+P groups outperformed the TI group on the posttest offline comprehension measures and that all three groups improved equally on the production measures. Results from the SPR task indicated that none of the groups processed the experimental sentences in a native-like way and that participants had difficulty comprehending sentences in this task. However, there is evidence to suggest that the PI and PI+P groups directed more attentional resources towards case markers after training. Furthermore, evidence from the SPR task suggests that the PI+P group did process object-first sentences with more effort after reading the entire sentence. Taken together, the pattern of results for these experiments suggest that Processing Instruction did have a significant impact on the acquisition of the target form. Furthermore, it appears that Processing Instruction does influence processing strategies, but that its effects are mediated by other factors (e.g., working memory, or the speed of lexical access) that prevent rapid and efficient integration of morphosyntactic information. These data suggest that lexical-semantic cues hinder the acquisition of morphosyntactic cues, but that prosodic cues can play a facilitative role in the acquisition and processing of these forms. These findings are discussed within the context of the Input Processing model (VanPatten, 2004a) and the Competition Model (Bates & MacWhinney, 1987), and pedagogical implications are discussed within the context of research on Processing Instruction (VanPatten & Cadierno, 1993).